Poster Abstracts

Poster Abstract
DEPRESSION AND QUALITY OF LIFE IN GLIOMA PATIENTS. Deborah Allen MSN, RN, APRN, BC, Duke Brain Tumor Center, Durham NC US; Susan Schneider PhD, RN, AOCN, Duke University School of Nursing, Durham NC US

The health state of patients with gliomas often deteriorates through the illness course, frequently overshadowing the identification of depression and quality of life (QOL) symptoms. Since gliomas are relatively rare, few studies have evaluated the incidence and treatment of depression and QOL symptoms.

The objective of this study was to determine what demographic, diagnostic, or treatment variables were associated with the presence of depression or QOL symptoms (fatigue, appetite changes, sleep disturbances, and sexual dysfunction) in patients with gliomas.

Based on the Chronic Illness Trajectory, the individual's perceived QOL should be considered in discussions regarding therapy and treatment outcome. The use of a screening tool to predict depression and QOL symptoms in the clinical evaluation may improve outcomes.

This study uses a retrospective exploratory design to evaluate the dependent variables of depression and QOL symptoms in adult patients. The Beck Depression Inventory-II, a commonly used instrument with established reliability and validity, was used to screen for moderate to severe depression and QOL symptoms. Fisher's Exact Chi-Square tests were performed to assess the relationship between a dichotomized indicator of depression and demographic, tumor, and treatment variables. Logistic regression was used to explain the joint effect of these factors.

There were 969 encounters between 09/2001 to 03/2002. A total of 502 BDI-II forms were completed with all items clearly marked for analyses. Most patients had high grade gliomas (n=379), mean age of 44.3 years, and were male (n=284). Only 66 patients indicated a score of 20 or greater for moderate to severe depression. There were 358 reports of sleep disturbances, 273 changes in appetite, 385 indicating fatigue, and 260 cases of sexual dysfunction. Patients taking decadron were more likely to have higher BDI-II scores for depression (p<0.05) and all four QOL symptoms (p<0.05). Patients on chemotherapy (p<0.001) or with high-grade tumors (p<0.05) were more likely to experience sleep disturbances, fatigue, or appetite changes but not sexual dsyfunction. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from the American Brain Tumor Association

Poster Abstract
WATCHFUL WAITING IN PROSTATE CANCER: WHERE CAN OLDER MEN FIND SUPPORT? Donald Bailey, Jr. PhD, RN, Duke University School of Nursing, Durham NC US; Janet Stewart PhD, RN, University of Pittsburgh School of Nursing, Pittsburgh PA US; Merle Mishel PhD, RN, University of North Carolina at Chapel Hill School of Nursing, Chapel Hill NC US

Watchful waiting has been proposed as a reasonable alternative for older men with localized prostate cancer. However, because this option is controversial men electing watchful waiting may encounter barriers to traditional sources of support. Family and support group members can pressure men to seek treatment based on the belief that cancer should be treated aggressively.

The purpose of this study was to identify the sources and types of support used by men electing watchful waiting.

Mishel’s Reconceptualized Uncertainty in Illness Theory (1990) guided the larger intervention trial of which this study was a part. Thirty-five men enrolled in a trial of the Watchful Waiting Intervention agreed to participate in a telephone interview at the completion of the larger study’s data collection. A semi-structured interview guide was used to elicit men’s reports of where they found support and how they used this support to live with the continual uncertainty of their disease. Interviews were tape recorded and transcribed verbatim.

Content analysis techniques were used to isolate data units that represented sources of support and the type of support exchange that took place. These were then organized into a categorical structure and frequencies were calculated for each category.

Participants were 86% Caucasian and 14% African American, with an average age of 75.4 years. Ninety percent were married. Nearly every man reported that his wife and/or adult child was a source of support. For 51% of the men, this was the only individual identified. Men had not confided in anyone else because of reluctance to share their feelings about their disease. The remaining 49% sought support from other family members in addition to their wives. Although 86% identified a connection with a church, only 20% had shared the news with their preacher, a religious leader, or members of their congregation and found support in prayer. No man identified membership in a cancer support group. These findings are similar to other investigators who found that men seek support from only one source. Nurses should assess sources of support in men electing watchful waiting and offer supportive intervention when indicated. Funding Sources: Post Doctoral Fellowship, The John A. Hartford Foundation Building Academic Geriatric Nursing Capacity, 2003-2005.

Poster Abstract
MEASURING CHEMOTHERAPY-INDUCED PERIPHERAL NEUROPATHY. Marie Bakitas MS, ARNP, AOCN, Dartmouth-Hitchcock Medical Center, Lebanon NH US; Jeffery Cohen MD, Dartmouth-Hitchcock Medical Center, Lebanon NH US; Camilo Fadul MD, Dartmouth-Hitchcock Medical Center, Lebanon NH US; Ellen Lavoie Smith MS, APRN-BC, AOCN, Dartmouth-Hitchcock Medical Center, Lebanon NH US

Chemotherapy-induced peripheral neuropathy (CIPN) is the result of damage to the structure and function of peripheral motor, sensory, and autonomic nerves, causing peripheral neuropathic symptoms and signs. The purpose of this presentation is to describe subjective and objective measures of CIPN currently used in clinical trials, comparing characteristics and available reliability and validity data. These comparisons will increase investigators knowledge to allow the selection of measures that will match the study purpose and the researchers' expertise and resources.

In the development of studies to understand CIPN researchers must have a clear understanding of the measures reliability, validity, purpose and data generated in order to select appropriate measures to meet study goals. Studies that evaluate CIPN have traditionally used clinician grading systems (e.g. NCI-CTC, ECOG), which lack inter-rater reliability and internal consistency. Other methods include objective neurological exam, quantitative neurological tests (e.g. vibration and nerve conduction studies), summary measures combining subjective symptoms, objective signs, and quantitative tests (e.g. Total Neuropathy Scale [TNS]), and self-reported symptom and quality of life scales. Many of these measurement techniques were originally developed to understand and measure diabetic neuropathy and have only recently been adapted for use in CIPN. Each measure has advantages and disadvantages related to cost (financial and researcher time), accuracy, comparability, and ability to describe the patient’s subjective experience.

Given the variety of measures used in clinical research it is difficult to describe and ascertain conclusions about the impact of CIPN on function and QOL. Conducting qualitative studies of the patient’s experience in conjunction with additional studies comparing objective measures may enhance the validity of CIPN measures.

CIPN is an understudied common side effect that presents research measurement challenges due to its variable character, course, and unresolved issues of measurement methods. Lack of measurement precision has led to under-recognition and under-treatment of this potentially dose-limiting toxicity. Standardizing measurement approaches of CIPN in descriptive studies and clinical trials will provide a greater understanding of the patient experience, will improve accuracy of incidence and prevalence estimates, and will increase ability to validly compare study results. Funding Sources: Department of Defense-Breast Cancer Research Program-DAMD17-03-1-0298

Poster Abstract
PREDICTORS OF HOSPICE LENGTH OF SERVICE: AN ANALYSIS OF THE 2000 NATIONAL HOME AND HOSPICE CARE SURVEY HOSPICE DISCHARGES. Marie Bakitas MS, ARNP, AOCN, Dartmouth Hitchcock Medical Center, Lebanon NH US; Mayur Desai PhD, MPH, Yale University School of Medicine, New Haven CT US

Determining factors associated with hospice length of service (LOS) can assist clinicians, administrators, and policy makers to identify changes in practice and policy that could improve hospice utilization.

The National Home and Hospice Care Survey (NHHCS) is nationally representative, underutilized data source for identifying national trends in hospice care.

The Anderson Model of Health Services Use and previous studies of factors influencing hospice LOS guided variable selection for the study. The discharge data file of the NHHCS for 2000 was used as it represents a complete episode of hospice care. Of the 6273 total patients in the database, analyses were conducted on the 2016 who were receiving hospice care, were discharged due to death and were over 30 years of age. Independent variables included: age, gender, marital status, race/ethnicity, primary source of payment, agency proprietary status, and two measures of dependency: ADL and IADL status. The outcome variable was length of service (LOS), measured in days for a complete episode of hospice care.

Bivariate analyses to determine mean LOS for all independent predictors were performed using ANOVA. Associations between length of stay and age, gender, marital status, race, dependency, primary payer, and agency proprietary status were determined. Linear regression modeling using a stepwise backward elimination strategy was used to determine the most parsimonious model. All analyses were performed using SAS software version 8. The mean age of the sample was 74 years (SD 12.8; range 32-107) and the median LOS was 19 days (mean 56; range 1-3297 days). The unadjusted shortest mean LOS occurred in adults who were least dependent, married, non-Hispanic Black in a not-for profit Medicare Certified Hospice agency, with private insurance. In exploratory multivariate analyses, after log transformation of LOS substantial dependency in activities of daily living (ADLs) was the most significant predictor of longer LOS. In the adjusted model Black race became a significant predictor of shorter LOS. LOS continues to be much shorter than the assumptions on which the Medicare Hospice Benefit was developed. Dying persons and their families are unlikely to achieve the full benefit of care when referrals occur late in the dying process.

Poster Abstract
A CLINICAL TRIAL OF GINKGO BILOBA FOR THE PREVENTION OF CHEMOTHERAPY -RELATED COGNITIVE DYSFUNCTION. Debra Barton RN, PhD, AOCN, Mayo Clinic College of Medicine, Rochester MN US; Charles Loprinzi MD, Mayo Clinic College of Medicine, Rochester MN US; Tom Fitch MD, Mayo Clinic Scottsdale, Scottsdale AZ US; Kay Hoff RN, BSN, Missouri Valley CCOP, Lincoln NE US; Cathy Basham RN, BSN, OCN, Missouri Valley CCOP, Omaha NE US; Paul Novotny MS, Mayo Clinic Rochester, Rochester MN US; Jeff Sloan PhD, Mayo Clinic Rochester, Rochester MN US; Lisa Carpenter RN, BSN, Mayo Clinic Rochester, Rochester MN US

Descriptive studies identify neurocognitive dysfunction ("chemo-brain"), consisting of short term memory loss, slowed reaction time and thought processing, as a symptom experienced by up to 40% of patients receiving chemotherapy and/or radiation for cancer. Little is known about the actual prevalence or trajectory of this symptom, or what can be done to prevent it.

One of the main limitations in the current research on cognitive dysfunction is that descriptive studies are cross sectional, usually after treatment is completed, thus, there are no baseline assessments. The purpose of this study is to test an herbal agent for efficacy in preventing cognitive changes during chemotherapy, and also to describe the trajectory of cognitive changes in women receiving breast cancer treatment.

The theoretical framework for this study comes from Dodd and colleagues at the University of California, San Francisco. The risk factor, that being chemotherapy, is evaluated with respect to subjective cognitive changes and objective measures of cognitive morbidity.

This study utilizes a two-arm, randomized, placebo-controlled, double-blind, phase III design to evaluate ginkgo biloba, 60 mg twice a day, versus placebo to prevent cognitive dysfunction in 220 women who are beginning adjuvant breast chemotherapy. The primary endpoint is the level of cognitive dysfunction as measured by the High Sensitivity Cognitive Screen. Secondary endpoints include mood and quality of life issues. Data are collected at baseline, at every chemotherapy cycle, after chemotherapy is completed, and at six month intervals out to two years post-treatment.

Baseline data will be presented for the first 76 patients enrolled on the trial. Descriptive, correlational analyses and t-tests were done to look at differences in demographic characteristics and baseline cognitive scores.

There is wide variability on baseline cognitive scores with 50% of the participants scoring outside of the normal range. Postmenopausal, compared to premenopausal, women had statistically significantly lower cognitive scores in some areas. The vigor/activity subscale of the Profile Of Mood States revealed low scores and patients perceived some loss of cognitive function at baseline. The variability of baseline scores raises questions about the available cross sectional cognitive dysfunction data. Funding Sources: CA37404-19 CCOP

Poster Abstract
ASSESSING FAMILIES NEEDS FOR INTERVENTION. Joan Bickes MSN, APRN, BC, Wayne State University, Detroit MI US; Margaret Falahee MSN, APRN, BC, Wayne State University, Detroit MI US; Julie Walker MSN, APRN, BC, University of Michigan, Ann Arbor MI US; Laurel Northouse PhD, RN, FAAN, University of Michigan, Ann Arbor MI US; Ann Schafenacker MSN, RN, University of Michigan, Ann Arbor MI US; Janet Harden PhD, RN, Wayne State University, Detroit MI US; Esther Bennett MSN, RN, Wayne State University, Detroit MI US; Darlene Mood PhD, Wayne State University, Detroit MI US; Elizabeth Galvin MSN, RN, Karmanos Cancer Institute, Detroit MI US; Susan Holmes MSN, RN, University of Michigan, Ann Arbor MI US

Families facing cancer often differ in their need for interventions that are offered within randomized clinical trials. Most studies, however, use "one-size-fits-all" interventions that are not targeted to the varying needs of families. In a time of cost constraint, more studies need to examine families’ needs for intervention so that programs of care are delivered to those in need of services.

The purpose of this study was to: 1) describe assessments of family dyads’ needs for intervention at the beginning of a randomized clinical trial, 2) determine if their initial needs changed over time, and 3) identify factors associated with varying levels of need.

A stress-coping framework, adapted from Lazarus et al., guided this study.

A secondary analysis of data obtained during a family-based randomized clinical trial was used to address study aims. The sample consisted of patients with advanced breast cancer and their family caregivers who participated in a family intervention (N=72 dyads) and whose need for intervention was assessed by Advanced Practice Nurses. Dyads were rated on a five-point Need for Intervention Scale during the initial intervention session and at four follow-up sessions. Participants completed standardized instruments: Current Concerns Scale, Symptom Scale, Personal Resource Questionnaire, Mishel Uncertainty in Illness Scale, Beck Hopelessness Scale, Appraisal of Illness and Caregiving Scales, and two quality of life measures (FACT G, MOS-SF 36), all with adequate psychometric properties.

Descriptive statistics and Pearson r correlation coefficients were used for analyses. At initial assessment, 31% of the dyads had low, 27% moderate, and 42% high need for intervention. Over time 41% had a decrease, 20% an increase, 28% no change, and 11% fluctuated in need for intervention. Patients with higher need for intervention scores had more hopelessness, uncertainty, symptoms, negative appraisal of illness, and lower education, support, and overall quality of life. Caregivers with higher need for intervention had more negative appraisal of caregiving and lower mental quality of life.

Implications suggest that families’ needs differ and that more effort needs to be directed toward targeting interventions according to dyads’ needs. Funding Sources: American Cancer Society

Poster Abstract
A PILOT TRIAL OF INNOHEP (TINZAPARIN) LOW MOLECULAR WEIGHT HEPARIN FOR PRIMARY PROPHYLAXIS OF VENOUS THROMBOEMBOLISM IN BRAIN TUMOR PATIENTS. Cindy Bohlin RN, MSN, Duke University Health System, Durham NC US; Kara Penne RN, BSN, OCN, Duke University Health System, Durham NC US; Peggy Lyons RN, BSN, Duke University Health System, Durham NC US; James Vredenburgh MD, Duke University Health System, Durham NC US

The incidence of venous thromboembolic complications in patients with malignant gliomas is significantly increased over the general population. Oncology nurses must be aware of this increased risk, not only in patients with malignant glioma, but any malignancy. In many cases it is the nurse who detects the initial symptoms of thromboembolic disease and brings the patient in for diagnosis and treatment.

This study will determine the safety of prophylaxis with Tinzaparin in primary malignant glioma patients. We will determine the incidence of deep vein thrombosis or pulmonary embolism in brain tumor patients who receive Tinzaparin as primary prophylaxis. Overall survival of patients with malignant glioma who recieve Tinzaparin will be analyzed.

Patients with malignant gliomas have a 5-25% risk of developing venous thromboembolic disease. The treatment of venous thromboembolic disease requires systemic anticoagulation and/ or mechanical barriers such as the inferior vena caval filter. Systemic anticoagulation carries the risk of CNS hemorrhage, particularly in patients with malignant glioma. Effective prophylaxis should decrease the complications associated with therapeutic systemic anticoagulation in the malignant glioma patient.

A single arm pilot trial will enroll forty patients with newly diagnosed pathologically confirmed WHO Grade III or Grade IV glioma. Patients will receive a single daily subcutaneous injection of Tinzaparin at 4500 IU beginning at least 48 hours after but no later than four weeks from craniotomy or stereotactic biopsy. A primary analysis will be conducted at six months and the safety will be determined by the incidence of clinically significant bleeding or CNS hemorrhage. If the patient does not have any hemorrhage, the Tinzaparin will be continued for an additional six months.

A 3-stage design will be used to evaluate the safety of Tinzaparin. A Kaplan and Meier analysis will be used to describe patient survival. The proportion of patients who experience thromboembolic disease will be tabulated.

The study is actively enrolling patients. At the time of this abstract, preliminary findings are not available. If fewer than 4 out of 40 patients experience unacceptable toxicity, Tinzaparin treatment will be considered for incorporation into a phase III trial.

Poster Abstract
FATHERHOOD AND SMOKING: HOW MEN JUSTIFY THEIR TOBACCO USE. Joan Bottorff RN, PhD, University of British Columbia, Vancouver BC CA; John Oliffe RN, PhD, University of British Columbia, Vancouver BC CA; Joanne Carey BA, Masters Candidate, University of British Columbia, Vancouver BC CA; Larry Mroz MSC, PhD Student, University of British Columbia, Vancouver BC CA; Cecilia Kalaw MA, University of British Columbia, Vancouver BC CA; Joy Johnson RN, PhD, University of British Columbia, Vancouver; Lorraine Greaves PhD, BC Centre of Excellence for Womens Health, Vancouver BC CA; Miriam Stewart RN, PhD, Unversity of Alberta, Edmonton AB CA

Tobacco use is associated with lung cancer and more than a dozen other types of cancer. Better interventions are needed to create the conditions for long-term smoking cessation and limit children’s exposure to ETS.

Although many women stop smoking during pregnancy, partner smoking status is one of the strongest predictors for smoking relapse. The purpose of this study was to examine men’s justifications of their smoking during their partners’ pregnancy and the postpartum period. The study was part of a larger project exploring the influence of couple interactions on women’s tobacco reduction.

Theoretical Framework: Symbolic interactionism In this grounded theory study, interviews with men who smoked were conducted following delivery and at 3-6 months postpartum. Although the interviews focused on the challenges the woman’s tobacco reduction posed for couples, the men also shared information about their own smoking.

Data were analyzed using constant comparative methods. Theories related to fatherhood and related concepts were reviewed to enhance theoretical sensitivity

The men constructed smoking as part of their identity by drawing on rights-based arguments and the pleasures of smoking as necessary for a good life. Although they drew attention to their active lifestyles to negate any ill effects from smoking, others acknowledged their addiction to reduce expectations for cessation. Smoking was presented positively in relation to fatherhood by providing a needed source of emotional stability and a stress reliever. They described smoking strategically to reduce the amount their pregnant partners smoked and as a way to support the child’s adjustment to the "toxins of the world."

Men’s resolve to continue to smoke, however, was challenged directly or indirectly by their pregnant partners’ cessation, by idealized notions of fathering, and by practicing fatherhood. Men reported their non-smoking partners were often jealous and angry about their continued smoking and resented the financial burden of their tobacco use. Men responded to these pressures with vague promises to quit, by minimizing the impact of their smoking and, in some instances, by supporting women’s return to smoking. The study findings indicate that nurses should take advantage of opportunities to address men’s smoking in the context of pregnancy and postpartum. Funding Sources: Canadian Institutes for Health Research

Poster Abstract
BREAST BIOPSY EDUCATION: PREPARING AFRICAN AMERICAN WOMEN FOR BREAST BIOPSY. Patricia Bradley PhD, RN, CS, Villanova University, Villanova PA US; Catherine Piccoli MD, Thomas Jefferson University, Philadelphia PA US; Ronald E. Myers PhD, Thomas Jefferson University, Philadelphia PA US

Breast cancer, the second leading cause of cancer death among African American women (AAW), is often diagnosed as a result of a biopsy procedure performed following an abnormal mammography result. Patient lack of knowledge, anxiety about undergoing a breast biopsy, and perception of inadequate social support has been associated with delay in help-seeking among women experiencing breast cancer symptoms.

Primary aims of the study were to: 1). develop an educational counseling intervention that involved a nurse educator and AAW who were recommended to have a breast biopsy. 2). pilot test the intervention and 3). assess the extent to which the intervention affected biopsy-related knowledge, stress, and perceived social support among participating women.

This study was conducted in two phases using self-regulation theory as a guide.

In phase one, an educational booklet and a counseling session protocol were developed and pre-tested using a focus group format with AAW (n=20) with a history of breast cancer. Phase two of the study used a one-group, pre/post-test design to pilot-test the efficacy of the educational intervention with 23 AAW who were recommended to have a breast biopsy. Eligible women provided consent, completed a baseline and endpoint survey, and participated in an educational counseling session and a post-biopsy follow-up telephone interview.

Frequencies were tabulated for the sociodemographic, baseline and endpoint variables. The non-parametric Wilcoxon test for paired data was used to assess pre/post differences of participant’s scores on knowledge, biopsy-related stress, and perceived level of social support measured on baseline and endpoint surveys.

Prior to biopsy, women expressed concern regarding the biopsy procedure and possible diagnostic outcome. After exposure to the intervention, biopsy-related knowledge, response efficacy and perceived support tended to increase. Participants rated the intervention as highly informative, useful and relevant. This study addressed the need to develop and test a theory-based intervention that can be used to prepare women for the breast biopsy experience. Findings from this study can be used to design a controlled trial to assess the efficacy of the intervention relative to knowledge, anxiety, social support and time to biopsy. Funding Sources: NCI Special Populations Network

Poster Abstract
ITEM ANALYSIS OF THE BLESCH MENS HEALTH BELIEF SURVEY. Carlton Brown RN, MSN, AOCN, Walter Reed Army Medical Center, Washington DC US

The purpose is to 1) present findings related to the item analysis of the Blesch Men’s Health Belief Survey (MHBS) 2) to present the Cronbach alpha coefficients from other studies using the MHBS, 3) to discuss particular issues related to the psychometric evaluation of item analysis. The MHBS is a tool which investigates knowledge of testicular cancer and the practice of testicular self-examination.

All measurement tools are subject to some form of measurement error, especially those that examine behavior. Random error results from chance variations in the measurement tool and from the occurrence of some chance extraneous factor that affects all measurements made with the same instrument. One cause of random error can be attributed to poor internal consistency which deals specifically with the homogeneity of a scale. Internal consistency can be equated using Cronbach’s coefficient alpha through item analysis.

Steps of item analysis will be presented. Item analyses were conducted on the four subscales (perceived susceptibility, perceived benefits, perceived susceptibility, and perceived seriousness) making up the MHBS. On the barriers subscale the Cronbach alpha coefficient was 0.7984, the benefits subscale .8424, the seriousness subscale .345, and susceptibility subscale 0.6361. Comparisons were also made to Cronbach alphas from other studies using MHBS in research.

There is concern in the low alpha in the seriousness scale. A literature review was done of other item analysis from other studies using the Blesch tool. Interestingly, the review revealed possible problems with the seriousness subscale consistently across studies. These low alphas suggest that the questions in this subscale may need revision. Caution should be taken when presenting significant findings using this current seriousness subscale because it doesn’t appear to be particularly reliable.

There is a significant requisite for a measurement tool that measures changes in practice. In order to use the MHBS in future studies, the seriousness subscale must be revised. Since the other three subscales (barriers, benefits, and susceptibility) have evidence of strong internal consistency and therefore, reliability, with revision to the seriousness subscale, MHBS may prove adequate in measuring changes in knowledge and practice in men at risk for testicular cancer. Funding Sources: Triservice Nursing Research Grant

Poster Abstract
BENCHMARKING AFRICAN AMERICAN RECRUITMENT TO CANCER CONTROL TRIALS WITH SOCIAL MARKETING AND DIRECT RESPONSE RADIO. Deborah Watkins Bruner RN, PhD, Fox Chase Cancer Center, Philadelphia PA US; Andre Konski MD, Fox Chase Cancer Center, Philadelphia PA US; Steven Feigenberg MD, Fox Chase Cancer Center, Philadelphia PA US; Natalie Dewberry-Moore MS, Fox Chase Cancer Center, Philadelphia PA US; Julia Goplerud BS, Fox Chase Cancer Center, Philadelphia PA US

Ethnic diversity in recruitment is a vital prerequisite to eliminating health disparities through cancer control trials.

The Prostate Cancer Risk Assessment Program (PRAP) is a research/screening program for those at high-risk for the disease. Eligibility is based on age (35-69 yrs), family history, and ethnicity (African American (AA), regardless of family history). AAs, having among the highest prostate cancer (PC) incidence and mortality in the world, are a focus of PRAP recruitment and account for 59% (296/498) of participants. Study aims were to develop and assess response to a theory-driven, ethnically appropriate direct response radio (DRR) message to promote information seeking and recruitment of AA males into PRAP.

A DRR advertisement for risk assessment and screening was designed guided by a conceptual framework that included Social and Relationship Marketing.

Focus groups were used to help develop messages. A 60-second ad aired for a total of 154-spots over 10 wks between 1/5/04 - 3/14/04. Assuming an average daily audience of 78,314 listeners on two primarily AA stations such that (as per the radio stations) 25% are males between 35 and 64 yrs and 86% are AA, estimated daily sample size of eligible AA men reached by the ads was 17,033.

103 telephone responses, most of whom were AA, were elicited from 1/5/04 through 5/31/04: 73 men requested information only, 5 who desired screening were ineligible, 25 were recruited (12 of these men did not keep their appointment [11 black, 1 white], and 13 were accrued [10 black, 3 white]). Assuming stable listenership over the airing period the results indicate a 0.5% response rate to the DRR ads.

An extensive search of the literature was unable to find comparable reporting. One marketing analysis showed that the best response to DRR related to retail sales was 9.4% of total audience. Due to obvious differences between retail sales and recruitment for PC control trials, one would expect a much lower return for the latter. Results of this study can serve as a benchmark for future comparisons of the success of DRR to improve information seeking and recruitment of AA men into cancer control trials.

Poster Abstract
SLEEP AS DESCRIBED BY BEREAVED CAREGIVERS. Patricia Carter PhD, RN, CNS, The University of Texas at Austin, Austin TX US

Sleep is a natural, yet poorly understood process. Nurses can help bereaved caregivers understand the importance of sleep in bereavement. Bereaved caregivers are often lost to services. Sleep quality may have a significant impact on caregivers' abilities to move through their bereavement. The purpose of this project was to elicit descriptions of sleep quality in bereaved caregivers and to understand how sleep quality related to movement through bereavement.

A phenomenological frame of reference was used to guide this study to seek the lived experience of bereaved caregivers as it related to sleep quality.

Bereaved caregivers were recruited from a local Hospice. All participants cared for their patient at home, were be able to speak, read and write English, and were at least 21 years old. Participants were interviewed at a location of their convenience by a PhD prepared nurse. Demographics, self-report sleep (PSQI) and depressive symptom (CESD) measurements were taken before the interview. Participants were asked to provide descriptions of their sleep during care and after the loss of the patient. Participants were also asked to describe their impressions of how their sleep impacted their physical and social health, quality of life, and bereavement. Interviews were transcribed verbatim and coded line by line to identify categories and themes by two PhD prepared nurses familiar with qualitative methods. Findings were brought back to participants for validation.

Descriptive analysis was performed on demographics, PSQI, and CESD data. Narratives were analyzed using open coding and line-by-line analysis to arrive at categories and themes to represent the data. Findings suggest that sleep is severely impaired in bereaved caregivers within 6 months of the loss of the patient. Further research is indicated to explore the patterns of sleep change over time in this population and to design behavioral sleep interventions to improve sleep quality in this vulnerable population.

Educational and practice implications include the need to educate nurses to understand the impact loss of a patient has on the surviving caregivers and how sleep may influence the caregivers' movement through bereavement. Funding Sources: Center for Health Promotion and Disease Prevention Research, The University of Texas at Austin, School of Nursing

Poster Abstract
RECRUITMENT RETENTION RELIABILITY: THE CHALLENGES OF CONDUCTING A RANDOMIZED CLINICAL TRIAL. Julie Chamberlain MS, RN, University of Nebraska Medical Center, Omaha NE US; Ann Berger PhD, RN, AOCN, University of Nebraska Medical Center, Omaha NE US

The purpose of this presentation is to provide a model to guide the PD in the development of strategies to meet the challenges of recruitment, retention and reliability within a RCT. A second purpose is to encourage dialogue and networking to foster growth in this nursing role.

The role of the Project Director (PD) is critical to the success of a randomized clinical trial (RCT) and presents many challenges that have no specific answers. Today with HIPAA regulations and IRB constraints, recruitment and retention are daily challenges. Reliability requires great attention to the operation of the RCT, delivery of the intervention and management of the database. Nurses have an excellent opportunity to be mentored through the research process and lead the research team, yet many of these day-to-day challenges have no written guidelines for practice and can consume large amounts of time and effort.

The center of the model contains the participant with the focus on recruitment. HIPAA regulations have made recruitment a challenging process that requires much effort and creativity in establishing clinical relationships and marketing the study. Two sections surround the center: the first section is retention of participants in either arm of the study. In addition to receiving the intervention or the attentional control, the participant may receive additional support and incentives. The second section is operational and the challenge is maintaining reliability of the study. Data collection, entry and management are included here. Team meetings continue, standardization of procedures is ensured and on going reports are sent to review boards that oversee the study.

Using this model can assist the PD in implementing plans and procedures to meet the challenges of recruitment, retention and reliability and promote networking and dialogue to foster growth in this nursing role. The multifaceted role of the PD overlies the entire model. The PD must continually communicate, coordinate, educate and evaluate in order to successfully manage and meet the challenges of a RCT. To facilitate professional growth, the PD will benefit from dialogue and networking with mentors in the field. Funding Sources: NIH/NINR RO1 NR007762-03

Poster Abstract
ASSESSING COGNITIVE FUNCTION IN WOMEN WITH AND WITHOUT BREAST CANCER USING FUNCTIONAL MAGNETIC RESONANCE IMAGING (FMRI). Bernadine Cimprich PhD, RN, FAAN, University of Michigan School of Nursing, Ann Arbor MI US; Barbara Therrien PhD, RN, University of Michigan, Ann Arbor MI US; Patricia Reuter-Lorenz PhD, University of Michigan, Ann Arbor MI US; Daniel Normolle PhD, University of Michigan, Ann Arbor MI US; Patricia M. Clark MSN, RN, University of Michigan, Ann Arbor MI US; Catherine Vincent PhD, RN, University of Michigan, Ann Arbor MI US; Daniel F. Hayes MD, University of Michigan, Ann Arbor MI US; Douglas Noll PhD, University of Michigan, Ann Arbor MI US; Robert Welsh PhD, University of Michigan, Ann Arbor MI US; James Nelson PhC, University of Michigan, Ann Arbor MI US

Cognitive impairment is a common side effect of adjuvant chemotherapy for breast cancer having detrimental effects on daily functioning, valued life roles, and quality of life.

The incidence of persistent cognitive deficits following adjuvant chemotherapy for breast cancer ranges from 17%-35%, although the underlying causative brain mechanisms are not well understood. The purpose of this pilot study was to determine the feasibility of using fMRI to assess attention and working memory in women newly diagnosed with early stage breast cancer.

The study focused on two basic cognitive processes, attention and working memory, because they are prerequisites for effective function in daily life.

Ten women with Stage I or II breast cancer (31-61 years) prior to adjuvant chemotherapy and 10 healthy controls (31-64 years) were recruited. Subjects completed a Verbal Memory Task (VMT) with four levels of task demand to assess brain activation patterns related to attention and working memory during fMRI.

Descriptive analyses of VMT reaction time and accuracy were performed. Brain image analyses were performed using the SPM99 analysis package.

Preliminary behavioral and imaging analyses are completed for 7 patients and 7 controls. Accuracy on the VMT for patients was 88%-98% and for controls, 92%-97%. Patients had significantly (p<.05) slower reaction times in the more demanding task conditions. High vs. low demand conditions showed group differences in brain function (using a statistical threshold of p<.01). Controls showed expected activation in the medial frontal cortex (MFC) in high demand conditions. In contrast, in the high demand conditions, patients showed increased activation not only in the MFC but also in the left inferior frontal gyrus, right lateral inferior frontal areas, left parietal and right parietal areas. Preliminary findings show a trend of slowed performance on a working memory task in a group of women with early stage breast cancer prior to adjuvant chemotherapy. Patients also showed recruitment of additional components of attention/working memory circuitry in both hemispheres in order to perform the more demanding task. Behavioral and imaging findings in this study suggest vulnerability in attention and working memory systems in middle age women with breast cancer prior to adjuvant chemotherapy. Funding Sources: Research Initiative Funds from University of Michigan fMRI Research Laboratory and School of Nursing

Poster Abstract
IN TUNE: PATIENTS FROM THREE ETHNIC GROUPS DESCRIBE NURSING EXPERTISE. Marlene Cohen RN, PhD, FAAN, The Univeristy of Texas Health Science Center at Houston School of Nursing, Houston TX US; Jane Mahoney, RN, DSN, The University of Texas Health Science Center at Houston, School of Nursing, Hou, Houston TX US; Daniel Lara none, The University of Texas Health Science Center at Houston, School of Nursing,, Houston TX US; Guadalupe Palos, RN, LMSW, DrPH, The University of Texas M.D. Anderson Cancer Center,, Houston TX US; Elizabeth Carlson DSN, MPH, RN, The University of Texas M.D. Anderson Cancer Center, Houston TX US

Understanding patients’ experiences with care is an important supplement to professional discussions of what is essential in effective nursing care. Much research has included too few persons of diverse ethnicity to know how experiences are similar and different for different groups. Descriptions of experiences of persons of diverse cultures who had blood and marrow transplantation (BMT) included descriptions about providers that revealed aspects of nurses’ behaviors and attributes that were both helpful and not helpful. The purpose of this presentation is to describe and illustrate aspects of expertise that nurses can use to provide purposeful patient centered care in a way that patients described as important and useful to them.

Hermeneutic phenomenological philosophy guided this study. Hermeneutic phenomenological research, the study of how people interpret their lives and make meaning of their experiences, guided this portion of a larger study. Interviews were completed with 20 Caucasians, 20 African Americans, and 20 Latinos patients (interviewed by bilingual English-Spanish staff) 5 times over the first 100 days of their BMT. Interviews continued until data were saturated.

Data were analyzed by a culturally diverse research team using hermeneutic phenomenological research methods. Transcripts were coded into logical categories, data in categories were compared, and the category of nurse expertise emerged as a key aspect of the experience.

These BMT patients described nurses’ expertise as including four aspects: a strong knowledge base, technical expertise, intuitive abilities to assess and intervene effectively, and interpersonal expertise, which included human connection, empathy, caring, and the intention of helping. The detailed examples patients provided illustrated that nurses who were "in tune" with them had an awareness that is described as a result of reflective practice. They were able to assess and intervene effectively because they understood both verbal and non-verbal cues. These data did not differ among the ethnic groups, indicating that culturally competent expert care does not differ for these groups. Nurses guided by these aspects of expertise will be more likely to note the aspects of care that patients valued that are sometimes overlooked. Funding Sources: This project was funded by grant RO1 NR05188-01A2 from the National Institutes of Health, National Institute of Nursing Research.

Poster Abstract
A RANDOMIZED TRIAL OF HOME-BASED EXERCISE FOR PATIENTS RECEIVING INTENSIVE TREATMENT AND EPOETIN ALFA FOR MULTIPLE MYLEOMA. Ann Coleman PhD, RNP, AOCN, College of Nursing, University of Arkansas for Medical Sciences, Little Rock AR US; Sharon K. Coon PhD, RN, AOCN, College of Nursing, University of Arkansas for Medical Sciences, Little Rock AR US; Kathy C. Richards PhD, RN, College of Nursing, University of Arkansas for Medical Sciences, Little Rock AR US; Elias J. Anaissie MD, MIRT, University of Arkansas for Medical Sciences, Little Rock AR US; C. Beth Stewart BS, College of Nursing, University of Arkansas for Medical Sciences, Little Rock AR US; Jennifer L. Shaw MPH, MAP, BA, University of Arkansas for Medical Sciences, Little Rock AR US

A home-based individualized exercise program for patients receiving intensive treatment may decrease cancer-related insomnia and fatigue. Approximately 50% of patients with cancer have insomnia and up to 100% have fatigue. At least 60% of patients with multiple myeloma (MM) are anemic (hemoglobin [Hb], 12 g/dL) at diagnosis. Nearly all become anemic during intensive treatment for MM and many require red blood cell (RBC) transfusions. Anemia can contribute to fatigue, the most common and often the most distressing symptom for patients with cancer. The purpose of this study is to determine if a home-based exercise program incorporating aerobic and strength resistance training will decrease fatigue and improve nighttime sleep and performance in patients receiving intensive treatment for MM and epoetin alfa therapy.

Winningham's theory of cancer-related fatigue provides the theoretical rationale for this study. Too much or too little rest and/or too much or too little exercise increases fatigue and results in decreased activity, which then leads to physiological deconditioning and increased fatigue. The resulting negative feedback loop results in a downward spiral toward disability. Aerobic exercise decreases fatigue for individuals with cancer and improves sleep for healthy individuals. Patients with cancer also experience skeletal muscle wasting; maximum benefit may be obtained when aerobic exercise is combined with strength resistance training.

This randomized controlled trial of a 15-week experimental period with 2 groups (exercise or usual care) with measurements at baseline, before stem cell mobilization, and after stem cell collection has a planned enrollment of 200 patients. Measurements, using instruments with established validity and high reliability, comprise fatigue (Profile of Mood States and Functional Assessment of Cancer Therapy-Fatigue), minutes of nighttime sleep (Actigraph), aerobic capacity (6 Minute Walk Test), muscle strength (weight lifted in pounds using dynamometer), Hb levels and RBC transfusions.

Available data from all patients completing study (currently 134) will be included in this preliminary analysis using descriptive statistics and repeated measures analysis of variance.

Preliminary analyses to date show trends towards increased nighttime sleep, better performance (aerobic capacity and muscle strength), and fewer RBC transfusions in the exercise group compared to the usual care group. Funding Sources: UAMS College of Nursing Intramural Grant, Arkansas Tobacco Research Fund, Ortho Biotech, Inc.

Poster Abstract
CHALLENGES TO DEVELOPING A SUPPORT INTERVENTION FOR AFRICAN-AMERICAN WOMEN WITH BREAST CANCER. Doris Coward RN, PhD, The University of Texas at Austin, Austin TX US

A needs assessment and recruitment strategies consistent with ONS Multicultural Outcomes (ONS, 1999) guidelines were used to document the need for, and to adapt, a support group intervention in a Central Texas urban area for African-American women newly diagnosed with breast cancer. The purpose of this poster is to describe challenges encountered in adapting, and recruiting for, the support group.

Differential treatment and decreased survival in African-American women with breast cancer may be related in part to lack of knowledge and assertiveness in newly-diagnosed women (Ashing-Giwa and Ganz (1997). Promotion of breast cancer advocacy includes empowering survivors to advocate for themselves and for other women. Although cancer support groups are traditional resources for the education and support that promotes empowerment, few minority persons participate (Barg & Gullatte, 2001). Another approach may be necessary to empower African-American women to effectively advocate for treatment information and appropriate care.

Local African-American breast cancer survivor leaders perceived a need for education and support for newly diagnosed women. However, few such women participated in local cancer education and support resources, including a Self-Transcendence Theory Breast Cancer Support Group Project developed by a CNS researcher specifically for newly diagnosed women. The CNS partnered with the survivor leaders to adapt the theory-based group specifically for local African-American women. While recruiting for support group participants, the CNS helped plan, and presented at, monthly cancer education programs sponsored by Sisters Network, a national African-American breast cancer survivors’ organization. However, over a two year time period not enough newly-diagnosed women were recruited to conduct and test a support group intervention specifically for African-American women.

Accruing participants from a relatively small minority population is problematic even when collaborating with population leaders. Starting early to build credibility and trust in that population, being flexible with eligibility criteria, starting with pre-experimental studies, and even paying participants may be required.

Although adapting and testing effective cancer education and support interventions for African-Americans is needed, nurse researchers may encounter barriers to conducting such projects. Lessons learned from this project may facilitate recruitment of participants and successful project completion in future projects. Funding Sources: Funded by NIH (NINR and NCCAM)

Poster Abstract
THE SYMPTOM CLUSTER OF PAIN, FATIGUE, DEPRESSION, AND SLEEP DISRUPTION IN WOMEN WITH BREAST CANCER. Jean Davis PhD, RN, Wayne State University, Detroit MI US; April Vallerand PhD, RN, Wayne State University, Detroit MI US; Kimberly Hart MD, Wayne State University, Detroit MI US

Patients with cancer rarely present with a single symptom. There is a need to assess multiple symptoms in patients experiencing pain because pain is frequently accompanied by depression, fatigue, and sleep pattern changes. There is evidence that patients with multiple symptoms have poorer outcomes. Interleukin 6 (IL-6), a cytokine related to sick behaviors and changes in sleep patterns, may play a role in modulating the symptom cluster of pain, fatigue, depression, and sleep disruption.

Although close to 50% of cancer patients receiving therapy reportedly suffer from symptoms of pain, fatigue, depression, and sleep disruption, little research has been done to study this symptom cluster. The purposes of this study are to: (1) determine the relationships among the symptoms of pain, depression, fatigue and sleep disruption as a symptom cluster in women diagnosed with breast cancer undergoing radiation therapy, (2) determine blood levels of IL-6 in these women, and (3) determine the relationships among IL-6 and the symptoms of pain, fatigue, depression, and sleep disruption.

The Conceptual Model of Symptom Management guides this study. The dimension of symptom experience will be explored in terms of a symptom cluster. A descriptive, correlational design will be employed.

Sixty pre-menopausal women 18 years of age and older diagnosed with DCIS Stage 1b or less, post-lumpectomy, and receiving radiation therapy will be enrolled. The following standardized instruments with adequate reliability and validity will be used: Brief Pain Inventory, Center for Epidemiological Studies - Depression Scale, Brief Fatigue Inventory and the Pittsburgh Sleep Quality Scale. IL-6 blood levels will be measured by ELISA radioimmunoassay techniques.

Informed consent will be obtained from eligible participants. Scales will be administered and IL-6 blood levels will be drawn during week 5 of radiation therapy.

Subject characteristics will be analyzed using measures of central tendency and dispersion Descriptive and appropriate correlational statistics will be employed to analyze the major variables.

Assessing symptom clusters, especially those of pain, fatigue, depression and sleep disruption, in cancer patients is essential to optimal symptom control from diagnosis throughout the course of the illness. Funding Sources: Detroit Medical Center/Wayne State University College of Nursing Scholars Award

Poster Abstract
TREATING CHEMOTHERAPY-INDUCED NAUSEA WITH ACUPRESSURE. Suzanne Dibble DNSc, RN, University of California, San Francisco, San Francisco CA US; Jill Israel BSN, University of California, San Francisco, San Francisco CA US; Misha Cohen OMD, LAc, Chicken Soup Chinese Medicine, San Francisco CA US; Judy Luce MD, University of California, San Francisco, San Francisco CA US

Although the newer antiemetic agents have controlled a notable portion of the vomiting associated with chemotherapy administration, delayed nausea continues to be a significant problem.

Therefore, the specific aims of the current randomized clinical trial (5R01CA084014-03) are to compare differences in the nausea experience and intensity among three groups (Total N=237) undergoing doxorubicin hydrochloride (Adriamycin®) and cyclophosphamide with or without fluorouracil chemotherapy for breast cancer. The groups are those receiving a) usual nausea care plus "Active Acupressure" via finger pressure on the nei guan point (P6), b) usual nausea care plus placebo acupressure and c) usual nausea care.

Acupressure is based upon the concept that the body is activated by "Qi" or a type of energy that travels along pathways at differing depths of the body known as "meridians". Along the meridians are points known as "acu" points in Chinese, which are the pumping stations for the "Qi". These meridians, similar to a river, can become deficient, excessive, stagnant, or wayward. Applying pressure (acupressure) or inserting a needle (acupuncture) into one of these points can relieve imbalances by either stimulating or easing energy flow. Acupressure is the older of the two techniques; a Chinese home remedy that gave rise to the more "technological" approach of acupuncture.

Using twenty -two oncology settings, participants were recruited who had experienced nausea with their previous chemotherapy treatment. They were randomly assigned to treatment groups. Stratification criteria included regimen and site. The interventions were conducted by carefully trained research assistants. All participants will be followed on a daily basis for a cycle of chemotherapy (a cycle is usually 21-28 days).

Analyses will be done using repeated measures analysis of variance and analysis of covariance, when baseline data is an appropriate covariate. A strength of this study is that it does not pit modern Western and Chinese medicine against each other to determine which is more effective. All participants will continue to receive their Western medical care, but the added value of acupressure will be explored. Preliminary results of this trial will be presented. Funding Sources: (5R01CA084014-03)

Poster Abstract
CARDIOVASCULAR FITNESS, BODY COMPOSITION, AND MUSCLE STRENGTH IN CANCER PATIENTS DURING AND AFTER CANCER TREATMENT. Marylin Dodd RN, PhD, FAAN, School of Nursing, University of California, San Francisco, San Francisco CA US; Patricia Painter Ph.D., UCSF School of Nursing, San Francisco CA US; Christine Miaskowski R.N., Ph.D., FAAN, UCSF School of Nursing, San Francisco CA US; Steven Paul Ph.D., UCSF School of Nursing, San Francisco CA US; John Duda R.N., UCSF School of Nursing, San Francisco CA US; Maria Cho R.N., Ph.D., UCSF School of Nursing, San Francisco CA US

Preliminary studies report the benefits of exercise on cancer treatment related fatigue, but not on: (1) cardiovascular fitness, measured by peak oxygen uptake (VO2peak); (2) body fat composition, measured by Dual X-Ray absorptiometry (DEXA); or (3) muscle strength, measured by isokinetic muscle testing.

The purpose of this preliminary analysis is to test for changes in cardiovascular fitness, body fat composition, and muscle strength during and after cancer treatment.

The data for this preliminary analysis are from a larger randomized trial testing the effectiveness of an individually tailored exercise intervention (projected sample is 100). The integrated Fatigue Model provided the framework. Sample. Eighty women with breast cancer (mean age = 49 years (SD=9.5)). Subjects were tested at three time points: at baseline, which was between their first and second chemotherapy treatments (T1), after completion of cancer treatment (T2), and a duration equal to treatment time following cancer treatment (T3). Measures: Symptom-limited treadmill tests measure oxygen uptake (VO2peak), DEXA scans assess body composition, and isokinetic muscle function testing measures muscle strength (quadriceps peak torque). The isokinetic test selected for this analysis is knee extension peak torque at 180 degrees/sec.

A repeated measures ANOVA was conducted to determine the differences in the average VO2peak scores, body fat percentage, and peak torque over the three time points.

Results. There was a significant main effect of time in: (1) the average VO2peak scores, F2,75=10.65,p<.01 (T1= 25.21 + 6.4; T2 = 24.19 + 6.1; T3 = 25.94 + 6.61 ml/kg/min); (2) the average body fat composition, F2,75=10.63, p<.01 (T1 = 35.64+ 8.9; T2 = 36.5 + 8.7; T3 =37.13+ 8.7); and (3) the average peak torque, F2,73=3.34, p<.05 (T1= 47.32 + 14.9; T2 = 46.54 + 14.7; T3 = 48.68 + 13.6 ft/lb). Maximal exercise testing and muscle strength testing are well tolerated in patients undergoing cancer treatment. Overall, subjects undergoing cancer treatment tolerated maximal exercise without problems. Subjects experience reduced cardiovascular fitness and increases in body fat percentage and muscle strength during their treatment (T2) and return to baseline fitness and muscle strength after treatment (T3), while body fat (%) remains elevated. Funding Sources: NIH, NCI CA 83316

Poster Abstract
THE ASSESSMENT OF BARRIERS TO SYMPTOM REPORTING FOR WOMEN WITH OVARIAN CANCER. Heidi Donovan PhD, RN, University of Pittsburgh School of Nursing, Pittsburgh PA US; Margaret Fields RN, BSN, University of Pittsburgh School of Nursing, Pittsburgh PA US; Yookyung Kim PhD, University of Pittsburgh School of Nursing, Pittsburgh PA US; Susan Sereika PhD, University of Pittsburgh School of Nursing, Pittsburgh PA US; Joseph Kelley MD, University of Pittsburgh Cancer Institute & Magee Womens Hospital, Pittsburgh PA US

Women with ovarian cancer experience multiple concurrent symptoms from disease and treatment, but many under-report their symptoms to health care providers.

Multiple studies have identified patients’ beliefs that act as barriers to reporting single symptoms (e.g. pain, fatigue), but there is no instrument designed to assess barriers to reporting multiple, concurrent symptoms. The purpose of this study is to conduct psychometric evaluations of two versions of the Symptom Management Barriers Questionnaire (SMBQ). Version 1 includes separate questions for each of a subject’s three "most noticed symptoms"; Version 2 includes global questions that address "overall symptoms".

Leventhal’s self-regulation theory which proposes that persons’ illness-related representations (beliefs) drive coping with health threats. This study follows a single group, descriptive, correlational, within-subjects design. Subjects are 60 women experiencing > 2 symptoms associated with ovarian cancer or treatment. Immediately after an appointment with a physician, subjects complete valid, reliable measures of symptom severity, symptom reporting, symptom interference with life activities, in addition to both versions of the SMBQ. Subjects are randomized to order of completion of the two SMBQs. Subjects also complete both SMBQs, in the same order, two weeks later to evaluate test-retest reliability.

Data collection is in progress. Twenty-one women have been recruited to date. The following findings will be presented: 1) Internal consistency and test re-test reliability of the two versions of the SMBQ (inter-item and item-total correlations, Cronbach’s alpha, intraclass correlation coefficient); 2) Within-subject comparisons between the scores on the "overall symptoms" SMBQ and mean scores from the three "most noticed symptoms" SMBQ (t-tests); 3) Predictive validity of both SMBQs, by determining whether barrier scores on each SMBQ are related to symptom reporting and symptom interference with life activities (correlation and regression analyses); and 4) patient preference between the two versions (descriptive statistics).

Development of an instrument to assess barriers to multi-symptom reporting is an important first step in optimizing symptom management for women with ovarian cancer. Funding Sources: Center for Research in Chronic Disorders, University of Pittsburgh

Poster Abstract
DEVELOPMENT OF AN INSTRUMENT TO MEASURE INDEPENDENT NURSING ACTIONS IN COOPERATIVE CARE. June Eilers PhD, APRN, BC, The Nebraska Medical Center, Omaha NE US; Judith A. Heermann PhD, RN, University of Nebraska College of Nursing, Omaha NE US; Margaret E. Wilson PhD, CPNP, University of Nebraska College of Nursing, Omaha NE US; Jan R. Attwood PhD, RN, University of Nebraska College of Nursing, Omaha NE US

Significance: Cooperative Care (CC) is a model of care based on a partnership between a transplant recipient/lay caregiver dyad and healthcare professionals. Preparation of the lay caregivers to assume acute care responsibilities is a key nursing intervention. Identification and measurement of independent nursing activities, which aim to improve outcomes, is critical to evaluate interventions designed to educate/prepare lay caregivers.

Problem/Purpose: Although lay caregivers administer direct care to acutely ill transplant recipients, nurses remain ultimately responsible for outcomes. The purpose of this portion of a multi-phase study is to develop an instrument that captures the nature and dose of indipendent oncology nursing actions in CC.

Theoretical/Scientific Framework: The nursing process provides the broad theoretical framework for this study that focuses on the independent nursing activities considered to be within nursing's scope of diagnosis and treatment.

Methods: The design of the multi-phase instrument development project is inductive and descriptive using both qualitative and quantitative approaches. In the first phase of the study, qualitative data provided by the 12 CC nurses during care episodes guided the inductively determined domains and categories of independent nursing actions. In this phase of the study, items are being generated and will be piloted tested for clarity, apparent internal consistency, and content validity using the CC nurses as "content expert" raters. In analyzing ratings for clarity, the proportion of "clear" ratings per item has to meet the criterion level of .75-.80 of the raters. Proportions are computed and treated in the same way for apparent internal consistency and content validity procedures.

Data Analysis: Interpretation of the data for all three rating procedures occurs simultaneously. The decisions about clarity, apparent internal consistency, and content validity will be made together. However, an item that does not meet criterion on one rating will not be retained in the instrument. Proportions for retained items will be recomputed, with revisions made until the instrument meets criteria.

Implications: The instrument being developed will aid nursing in communicating CC nursing to others. Potential uses of the tool include testing of interventions designed to educate/prepare caregivers for their role as lay caregivers. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Bristol-Myers Squibb Oncology

Poster Abstract
RELATIONSHIPS AMONG OBJECTIVE SLEEP PARAMETERS, OBJECTIVE HOT FLASHES AND SUBJECTIVE HOT FLASHES IN BREAST CANCER SURVIVORS. Julie Elam RN, MSN, OCN, Indiana University, Indianapolis IN US; Janet S. Carpenter RN, PhD, Indiana University, Indianapolis IN US

Sleep disturbances are an ONS research priority. Findings suggest sleep disturbances are common in menopausal breast cancer survivors. Previous research found that the majority of breast cancer survivors (BCS) having hot flashes reported poor sleep quality and high sleep disturbance on subjective questionnaires (73%). However, the relationship between sleep and hot flashes has not been validated with objective measures in BCS. Objective sleep quality has also not been fully examined in relation to subjective hot flashes in BCS. To examine relationships between objective sleep parameters (sleep efficiency, wake bouts, and mean wake bout time) and objective and subjective hot flashes. The Lenz Theory of Unpleasant Symptoms served as the study framework. The three interrelated components of the model (symptoms experienced (number of wake bouts), influencing factors that cause the symptoms (hot flashes), and consequences of the symptoms (poor sleep efficiency, high number of wake bout minutes)) are studied here.

Two-week baseline data from a randomized, placebo-controlled, cross-over study testing the efficacy and side effects associated with use of an SNRI for hot flashes, mood, fatigue, sleep, and quality of life in breast cancer survivors. Data will be analyzed from 2 nights wrist actigraphy (objective sleep), two nights sternal skin conductance monitoring (objective hot flashes), and diary entries (subjective hot flashes).

Sample demographics will be analyzed using frequencies and descriptive statistics. Correlations between sleep (percentage sleep efficiency, number of wake bouts, and mean time of wake bouts in minutes) and hot flashes (objective and subjective frequency) will be reported.

Expected findings between sleep and hot flashes include: (1) negative correlation between % sleep efficiency and hot flash frequency (objective and subjective), (2) positive correlation between number of wake bouts and hot flash frequency (objective and subjective), and (3) positive correlation between mean wake bout time and hot flash frequency (objective and subjective). Findings will provide increased knowledge regarding the relationship between sleep and hot flash frequency in BCS and may be useful for guiding future descriptive or intervention research. Funding Sources: Funding Supported by NINR/NIH grant R01 NR05261

Poster Abstract
HEALTH CARE ACCESS: POST-PROSTECTOMY SURVEILLANCE. Elizabeth Ercolano, Yale University School of Nursing, New Haven CT US

Research has noted that disparaties in access for cancer care exist with few studies about determinants to access during cancer post-treatment follow-up.

A descriptive study was designed. The major aims were to describe patient, provider, institution, and health plan characteristics suspected to influence access to cancer follow-up services. Additional aims were to explore the relationships among these characteristics to the utilization of follow-up services and to examine the fit of this follow-up practice with National Comprehensive Cancer Network (NCCN) and American Cancer Society (ACS) guidelines.

A review of the empirical literature about cancer care access was done, noting gaps in the research about access to follow-up services. A framework of health care access was adapted to cancer care and used to guide the study design.

A descriptive study was undertaken and a survey was created specific to post-treatment follow-up, since none existed. Content validity was established for the survey. The subjects were men, treated by radical prostatectomy, in survival for 3-5 years. Data were collected by phone and written interviews.

The patient, provider, institution, and health plan characteristics were described. The follow-up services used by the men were also described. Chi-square and Anovas were performed to explore the relationships of these characteristics to each other and to the utilization of follow-up services. Each patient's care was evaluated for "fit" with the national guidelines. Complete data was obtained for 57 men, the majority who were Caucasian. All men had access to regular follow-up care done principally by a urologist associated with a cancer center. The follow-up consisted of ongoing physician visits and use of surveillance services such as PSA testing and digital rectal examinations. All men had health insurance except one who was self-pay. Health plans used referrals, co-pays and deductibles which did not deter access. Access to providers resulted in the monitoring of late-term effects such as incontinence and erectile dysfunction experienced by more than half of the men. The majority of the men's care was consistent with national guidelines. Future research should include longitudinal and prospective studies about access to follow-up services for other cancers, both genders, different health systems, ethnically diverse populations, and use of national guidelines. Funding Sources: Columbia University School of Nursing Alumnae Association-partial funding

Poster Abstract
SLEEP PATTERNS OF WOMEN RECEIVING ADJUVANT CHEMOTHERAPY FOR BREAST CANCER. Jeanne Erickson RN, MSN, AOCN, University of Virginia, Charlottesville VA US; Debra Lyon PhD, RN, University of Virginia, Charlottesville VA US; Diana Taibi PhDc, RN, University of Virginia, Charlottesville VA US; Susan Kennel PhDc, RN, University of Virginia, Charlottesville VA US

Women receiving adjuvant chemotherapy for breast cancer experience a number of troubling side effects during the treatment period. Fatigue, sleep disturbances, anxiety, and depression are common symptoms in this population and contribute to a lower quality of life.

While progress has been made to prevent and manage these problems, patients often report incomplete relief of these distressing symptoms. This study is a preliminary exploration of sleep data from a randomized controlled trial to test the effects of microcurrent cranial stimulation on distressing symptoms in women receiving adjuvant chemotherapy for breast cancer.

Sleep is an active process that serves several physiological functions, including energy restoration, rejuvenation, and memory consolidation. Problems related to sleep and sleep disturbances are troubling symptoms for patients with diverse illnesses, including breast cancer. Women receiving chemotherapy for breast cancer report difficulty falling asleep, increased night-time awakenings, and difficulty falling back asleep, resulting in lower sleep efficiency. Sleep disturbances are associated with higher levels of fatigue, which compromise quality of life in this population. The larger study is a longitudinal, repeated measure design, with a randomized sample of 50 women receiving adjuvant chemotherapy for breast cancer. Sleep behaviors are measured weekly using the General Sleep Disturbance Scale (GSDS) and at three intervals using actigraphy over three cycles of chemotherapy. Actigraphy is a valid and objective measure of sleep, and the GSDS has evidence of reliability and validity, particularly in women.

Longitudinal scores on the GSDS and actigraphic measures of sleep will be graphically displayed for an initial sample of at least 4 women to show their patterns of sleep and sleep symptoms over 3 cycles of adjuvant chemotherapy. Future analyses will include correlations to explore differences in measures over time, and effect size will be examined via pair-wise comparisons of selected variables.

These preliminary findings will add to knowledge about sleep disturbances experienced by women over a course of adjuvant chemotherapy for breast cancer. Clinicians can incorporate this information into their practice as they implement interventions to improve sleep in this population. Funding Sources: American Cancer Society and National Cancer Institute

Poster Abstract
OSTEOPOROSIS RISK, SYMPTOM DISTRESS, AND INTEREST IN STRENGTH TRAINING IN MEN RECEIVING ANDROGEN ABLATION THERAPY. Melissa K. Fulton RN, MSN, Urology, PC, Lincoln NE US; Carol Ott RN, OCN, PhD, University of Nebraska Medical Center College of Nursing, Kearney NE US; Melissa K. Fulton RN, MSN, Urology, PC, Lincoln NE US

While androgen ablation therapy (AAT) has decreased death rates from prostate cancer, it also increases the risk of osteoporosis. Purposes are to: a) describe the prevalence of risk factors for osteoporosis in men receiving AAT; b) compare functional status and symptom distress between those interested and not interested in a future strength training (ST) study; and c) examine relationships among self-efficacy for ST, functional status, and symptom distress.

Bandura's Self-Efficacy Theory (1997) was applied to the behavior change of future ST.

This descriptive and correlational study includes 40 men (mean age=75.8 yrs) with locally advanced prostate cancer, capable of self-care, and receiving AAT. EORTC QLQ-C30 (with established reliability and validity) and the EORTC QLQ PR25 (in Phase 4 development) were used to measure functional status and symptom distress.

Several risk factors for osteoporosis were present in >40% of the men while only 18% recalled bone density testing. 68% were interested in a future ST study. Analyses for group differences of symptom distress between those interested and not interested (Mann Whitney U) indicated no significant differences for selected functional status and symptom distress variables. There were significant positive relationships of self-efficacy for learning and overcoming barriers to ST with most functional status variables (r = .41 to 62; p < .05), and significant negative relationships with pain (r = -.48 to - .52; Funding Sources: UNMC College of Nursing Mini Grant

Poster Abstract
A PILOT STUDY TO DETERMINE THE NUTRITIONAL NEEDS OF RADIATION PATIENTS. Tracy Gosselin RN, MSN, AOCN, Duke University Health System, Durham NC US; Rhonda Tinnen , Duke University Health System, Durham NC US

Nutrition services are a critical component of any cancer program. Patients today are receiving more services and treatments in the outpatient setting and use of combination therapy has increased. Patients are often faced with appetite changes and alterations in gastrointestinal (GI) function that lead to weight loss. Nurses are in a pivotal position to develop educational materials, educate patients and advocate for patient needs. The purpose of project was determine how many patients receiving radiation therapy or combination therapy needed to be evaluated by a dietitian. The radiation clinic currently receives dietary support from the inpatient dietitian one morning a week. With operational hours from 7-5pm patients have limited access to the dietitian and may only see the RN for nutritional guidance.

The Scored Patient-Generated Subjective Global Assessment (PG-SGA) was selected by the RN’s as the patient assessment tool. This tool was selected because it has multiple levels of nutritional triage recommendations and established reliability and validity. Another tool was developed to note the treatment field, current dose and total dose of radiation, if receiving combined treatment, what agents were being used, their frequency and if they had seen a dietitian. The nurses randomly sampled 88 patients using both the PG-SGA and the tool.

The assessment tools were turned in and entered into an Access database. Of the 88 patients surveyed it was determined that 60% of those patients receiving radiation alone needed to see a dietitian based upon the triage guidelines. 83% of those patients receiving combined therapy needed to see a dietitian based upon the triage guidelines. Of the disease sites reviewed breast, gastrointestinal, head & neck and lung cancer patients were the ones who predominately needed to see a dietitian.

Based on the above data, presentations have been made at both the department and service line level. The data has also been shared with the Director of Nutrition Services who is benchmarking with other programs. A work group is being convened to establish screening criteria as well as to ensure that educational materials and sample diets based upon specific side effects are up to date.

Poster Abstract
PATIENTS ASSISTING LUNG SURVIVORS (PALS PROGRAM): AN INTERACTIVE SUPPORT APPROACH TO PATIENTS WITH LUNG CANCER. Marilyn Haas PhD, RN, CNS, ANP-C, Mountain Radiation Oncology, Asheville NC US; Diana Wortham MSN, OCN, Mission Hospital, Asheville NC US

Research links positive outcomes for "support interventions" including enhanced knowledge and improved physical outcomes (Holland, 2004). Nurses can promote utilizing non-traditional telephone support groups.

Lung cancer (LC), the "unspoken" or ignored disease, is the leading cause of all gender related mortality. National attention, research studies, and funding are lagging behind (CARMA, 2000). Underpinned by a lack of social empathy, LC does not elicit the same emotional feelings, social reform, or legislative platform from healthcare professionals (Haas, 2002). Changes in healthcare systems have necessitated LC patients to enhance support outside hospital settings. Patient-centered-care redefines the relationship between providers and consumers by shaping healthcare delivery approaches to empower and foster independence, and build on individual/family strengths (Gerteis, et al. 1993). The purpose is to determine the perceived support and overall satisfaction with using a formal/informal telephone support program for LC patients. Utilizing Erickson’s Role Modeling Theory, the oncology nurse’s role and interventional aims are: 1) promoting trust 2) client control 4) strength 5) setting health directed mutual goals (Erickson, et al., 1983).

A prospective, descriptive study of 20 LC patients, self-identified from inpatient/outpatient oncology settings, was conducted from 9/03- 3/04. Two methods of networking were utilized: ALCASE Phone Buddy program (peer-to-peer LC support network) and 120 minute pre-paid phone card enabling familial and social support was given to patients during teaching by oncology nurses. Consents were obtained for follow-up utilizing the FACT-G, Multidimensional Scale of Perceived Support and a questionnaire exploring patterns/outcomes of intervention utilization. LC patients utilized phone cards anytime to talk with family, friends, LC survivors.

Of the 20 LC patients, 17 patients participated in the follow-up questionnaire. Patients were Stage III (82%) and IV (12%) and evenly represented both genders. While 100% utilized the 120-minute phone card and 18% ALCASE Phone Buddy card, 100% felt telephone support was helpful and provided emotional support, especially off hours. The FACT-G supported the same findings.

Results can be used to better understand patient perceptions and methods for information seeking. Summary ideologies may contribute to conducting future research to evaluate how nursing can develop and implement support programs for LC clients. Funding Sources: educational mentorship for one novice researcher

Poster Abstract
PRIMARY ADJUVANT TREATMENT CONSULTATIONS IN BREAST ONCOLOGY: EXPLICATING CONTENT AND IDENTIFYING PREDICTORS OF PSYCHOSOCIAL ADJUSTMENT. Tom Hack Ph.D, Faculty of Nursing, Winnipeg MB CA Tom Pickles , British Columbia Cancer Agency, Vancouver BC CA; Barry Bultz , Tom Baker Cancer Centre, Calgary AB CA; Dean Ruether , Tom Baker Cancer Centre, Calgary AB CA; Lorna Weir , British Columbia Cancer Agency, Vancouver BC CA; Lesley Degner , University of Manitoba, Winnipeg MB CA; John Mackey , Cross Cancer Institute, Edmonton AB CA

The empirical literature demonstrates that a significant proportion of women with breast cancer are dissatisfied with the kind and/or amount of biomedical and psychosocial information received from health caregivers. Patients are known to be highly anxious during the primary adjuvant treatment consultation and to have poor recollection of information received during this consultation.

A systematic review of a representative subset of consultation audiotapes was conducted for the purpose of identifying patient and oncologist characteristics - and other communication factors associated with the primary adjuvant treatment consultation - that are predictive of satisfaction with communication, information recall, mood, and cancer-specific quality of life. An empowerment theoretical model was used to describe the setting of goals and agenda items for the consultation, the examination of patient values during the consultation, and the physician’s role during the consultation.

The consultation audiotapes were coded using the Medical Interaction Process System (MIPS); a system developed for use in oncology that provides an objective, reliable, and valid means of rating consultations. The MIPS classifies patient-physician exchanges in terms of ‘content’ and ‘mode’ of exchange. Coding reliability checks were used during rater training until a mean reliability estimate of .95 was achieved.

Univariate statistics (t tests, Mann-Whitney U tests, and chi square statistics where appropriate) were used to assess the association between sociodemographic and illness variables, patient and oncologist consultation characteristics, and the outcome variables. Significant variables at the univariate level were then entered as independent variables in a stepwise, linear regression analysis.

A total of 78,983 utterances were generated from 180 patient-oncologist consultations. Clinician utterances accounted for 73% of the consultation, while patient utterances made up 27%. Biomedical content was predominant, and accounted for 93% of all utterances, while psychosocial content made up 7% of the consultation. Biomedical and psychosocial content accounted for 89% and 11% of patient utterances, respectively, and 94% and 6%, respectively, of physician utterances. Three ratios (patient directedness, patient centeredness, and psychosocial focus) were generated, and the statistical relationships between these ratios and the outcomes of interest will be presented. Funding Sources: Canadian Breast Cancer Research Alliance

Poster Abstract
EFFECTS OF A MUSIC THERAPY INTERVENTION FOR WOMEN WITH METASTATIC BREAST CANCER. SUZANNE HANSER EdD, Berklee College of Music, Boston MA US; SUSAN BAUER-WU DNSc, DANA-FARBER CANCER INSTITUTE, BOSTON MA US; LORRIE KUBICEK BS, CMT, DANA-FARBER CANCER INSTITUTE, BOSTON MA US; MARTHA HEALEY MS, FNP, DANA-FARBER CANCER INSTITUTE, BOSTON MA US; CRAIG BUNNELL MD, DANA-FARBER CANCER INSTITUTE, BOSTON MA US; JUDY MANOLA MS, DANA-FARBER CANCER INSTITUTE, BOSTON MA US

Hundreds of thousands of women are living with metastatic breast cancer (MBC) with its many psychological and physical symptoms. Studies are needed to identify effective interventions to improve symptoms and enhance coping and quality of life (QOL) in MBC.

Various psycho-behavioral interventions have been shown to benefit MBC patients, and considerable descriptive data support the use of music therapy (MT); but no prospective study has evaluated MT in this population. This study examined the effects of MT (immediate and over-time) on in MBC patients’ psychological functioning and QOL.

MT, a cognitive-behavioral intervention whereby patients use music strategies to cope with cancer-related stressors, is based on a transactional stress-coping framework.

Using a longitudinal, experimental design, 70 MBC patients were randomized to either a MT or usual care group. The MT consisted of three 45-minute individual sessions (3-5 weeks apart) with a music therapist, including techniques to use music as therapy, at home. Psychological symptoms were measured at baseline, 6 weeks, and 3 months with the Hospital Anxiety and Depression Scale (HADS) and QOL with the Functional Assessment of Cancer Therapy-Breast (FACT-B) plus Spirituality subscale. Visual analog scales and heart rate were assessed in the MT group immediately before and after individual sessions.

Paired t-tests and Wilcoxon rank-sum tests compared change scores. Content analysis was done on written patient feedback elicited at the end of the third MT session.

Significant improvement in symptoms and decreased HR were observed after each of the three MT sessions: relaxation, P=<.0001; comfort, P=<.0001; happiness, P=<.01; HR, P=.02, although no significant differences between groups were found over time. High attrition and small sample size may explain the lack of significant group differences. MT completers (n=23) reported great satisfaction and frequent use of music to cope with cancer. While MT produced the immediate benefits of enhanced relaxation and comfort for MBC patients, additional studies with larger sample sizes are needed to assess over-time effects. The high attrition rate underscores the complexities inherent in conducting intervention research with MBC patients.

Poster Abstract
QUALITY OF LIFE BREAST CANCER SURVIVORS WITH LYMPHEDEMA. Sue Heiney PhD, RN, CS, FAAN, Palmetto Health SC Cancer Center, Columbia SC US Camelia Vitoc Vitoc MD, Arnold School of Public Health, University of South Carolina, Columbia SC US Joan Cunningham PhD, Arnold School of Public Health, University of South Carolina, Columbia SC US Kel Jansen R.O.T., Palmetto Health, Columbia SC US

Advances in treatments have significantly increased survival for breast cancer patients (BrCA). Unfortunately, 28-38% of these survivors develop secondary lymphedema (SLE). Physical effects include pain, reduced motion, and risk for infection. Knowledge is particularly needed to better understand the impact of QOL in rural white Southern and African American (AA) BrCA patients in an environment of major health disparities.

BrCa treatment and the development of SLE are known to affect quality of life (QOL). Differences in QOL between patients with and without SLE have not been well documented. The primary purpose of the After Breast Cancer (ABC ) study is to understand women’s knowledge of and experiences with SLE. The secondary aim and focus of this abstract is to compare QOL of BrCA patients with and without SLE.

The QOL component of the study is based on Ferrell's 4 dimensional model of QOL (physical, psychological, social and spiritual). The spirituality dimension is particularly relevant to a sample drawn from the Bible belt, where spirituality is a major cultural component.

A correlational design is used. Packets with consents, the ABC survey, QOL and demographic data instruments and a postage paid mailer were sent to 957 BrCA patients. The QOL component of the study used the valid and reliable Quality of Life – Breast Cancer Version instrument which measures 4 dimensions of QOL, physical, psychological, social and spiritual well-being.

Descriptive and correlational statistics (ANOVA, linear correlation and regression) will be used.

Currently, 314 surveys have been returned; 49 women refused; and 20 women were deceased; 96 had SLE. Ethnicity is 82% European American, 17% AA and 1% other. Preliminary data analysis for the first 216 participants found significant differences in social well-being (p<.001) and overall QOL (p=.01). Preliminary results document significant QOL differences between women with and without SLE. Our sample includes more AA than previous studies. The 36 % return rate is especially good for a mailed survey to a sample known to have low literacy. These results provide evidence for the differences between QOL in women with and without SLE, especially the impacts to physical and social well-being. Also, the study points to critical areas that should be addressed in helping women cope with SLE. Funding Sources: Bonner Family Fund, Palmetto Health Foundation

Poster Abstract
QUALITY OF LIFE OF PATIENTS NEAR THE END OF LIFE. Carla Hermann PhD, RN, University of Louisville, Louisville KY US

The primary objective of care for dying patients is to improve their quality of life (QOL). There are major gaps, however, in knowledge regarding all issues related to end of life (EOL) care, including how to enhance dying patients' QOL. The few available studies have focused on hospice patients, but less than twenty percent of individuals in the United States experience EOL receiving hospice care. Research with a focus on identifying determinants of dying patients' QOL is needed.

The purpose of this study was to increase understanding of QOL at the EOL. Specific aims were to: (1) describe the QOL of patients near the EOL; (2) examine symptom frequency, symptom severity, symptom distress, functional status, anxiety, and depression in patients near the EOL; and (3) identify determinants of QOL of patients near the EOL.

Many patients near the EOL, such as advanced lung cancer patients, are actively seeking treatment. A high number of uncontrolled symptoms and a great deal of physical and psychological distress have been reported for patients near the EOL. Little is known about their QOL. QOL, viewed multi-dimensionally, encompasses four different but overlapping dimensions: physical, psychological, social and spiritual.

Eighty adults diagnosed with stage IIIB or IV lung cancer within the previous month participated in this longitudinal, cross-sectional study. Fifty-five subjects participated at the 2-month follow-up and 41 at the 4-month follow-up. Subject attrition was due to patient death for all but one subject. Subjects were interviewed for responses to instruments measuring QOL, symptoms, anxiety, and depression.

Longitudinal data were summarized using "area under the curve". Randomized regression models were used to make use of all available data. Although most subjects received their initial cancer diagnosis within thirty days of entering the study, only half were alive for the 4-month follow-up. Symptom distress was the most important correlate of QOL. Significant downward trends were present with functional status, anxiety and symptom experience. Advanced lung cancer patients are often near the EOL upon diagnosis. Interventions that reduce symptom distress may increase QOL of patients near the EOL. Further study should include testing interventions aimed at decreasing symptom distress. Funding Sources: National Institute of Nursing Research

Poster Abstract
PREDICTORS OF PAIN, FATIGUE, AND NAUSEA IN NEWLY DIAGNOSED PATIENTS WITH LUNG CANCER PRIOR TO CHEMOTHERAPY. Amy Hoffman M.S.N., R.N., Ph.D. Student, Michigan State University, East Lansing MI US; Barbara Given Ph.D., R.N., FAAN, Michigan State Univeristy, East Lansing Michigan MI US; Alexander von Eye Ph.D., Michigan State Univesity, East Lansing MI US; Audrey G. Gift Ph.D., R.N., FAAN, Michigan State University, East Lansing MI US; Charles Given Ph.D., Michigan State University, East Lansing MI US

Lung cancer is the second most common malignancy and the foremost cause of cancer mortality in both men and women in the United States. Most patients with lung cancer present at diagnosis with advanced disease and suffer from multiple symptoms that continue throughout their treatment and disease progression. Each additional symptom places greater burden on the patient with lung cancer but there is little to identify those patients most at risk for symptom burden.

This study examined the patient and disease characteristics predicting symptom burden from pain, fatigue, and nausea in newly diagnosed patients with lung cancer prior to receiving chemotherapy.

The Theory of Unpleasant Symptoms (TOUS) was used to guide the study. The TOUS conceptualizes symptoms as occurring together and identifies categories of antecedents to symptoms. Symptom burden for this study includes symptom severity, limitation, and bother.

Secondary data analysis from the baseline observation of a randomized control trial was performed on 82 patients with lung cancer using only their pre-chemotherapy information. The burden associated with each symptom (i.e., pain, fatigue, and nausea) was calculated by summing each subject's responses to severity, limitaion, and bother scores for each symptom and dividing by three to standardize total symptom burden to a 10-point scale.

Fatigue (94%), pain (67%), and nausea (51%) were the most prevalent symptoms, but the rank order for burden was fatigue, nausea, and pain. Mean symptom burden score increased as the number of symptoms climbed from 0 to 3. Best subset regression identified a model with seven statistically significant predictors (i.e., age, anxiety, depressive symptomatology, co-morbid conditions, being on disability, not being able to work, and having health insurance) of symptom burden accounting for 43.8% of the variance. Each of the predictors independently contributed to the symptom burden.

Fatigue, pain and nausea are symptoms experienced by most patients with lung cancer. Patients most at risk for these symptoms can be identified early and interventions developed to more adequately alleviate the symptom burden. Funding Sources: Grant R01 NR/CA01915--The Family Home Care for Cancer--A Community-based Model

Poster Abstract
FEASIBILITY, RELIABILITY, AND VALIDITY TESTING OF A COMPUTER-GENERATED QUALITY OF LIFE ASSESSMENT AND SCORING PROGRAM FOR PATIENTS WITH LUNG CANCER (LCSS-QL). PJ Hollen PhD, RN, University of Virginia, Charlottesville VA US;RJ Gralla MD, Lung Cancer Alliance, New York NY US; N Leighl MD, Princess Margaret Hospital, Toronto CA; B Landen RN, COMET Group, Toronto CA; H Kimber RN, COMET Group, Toronto AL CA; V Buttu RN, COMET Group, Toronto AL CA

This study tests an electronic version (LCSS-QL) of the Lung Cancer Symptom Scale, converting the validated paper measure to a format using a hand-held computer. The device provides immediate graphics of scores and change over time. Objectives include to: 1) determine correlation of the LCSS-QL with the paper version, 2) measure completion times, 3) assess acceptability of the LCSS-QL by patients, nurses, and physicians. Testing was based on the LCSS model (Hollen, 1994).

This methodological study evaluates feasibility, reliability, and validity of the LCSS-QL. Patients are entered in the community settings of 10 COMET clinics in Ontario. All patients have NSCLC, KPS >60, no prior chemotherapy, and receive initial courses of docetaxel+platinum. 200 patients are planned; the initial 80 complete both the paper and electronic forms (pretreatment, and with the next 4 chemotherapy cycles - paper version every other cycle). The final 120 complete only the LCSS-QL. Multiple agreement methods are used; Pearson r does not correct for systematic bias. Analyses include: a) general agreement between electronic and paper forms (intraclass correlation, Lin’s concordance correlation; Bland-Altman plots characterizing bias); c) Cronbach alpha, and d) completion times.

31 patients to date include: 52% women; KPS (median 80%; range 60% - 100%); age (median 67; range 46-78); Stage IV: 77%. Acceptance of this simple electronic QL instrument by patients, nurses, and physicians has been excellent. Mean completion time is 2.9 minutes (SD 2.03). To date, the agreement is promising. Example coefficients for global QL item are Pearson r, 0.77; ICC, 0.86; Lin’s CCC, 0.86; alpha (paper, 0.84; electronic, 0.85). By the February 2005 meeting, all 80 initial patients will be entered. If convergence between the two formats is high, then existing psychometric properties for the original measure can be generalized to the new presentation. If feasibility, reliability, and validity testing confirm these results, this electronic format provides a practical method to bring QL evaluation to patient management and to clinical trials.

A 2002 international quality of life (QL) conference concluded that improved patient completion rates with less measurement error are key to enhancing QL use. Funding Sources: Supported by a grant from Aventis Canada.

Poster Abstract
DETERMINANTS OF EXERCISE FOR BREAST CANCER SURVIVORS IN TAIWAN. Hsin-Tien Hsu RN, PhD(c), University of California at San Francisco, San Francisco CA US; Marylin J. Dodd RN, PhD, FAAN, University of California at San Francisco, San Francisco CA US; Kathryn A Lee RN, PhD, FAAN, University of California at San Francisco, San Francisco CA US; Geraldine V. Padilla RN, PhD, FAAN, University of California at San Francisco, San Francisco CA US; Noreen C. Facione RN, PhD, FAAN, Loyola University Chicago, Maywood IL US; Shiow-Li Hwang RN, PhD, National Taiwan University, TW

Breast cancer is the second leading cause of cancer in Taiwanese women. Cancer-Related Fatigue (CRF) has been recognized as a universal side effect of cancer treatment. Exercise is one of the few interventions suggested to prevent or decrease CRF.

No information is available about the determinants of exercise for breast cancer survivors in Taiwan. It is important to identify major determinants of exercise and develop a theoretical model to direct intervention research. Therefore, the purpose of this study is to examine the relationship between fatigue, motivating factors of exercise, and exercise behavior among breast cancer survivors.

Social Cognitive Theory

This report is the baseline data of a prospective, longitudinal, repeated measures on going study.

Descriptive analyses, correlation, t-test, chi-square and ANOVA Findings: A total of 121 women completed questionnaires at one month after treatment finished. Characteristics of subjects were: mean ages were 48.43 ¡Ó 9.935 years, 71.7% were married, 43.8% were homemakers, 73.6 % were from Fujian Province, 41.2% were Buddhist, 34.5% were college educated. The majority of the women had received a stage I (27.1%) or stage II (51.7%) breast cancer diagnosis. All subjects had surgery and received chemotherapy (40.5%), radiotherapy (8.3%), or both (47.1%). Average days for chemotherapy were 132 days and 40 days for radiotherapy.

For the current activity level, 58.7% of subjects reported being able to carry on normal activities but having minor signs or symptoms related to treatment. Overall, 28.3% intended to exercise, 20.8% exercised regularly less than 6 months, and 15% exercised regularly more than 6 months. Results shows that age (p< .001), exercise history (p< .001), social support for exercise (p< .002), exercise self-efficacy (p< .000), exercise outcome expectancy (p< .011) and perceived barriers to exercise (p< .000) significantly influence exercise status among breast cancer survivors.

Implication for Nursing Practice: Preliminary data from our research indicates that there is abundant information related to culturally different exercise behavior on the breast cancer survivors. Findings from this study will contribute significantly to the literature on psychosocial and exercise aspects of breast cancer survivors in Taiwan. Funding Sources: Funded by Department of Defense, Breast Cancer Research Program, DAMD17-03-1-0521

Poster Abstract
POVERTY AND THE COMPLETION OF ADVANCED DIRECTIVES: A META-ANALYSIS. Anne Hughes RN, MN, AOCN, Laguna Honda Hospital and Rehabilitation Center, San Francisco CA US

Poverty increases the burden persons endure when living with cancer. Poor people face substantial barriers to quality cancer care, experience more pain and suffering, and are more fatalistic about the disease (Freeman, 2004). Federal and state laws mandate that health care facilities inquire about a patient’s advance directives in the event of critical illness. Advance directives are believed the cornerstone of good end of life care. The purpose of this meta-analysis is to describe the extent to which poverty influences the completion of an advanced directives.

Meta-analysis is a standardized process of systematically identifying research, summarizing findings and creating a common metric to compare results across studies to answer a specific research question.

Three approaches were used to identify studies for inclusion. Electronic bibliographic databases i.e., PubMed and Cochrane Collaboration were searched, reference lists from identified studies were reviewed and leading researchers were contacted for potential unpublished research. Inclusion criteria were developed: English language, published or unpublished, quantifiable measure of poverty, comparison or control group, and some measure of advance directive.

Each study was evaluated for quality by the single reviewer who extracted data in an unblinded fashion. Effect sizes were calculated for each study included.

Six studies, from over 144 studies identified, met the inclusion criteria. Four were randomized control trials and two cross-sectional design. The quality of the studies was variable. Effect sizes were quite small and ranged from -.28 to +.17. Three effect sizes were non-significant. Of the remaining three, two studies were negatively significant, i.e. lower income was not associated with completion of advanced directive and one positively significant, i.e., lower income was associated with advanced directive completion. Comparing studies was complicated by different annual income cutoffs for poverty, and different years when data was collected so buying power of same annual income may differ. In conclusion, this meta-analysis was unable to determine the influence of poverty in completion of advanced directives. In the absence of an evidence base, nurses need to explore with each patient, their preferences for articulating in advance their wishes for treatment and a surrogate decision-maker. Funding Sources: American Cancer Society Doctoral Scholarship in Nursing, DSCN-01-202-01-SCN, National Institute of Nursing Research, NIH, F31NR079923.

Poster Abstract
DEPRESSION, SATISFACTION AND QUALITY OF LIFE (QOL) IN CANCER PATIENTS UNDERGOING PAIN MANAGEMENT. Shirley Hwang RN, MS, VA New Jersey Health Care System, East Orange NJ US; Victor Chang MD, VA New Jersey Health Care System, East Orange NJ US; Qi Xia MS, VA New Jersey Health Care System, East Orange NJ US; Donald Hoover PhD, Rutgers University, Piscataway NJ US; Shanti Srinivas MD, VA New Jersey Health Care System, East Orange NJ US; Basil Kasimis MD, VA New Jersey Health Care System, East Orange NJ US

We find that depression strongly predicts QOL and satisfaction in cancer pain patients, which suggests that treating depression may improve these outcomes.

The relationship of satisfaction and QOL to pain management is unclear. We previously found that QOL and satisfaction may be independent outcomes of pain management and that the timing of assessment may be important(Hwang,JPSM;2002:190-200). The purpose is to develop a multidimensional longitudinal QOL and satisfaction cancer pain outcome model.

A multidimensional QOL framework categorized the predictor variables into six dimensions: individual characteristics, pain, symptoms, psychological, function and health. The outcomes were QOL and satisfaction. This IRB approved longitudinal survey studied 195 cancer patients with baseline worst pain > 4/10(median age 68 yrs, range 44-87) who were managed according to AHCPR guidelines. All patients completed Functional Assessment of Cancer Therapy(FACT-G), satisfaction, Brief Pain Inventory, Memorial Symptom Assessment Scale–Short Form, Geriatric Depression Scale and health status - at 4 weekly time points . KPS was assessed weekly.

Changes in pain, QOL and satisfaction over time were estimated by mixed effect models. To identify independent predictors of QOL and satisfaction, multiple linear regression analyses for cross-sectional data and mixed effect model for longitudinal data were performed.

There were significant improvements over the three weeks in mean worst pain(8.4 to 5.1,p<0.0001), pain interference(38.1 to 17.0,p<0.0001), FACT-G QOL(62.7 to 71.6,p<0.0001) and satisfaction(1.42 to 2.99,p<0.0001). Different sets of predictors were identified at each time point for QOL and for satisfaction, but depression predicted QOL and pain relief predicted satisfaction consistently over time. The mixed effect model showed that depression(p<0.0001), pain interference(p=0.0005), KPS(p=0.0005) and health(p=0.04) each predicted QOL. Pain relief(p<0.0001), depression (p=0.0006), worst pain(p=0.005), age(p=0.04) and timing of assessment (p<0.0001) each predicted satisfaction. These results demonstrated the dynamic and multidimensional nature of QOL and satisfaction outcome predictors in cancer pain patients. Depression was important to both QOL and satisfaction. Screening and treatment of depression may improve QOL and satisfaction outcomes. Funding Sources: VA HSR&D PCC 98068

Poster Abstract
A FEMINIST ANALYSIS OF INTERNET CANCER SUPPORT GROUPS. Eun-Ok Im PhD, MPH, RN, CNS, The University of Texas at Austin, Austin TX US; Wonshik Chee PhD, The University of Texas at San Antonio, San Antonio TX US; Hsiu-Min Tsai MSN, The University of Texas at Austin, Austin TX US; Li-Chen Lin MSN, The University of Texas at Austin, Austin TX US

The purpose of this presentation is to analyze Internet Cancer Support Groups (ICSGs) searched through Google.com, Yahoo.com, MSN.com., AOL.com., and ACOR.org using a feminist perspective, and explore issues in using ICSGs as a research setting or recruitment resource.

A total of 317 general ICSGs and 229 ethnic-specific ICSGs were analyzed. Research staff recorded issues as the issues arose and wrote memos regarding the issues from a feminist perspective. The written memos and records were reviewed and analyzed using the content analysis by Weber (1990). The analysis indicated that inconsistent terms were being used to refer to ICSGs, and there was no way to validate that the participants of ICSGs were real cancer patients. Most of the ethnic-specific ICSGs retrieved through the Internet search were actually general ICSGs whose members were dominantly Whites. Female participants tended to share their stories and seek for emotional support. ICSGs for both men and women hardly represented womens voices and experiences. Messages in most bulletin boards and chat groups tended to be dominated by a few dominant participants. Most of the information resources provided by ICSGs were not validated by health professionals. Most of the websites did not ensure confidentiality of the interactions.

The issues discussed in this presentation will provide directions for future Internet research, especially for Internet research using ICSGs as a research setting or a data collection method.

The findings suggest the needs for development of ethnic-specific support groups for ethnic minorities, education programs for moderators and/or facilitators, standards and/or policies regulating health-related resources provided in ICSGs, and quota sampling in Internet recruitment.

When ICSGs are used in research to support the predominant androcentric views and interests, those who are not part of this dominant group (e.g., ethnic minorities, women) are marginalized, and their issues are either not considered relevant for study or not reflected accurately in research and health care practice. Funding Sources: NIH, NINR (1 R01 NR007900-01A1).

Poster Abstract
ISSUES IN INTERNET RECRUITMENT OF RESEARCH PARTICIPANTS. Eun-Ok Im PhD, MPH, RN, CNS, The University of Texas at Austin, Austin TX US

The purpose of this paper is to provide future directions for recruitment of research participants through the Internet based on issues raised in recruitment process in two Internet studies one among cancer patients (Study 1); and the other among international oncology nurses (Study 2).

Throughout the research process, research staff recorded recruitment issues as they arose and wrote memos regarding the issues and their possible reasons. Weekly group discussions were conducted, and written records of these discussions were kept. Then, the written memos and records were analyzed using the content analysis by Weber. The analysis indicated a very low response rate (2). Most participants of Study 1 were a select group of high income, well-educated White males. The participants of Study 2 tended to be young, highly educated, and high income nurses. In Study 1, a low response rate in winter and summer time was noticeable. In Study 2, major holidays were identified by the participants as a major reason for the dramatic decrease of their responses. There was no way to ensure that the participants were real cancer patients in Study 1. Study 2 posed a problem for international participants who needed to spend more time to fill out the questionnaire and participate in email group discussions because English was their second language.

The issues discussed in this presentation will provide directions and guidelines for future Internet research in oncology nursing. This presentation has a limitation in generalizability of the findings because only two Internet studies were analyzed.

Based on the discussion on these issues, the followings are proposed for future Internet research (a) creative motivation strategies, (b) sensitivity to gender, ethnicity, and socioeconomic status of potential participants, (c) appropriate timing of data collection, and (d) sensitivity to potential ethical issues.

With the lack of knowledge on Internet research in general, issues in Internet recruitment have been rarely explored and discussed in oncology nursing. Thus, it is imperative to explore issues in Internet recruitment and provide future directions for Internet recruitment at this beginning stage of Internet research in oncology nursing. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from SmithKline Beecham

Poster Abstract
EFFECTS OF MASSAGE ON PAIN INTENSITY, ANXIETY, AND PHYSIOLOGICAL RELAXATION FOR TAIWANESE PATIENTS WITH METASTATIC BONE PAIN. Sui-Whi Jane, University of Washington, Seattle WA US; Diana J. Wilkie , University of Illinois at Chicago, Chicag IL US; Randal D. Beaton , University of Washington, Seattle WA US; Betty B. Gallucci , University of Washington, Seattle WA US; Hsiu-Ying Huang , University of Washington, Seattle WA US

50-74% of patients with metastatic cancer are more likely to report pain compared to 15-17% patients with non-metastatic pain. Bone metastases are the commonest cause of pain that can't be completely relieved by medications alone. The existing studies demonstrate an immediate or short-term effects of massage on general cancer-related pain but not specifically for bone pain. Whether massage is also effective for this specific population and to what the extent of the appropriateness of the interval for measuring outcome variables is not yet established. Therefore, it becomes important for oncology researchers to prove that massage really serves as an effective strategy to optimum bone pain relief with more rigorous approaches.

To validate the duration of massage effects on pain intensity, anxiety, an physiological relaxation over time.

A modified Gate Control Theory of Pain proposed by Mezlack & Wall (1965) will serve as a theoretical framework for exploring the underlying mechanism of massage therapy.

The investigator proposes a quasi-experimental design with repeated measures to examine the duration of massage effects over time on pain intensity, anxiety, and physiological relaxation in 30 Taiwanese patients with metastatic bone pain during an in-patient hospitalization in a 3600-bed Taiwanese teaching hospital. Time effects of the massage therapy will be evaluated with present pain intensity as measured with a single-item PPI-VAS, anxiety as measured with a single-item Anxiety-VAS, and physiological relaxation as measured with heart rate (HR)and blood pressure (BP).

Each value of post-intervention measures on PPI-VAS, Anxiety-VAS, and HR and BP will be separately compared to the value of the baseline time point by using paired t-tests.

The amount of change of massage effects over time will allow investigators to identify the best time points in time at which the outcomes should be assessed in order to demonstrate the magnitude of massage benefits in future randomized clinical trials. Clinically, these results can establish scientific knowledge of how to appropriately implement this non-pharmacological nursing intervention concurrently with pharmacological treatment optimally aimed at improving metastatic bone pain management. Funding Sources: McLaws Nursing Scholarship and Benolie Fund from the School of Nursing, University of Washington.

Poster Abstract
THE EFFECTS OF PROGRESSIVE MUSCLE RELAXATION AND GUIDED IMAGERY IN REDUCING CANCER PAIN: A SYSTEMATIC REVIEW. Sui-Whi Jane Doctoral Candidate, University of Washington, Seattle WA US; Jyhun-I Chen Lecture, Chang Gung Institute of Technology, Tao-Yuan, Taiwan TW

Despite extensive progress in the scientific understanding of pain, 51-77% of patients with cancer experience moderate to severe intense pain at some time in their illness trajectory. Theoretically, 90% of cancer pain can be adequately relieved with relatively simple medical interventions; however, in practice, less than 50% of cancer patients actually achieve effective pain relief due to the multidimensional nature of pain phenomena and tolerance and unexpected side effects of analgesics. Attention to the psychosocial variables modulating pain through non-pharmacological interventions such as progressive muscle relaxation (PMR) and guided imagery (GI) provide an additional and viable avenue for managing cancer pain. Within scientific realms, it is crucial for researchers and clinicians to validate that these two techniques serve as effective strategies for pain management with rigorous evaluation processes. More importantly, none of the existing published meta-analyses specifically examining the effects of PMR and GI on cancer pain. Thus, the aims of this paper are to 1) systematically review the efficacy of PMR and GI in managing adult cancer pain; 2) critically analyze existing methodological issues derived from these reviewed articles; 3) furthermore make recommendations for future research.

Despite the existing methodological issue, PMR and GI are considered to be easy-to-learn and less time-consuming interventions and may provide patients with additional benefits for the management of cancer pain. In practice, thus, it is vital for healthcare providers to appropriately employ them in conjunction with pharmacological treatment, thereby enhancing pain management.

To provide clinicians with evidence-based nursing interventions, the focus of future studies for PMR and GI will be on defining their optimum strength, underlying mechanisms of action from the psycho-neuro-immunological perspective, appropriate intensity of protocol, and cost-effectiveness. Additionally, potential mediators and moderators that may hinder or augment the therapeutic effects are needed to explore.

This systematic review examined seven published studies with full text from 1985 to 2003 in English upon the inclusion criteria. The effects of PMR and GI in this review have yielded inconsistent across studies, partially due to variations in sample sizes, the characteristics of populations, the intensity and duration of intervention protocols, and the selection of outcome variables and measurements. Additionally, several methodological flaws may contribute to this inconsistence, including the lack of a consistent theoretical framework, rigorous inclusion criteria for ensuring the homogeneity of subjects, and a standardized intervention protocol, sensitivity of selected measures, adequate statistical power, consistent effect sizes, and consideration of potential confounding variables and placebo effects. Five of seven studies concluded that there was no significant difference between the control and intervention groups on decreasing pain intensity and one study even found that the treatment group reported increasing in pain intensity. Whereas only two studies revealed group effects on ability of control pain and decreasing pain distress and anxiety, but the group effect didn't exist among intervention groups and the increased complexity of the intervention seemed not to be additive to intended beneficial outcomes. Overall, evidence from this review fails to prove positive effects of PMR and GI for cancer pain management. The focus of future study will be on taking the above methodological flaws into account to ensure the reliable evidence-based findings of PMR and GI.

Poster Abstract
HEMATOLOGICAL TOXICITIES, SYMPTOM BURDEN, AND QOL IN ADJUVANT AND NEO-ADJUVANT BREAST CANCER PATIENTS RECEIVING DOSE DENSE CHEMOTHERAPY (DDC). Gina Johnson MSN, APRN, BC, Accelerated Community Oncology Research Network, Memphis TN US; Gina Johnson , Accellerated Community Oncology Research Network, Memphis TN US; Lee Schwartzberg MD, The West Clinic, Memphis TN US; Kurt Tauer MD, The West Clinic, Memphis TN US; Qian Mao , Accelerated Community Oncology Research Network, Memphis TN US; Barry Fortner PhD, Accelerated Community Oncology Research Network, Memphis TN US

CIN and CIA are significant chemotherapy-induced toxicities that require careful clinical attention. Understanding the symptom and QoL impact of DDC is important to anticipating appropriate patient education and supportive interventions, including erythropoietic agents and growth colony stimulating factors, to enhance clinical outcomes in evidence based medicine. This study evaluated changes in ANC, Hgb, QoL, and symptom burden during DDC. DDC is a promising approach for adjuvant and neo-adjuvant breast cancer patients (pts). However, shorting the chemotherapy cycle increases the potential for chemotherapy-induced toxicities.

Consecutive pts with Stage II/III breast cancer receiving Q 14 day AC followed by a taxane were retrospectively identified at a large community oncology practice. Twenty (20) weeks of chart data were reviewed for disease, chemotherapy, anemia, neutropenia, symptom burden, and QoL parameters. Symptom burden and QoL was measured by the Cancer Care Monitor. Anemia and neutropenia specific symptoms and index scores for physical, psychological, and functional status are being examined within subjects across DDC and between subjects in relation to anemia and neutropenia grade. Symptom burden and QoL is also being explored in relation to growth factor support and intervention. Dose reductions, delays, and discontinuation are also being explored in relation to toxicities.

73 pts (median age 50 [range 30 -69], 34% Stage III) were treated with adjuvant (73%) or neoadjuvant (27%) DDC. 53 pts received 8 cycles, 14 pts received 5 -7 cycles, and 6 pts received 4 cycles total. Mean Hgb fell by 1.5 g/dl after 2 cycles and 2.1 g/dl after 4 cycles of AC. The mean Hgb at end of study was 11.5 g/dl. Epo was administered to 50/73 pts (68%) with 28 beginning therapy by week 7. 2 pts (3%) received PRBC transfusions at week 5 and 11 respectively. Worst CTC anemia grade (number of pts) recorded during all cycles was: Grade 1 (41), Grade 2 (26), and Grade 3 (3). Worst CTC neutropenia grade (number of pts) during all cycles was: Grade 0 (48), Grade 1(8), Grade 2(7), Grade 3 (2) and Grade 4 (8). No pt developed febrile neutropenia or was hospitalized for neutropenia complications. Funding Sources: This study is based in part on a study funded by Amgen Inc.

Poster Abstract
WHEN DO EXPERIENCES WITH AFFECTED FAMILY MEMBERS AND FRIENDS, AND PERSONAL EXPERIENCES WITH ABNORMAL BREAST SYMPTOMS INFLUENCE PERCEIVED BREAST CANCER RISK? Maria Katapodi RN, MSN, PhD, University of California San Francisco, School of Nursing, San Francisco CA US; Marylin Dodd RN, PhD, FAAN, University of California San Francisco, San Francisco CA US; Noreen Facione RN, PhD, FAAN, Loyola University, Chicago IL US; Lee Kathy RN, PhD, FAAN, University of California San Francisco, San Francisco CA US; Bruce Cooper PhD, University of California San Francisco, San Francisco CA US; Janice Humphreys RN, PhD, University of California San Francisco, San Francisco CA US

Although experience with affected family members increases perceived breast cancer risk, it is less clear whether women with a positive family history perceive higher breast cancer risk because they understand that they share genetic material with their family members or because they worry.

The study aimed to examine 1) whether having experiences with affected family members and friends, and experiences with abnormal breast symptoms increased perceived breast cancer risk and 2) whether knowledge of breast cancer risk factors and worry moderated the relationships between these experiences and perceived risk.

The Adoption Precaution Process (Weinstein, 1988) suggests that personalized information about risk factors and personal experience leads to a more accurate perception of susceptibility to disease.

Thsi cross-sectional survey recruited 184 women from community settings (43% White, 26% Black, 17% Asian, 14% Hispanic) to complete a questionnaire in English. Participants have never been diagnosed with cancer and were between 30 and 85 years old (X=46„b12). Most (49%) were college graduates and had a median annual income $30,000 to $40,000. A Principal Component Analysis of three probability measures created a new measure of Perceived Risk (Cronbach £\ =.75). We used the Breast Cancer Risk Factor Knowledge Index (Cronbach £\ =.80), a four-item Worry scale (Cronbach £\ =.85), and four indicators of experiences with abnormal breast symptoms.

Descriptive statistics and hierarchical regression analysis. Moderator effects were tested with established methods (Baron & Kenny, 1986).

Having a family member and a friend with breast cancer accounted for 6% and 2% respectively of the variance in perceived risk (p<.05). Experiences with affected family members and friends, and abnormal symptoms influenced perceived risk through knowledge of risk factors, worry, and heuristic thinking. Knowledge of risk factors moderated the relationship between family history and perceived risk, and worry moderated the relationship between abnormal symptoms and perceived risk. Educational interventions should be aimed at increasing women’s knowledge about risk factors, addressing their worries, and take into account cognitive mechanisms of information processing. Funding Sources: Department of Defense, Medical Research, Clinical Nurse Researcher Award No DAMD17-03-1-0356

Poster Abstract
PARENT DECISION MAKING: WEB BASED DATA COLLECTION PILOT. Katherine Kelly RN, MN, University of Missouri Health Care/Children's Hospital, Columbia MO US; Kimberly Pyke-Grimm RN, MN, Children’s Hospital of San Diego, San Diego CA US; Janet Stewart RN, PhD, University of Pittsburgh School of Nursing, Pittsburgh PA US

Supporting parental treatment decision-making (TDM) is a central pediatric oncology nursing function. Pediatric oncology multi-site research mandates validating novel methods for centralized remote data entry such as web-based surveys.

Parental TDM is an important area of nursing research. Increased knowledge about parents’ TDM could inform interventions designed to improve TDM, reduce emotional distress, and improve families’ adjustment to childhood cancer. On line data collection could facilitate the multi-site research necessary to obtain adequate sample sizes. The purpose of this study is to 1) determine the feasibility of internet-based, multi-site data collection using electronic instruments accessed via an interactive web site, and 2) determine the acceptability of this method to parents.

The conceptual model (Stewart, Pyke-Grimm & Kelly, in press) that informed selection of instruments for this study incorporates elements from Degner and Beaton’s (1987) control preferences model, O’Connor’s (1997) decisional conflict model, Thorne and Robinson’s (1989) model of health care relationships and Chu & Power’s (1995) interactive model of synchrony. In this descriptive pilot study we will ask 20 parents from 3 sites to complete the web survey, two instruments in the traditional pencil and paper format, and an interview about their experience within two months of making a decision regarding their child’s participation in a clinical trial. Survey instruments have been previously used and validated in treatment decision-making research.

Descriptive statistics and internal consistency estimates will be calculated for each instrument. Paired sample t-tests will be used to test the effect of survey method on completion time and score differences. If more than 50% (95% CI) of the parents prefer web-based method or have no preference, the web-based method will be considered acceptable. Parent interviews and investigator field notes will be subjected to content analysis. To date the web survey has been constructed and initial beta testing completed. Data collection will begin this summer and completed in October 2004. Final results will be available for this presentation. Results from this study will provide proof of concept for the data collection strategy and will also be used to estimate sample sizes for successor multi-site studies. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Aventis Pharmaceuticals.

Poster Abstract
THE EFFECT OF A PILOT EXERCISE INTERVENTION ON BONE REMODELING IN BREAST CANCER SURVIVORS. M. Tish Knobf RN, PhD, AOCN, Yale School of Nursing, New Haven CT US; Karl Insogna MD, Yale School of Medicine, New Haven CT US; Kristopher Fennie MPH, PhD, Yale School of Nursing, New Haven CT US; Loretta DiPietro PhD, Yale School of Medicine, New Haven CT US; A. Siobhan Thompson MPH, Yale School of Nursing, New Haven CT US

Significance: Estrogen deficiency from chemotherapy-induced menopause or withdrawal of supplemental estrogen puts breast cancer survivors at risk for accelerated bone loss.

Problem and Purpose: Breast cancer survivors represent 22% of the 9 million cancer survivors and one of the key areas identified by the Office of Cancer Survivors is to test interventions to minimize adverse outcomes and promote positive health practices. The purpose of this pilot study was to evaluate the effect of a weight loaded exercise intervention on bone turnover, body composition, physical functioning and symptom distress. This interim analysis will report on the primary outcome, the rate of bone turnover. Scientific Framework: Winningham's Psychobiological-Entropy Model of Functioning (1999) and the Error Strain Distribution Hypothesis (Lanyon 1996). Methods: A one-group pre and posttest design was used to test a 3 times per week, 16 week supervised exercise intervention. A weight loaded waist belt progressed from 1 lb to 5 lbs over the first five weeks which was then maintained. Women with breast cancer who experienced chemotherapy induced menopause, or were within two years of menopause or discontinuance of hormone replacement therapy were eligible. Bone turnover was assessed by measuring serum concentrations of the N-terminal propeptide of type I collagen (NTX) and osteocalcin.

Wilcoxin signed rank test and repeated measures were used for analysis.

Findings and Implications: This interim analysis reports on the first 20 subjects who completed baseline and 16 week measures. The majority of subjects were married, well educated and had a mean age of 51.5 yrs (SD=6.2). Sixty-five percent of subjects received chemotherapy, 45% were taking Tamoxifen and 35% an aromatase inhibitor. Subjects taking Tamoxifen had significantly lower levels of serum NTX (p=.02). In the group as a whole, there was a trend towards decreasing NTX values with time although not statistically significant. The intervention was feasible with high adherence rates (87.5%-90.5%). The ability to detect a change at 48% power with 20 subjects indicates a potential effect of the intervention to reduce rates of bone turnover and bone loss, which will require confirmation in analysis of the data from the completed study. Funding Sources: American Cancer Society Professor Oncology Nursing; NINR P20 Center for Self Management, Yale School of Nursing, Intramural Funds, Yale Cancer Center

Poster Abstract
PRELIMINARY PSYCHOMETRIC TESTING OF A RELAXATION ABILITY SCALE. Kristine Kwekkeboom PhD, RN, The University of Iowa College of Nursing, Iowa City IA US

Unrelieved pain and distress remain a problem for persons with cancer despite analgesic therapy.

Relaxation strategies may be useful adjunctive interventions in managing cancer pain and distress, however, some individuals experience a paradoxical increase in anxiety and become very uncomfortable with relaxation interventions (Heide & Borkovec, 1983; Smith, 2001). Nurses may be able to use a measure of relaxation ability in identifying patients likely to benefit from relaxation interventions. Thus, the purpose of this study was to assess initial psychometric properties of a measure of relaxation ability.

The Model for Cognitive-Behavioral Interventions (Kwekkeboom, 1999) suggests that skill or ability with a specific intervention moderates success with that treatment.

Using a correlational design, a convenience sample of 27 healthy adults completed the relaxation ability scale twice, at a 3-week interval, in addition to measures of past relaxation use and outcome expectancy for relaxation interventions. The relaxation ability scale was comprised of 9 items taken from the relaxation item pool of the Wisconsin Experience Questionnaire (Kihlstrom et al., 1989; Nadon et al., 1991). Items selected for the scale reflect the process of becoming relaxed.

Cronbach’s alpha, t-tests, Mann-Whitney U, and correlation coefficients were calculated.

Internal consistency of the scale was alpha=0.87 at time 1 and alpha=0.85 at time 2. Test-retest reliability was r=0.81 (p<.01). Relaxation ability scores appeared greater among persons who used relaxation for symptoms successfully in the past (M=2.33, SD=0.67) compared to persons who had never used relaxation or who were unsuccessful in past attempts (M=2.26, SD=0.73), however, the difference was not significant in this sample. A difference in outcome expectancy was significant, with persons who had successfully used relaxation in the past reporting more positive outcome expectancy. Relaxation ability was correlated with the belief that relaxation works (r=.49, p<.01). While these preliminary data support internal consistency and test-retest reliability, future research will need to confirm validity by testing the scale in a prospective study, implementing a relaxation intervention in a large sample of persons with cancer-related pain.

Poster Abstract
WORRY AND COGNITIVE REPRESENTATIONS OF ILLNESS IN INDIVIDUALS TREATED FOR LUNG CANCER. Rebecca Lehto Ph.D, OCN, Ann Arbor VA Health Care System, Ann Arbor MI US; Bernadine Cimprich Ph.D., FAAN, University of Michigan, Ann Arbor MI US

The study examined the impact of worry on formation of cognitive representations of illness among individuals newly diagnosed with suspected lung cancer. Worry, a cognitive process driven by anxiety and fears, is common among newly diagnosed cancer patients, but little research has explored how worry influences early illness perceptions. Cognitive representations are learned knowledge structures that guide thoughts and behavior and influence illness adjustment. Worry may affect illness representations by biasing perceptions and facilitating formation of threat-laden thoughts that contribute to further anxiety and poorer adjustment.

This study examined: 1) the influence of worry on cognitive representations of illness over time, and 2) the relationship of selected patient factors and worry in persons with suspected lung cancer. A theoretical framework of person-environment compatibility from a cognitive map perspective.

Using repeated measures, forty-two volunteers (27 males, 15 females) aged 37 to 83 years with suspected lung cancer were assessed in the early post-diagnostic period, and again three weeks following surgery. Measures of worry (Penn State Worry Questionnaire, 3CM Worry rates), the cognitive representation of illness (Illness Perception Questionnaire-Revised, Conceptual Cognitive Map), and measures of patient factors; state-trait anxiety, the illness experience, cognitive functioning, social support, and optimism were used.

Repeated measures ANOVA determined changes in worry and representations of illness measures over time. Correlation and regression analyses determined relationships between worry, illness representations, and possible covariates.

Higher worry was significantly (p<.05) related to negative dimensions of the cognitive representations of illness. The high worriers had higher emotional distress, lower perceived control, and lower perceptions of illness coherence than low worriers over time. Regression analyses showed that state anxiety was a significant (p<.05) predictor of cancer-related worry when controlling covariates. Four variables, state anxiety, an unsatisfying illness experience, less optimism, and ineffective cognitive functioning predicted negative dimensions of cognitive representations of illness. The findings provide new information about detrimental effects of worry on illness representations and indicate need for nursing interventions to counteract negative effects of worry on cognitive representations of illness. Funding Sources: Research supported by NINR, 1 F31 NR07695-01A1.

Poster Abstract
CHALLENGES IN CONDUCTING A MULTI-STATE ONCOLOGY NURSING CLINICAL TRIAL. Frances Lewis R.N., M.N., Ph.D., University of Washington, School of Nsg, Seattle WA US; Marcia Grant R.N., D.S.N., City of Hope Medical Center, Duarte CA US; Joan Haase R.N., Ph.D., Indiana University/Purdue Universiety, Indianapolis IN US; Arlene Houldin R.N., Ed.D., University of Pennsylvania, Philadelphia PA US; Janice Post-White R.N., Ph.D., University of Minnesota, Mpls MN US

The purpose of this presentation is to analyze five methodological challenges and their management that were experienced by a team of nurse scientists when conducting a multi-state clinical trial, the Enhancing Connections Program (alphabetical): Arizona, California, Indiana, Minnesota, Pennsylvania, and Washington.

The purpose of this presentation is to analyze 5 major challenges that a team of nurses experienced in conducting a 6-state clinical trial of a behavioral nursing intervention. i-state clinical trial of a behavioral nursing intervention, the Enhancing Connections Program involving (alphabetical): Arizona, California, Indiana, Minnesota, Pennsylvania, and Washington. The principal investigator and the state’s principal investigators will present both the challenges, ways they were managed, and recommend ways to manage them in future clinical trial studies. Challenges in the clinical trial involved: 1) maintaining intervention integrity across multiple interventionists and states; 2) dealing with tolerable limits of variability in the delivery of the intervention while allowing for nurse-interventionist discretion; 3) facing recruitment challenges, including maintaining high morale in intermediaries; 4) maintaining each state team’s focus and cohesion; and 5) maintaining intermediaries’ commitment for recruiting a highly homogeneous sample over a long accrual period.

Multiple strategies and structures were instituted to manage and minimize the five challenges, including: intensive initial and follow up training with extensive printed materials; performance checklists for intervention monitoring; review of intervention integrity across multiple states; custom-designed spreadsheet tracking forms; processes for tracking timeliness; developing and sustaining nurturant, interactive links with a network of site intermediaries; and holding regular conference calls across the states, among other strategies. Each challenge was managed and our experience enables us to recommend ways to prevent, minimize or address such challenges in future clinical trial studies in oncology nursing. Our goal is to facilitate both rigorous science in oncology nursing as well as offer informed, practical solutions to real-life challenges that will likely surface in future clinical trial studies with behavioral nursing interventions.

Future behavioral research studies in oncology nursing can benefit by addressing these specific challenges early on in the planning process. The conduct of behavioral interventions in oncology nursing within a clinical trial design requires much more than knowing the threats to internal validity; it requires implementing new structures and processes across a multi-state trial. Funding Sources: National Cancer Institute, NIH

Poster Abstract
DOMAINS OF SPOUSAL BREAST CANCER-RELATED CONCERNS: A META-SYNTHESIS OF QUALITATIVE RESEARCH. Frances Lewis PhD, RN, University of Washington, Seattle WA US; Barbara Cochrane PhD, RN, Fred Hutchinson Cancer Research Center, Seattle WA US; Ellen Zahlis MN, University of Washington, Seattle WA US; Mary Ellen Shands MN, RN, University of Washington, Seattle WA US

The increased psychosocial distress of spouses of women with breast cancer and its negative consequences for the diagnosed wife have been documented extensively in previous research.

Both cross-sectional and longitudinal studies in couples with breast cancer provide extensive evidence of spousal distress. However, it is one thing to document the level of morbidity in a spouse; it is another to describe the spouses’ perceived illness-related concerns that impact his psychosocial functioning. The purpose of this study was to synthesize previous qualitative research with spouses in order to derive conceptual domains of their illness-related concerns.

The current study is based on a relational model of adjustment to breast cancer in which members of the household, including spouses, directly experience ill-related concerns and are impacted by the diagnosed wife’s affect, mood, and behavioral changes related to the cancer.

Secondary analyses were conducted on results obtained from five case-intensive elicitation interview studies involving over 300 spouses. All wives had been diagnosed with breast cancer an average of 6-12 months earlier.

Interview data were audio-taped, transcribed verbatim, and inductively coded to identify the conceptual domains of illness-related spousal concerns.

Results across all studies were integrated to identify overarching domains of the spouses’ illness-related concerns. Resulting domains were cross-validated with original transcription text and by peer confirmation of the domains, definitions, and exemplars identified.

Seven domains of spousal illness-related concerns were identified, the spouses' own words: Being overwhelmed with her vulnerability and the unpredictable illness; Riding the roller coaster of her emotions; Managing disruption in lifestyle and routines; Stuffing personal emotions; Struggling to read and support her; Dealing with her changed body; and Feeling inadequate to protect her. An understanding of these domains is important to advance intervention research and evidence-based oncology nursing practice. Intervention research is needed that appropriately targets these priority concerns of spouses. Funding Sources: National Cancer Institute, National Institutes of Health

Poster Abstract
THE MEANING OF PROSTATE CANCER TREATMENT-RELATED ERECTILE DYSFUNCTION: LOW-INCOME LATINO MEN. Sally L. Maliski PhD, RN, UCLA Department of Urology, Los Angeles CA US; Mark S. Litwin MD, MPH, UCLA Department of Urology, Los Angeles CA US

This study will provide oncology nurses with insights into the meaning of prostate cancer treatment-related erectile dysfunction (ED) among low-income Latino men.

Because most prostate cancer treatments affect sexual function, it is necessary understand its meaning within cultural and treatment contexts to design culturally appropriate interventions. However, little is known about ED among low-income Latino men. Our purpose is to describe meanings of prostate cancer treatment-related ED among low-income Latino men. The specific aims are to elicit Latino men’s perspectives on prostate cancer-related ED, identify emergent themes, and describe concepts of meaning and their underlying structure.

This study is grounded in the assumption that cultural beliefs and socioeconomic factors influence the meaning made of cancer-related symptoms. Furthermore, cultural concepts of masculinity are assumed to affect perceptions of ED.

The design employs ethnomethodology that combines ethnographic perspective (cultural) with phenomenologic techniques (meaning) to develop an in-depth understanding of erectile dysfunction through personal interviews. Interviews are conducted by male, bilingual interviewers in the man’s home or by telephone depending on distance to home and the participant’s preference using an open-ended guide. Interviews are audiotaped and transcribed verbatim. Spanish transcripts are translated using a forward-backward-forward process. Follow-up telephone interviews are conducted 3 months later to clarify questions and confirm emerging meanings.

The PI reads all transcripts in their entirety. The sentence is the unit of analysis for data management using NVivo. Unit-by-unit coding is done, identifying the major thought in each unit. For each transcript, themes will be identified from the codes. Emerging codes and themes will be constantly compared to previous ones. Concepts will be identified from the categories, derived from clustering themes across transcripts. These concepts will be dimensionalized to show the range of themes supporting the concepts and the surrounding variations within themes evidenced in codes. Themes and concepts will be examined back through the unit-by-unit coding for expressions of culture-influenced beliefs.

Initial coding is revealing multiple factors affecting masculine identity, acceptance of current ED if temporary, and reluctance to use "unnatural" means of achieving erections. Funding Sources: Department of Defense Prostate Cancer Research Grant

Poster Abstract
UNSOLICITED COMMENTS ON QUANTITATIVE SURVEYS: AN UNTAPPED DATA SOURCE. Sally Maliski PhD, RN, UCLA Department of Urology, Los Angeles CA US; Mark S. Litwin MD, MPH, UCLA Department of Urology and School of Public Health, Los Angeles CA US>

Our purpose is to describe a methodology developed to analyze unsolicited comments written on longitudinal quantitative Health Related Quality of Life (HRQOL) surveys.

Often, people write comments on surveys, perhaps needing to convey thoughts not captured by the items and their choices. Surveys do not allow for individuals’ tendancy to narrativize to make meaning from situations. Thus, these insights go unaccessed.

Numerous HRQOL surveys among men treated for prostate cancer exist. These studies have investigated both general and prostate cancer-specific HRQOL. Few have been longitudinal. To our knowledge, none have reported an analysis of unsolicited comments on these HRQOL surveys. Therefore, the information in these comments has remains unexplored.

We developed a methodology to qualitatively analyze and quantitatively display unsolicited comments on a survey measuring general and prostate cancer-specific HRQOL, urinary symptoms, fear of recurrence, and decisional regret administered over a 2-year period. Our goals were to develop and apply a methodology for analysis and display of this type of data, and identify patterns of concerns expressed in these comments over time.

Unsolicited comments written on these surveys were abstracted from the parent study database for pre-treatment, 1, 2, 4, 8, 12, 18, and 24 months post-treatment. Comments were grouped by ID number by date. We read through all comments. Not all subjects wrote comments at each time point. Thus the number of subjects at each time point varies. Investigators coded the main idea expressed by each statement in each comment. We grouped codes into categories qualitatively and then calculated the number and percent of subjects with codes falling within each category at each time point. Categories with number and percent of times that category appeared at each time point were entered into an Excel spreadsheet. Graphs displaying longitudinal patterns for the most common prostate cancer treatment-related symptoms were created using the percents.

This method allows for descriptive display of data. Its usefulness lies in suggesting explanations of quantitative findings, uncovering areas for exploration, and opening a new data source.

This method allows for analysis of data not originally solicited, but which can provide valuable insights into issues of importance.

Poster Abstract
INTRODUCING THE MUC16 GENE. NEW NURSING IMPLICATIONS FOR EARLY DETECTION OF OVARIAN CANCER. Monica McLemore MPH, RN, University of California, San Francisco, San Francisco CA US; Bradley Aouizerat Ph.D., University of California, San Francisco, San Francisco CA US; Christine Miaskowski Ph.D., RN, FAAN, University of California, San Francisco, San Francisco CA US

The CA125 biomarker is used to monitor progression and regression of ovarian cancer and has not been recommended for population screening as it is not specific nor sensitive to meet screening standards. Surgery, changes in chemotherapeutic agent, and initiation of palliative care are made based on CA125 changes in sera. Understanding tumor markers and their use aids nurses to help patients with decision-making and answer important questions about interpreting these markers. CA125 protein is encoded by MUC16 and is subject to splicing variation. There are 4 known isoforms of CA125 with varying number repeat domains, which house epitopes that identify CA125 in 2 commonly used assays, which do not take into consideration which isoform is being quantified.

Recently, data from molecular geneticists have provided information about the MUC16 gene that have implications including explanations about why CA125 is highly variable across and within populations and why serial measurement is more reliable than single measurement.

The CA125 biomarker cannot distinguish ovarian cancer patients because the current method used to assay the protein. Ethnic and genetic variation among individuals has been shown to impact how people metabolize and respond to drug therapy. This framework can be applied to screening and surveillance biomarkers and should be used to establish ethnicity or genotype specific cutoffs for these markers.

Case and control banked samples will be used in a pooled DNA analysis to: a.) Construct genomic maps of MUC16 in diverse populations to identify single nucleotide polymorphisms-SNPS, b.) Population frequencies of SNPs found in step A will be examined for specific cutoffs and patterns, c.) Information about variation within the biomarker will be compared to clinical data including historical CA125 levels, stage/grade of disease, and chemotherapeutic regimen.

Given the case control design, appropriate parametric and non-parametric statistical analysis will be performed to assess differences between control and case populations on genetic, clinical and demographic variables.

The findings of this study may allow for new CA125 level guidelines to allow for population based screening for Ovarian Cancer. Funding Sources: National Institute of General Medical Sciences

Poster Abstract
FAMILY COMMUNICATION AND DECISION MAKING REGARDING INHERITED BREAST/OVARIAN CANCER RISK INFORMATION. Suzanne Mellon, University of Detroit Mercy, Detroit MI US; Lisa Berry-Bobovski BA, Karmanos Cancer Institute, Detroit MI US; Robin Gold MS, CGC, Karmanos Cancer Institute, Detroit MI US; Nancy Levin BS, Karmanos Cancer Institute, Detroit MI US; Michael Tainsky PhD, Karmanos Cancer Institute & Wayne State University, Detroit MI US

Dramatic advances in cancer genetics and identification of germline mutations in cancer genes such as BRCA1 and BRCA2 have led to new options in genetic risk assessment for families with histories of breast and ovarian cancer. However, little research has been carried out with individuals and their families regarding how cancer risk information is communicated within families and factors that may affect individuals and family members making informed decisions about their health.

The purpose of this study was to explore participants' knowledge of cancer risk, their perceptions and concerns regarding inherited cancer risk information, family communication patterns, and factors that may affect decision-making about inherited cancer risk information.

A family stress model, based on McCubbin and McCubbin's resiliency model, guided this research.

Nine focus groups of family dyads were conducted (N=39) consisting of breast or ovarian cancer patients and close female relatives. Three types of focus groups were used: family dyads together, cancer survivors only, and family members only. A diverse sample of Caucasian (N=27) and African-American (N=12) participants was represented. A semi-structured interview guide was used in all groups.

Content analysis of transcribed focus group interviews was done using the qualitative software package NVivo. Coding of data was done independently by the investigators with verification procedures throughout the analysis.

Analysis revealed six major thematic categories: knowledge and uncertainty of cancer risk, meaning of inherited cancer risk to the family, vigilance in watching out for cancer, barriers in communicating inherited cancer risk information, diversity in family communication patterns, and the family's influence on decision making. There was a pervasive sense of worry among survivors and family members and a heightened perceived risk for their families. Misinformation about genetic cancer was often present as well as vigilance in watching each other's health. Diversity in communication patterns was present across and within families. A majority of women also viewed their families as critical in decision making about risk information. Results from this study support the inclusion of family members in addressing inherited cancer risk information and the importance of contextual family factors to consider when intervening with high risk families. Funding Sources: This work was supported by the Barbara and Fred Erb Endowed Chair in Cancer Genetics to M.A. Tainsky and research funds from the Karmanos Cancer Institute

Poster Abstract
SCREENING RELATED BELIEFS BY STAGE OF BEHAVIOR ADOPTION. Usha Menon PhD, RN, University of Utah College of Nursing, Salt Lake City UT US; B. Erin Witter PhD, University of Utah College of Nursing, Salt Lake City UT US; Victoria L. Champion DNS, RN, FAAN, Indiana University School of Nursing, Indianapolis IN US; Celette Sugg Skinner PhD, Duke University Comprehensive Cancer Ctr, Durham NC US

Colorectal cancer (CRC) remains the third leading cause of cancer death in the United States despite increased attention to early detection and prevention behaviors. Interventions to increase screening for cancer often focus on impacting behavior change by altering beliefs and knowledge.

This study combined two strong behavior change models (TTM and HBM) to provide a basis for tailoring education to increase CRC screening behavior. The primary aim of this study is to test the effectiveness of tailored and non-tailored interventions designed to increase the use of FOBT and sigmoidoscopy among non-adherent men and women aged 50 years or older.

According to the Transtheoretical Model (TTM), behavior change occurs in series of stages rather than as a dichotomous event. Each stage is incremental and effective interventions must focus on altering beliefs at each stage to impact behavior change. The Health Belief Model (HBM) provides guidance for tailoring education at each of the stages for behavior change.

In this prospective, randomized intervention study participants (N=206) were randomly assigned to one of 3 groups: 1) usual care, 2) tailored print communication, and 3) non-tailored print communication. Data were collected via telephone at baseline and 2 months post-intervention. The sample was primarily Caucasian (81%), female (57%), middle class, reported at least a high school education (64%), with a mean age of 60.

Several beliefs differed by stage of behavior adoption. Knowledge as well as beliefs about perceived susceptibility, benefits, barriers, and self-efficacy were assessed for each screening test. For stool blood test, there were significant differences in barriers (F = 9.48, p< .001), benefits (F = 3.17, p < .001) and self-efficacy (F = 2.58, p<.05) were significantly different by stage; precontemplators had significantly higher barriers and lower self-efficacy than individuals in contemplation and action.

Results from this study indicate that individuals differ significantly in beliefs according to their stage of screening adoption. These results could guide future research to develop interventions tailored to individual beliefs and stage of behavior adoption. Funding Sources: National Cancer Institute

Poster Abstract
TELEPHONE LINKED CARE: IMPROVING COMMUNICATION ABOUT UNRELIEVED SYMPTOMS. Kathi Mooney RN, PhD, AOCN, FAAN, University of Utah, Salt Lake City UT US; Susan L. Beck RN, PhD, AOCN, FAAN, University of Utah, Salt Lake City UT US; William N. Dudley PhD, University of Utah, Salt Lake City UT U;S Ramesh Farzanfar PhD, Boston Unversity and Boston Medical Center, Boston MA US; Robert H. Friedman PhD, Boston Unversity and Boston Medical Center, Boston MA US; Camille Broadwater MPH, University of Utah, Salt Lake City UT US

Cancer chemotherapy causes distressing side effects. Since many patients are at home and do not report symptoms, providers do not know that symptoms are unrelieved and can not intervene.

The purpose of this study was to test a computer-based telecommunication system, Telephone Linked Care (TLC-Alert), designed to increase patient/provider communication about treatment-related symptoms. The study was guided by Communication Theory and concepts from The Theory of Unpleasant Symptoms.

Using an experimental design, chemotherapy patients with at least one moderately severe symptom on first cycle were randomized to TLC-Alert or usual care. Eleven symptoms were monitored daily for the second and third cycles. TLC utilizes automated telephone conversations to monitor patients’ symptom experience. It then alerts providers (physician and nurse) by fax or email about moderate to severe symptoms, allowing providers to respond in any manner. Symptom data were measured with standard 1- 10 severity ratings. A semi-structured exit interview documented provider perspective on TLC as a communication tool.

Descriptive statistics, chi square analysis and repeated measures ANOVA were used to describe the sample, their symptom experience and to test differences between the groups.

250 patients participated (126 TLC-Alert; 124 usual care). The participants were primarily female (75.6%), white (92.6%), married (72.5%), with breast cancer (41.3%). On average participants made 74% of daily calls. The highest prevalence of symptoms at moderate to severe levels were: trouble sleeping (76%), fatigue (74%), distress about appearance (74%), depressed mood (62%), pain/discomfort (58%), and nervous/anxious (57%). There were no differences between groups on symptom severity. However, TLC-Alert had more provider-initiated contacts (28.3% versus 14.0%; p=0.05). Provider interviews demonstrated mixed responses to TLC. Some providers reported that TLC provided too much information with little time to respond and patients becoming overly focused on their symptoms. They viewed many symptoms as subjective (fatigue, sleep) with useful information coming only from quantifiable symptoms (fever, vomiting). While provider-initiated calls doubled with TLC, the increase was modest, resulting in no significant symptom improvement. TLC was effective in increasing communication, yet providers may under value patient-reported symptoms, representing a significant barrier to improved symptom relief with both clinical and research implications. Funding Sources: Funded by The National Cancer Institute-1 R01 CA89474-01

Poster Abstract
RISKS OF CHEMOTHERAPY EXPOSURE FROM TRANSPORTING CHEMOTHERAPY PATIENTS TO RADIATION ONCOLOGY DEPARTMENTS. Katen Moore MSN, APRN, BC, NJ Healthcare System Department of Veterans Administration, East Orange NJ US; Marilyn Haas PhD, ANP-C, Mountain Radiation Oncology, Asheville NC US; Peter Goyer MS, RSO, NJ Healthcare System DVA, East Orange NJ US

The risk of chemotherapy exposure by radiation oncology (RO) nursing is becoming critical as more patients are being transported with chemotherapy to RO departments without evidence-based guidelines. The purpose was to assess available protocols and educational guidelines concerning the risk of occupational chemotherapy exposure not routinely cited for "being at risk", i.e. Radiation Therapy (RT) Departments. While annual surveillance recommendations and guidelines for hospital medical oncology and pharmacy staff exists, workplace hazardous substance (i.e. chemotherapy) guidelines are not found in RT Departments, nor given the same attention as their counterparts. There are no published recommendations or existing guidelines for RT departments, leaving RT personnel unaware of their risks and without knowledge about the appropriate actions necessary in the event of a chemotherapy spill. Concurrent Continuous Intravenous Chemotherapy ("CCIVC") is too ubiquitous to ignore in this staff population and was the reason for this research. A Pub Med literature search did not identify any articles specifically addressing chemotherapy exposure outside the medical oncology or pharmacy settings (4/04).

To begin quantifying the need for action, seven e-mails were sent to RT sites treating patients with CCIVC. Five of seven sites reported having no policies regarding transportation of CCIVC patients to and from RT departments. Subsequently, a non-random CCIVC questionnaire was piloted to RT nurses at the 2004 ONS Congress RT SIG meeting.

Thirty-one questionnaires were completed, representing 16 US states and Alberta, Canada and 50% working in outpatient hospital-based RT settings. Over 50% of the respondents were staff nurses and 33% department managers. Though 88% indicated there were CCIVC patients coming through their departments, 50% were unaware of having any hospital policies regarding transporting CCIVC patients and over 70% had no known policies regarding travel to RT clinics. Furthermore, only 64% had chemotherapy spill kits in RT departments, 15% of RT nurses evaluate daily patency of infusion sites, and only15% of radiation therapists had training to assess IV sites.

Current safety measures to protect RT staff appear to stop at the borders of medical oncology and pharmacy settings. Though the actual exposure risk may be small now, with more patients on CCIVC regimens, the risk will only increase in the future. Without precautions to guide staff, RT personnel will remain at high risk.

Poster Abstract
AFRICAN AMERICAN COUPLES COPING WITH BREAST CANCER. Phyllis Morgan PhD, APRN, BC, Fayetteville State University and Johns Hopkins University School's of Nursing, Fayetteville NC US

Diagnosis of breast cancer is a stressful life event that requires women and their partners to develop coping strategies for adaptation. Despite the known relevance of coping strategies on survival and adjustment to breast cancer, and the influence that cultural values may have on coping behaviors, the process of coping with breast cancer among African-American couples has not been well-studied.

The purpose of this qualitative study utilizing grounded theory methods was to explore the processes by which African American couples cope with breast cancer.

The concepts that were derived evolved naturally without utilizing a pre-existing conceptual model or theoretical framework (Strauss & Corbin, 1998).

Interviews were conducted within the mid-Atlantic United States at a safe and convenient place for participants. African American women who were within 6 months to 2 years post-diagnosis participated in the study. Theoretical sampling was used to recruit additional participants for the study until the phenomena of interest was fully explored.

African American couples were asked to complete a demographic data sheet and participate in a 1 to 1-1/2 hour semi-structured interview. Interviews were tape-recorded and transcribed verbatim.

Qualitative data was reported based on identified themes from the constant comparative method. Demographic data was analyzed utilizing descriptive statistics.

African American women and their male partners reported difficulty expressing their needs and concerns with each other in relation to breast cancer. Coping was described as a day-to-day process. Intimacy was one of the most difficult challenges for couples to maintain especially during the treatment phase of the disease. Prayer and spirituality were identified as integral components of coping with the breast cancer experience for African American couples. Males often discussed how they felt helpless and left out of the process. Supportive networks were found to be essential for couples. It is imperative for nurse practitioners to have an understanding of the impact of breast cancer among African American couples. Nurse practitioners should develop culturally sensitive interventions that are geared towards assisting African American couples to communicate more effectively throughout the breast cancer experience. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Sigma Theta Tau International; and American Academy of Nurse Practitioners

Poster Abstract
PREVALENCE OF CANCER CARE BY PRACTICING REGISTERED NURSES IN GREATER LOUISVILLE. Mary Mundt PhD, RN, University of Louisville, Louisville KY US; Carla Hermann PhD, RN, University of Louisville, School of Nursing, Louisville KY US; April Conner MSSW, University of Louisville, School of Nursing, Louisville KY US

The Center for Cancer Nursing Education and Research (CCNER) at the University of Louisville met with community practice partners to discuss a community-wide- educational needs assessment. The group agreed that registered nurses encounter persons with cancer and cancer- related health concerns in all nursing practice and community settings. There was little knowledge about where and with what frequency this occurs. This information is needed to advance cancer nursing practice to respond to cancer as a continued and serious public health problem.

The purpose of the study was to document the prevalence of registered nurse encounters with cancer patients according to the variables of frequency of contact, practice setting, educational level, and experience. The information will be used by CCNER for planning community-wide interventions to improve cancer nursing care.

This study uses a population-based model of describing the occurrence of an event/s to establish prevalence. The population of interest is all licensed registered nurses in a five county region surrounding Greater Louisville (N=9,971). The events documented are: a) frequency of nurse encounters with cancer patients, b) practice setting, c) years of experience as an RN and d) educational preparation of the nurse.

A random sample of 1,180 was selected from the registration list of the Kentucky Board of Nursing. A mailed survey format was used. Subjects received an introductory letter and a survey printed on a stamped, addressed postcard. The survey contained 7 questions-5 forced-choice and 2 open-ended. A second mailing was sent to all non-responders after five weeks.

The analysis will include frequency and percent of response to each question. Chi square analysis will be used to determine differences according to the variables of practice setting, education, and experience of respondents.

To date 46% of the sample is accounted for with 499 (42%) useable returns and 4% undeliverable/new addresses. Data collection continues. The findings of this study will be important to improve knowledge about the self-reported experience of registered nurses in caring for cancer patients. It is imperative that all nurses increase awareness of the needs of cancer patients and their ability to provide high quality cancer nursing care. Funding Sources: Dept. of Health and Human Services, Health Resources and Services Administration

Poster Abstract
DESCRIBING THE EXPERIENCE OF HAIR LOSS WITH CANCER CHEMOTHERAPY. Lillian Nail PhD, RN, FAAN, OHSU, Portland OR US; Frances Lee-Lin RN, MN, OCN, CNS, OHSU, Portland OR US; Jennifer Scherer MPH, OHSU, Portland OR US

Hair loss (HL) is consistently rated as highly distressing to cancer patients, may signal others that someone has cancer, and disrupts self-image. Although HL is a common side effect of chemotherapy and affects approximately 250,000 American adults annually, there is little information about the experience of chemotherapy-induced HL. No published studies capture the extent to which various strategies suggested for dealing with HL are adopted, how HL influences the individual’s daily life, and the extent to which the "new" hair differs from the original scalp hair. The information currently provided to patients about the HL experience is largely based on anecdotal reports, opinion, and speculation.

The purpose of this retrospective descriptive survey is to describe selected aspects of the HL experience in adult cancer patients and to explore relationships of personal characteristics (age, gender, race/ethnicity, and hair characteristics) to variation in hair regrowth, responses to HL, and strategies used to deal with HL.

Cognitive appraisal theory was used to guide the selection of study variables for this exploratory survey. Key variables include recollection of appraisal of HL, selection of coping strategies, distress and disruption in function due to HL, and the match of expectations to actual experience. The anonymous survey was developed by the investigators to reflect the theoretical framework and variables discussed in the research, clinical, patient education, and lay literature. Content and face validity were established through review by clinical experts. Clarity and feasibility were established by review of responses of the first 20 subjects. Subjects (N=200) are being recruited from clinical facilities, cancer advocacy groups, and cancer support groups.

Descriptive statistics, correlations, and cross-tabulations will be used to analyze the results. Relationships between responses to HL and personal characteristics will also be explored.

The findings of this preliminary study will provide essential information about people’s experiences with HL due to chemotherapy. The information is critical to developing accurate informational materials to support coping with cancer treatment and understanding the influence of personal characteristics on responses to HL. Furthermore, the findings will contribute to the development of further research on coping with side effects of cancer treatment. Funding Sources: National Cancer Institute, K24 CA98442

Poster Abstract
SLEEP/WAKE PATTERNS VIA 42-CONSECUTIVE HOUR AMBULATORY POLYSOMNOGRAPHY IN ADVANCED CANCER PATIENTS. Kathy Parker PhD, RN, FAAN, Nell Hodgson Woodruff School of Nursing, Atlanta GA US; Jo Ann Dalton EdD, RN, FAAN, Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta GA US; Donald Bliwise PhD, Department of Neurology, Emory University, Atlanta GA US; Sanjay Jain MD, PhD, Hematology/Oncology, Emory University, Atlanta GA US; Mary Kay Kohles-Baker MSW, RN, Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta GA US; Cathy Vena PhDc, Nell Hodgson Woodruff School of Nursing, Emory University, Atlanta GA US

Patients with advanced cancer frequently report sleep/wake disturbances but polysomnographic (PSG) measures of these patterns have never been descried.

To describe polysomnographic measures of sleep/wake patterns of patients with advanced cancer.

According to the Two-Process Model of Sleep Regulation, irregular nocturnal and daytime sleep can interact to disrupt control of the sleep/wake cycle.

The sample included 31 subjects with metastatic solid tumors recruited during the initial two years of this study. The mean age was 56.7 (8.3); 18 were male and 13 were female. Twelve subjects were White and 18 subjects were Black.

Subjects underwent continuous, ambulatory PSG for 42 hours in their home environment. Analysis included data from two nights (pooled) and the intervening day. Sleep parameters obtained included total sleep time (minutes), sleep efficiency (time asleep/time in bed X 100,%), sleep latency (minutes), latency to the onset of rapid-eye-movement sleep (REM) sleep (minutes), and the percentages (%) of stages 1 through 4 non-rapid-eye-movement (NREM) and REM sleep.

Descriptive statistics were used to summarize sleep parameters. The amount of sleep and waking obtained across the study period was also summarized in 1-hour bins.

Subjects obtained an average of 407.7 (77.7) minutes of nocturnal sleep. The sleep latency was normal at 24.4 (36.0) but the sleep efficiency was low [78.5 (10.3)%]. Most sleep was light NREM Stage 1 [11.5 (5.6)%] and Stage 2 [74 (8.7)%]. An absence of deep NREM Stage 3 [0.3(0.80)%] and Stage 4(0.00)% was noted in most subjects. REM sleep was reduced [13.4 (8.1)%] and the REM latency was prolonged [117.4 (86.0)minutes].

During the daytime, subjects slept an average of 92.6 + 115.5 minutes; 12.4 + 7.3% - Stage 1; 75.0 + 9.3% - Stage 2; 0.3 + 1.3 - Stage 3, 0% - Stage 4; and 12.4 + 9.1% REM.

Hourly analysis across the 42-hours revealed a poorly consolidated sleep/wake pattern.

Results suggest that advanced cancer patients have dysregulation of sleep/wake patterns and that clinical interventions designed to improve both the quality of nocturnal sleep and daytime alertness are warranted. Funding Sources: National Institute of Nursing Research RO1 008125, P20 07798

Poster Abstract
INFORMATION NEEDS OF ADOLESCENTS WITH CANCER. Celeste Phillips MSN, RN, Indiana University School of Nursing, Indianapolis IN US; Joan Haase PhD, RN, Indiana University School of Nursing, Indianapolis IN US; Carol Decker MSW, PhD(c), Indiana University School of Social Work, Indianapolis IN US

Adolescents with cancer (AWC) have much poorer outcomes than younger children or adults: a 30% increase in diagnosis compared to a 10% increase for younger children and lower rates of decreasing mortality and lower 5-year survival rates than younger children.

Although provision of cancer-related information can improve outcomes, little is known about the types of information AWC need and want. The purpose of this paper is to describe AWC perspectives on cancer-related information. Specific aims are to describe the AWC-rated importance of various cancer-related information topics, analyze gender and age group differences in need for information and compare information needs of two different time-since-diagnosis groups of AWC.

This report is drawn from a larger study of resilience and quality of life.

The sample consisted of adolescents with newly diagnosed cancer (n=74) and those one to three years from diagnosis (n=39). The Information Preferences of Adolescents scale was used to measure the adolescents’ information ratings. The scale asks the respondent to rate on a Likert-type scale each of 18 cancer-related items for importance.

The scores on each of the items and the total score were analyzed for each of the two time-since-diagnosis groups. Both groups of AWC rated the need for information as high, although there were differences in specific items rated highest. While there were no significant age group differences, the females had significantly higher item means and total scores in the newly diagnosed group. The qualitative analysis of the additional write-in items generated four themes.

The overall high ratings from the adolescents underscore the need to include AWC in initial and on-going discussions related to their diagnosis and treatment as well as the need for support from the healthcare team (including the MDs, RNs, social workers, etc) to address their psychosocial needs. The findings from this study can serve as a guide for the opening discussions with AWC on the implications of the diagnosis, treatment, and psychosocial aspects their cancer. The initial discussion at diagnosis will serve to open discussions on cancer-related topics, but on-going probing for concerns is required as new information needs arise. Funding Sources: NIH/NINR R29 NR03882-01A1

Poster Abstract
THE EXPERIENCE OF PARTNERS OF WOMEN WITH OVARIAN CANCER. Julie Ponto RN, MS, AOCN, CNS, AL US; Debra Barton RN, PhD, AOCN, Mayo Clinic Cancer Center, Rochester MN US

Ovarian cancer is the second most common gynecologic cancer in the US and has the highest mortality rate of all cancers of the female reproductive system. Approximately 70% of women have stage III or IV disease at the time of diagnosis where the five-year survival rate is less than 30%. These factors pose unique challenges for the woman with cancer and her partner including intensive and prolonged treatment, it’s associated morbidity and significant uncertainly for the future.

The partner’s experience of ovarian cancer is largely unexplored. Studies conducted in families with cancer have found that husbands of women with cancer may experience as much stress related to the cancer experience as the patients themselves. The purpose of this study is to describe the experience of the partners of women with ovarian cancer.

Husserlian phenomenology provides the philosophical framework for this study which will explore the essences and meanings of the partner’s experience. Phenomenologic reasoning assumes that individuals are self-reflective beings who incorporate into their life experiences the effects of spatiality, temporality, corporeality and relationality. The individual’s experience is unique and paramount.

A qualitative, exploratory, descriptive research design was undertaken. Criterion sampling was used to identify partners of women with ovarian cancer who were at least 18 years of age, able to speak and read English, and able to participate in a one-to-one telephone interview. Data was collected through telephone interviews, which provide participants the opportunity to share their ideas, thoughts and perceptions during individual interviews. The interviews were audiotaped and transcribed verbatim. The interviewer reviewed the transcriptions to verify transcription and add commentary regarding the tone, flow and character of the interview. Analysis will be performed by two members of the research team to extract essences and meanings of the experience. These will be clustered into categorizes representing the experience. Discrepancies will be discussed between the coders until agreement is reached. Results of the analysis will be presented.

The experiences of partners of women with ovarian cancer remain largely unexplored and warrant investigation. The results of this descriptive work should yield meaningful information that will guide future research in this population. Funding Sources: Mayo Clinic Cancer Prevention and Control

Poster Abstract
TESTING THE QUALITY HEALTH OUTCOMES MODEL IN CANCER CARE. Laurel Radwin RN, Ph.D., University of Massachusetts Boston, Boston MA US; Gail Wilkes RNC, MS, AOCN, Boston Medical Center, Boston MA US Joanne M. Garvey Ph.D (c), R.N., University of Massachusetts Boston, Boston MA US; Wendy Hubenthal M.Ed., M.S., University of Massachusetts Boston, Boston MA US; Mary Philpot RN, BSN, University of Massachusetts Boston, Boston MA US; Caitlinn Mann B.S., University of Massachusetts Boston, Boston MA US; Jill ODonnell BS, RN, Beth Israel Deaconess Medical Center, Boston MA US; Susan Whalen BS, RN, Brigham and Womens Hospital, Boston MA US; Weibo Lu MS, RN, Second Military Medical University, Shanghai CN

A convincing body of evidence verifies the relationship between nurse staffing levels and poor patient outcomes. The relationship between the quality of cancer nursing care and desired outcomes for hospitalized patients deserves similar attention.

The purpose is to determine relationships among the quality of patient-centered cancer nursing care, desired health outcomes, patient trait and state characteristics, and nursing-focused healthcare system characteristics.

The Quality Health Outcomes Model (QHOM) posits relationships between interventions and outcomes that are affected by patient characteristics and health care system characteristics. In this study, the intervention component of the QHOM is conceptualized as patient-centered nursing care; the outcome component as desired health outcomes; the healthcare system component as nursing focused characteristics; and patient characteristics as trait and state characteristics.

A correlational longitudinal design is used. The sample will include 240 hospitalized cancer patients, and approximately 30 nurses who provide direct patient care. Patient participants will complete three self-administered questionnaires during hospitalization, a telephone interview one week after discharge, and a questionnaire at the first clinic visit. Nurse participants will complete monthly questionnaires. Estimates of reliability and validity of the questionnaires are acceptable.

Patient questionnaires and interviews measure four interventions representing the quality of patient-centered nursing care: responsiveness, individualization, coordination, and proficiency. Patient questionnaires also measure desired health outcomes: sense of well-being, cancer patient optimism, authentic self-representation, trust in nurses, fortitude, and functional status. Nurse questionnaires measure nursing-focused healthcare system characteristics: nursing experience, nursing expertise, nurse-physician collaboration, nurse control over practice, and nurses’ advocacy for patients. Nurse staffing will also be measured.

A patient demographic questionnaire measures patient trait characteristics. Nursing acuity data and medical diagnosis variables measure patient state characteristics.

Exploratory data analyses will be conducted before analyses testing the specific study hypotheses. Structural equation modeling will be used to examine relationships among the variables. Two models will be tested; one examines the moderating effects of system and patient characteristics on the relationship between interventions and outcomes; the other examines the mediating effects of the characteristics.

Preliminary findings will be presented and implications will be described. Funding Sources: Mentored Clinical Scientist Development Award from the Agency for Healthcare Research and Quality (K08 HS 11625) to Dr. Radwin

Poster Abstract
DEVELOPMENT OF A TAILORED PRINT INTERVENTION TO PROMOTE COLON CANCER SCREENING. Susan Rawl PhD, RN, Indiana University School of Nursing, Indianapolis IN US; Victoria Champion DNS, RN, Indiana University School of Nursing, Indianapolis IN US; Celette Skinner PhD, Duke University Medical Center, Durham NC US; Linda Scott MS, Indiana University School of Nursing, Indianapolis IN US; Karen Schmidt RN, MS, Indiana University School of Nursing, Indianapolis IN US; Leslie Weaver MSW, Indiana University School of Nursing, Indianapolis IN US; Samantha Steen , Indiana University Cancer Center, Indianapolis IN US; Patrick Monahan PhD, Indiana University School of Medicine, Indiana polis IN US;

First-degree relatives of colorectal cancer (CRC) survivors have a two- to threefold increased risk of developing the same disease. Screening which leads to removal of precancerous polyps has been shown to decrease CRC incidence by 75-90%.

Despite their increased risk, participation in CRC screening among first-degree relatives is low. Tailored print interventions based on theories of behavior change have demonstrated considerable promise as interventions for promoting health behaviors. The primary purpose of this study was to develop and test a tailored print intervention to increase participation in screening among first-degree relatives of CRC survivors.

The interventions contains messages individualized for each recipient based on assessment of each person’s demographics, CRC risk factors, health beliefs (perceived risk, benefits, barriers, self-efficacy), and stages of adoption for CRC screening.

An ongoing randomized trial is being conducted to determine efficacy of this tailored print intervention; participants are randomly assigned to receive the tailored print intervention (experimental) or a non-tailored American Cancer Society brochure (control). Data regarding readership, relevance and satisfaction with both the tailored and nontailored interventions were collected via structured telephone interviews at one week post-intervention. This presentation will describe the process used to develop the tailored print intervention and results of comparisons between experimental and control groups’ perceptions of the readership, perceived relevance and satisfaction with the print materials.

Analyses will be conducted using descriptive statistics and chi-square tests of independence.

Research implications for development of tailored print interventions and assessment of readership, relevance and satisfaction will be presented. Funding Sources: National Cancer Institute

Poster Abstract
FATIGUE IN BREAST CANCER SURVIVORS: A PSYCHOMETRIC ANALYSIS. Kristina Reuille MSN, RN, Indiana University School of Nursing, Indianapolis IN US; Janet Carpenter PhD, RN, Indiana University School of Nursing, Indianapolis IN US

Fatigue is a well-known and distressing side-effect of cancer treatment. For some survivors of cancer, fatigue may be a persistent problem. Although reliable and valid measures of fatigue have been identified for patients undergoing treatment, the psychometrics of these measures have not been widely evaluated in cancer survivors.

Fatigue may persist after treatment has ended. However, measures of fatigue have not been widely evaluated for reliability and validity in cancer survivors. Therefore, the purpose of this study is to analyze the psychometrics of several instruments used to measure fatigue in breast cancer survivors (BCS), with the goal of determining which instruments are the best unique measures of fatigue in this population.

Psychometric theory and a symptom experience model based on the Common Sense Model will guide this analysis.

Baseline data are being collected from 55 BCS (Mean age=51; SD=9) participating in a randomized clinical trial testing the efficacy of, and side effects associated with, a medication to treat hot flashes in BCS. The majority of women are married (84%), employed (75%), and Caucasian (92%). Mean time since treatment is 37 months (SD=44) and 51% of the women are taking tamoxifen.

Fatigue measures include: 1) F_POMS-sf; 2) Piper Fatigue Scale); 3) Circumplex Octant 6 (low activation, negative affect), and 4) the MOS-SF36 Vitality Subscale. Construct validation measures include: 1) CES-D; 2) STAI; 3) PANAS; 4) Ham-D; 5) MOS-SF36 physical functioning and role-Physical subscales; 6) Marlowe Crowne Social Desirability Scale (MCSDS) and 7) the Pittsburgh Sleep Quality Index.

Planned analyses include 1) Cronbach Alpha reliabilities; 2) content validity evaluation assessing fatigue dimensions (quality, intensity, duration, distress); and 3) construct validity including convergent, discriminant and known groups validity using Pearson correlations and t-tests.

Preliminary hypotheses are that fatigue measures will: a) correlate positively with measures of depression, anxiety, and negative affect; b) correlate negatively with the physical subscales of the MOS-SF36; c) correlate weakly with the MCSDS; and d) be significantly higher for those with poor sleep quality (using the PSQI cutoff score).

Findings will be used to guide selection of fatigue measures in future studies of breast cancer survivors. Funding Sources: Supported by NINR/NIH grant R01 NR05261 and PHS grant 5T32NR07066.

Poster Abstract
PSYCHOMETRIC TESTING OF THE CANCER TREATMENT-RELATED FATIGUE REPRESENTATION (CTRFREP) SCALE. Kristi Reuille MSN, RN, Indiana University School of Nursing, Indianapolis IN US; Victoria L. Champion DNS, RN, FAAN, Indiana University School of Nursing, Indianapolis IN US; Joan E. Haase PhD, RN, Indiana University School of Nursing, Indianapolis IN US

Cancer treatment-related fatigue (CTRF), a distressing symptom of cancer treatment, affects up to 100% of patients. The relationship between CTRF intensity and its associated distress is not direct; some who experience high levels of intensity experience little distress, whereas for others, low-level intensity may result in high levels of distress.

CTRF representation – the beliefs, thoughts and associated emotions of the CTRF experience – may relate to development or absence of CTRF distress. A valid, reliable measure of representation is needed to test models of CTRF experience (i.e. intensity and distress) including representation. The purpose of this study is to test a measure of CTRF representation adapted from existing illness representation measures.

The Common Sense Model (CSM) and symptom experience literature guided the development of a model of CTRF experience including representation. This model and psychometric theory are guiding the study.

To develop the CTRFRep, items were added to existing illness representation measures to reflect CTRF representation as described in research. Content validity was established using 4 content area/measurement experts. The CTRFRep (item N=96) has subscales reflecting Identity (N= 19), Timeline (N=10), Consequences (N=11), Controllability (N=20), Coherence (N=5), Emotional Representation (N=8) and Cause (N=23).

Adult patients are eligible for the study if 1) receiving > 4 weeks of radiation treatment, 2) not receiving brain radiation and 3) without known brain metastases. Data collection occurs via in-person interview 4 weeks following radiation treatment. The projected sample size is 100.

Construct validation measures include the: 1) F_POMS-sf; 2) MFI-20; 3) CES-D; 4) PANAS and 5) Cancer Related Fatigue Distress Scales (CRFDS). Planned analyses include: 1) internal consistency reliability; 2) construct validity of CTRFRep subscales – convergent and discriminant – using Pearson correlations; and 3) exploratory factor analysis to determine the factor structure of the overall scale. The goal of the analysis will be to generate a parsimonious measure of CTRF representation.

A reliable and valid measure of CTRF representation can be used to test models of the CTRF experience in which representation is hypothesized to mediate the relationship between CTRF intensity and its associated distress. Funding Sources: This research is supported by PHS grant 5T32NR07066.

Poster Abstract
IN-HOME PAIN EDUCATION FOR CANCER PATIENTS-A PILOT STUDY. Mary Anne Hales Reynolds RN, PhD, CS, Idaho State University-Dept. of Nursing, Pocatello ID US

The World Health Organization estimates 3.5 million people suffer from cancer-related pain daily and, in theory, poorly managed pain impacts negatively quality of life for the cancer patient and their caregiver. The purpose of this study is to evaluate the effectiveness of a well-established pain management education protocol (PMEP), developed at the City of Hope (COH)in a rural Eastern Idaho community (E. Idaho). Specific aims are to: (1) examine the effect of PMEP on quality of life, pain knowledge and attitudes, and pain scores; (2) compare demographics and outcome scores between COH and E. Idaho; and (3) generate information for future PMEPs and research.

Quality of life is conceptualized as multidimensional including: physical, psychological, social, and spiritual well-being domains. Poorly managed pain interrupts all four domains and is seen as a loss of function, interferes with normal roles, and may result in unnecessary suffering. A convenience sample of twenty-five E. Idaho home health/hospice adult cancer patients and caregivers will receive two, one hour PMEP sessions, set in their home, within a two-week period of time. The PMEP includes general information about pain and pharmacological and non-pharmacologic pain management. Teaching methods include verbal instruction by the nurse, a patient-education booklet, and two audiocassettes. All scripts were developed by the COH. Using a pre-post test/time-series design, measurements will be obtained before the first PMEP session and one week after the last PMEP. Instruments include: Quality of Life Patient and Family Versions; Patient and Family Pain Questionnaire; and Self-Reported and Caregiver Reported Pain Scores. All instruments have established reliability and validity data.

Data analysis will include aggregated descriptive profiles of the E. Idaho study population compared to COH populations. Descriptive statistics will be used to evaluate outcome measures over time using repeated/measures MANOVA to examine the PMEP effectiveness.

Clinical implications from this study relate to evaluating the effectiveness of in home patient education by nurses, and the refining, adapting, and tailoring the transferability of the COH-PMEP. This information can then be used in developing future patient education and research programs that will address the unique needs of oncology patients. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Knoll Pharmaceutical

Poster Abstract
LYMPHEDEMA EDUCATION AND IDENTIFIED EDUCATIONAL RESOURCES IN BREAST CANCER PATIENTS. Sheila Ridner PhD, RN, ACNP, Vanderbilt University School of Nursing, Nashville TN US

Significance: Approximately 15% to 28% of American breast cancer survivors develop chronic lymphedema after breast cancer treatment.

Problem and Purpose : Breast cancer survivors report dissatisfaction with information received about the risk of getting lymphedema. Survivors with lymphedema verbalize distress towards healthcare professionals for failure to educate them about lymphedema. The purpose of this retrospective study was to compare lymphedema and risk reduction strategy education prior to breast cancer treatment between women with and without subsequent lymphedema.

Theoretical/Scientific Framework: Self-regulation theory proposes that individuals cope based upon their understanding of an experience. Inherent in this theory is the need for patients to have adequate information about the experience. Failure by healthcare professionals to educate about the lymphedema risk experience may impair patients’ ability to manage this risk.

Methods: Breast cancer survivors (74 with lymphedema, 75 without) were asked: (1) Prior to having breast cancer treatment did anyone talk to you about your risk for lymphedema? If yes, who?; (2) Prior to having breast cancer treatment did anyone talk to you about ways to decrease your risk for lymphedema? If yes, who?; and (3) If you want to learn more about lymphedema occurring after breast cancer treatment who would you ask or where would you look for information?

Data Analysis: Chi-square was used to analyze results. Frequency data were calculated to identify educational resources.

Findings and Implications: Significant group differences were noted: (1) 81% with lymphedema did not recall receiving lymphedema risk education compared to 60% without lymphedema (p=.005). (2) 88% with lymphedema did not recall being told of risk reduction strategies compared to 71% without lymphedema (p=.01). Surgeons, nurses, oncologists, and friends provided pretreatment information. The internet was the most cited resource. These data suggest that breast cancer patients recall little education about lymphedema being given pretreatment and those with lymphedema recall the least amount being given. Although some healthcare professionals provide information pretreatment, the internet is viewed as the greatest resource. Healthcare professionals should consistently provide this education. Research concerning the influence of pretreatment education on lymphedema risk reduction and optimal methods of educational delivery is indicated. Funding Sources: NRSA Grant 5 F31 NR07854, Iota Chapter Sigma Theta Tau,Vanderbilt Disseratation Enhancement Award

Poster Abstract
PICC LINE INSERTION COMPLICATIONS UPPER ARM VS. LOWER ARM. Philip Rosati RN, OCN, Sunrise Hospital & Medical Center, Las Vegas NV US Regina Lee RN, CCRC, Sunrise Hospital & Medical Center, Las Vegas NV US Christine Belle RN,BSN,CCRP, Sunrise Hospital and Medical Center, Las Vegas NV US

Growing numbers of oncology patients have PICC (peripherally inserted central catheter)lines placed as the VAD (venous access device)of choice. The potential complications of these devices can be profound.

Common complications of antecubital space PICC placement include deep vein thrombosis, mechanical phlebitis, infection and mechanical occlusion. There are no existing clinical studies that specifically compare complication rates of PICCs when inserted into the upper arm versus the traditional insertion method of antecubital space placement. The hypothesis suggests PICC complication rates can be decreased, and patient outcomes increased utilizing the upper arm insertion method. The use of high tech tools and advanced practice techniques such as ultrasound guidance and micro introduction facilitate placement in the upper arm where blood flow is greatest, thereby potentially reducing complications.

PICC catheters placed in the upper arm (above the antecubital fossa) using advanced placement techniques will have fewer post placement complications than those placed in the lower arm (below the antecubital fossa), utilizing the same PICC catheter and advanced placement techniques. The Groshong NXT will be the only catheter inserted for purposes of this study.

This is a randomized parallel group study to evaluate the placement of a Groshong NXT PICC by comparing post insertion complication rates between upper and lower arm placement. Patients will be randomly assigned to the lower arm or upper arm group. A total of 250 patients will be enrolled from each of four participating facilities for a significance level of alpha0.05 and 80 power to detect a 5 difference.

Analysis and summaries will be performed using the following two datasets 1. All randomized patients. 2. Completed patients. Analysis will compare the incidence of complications between the two groups at the end of the first seven days of treatment including the demographics and other pretreatment characteristics. The denominators for the calculations will be based on the number of patients in each group. The Cochran-Mantel-Haenszel formula will be utilized.

This study will provide the rationale to protect the compromised oncology patient population with outcomes-based practice guidance for PICC insertions. Funding Sources: BARD Access Systems and Sunrise Hospital & Medical Center

Poster Abstract
ENHANCING THE LIKELIHOOD OF PRACTICE CHANGE IMPLEMENTATION: SELECTION OF TEAM MEMBERS. Dana Rutledge RN, PhD, California State University, Fullerton Department of Nursing, Fullerton CA US; Cindy Idell RN, MSN, AOCN, City of Hope National Medical Center, Duarte CA US

Implementing evidence-based practices is a healthcare imperative. Processes that maximize the success of implementation are little-studied. This project discusses the potential impact of team member selection using experiences with an ongoing project.

Knowledge utilization frameworks (diffusion of innovations, CURN, Iowa Model) support evidence-based practice (EBP) and research utilization strategies (linkage agents, pilot work), yet require specific knowledge and experience sets for persons implementing practice changes. To date, no one has evaluated use of theory-based selection processes for team members in an EBP project. Rutledge (1995; in press) describes potential roles of nurses and others in EBP processes (e.g., information brokers, systems savvy providers, change agents, practice adopters, educators). Purposeful work team selection - maximizing inclusion of persons with specific skill sets - increases the likelihood of success in implementation/maintenance and decreases barriers such as difficulty understanding research or lack of authority to implement changes.

In 2002, Idell and colleagues, in a change project focused upon increasing pain re-assessment in nurses, put together a work team made of persons based on anticipated role fulfillment. As expected, many of the 23 team members have served more than one role in the project. Team attrition over 2 years (4 members lost) was unrelated to role or position. During this continuing practice change effort, the most common roles are information brokers (47%), interested clinicians (37%), systems savvy members (37%), change agents (32%). Lesser roles include educators (21%), early new practice adopters (16%), evidence retrievers (11%).

Use of theory-based team selection in an EBP project may maximize process/clinical outcomes while decreasing barriers. Purposeful team role diversity has helped facilitate project success. Including information brokers, retrievers, and critiquers ensures selection of EBP strategies by interested RNs and early practice users rather than sole reliance on clinical judgment. Including managers, physicians and other systems savvy members (Pharmacy, Rehab, Research) enhances stakeholder buy-in, avoids replaying past mistakes, and illuminates driving/restraining forces. Authors recommend that future groups include fewer information brokers and more educators/early new practice users (proportionally) to encourage earlier and smoother implementation. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Ortho Biotech Products, L.P.

Poster Abstract
VIRTUAL REALITY INTERVENTION FOR CHEMOTHERAPY SYMPTOMS. Susan Schneider PhD, AOCN, Duke University School of Nursing, Durham NC US; Linda Hood RN, MSN, AOCN, Duke University School of Nursing, Durham NC US; Marcia Grant RN, DNSc, FAAN, City of Hope, Duarte CA US

Successful completion of chemotherapy offers a greater chance of tumor non-recurrence and long-term quality of life. However, many patients have difficulty adhering to the prescribed regimen because of related symptoms. Virtual reality (VR) provides a distracting, immersive environment, which blocks out competing stimuli, ameliorates chemotherapy symptoms, and thus, helps patients tolerate their chemotherapy regimen.

This study explored VR as a distraction intervention to relieve symptom distress in outpatients receiving chemotherapy and to determine the post-treatment effect on symptom distress after 48 hours.

Lazarus and Folkman’s Stress and Coping Model identifies interactive distraction as an emotion-focused coping strategy utilized by individuals experiencing a threatening situation. VR is an immersive and interactive intervention, which engages several senses simultaneously. The individual wears a headset that projects an image with the corresponding sounds of the environment. The sense of touch is involved through a computer mouse that allows image manipulation.

A crossover design was used to determine whether VR was effective in reducing chemotherapy-related symptom distress in patients and whether the effects last for two days.

Over 100 adults receiving chemotherapy for breast, colon, or lung cancer at Duke University Health System were randomly assigned to receive VR during one chemotherapy treatment and no VR (control) during an alternate treatment. The Adapted Symptom Distress Scale-2, the Revised Piper Fatigue Scale and the State Anxiety Inventory measured aspects of symptom distress for two matched chemotherapy treatments at one pre-test and two post-test data points. The Presence Questionnaire and an open-ended evaluation were used to evaluate the subjects’ VR experience. The influence of age, gender, coping style, and immersive tendency on the effectiveness of the VR intervention were explored. All instruments have demonstrated reliability and validity in this population.

Ninety-two subjects have been enrolled, with preliminary results demonstrating positive outcomes following VR use. Paired t-tests. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Ortho Biotech Products, L.P.

Poster Abstract
DETERMINANTS OF HOSPITAL NURSES INTEGRATION OF TOBACCO REDUCTION INTO PRACTICE: A PATH ANALYSIS. Annette Schultz MN PhD Candidate, RN, School of Nursing, University of British Columbia , Vancouver BC CA

Reduction of tobacco use is the single most effective means to reduce cancer related morbidity and mortality. Nurses are encouraged to integrate tobacco reduction (TR) into practice, which provides hospitalized smokers with an opportunity to talk about tobacco use and stopping. Even though successful stopping is unlikely during hospitalization, this opportunity would educate and engage tobacco users in the process of stopping, which could be readdressed at post-discharge health care visits.

The purpose was to investigate acute care nurses' integration of TR into practice. This presentation discusses results from testing a hypothesized path model that defines relationships among intrapersonal and interpersonal variables associated with nurses' integration of TR.

This mixed-methods study investigated the practice, perspectives, and work context of nurses working on 16 adult in-patient hospital wards in Western Canada. An ecological perspective framed this study, which theorizes that behavior is a product of both individual and contextual factors.

A survey was completed by nurses (n=214, response rate 54%), which measured the following variables: age, professional demographics, smoking status, workplace environment, perception of organization resources, perception of other nurses' TR activities, attitude toward a TR role, perception of barriers, and their own TR activities.

The analysis was guided by the theoretical model that hypothesized the relationship among the variables. A series of multiple regression analysis were conducted testing each path in the model.

Enhancing our understanding of nurses' integration of TR into practice could inform health policy and programming. Nurses' perception of other nurses' activities and organizational resources were the most influential variables. There are three barrier dimensions: professional, concern for patient, tobacco related health issues. The perception of professional barriers appears to be the strongest influencing dimension. Less influential and still statistically significant was nurses' attitude toward providing TR. Smoking status was statistically significant, but only in relation to perception of professional barrier and in shaping nurses' attitude toward TR. Perceived innovativeness of the ward was also significantly related to level of TR activities. These findings suggest that support for TR champions partially through incorporation of organizational resources will influence all nurses' integration of TR activities. Funding Sources: University of British Columbia, School of Nursing, Katherine McMillan Directors Discretionary Fund

Poster Abstract
PSYCHOMETRIC TESTING OF THE REVISED FAMILY CAREGIVING SKILL PROFILE. Karen Schumacher RN, PhD, University of Nebraska Medical Center College of Nursing (as of 11/01/04), Omaha NE US; Barbara Stewart PhD, Oregon Health & Sciences University, Portland OR US; Susan Beidler RN, PhD, Florida Atlantic University, Boca Raton FL US; Patricia Archbold RN, PhD, Oregon Health & Sciences University, Portland OR US

This poster describes the third in a series of studies designed to develop and test a new instrument called the Family Caregiving Skill Profile (FCSP).

The development of caregiving skills is integral to the process of taking on the caregiving role when a family member begins treatment for cancer. Clinical experience suggests that family caregiving skill may be related to both patient outcomes (e.g. symptom distress, health services utilization) and caregiver outcomes (e.g. strain, anxiety). However, lack of an instrument with which to measure caregiving skill limits research in this area. The FCSP is designed to fill this gap. The first step in instrument development was a concept development study using qualitative research methods. The initial version of the FCSP was based on this study. The second step was the initial testing of the instrument, which documented reasonable internal consistency reliability and concurrent validity for a new instrument, and indicated revisions needed in FCSP items, scoring, and administration. The current study aims to test the revised FCSP with a sample of 50 family caregivers of adults receiving chemotherapy for solid tumors or lymphoma. To administer the FCSP, a semi-structured interview is conducted with the family caregiver (and patient when he or she is able and willing to participate). Interview data are rated for the extent to which 68 indicators of skill across 8 subscales are present. Mean scores are calculated for each subscale and the FCSP as a whole. Internal consistency and interrater reliability will be calculated using Cronbach’s alpha and Cohen’s kappa respectively. Concurrent validity will be assessed using bivariate correlations between FCSP scores and theoretically related caregiver and patient variables. Predictive validity will be assessed by correlating FCSP scores with patient health service utilization during the subsequent month.

The results of this study will provide a basis for larger scale testing of the FCSP using multivariate psychometric techniques.

The ability to measure family caregiving skill would make it possible to include family caregiving skill as a variable in research. Funding Sources: National Institute of Nursing Research, (RO1 NR04685)

Poster Abstract
DIFFERENTIAL EFFECTS OF A COGNITIVE BEHAVIORAL INTERVENTION ON SYMPTOM MANAGEMENT. Paula Sherwood RN, PhD, CNRN, Michigan State University, East Lansing MI US; Charles W. Given PhD, Michigan State University, East Lansing MI US; Barbara A. Given RN, PhD, FAAN, Michigan State University, East Lansing MI US; Victoria L. Champion RN, DNS, FAAN, Indiana University, Indianapolis IN US

Symptom management is essential to quality of life for cancer patients, and is a national priority for oncology nurses. Cognitive behavioral interventions (CBIs) can decrease the severity of treatment-related symptoms for cancer patients. However, the effectiveness of CBIs for subgroups of patients is unknown, as samples are rarely large enough to determine significant differences. The purpose of this study was to examine the effectiveness of a CBI on symptom severity for specific subgroups of patients.

Cognitive behavioral theory guided the intervention, helping patients form a positive orientation to symptom management, understand the nature of symptoms, identify realistic goals, implement strategies to decrease symptoms’ severity, and verify the effectiveness of strategies.

Data from two randomized clinical trials, which used an identical CBI, (N=265 and N=124) were combined for the analysis. Both studies recruited patients over 21 with a solid tumor cancer or non-Hodgkin lymphoma who were undergoing chemotherapy. Patients in the control group received standard supportive care. Patients in the experimental group received a 5-contact, 10-week nurse-directed CBI. Symptom severity was assessed at baseline and at 10-weeks by asking the patient to rate the severity of eleven symptoms on a scale from 0 to 10, then summing individual severity scores to generate an overall symptom severity score. Reliability analysis of severity scores were not appropriate, as they were based on differing symptoms for each patient. Regression analyses were used to determine the effects of symptom severity and depression at baseline, age, gender, study, site, stage, and type of cancer on symptom severity at 10-weeks.

The only significant predictors of symptom severity at 10-weeks were group (p=.00) and the interaction between symptom severity at baseline and sex (p=.01). Further analyses revealed that the intervention improved and/or maintained symptom severity for males with all levels of symptom severity at baseline and for females with moderate symptom severity at baseline.

Regarding females with low and high symptom severity at baseline, the intervention did not improve or maintain symptom severity. Before significant financial resources are invested in interventions, subtle differences such as these must be further explored to tailor interventions toward those who are amenable to intervention. Funding Sources: Mary Margaret Walther Cancer Institute, National Cancer Institute and the National Institute for Nursing Research (2 R01 NR/CA01915)

Poster Abstract
PAIN KNOWLEDGE IN HOSPICE AND INTERNAL MEDICINE NURSES IN TAIWAN. Chen Shu-Ching, Taipei Medical University/Chang Gung Institute of Technology, Taoyuan WLai Yeur-Hur, Taipei Medical University, Taipei TW

Pain is a major care problem in terminal cancer patients. Due to the limited space in hospice-care institutions in Taiwan, some terminal cancer patients receive care in internal medicine wards. To provide better pain control for terminal cancer patients, it is necessary to identify how well prepared nurses are to manage cancer pain.

The purposes of this study were to examine and compare the level of knowledge about cancer pain management in nurses working in hospice and internal medicine.

The Nurses’ Pain Knowledge and Attitudes Survey Chinese version was used cross-sectionally to assess nurses’ knowledge about pain. Of 150 surveys distributed, 141 were returned with complete responses (71 from internal medicine nurses and 70 from hospice nurses; 94.0% response rate).

Data were analyzed by descriptive analysis, chi-square and independent t-test.

The results showed that (1) nurses working in both settings had insufficient knowledge of pain management, in particular, misconceptions and lack of knowledge about opioids; (2) nurses in both settings had more problems handling questions related to case study; (3) nurses in both groups had better knowledge about pain assessment than about pain management; and (4) nurses working in hospice had a significantly higher correct answer rate to survey items (74%) than did nurses working in internal medicine wards (47.1%).The results strongly suggest that nurses¡¦ knowledge about pain management still needs to be strengthened, especially for nurses working in non-cancer care specific internal medicine wards. More discussions of case studies about pain management are suggested to enhance nurses¡¦ abilities in integrating pain knowledge into a complex pain management situation.

Poster Abstract
THE RELATIONSHIP OF ATTITUDE, SUBJECTIVE NORM, PERCEIVED BEHAVIORAL CONTROL, AND PERCEIVED THREAT ON MAMMORGRAPHY BEHAVIOR. Susan Steele DNS, RN, AOCN, OLOL College, Baton Rouge LA US

Breast cancer is a life threatening health problem of concern for many women. Early detection with mammography is a woman’s best chance of surviving the disease, but unfortunately many women are not active participants in their own health.

The purpose of this study was to test the expanded Theory of Planned Behavior in order to determine the relationship between Southeastern Louisiana women’s beliefs, attitudes, subjective norm, perceived behavioral control, and perceived threat to mammography intention.

The sample consisted of 302 adult women, between the ages of 40-74, who resided in one of the targeted parishes, were able to read English, respond to a written questionnaire (Chronbach’s alpha of .7521 to .9242 and a CVI of .945), and without a prior history of breast cancer. Data was collected over a six-week period from community agencies in East Feliciana, Iberville and St. Helena parishes.

A correlation study design with multiple regression for path analysis was used.

The findings indicated that the perceived threat of breast cancer development and breast cancer susceptibility were not related to mammography intention. The addition of breast cancer severity to the Theory of Planned Behavior explained 28% of the variance. Based on stepwise regression of the significant variables that independently contributed to prediction of intention to have a mammogram included: "My doctor orders one" (â = .292; p < .001), "Having time to go" (â= .232; p < .001), "The thought of breast cancer scares me" (â = .186; p < .001) and "Will cause me to have pain" (â = .121; p = .019). When a physician discussed mammography with a woman and the woman was above the poverty level, she was twice as likely (odds ratio = 2.06) to have a mammogram. Therefore, healthcare providers should encourage an annual screening mammogram, provide convenient access, and decrease the myths surrounding mammography. Funding Sources: OLOL College: Endowed Professorship

Poster Abstract
BALANCING RELATIONSHIPS WITHIN A DISCORDANT WORLD: THE SOCIOCULTURAL CONTEXT OF BREAST CANCER SCREENING AMONG KOREAN IMMIGRANT WOMEN. Eunyoung Suh PhD, RN, San Diego State University, San Diego CA US

Korean immigrant women (KIW), who make up one of six major sub-populations in Asian American and Pacific Islanders in the U.S., are reported having far lower rates of breast cancer screening than that of women in general.

Sporadically reported barriers to the screening practices failed to illustrate the broader sociocultural contexts, within which women in this population seek help for breast health. Thus, the purpose of this study was to explore and interpret the cross-generational and sociocultural processes of breast cancer screening among KIW.

Three theoretical underpinnings of this study include symbolic interactionism, the meta-concept of cultural competence, and a concept of Korean womanhood.

The Grounded Theory methodology was utilized. Twenty KIW, age between 20 and 81, participated in a set of two consecutive qualitative interviews, which are conducted in Korean language.

The qualitative data was transcribed and analyzed using the constant comparison technique. The open coding was carried out in Korean to preserve Korean cultural values and nuance embedded in the text. Translation into English occurred from the axial coding constantly comparing the contextual meanings between two languages.

Balancing relationships within a discordant world is the core concept of the context of breast cancer screening among KIW. There are sociocultural discords in perceptions of breast cancer and screening procedures between Asian ways of thinking and Western biomedical premises. The existential conditions of the participants include being a Korean woman and living as an immigrant. The metaphysical dialectics among the participants encompass the traditional Korean health and illness beliefs, and predetermined life. The three contextual domains of breast, breast cancer, and breast cancer screening illustrated the unique perspectives from the participants in reference to their body, illness, and the Western medical procedures. The grounded theory illustrates how a group of KIW interacts with Western medical procedures that evoke multiple dissonances with their cultural reasoning. The voices of the participants shed lights on what health care professionals should incorporate in approaching KIW associated with their practices of breast cancer screening. Funding Sources: Sigma Theta Tau Xi chapter research grant

Poster Abstract
QUALITY OF LIFE AND WELL-BEING OF CAREGIVERS OF LEUKEMIA PATIENTS: A PRELIMINARY REPORT. Geline Tamayo MSN, RN, CS, The University of Texas M.D. Anderson Cancer Center , Houston TX US Marlene Cohen RN, PhD, FAAN, The University of Texas Health Science Center at Houston, School of Nursing, Houston TX US; Anita Broxson RN, MSN, The University of Texas M.D. Anderson Cancer Center, Houston TX US; Sotea Bauri BSN, RN, The University of Texas M.D. Anderson Cancer Center, Houston TX US; Veronica Smith MSN, RN, FNP, The University of Texas M.D. Anderson Cancer Center, Houston TX US; Mark Munsell MS, The University of Texas M.D. Anderson Cancer Center, Houston TX US

Primary caregivers must manage increasingly difficult care because cancer patients are increasingly receiving complex and aggressive treatments as outpatients. Most literature about caregivers of cancer patients addresses issues in hospice or palliative care settings. In contrast, this study focuses on caregivers of cancer patients who are receiving outpatient chemotherapy. This project will describe and identify strategies to promote quality of life (QOL) and well-being of these caregivers.

The investigators developed a conceptual framework based on the literature to guide examination of caregivers’ needs.

Data were collected with the Caregiver Well-Being Scale and the Caregiver Quality of Life Scale-Cancer, both reliable and valid tools used with cancer patient caregivers in other settings, and a learning needs questionnaire developed by the research team. Participants are primary caregivers identified by patients with leukemia receiving outpatient chemotherapy at a large comprehensive cancer center in the southwestern United States. Ninety-eight participants have completed the survey to date, and data collection will continue until 200 caregivers provide data.

Descriptive statistics summarized the preliminary data. Statistical modeling and testing will be done when data collection is completed.

Preliminary analysis indicates that, on the well-being scale, the least-met needs were related to attendance to personal physical needs and expression of personal feelings and the best-met needs were related to security and self-esteem. The least-met need of activities of daily living was time for leisure activities. Primary concerns so far revealed in the quality of life scale are that patient care is a burden and disrupts caregivers’ lives. According to the learning needs questionnaire, 72% of caregivers want more information about giving medications and 80% want more information about side effects and symptom management. The findings thus far indicate a need for more communication between professional staff members and caregivers. The QOL and well-being scales also indicate a need to assist caregivers in promoting their own physical and emotional health. Many caregivers need to identify ways to meet their own needs in order to remain able to assist family members. Future research may need to focus on teaching interventions, particularly on caregiver management of patients’ symptoms and side effects.

Poster Abstract
PHILOSOPHICAL APPROACHES TO THE STUDY OF SPIRITUALITY AMONG FEMALE MEMBERS OF FAMILIES WITH HBOC. SUSAN TINLEY R.N., M.S., C.G.C., CREIGHTON UNIVERSITY, OMAHA NE US

The purpose of this presentation is to identify the most appropriate philosophical approach to the study of spirituality among female members of famlies with hereditary breast/ovarian cancer.

There is little in the literature on this topic, but anecdotally it is common to hear these women indicate that spirituality helps make sense of the mystery of their experiences with cancer risk, illness and deaths. Studies among other populations who are dealing with illness in themselves or family members indicate that spirituality has many different meanings and experiences making it an elusive concept for study. For some spirituality is intimately entwined with their religious convictions and practices; for others it involves their relationship with a supreme being or God outside of any faith based tradition, and for still others with no relationship with a higher being, spirituality relates to their relationships with themselves, with others and with the environment. This elusive and immaterial nature of spirituality presents challenges for the identification of the most appropriate philosophical approach for its study. In the past decade, nurses have approached the study of spirituality from multiple philosophical paradigms.

This presentation provides an examination of three such approaches in order to identify the most appropriate for the study of spirituality among women in HBOC families. The three philosophical approaches include empiricism, poststructuralism and Heideggerian phenomenology. Each philosophy has assumptions about the derivation of knowledge. Empiricism assumes that concepts have universal definitions and knowledge is derived from value free observations that can be quantified. Poststructuralism assumes that new realities can emerge from deconstructing cultural and historical discourses about the concept of interest. Heideggerian phenomenology assumes that knowledge is derived from participant experience and meaning interpreted within the social and historical context. A representative study of spirituality from the nursing literature for each approach is reviewed in terms of the assumptions, the object of inquiry, the outcomes, and the advantages and disadvantages of the philosophical approach. Each philosophical approach serves a legitimate purpose in expanding nursing knowledge about spirituality.

When studying a concept with diverse personal and cultural meanings, such as spirituality, among a population in which it has received little research attention, such as female members of families with HBOC, Heideggerian phenomenology provides the most fitting philosophical perspective.

Poster Abstract
TREATMENT-RELATED SYMPTOMS AND FUNCTIONAL STATUS IN WOMEN WITH ADVANCED OVARIAN CANCER. Lorraine Tulman DNSc, RN, FAAN, University of Pennsylvania School of Nursing, Philadelphia PA US; Deborah McGuire PhD, RN, FAAN, University of Maryland, Baltimore MD US; Karen Oleszewski MS, RN, AOCN, Franklin Square Hospital Center, Baltimore MD US; Jean Flack BSN, RN, OCN, Franklin Square Hospital Center, Baltimore MD US Bernadette Trujillo BSN, RN, OCN, Franklin Square Hospital Center, Baltimore MD US; Brooke Slater BSN, RN, University of Pennsylvania, Philadelphia PA US

Research on chemotherapy-related symptoms in the 25,000 women annually diagnosed with ovarian cancer has focused on the description, incidence, and management of these symptoms but has failed to examine their effects on functional status, including performance of household, social, community, personal care, and occupational activities. This lack of knowledge prevents clinicians from proactively counseling women on barriers to maintaining functional status or hazards that may be encountered during their usual activities (for example, diminished sensation and motor function due to peripheral neuropathy).

The aim of this cross-sectional pilot study was to examine the relationship of treatment-related symptoms to functional status following initiation of first-line chemotherapy for Stages III and IV ovarian cancer, the stages in which 70% of women are diagnosed.

The study was based on the Theory of Unpleasant Symptoms.

Eighteen women at two clinical data collection sites who were about to begin or who were 4, 8, or 12 months post initiation of first-line chemotherapy were assessed with the following valid, reliable instruments: the Memorial Symptom Assessment Scale, the Peripheral Neuropathy Scale, the Inventory of Functional Status Scale, the GOG Performance Scale. Data collection occurred through personal interview and chart review. Women just beginning chemotherapy had no chemotherapy-related symptoms, but had physical symptoms related to surgery and psychological symptoms related to their cancer diagnosis. Most women at 4 months (approximately halfway through chemotherapy) had begun to experience the detrimental effects of fatigue, peripheral neuropathy, and insomnia on functional status. Women at 8 monts reported fewer symptoms and improvement in functioning but peripheral neuropathy symptomatology continued to worsen. Women who were 12 months post initiation of chemotherapy exhibited about the same levels of symptomatology and functional status as the women at 8 months.

During and following first-line chemotherapy, a variety of significant physical and psychosocial symptoms appear to be associated with substantial impairment of functional status. Longitudinal research is needed to further determine the nature of these relationships in order to provide a foundation for developing and testing interventions to help women maintain their functional status as they battle this disease.

Poster Abstract
SLEEP-WAKE DISTURBANCES AND ASSOCIATED FACTORS IN PATIENTS WITH ADVANCED CANCER. Catherine Vena RN, MSN, PhD(c), Emory University, Atlanta GA US; Kathy Parker RN, PhD, FAAN, Emory University, Atlanta GA US

While, sleep-wake disturbances play a significant role in well-being and functioning, few studies have focused on sleep and related factors in late stage cancer populations. This information is important for oncology nurse’s practice and further research.

Sleep-wake disturbances, common complaints for oncology patients, have not been well described in all cancer populations. Therefore, we are examining subjective nocturnal sleep quality, daytime sleepiness, and related demographic, psychosocial, clinical, and treatment factors in persons with specific late-stage cancers.

According to the Two-Process Model of Sleep Regulation, both homeostatic and circadian processes interact to regulate normal sleeping and waking. People with cancer experience many clinical and treatment-related factors that interfere with these processes.

The sample included 80 medical oncology outpatients (48%female) with either stage IV colorectal (n=37) or either stage IIIB-IV non-small cell or extensive small cell lung cancer (n=43). Subjects completed the Pittsburgh Sleep Quality Index (PSQI) and the Epworth Sleepiness Scale (ESS). Demographic, clinical, psychosocial, and treatment factors were obtained via interview/record review, the Geriatric Depression Scale Short Form (GDSS), Karnofsky Performance Status (KPS), and Medical Outcomes Study Short Form 36-version 2 (SF36v2).

Group differences in sleep-wake variables were analyzed with ANOVA and Mann Whitney U tests. Bivariate correlations and multiple regression analyses between sleep variables and related factors were conducted. Sixty percent of the participants had moderate to severe nocturnal sleep disturbance and 43.8% had excessive daytime sleepiness (EDS) as evidenced by PSQI and ESS scores. EDS was associated with poor nocturnal sleep quality (r=.415;p=.015). Group differences in sleep quality and EDS were not statistically significant. There was no significant association between demographic variables and nocturnal sleep or EDS. There were significant associations between overall energy level, mental health, use of HT3 blockers and nocturnal sleep quality (Adj.R2=.364;p=.008). EDS was significantly associated with the number of comorbidities, the number of medications, physical function, and pain (Adj.R2=.193; p<.0001). These findings indicate clinical and treatment factors may play a role in the dysregulation of sleep-wake patterns. Furthermore, daytime sleepiness may significantly impact nocturnal sleep. Further research in this area should focus on development of interventions to improve nocturnal sleep quality and daytime alertness. Funding Sources: Virginia Henderson Clinical Research Grant, Sigma Theta Tau International; Emory University, Graduate School of Arts and Sciences Quadrangle Research Fund

Poster Abstract
IMPACT OF CANCER TREATMENT WITH PLATINUM-CONTAINING REGIMENS ON PERIPHERAL NERVE FUNCTIONING IN OLDER CANCER PATIENTS WITH DIABETES: AN EXPLORATORY STUDY. Constance Visovsky PhD, RN, ACNP, Case Western Reserve University, Cleveland OH US

Cancer remains a challenging problem for elders, as both cancer and comorbid conditions are common in this age group, and can result in increased morbidity from toxicities associated with cancer treatments. The interaction of comorbid conditions and cancer treatment remains largely unquantified. Knowledge concerning the effects of diabetes mellitus in conjunction with cancer treatment is very limited. Diabetics may be most at risk for the additive effects of neurotoxic agents, in addition to the effects induced by the chemotherapeutic agents alone. The purpose of this prospective, longitudinal study is to characterize peripheral nerve function in older diabetics undergoing cancer treatment for solid tumors with taxane or platinum-containing regimens. The specific aims are: 1) to determine the change in peripheral nerve function of diabetics occurring during and after treatment with neurotoxic taxane or platinum-containing chemotherapy regimens, and 2) to determine the extent to which age, pre-existing diabetes mellitus and level of glycemic control predict a differential pattern in outcome beyond the effect from the drugs; and 3) to develop a clinical assessment that would predict whether or not patients with diabetes mellitus are more or less susceptible to the neurotoxic effects of chemotherapy.

A physiologic framework of drug-induced neuronal degeneration was used.

A convenience sample of 50 participants with diabetes diagnosed with solid tumor cancer receiving treatment with taxane and/or platinum-containing regimens will be recruited. The outcome measures assessed are: sensation, vibration, gait, balance, muscle strength, sensory and motor nerve conduction and orthostatic blood pressure levels. Data will be collected at baseline, and every 3 weeks of treatment (4 cycles), and two months after treatment completion. Moderating variables are diabetes type (I or II) and level of glycemic control.

Descriptive statistics will be used to describe the peripheral nerve function. Data will be analyzed using plots and regressions slopes to determine change scores in the dependent variables over time. ANCOVA will be used to determine predictors controlling for cumulative drug dose.

Consistent, comprehensive measurement of progressive neuropathy in older diabetics will generate data useful in designing nursing interventions to reduce the impact of peripheral neuropathy on safety, activities of daily living, and quality of life. Knowledge about the effects of comorbidities and cancer treatment can assist in optimum treatment for debilitating neuropathies. Funding Sources: National Cancer Institute

Poster Abstract
WHAT CONSTITUTES A DIGNIFIED DEATH? THE VOICE OF ONCOLOGY ADVANCED PRACTICE NURSES. Deborah Volker RN, PhD, AOCN, The University of Texas at Austin School of Nursing, Austin TX US

Collectively, the lay public and health care professionals share an interest in improving care of the dying and promoting dignified dying. Oncology nurses play a pivotal role in providing quality care at the end of life, and helping patients and their families to achieve their preferences for control over this experience.

ONS has identified implementation of end-of-life care practices in the clinical setting as a research priority, yet measurable indicators that are specific to dignified dying have not been well explored. Although the Institute of Medicine (IOM) posed a definition of a good death, the definition was based on personal, professional, and philosophical perspectives. Hence, the issue of what constitutes a good or dignified death remains in question from an empirical viewpoint. The purpose of this study was to explore the concept of dignified dying from the perspective of oncology advanced practice nurses (APNs). APNs were targeted because of their pivotal role in influencing both patient outcomes and generalist nursing practice.

Lewis’ conceptual typology of control is the study's philosophical framework. The study rationale is based on the need for better conceptualization of outcomes meaningful to dying people and their families (IOM, 1997). Thus, an understanding of what constitutes a dignified death is warranted.

A naturalistic, hermeneutic approach was used to interview the study participants. A statewide sample of 19 oncology APNs was obtained by combining two data sets from studies of patient control and end-of-life care. To promote trustworthiness of study findings, research team members with expertise in qualitative methods and end-of-life care will review transcripts, field notes, and analytic decisions.

Audiotaped interviews are transcribed verbatim and are being analyzed using Cohen, Kahn, and Steeves’ approach to phenomenological analysis.

Preliminary analysis suggests that dignified dying is an experience in which the patient’s existential and physical comfort needs are met, and the patient’s personal choices are respected. The poster will present final results. Findings can be used to better understand the oncology APN perspective of patient experiences at the end of life, inform academic programs that prepare oncology APNs, and improve clinical care for people who are dying. Funding Sources: Funded by the ONS Foundation through an unrestricted grant from Genentech, Inc.; and the National Institute of Nursing Research (1 R15 NR007986-01)

Poster Abstract
QUALITY OF LIFE AND SYMPTOM EXPERIENCE OF SPOUSES OF MEN WITH PROSTATE CANCER. Mary Vonderheide RN, MSN, OCN, Sharp Grossmont Hospital, La Mesa CA US; Susan Rawl PhD, RN, Indiana University, Indianapolis IN US; R. Brian Giesler PhD, IN US; Victoria L. Champion DNS, RN, FAAN, Indiana University, Indianapolis IN US; Debra Burns PhD, IN US; Joanne K. Daggy MS, IN US Sally Weinrich PhD, RN, IN US; Barbara Given PhD, RN, FAAN, MI US; Charles Given PhD, MI US

The impact of a prostate cancer diagnosis and its subsequent treatment are not limited to the male patient; they have the potential to affect the quality of life (QOL) for both the cancer patient and his spouse. The needs of the spouses are often overlooked while healthcare professionals focus primarily on the needs of the cancer patient.

To explore relationships between prostate cancer treatment, post-treatment symptoms, and spouses' global and disease-specific QOL.

The conceptual framework that guided this study proposes that QOL can be conceptualized as encompassing three distinct domains (global – overall well-being, general – SF-36 & CES-D, and disease-specific – fatigue, nausea) that are hierarchical and theoretically related.

Cross-sectional analyses of baseline data from a randomized trial testing a physcho-educational intervention to improve QOL for prostate cancer patients and their spouses. Spouses of patients who recently completed treatment for prostate cancer were recruited from five cancer centers located in the Midwest. The majority of spouses were Caucasian, married, and living with the patient. Data were collected from patients and spouses separately within six weeks of the completion of treatment by trained interviewers using a computer-assisted telephone interviewing system.

Analysis of variance were used to examine relationships between spouses' disease-specific and general QOL with type of prostate cancer treatment received by their partners. Multiple regression was used to examine the influence of disease-specific QOL indicators on global QOL indicators within each treatment type.

Differences in spousal disease-specific and global QOL scores were observed across treatment types. After controlling for age, spouses of men who had surgery had lower urinary and sexual function scores than spouses of men who received radiation. Spouses of radiation patients had lower mental health scores than either spouses of men who had surgery or brachytherapy. Urinary, sexual, and bowel QOL indicators were significant predictors of both CES-D and SF-36 Mental Health Index Scores. The results of this study give support to the significant impact which men's prostate cancer treatment symptoms have on their spouses' QOL. Spouses need to be supported and educated about prostate cancer, its treatment, and the symptoms that their husbands are experiencing.

Poster Abstract
BREAST CANCER TREATMENTS AND BMD. Nancy Waltman PhD, APRN, University of Nebraska College of Nursing, Lincoln NE US; Carol Ott RN, PhD, OCN, University of Nebraska College of Nursing, Kearney NE US; Gloria Gross RN, PhD, University of Nebraska College of Nursing, Scottsbluff NE US; Janice Twiss PhD, APRN, University of Nebraska College of Nursing, Omaha NE US; Ada Lindsey RN, PhD, FAAN, Dean Emeritus, University of Nebraska College of Nursing, Omaha NE US; Timothy Moore MD, University of Nebraska Medical Center, Omaha NE US

Thirty percent of postmenopausal women in the U.S. have osteoporosis (decreased bone mineral density [BMD] that leads to fractures). Postmenopausal breast cancer survivors (BCS) are at greater risk for osteoporosis than other women.

This presentation examines relationships between BMD in women and treatments for breast cancer, including cytotoxic chemotherapy, aromatase inhibitors, and tamoxifen.

Treatments for breast cancer such as cytotoxic chemotherapy and aromatase inhibitors may accelerate bone loss in postmenopausal women, most likely by further depleting estrogen levels. Women with estrogen receptor-positive cancer are often treated with the hormone tamoxifen, and while tamoxifen is an estrogen antagonist in breast tissue, the effects of tamoxifen on BMD have been inconsistent.

This is a secondary analysis of a larger intervention study on prevention of osteoporosis in BCS. Baseline data for the first 100 women are described in the abstract, and results for all 240 women will be presented at the conference.

In this study, BMD was measured using DEXA, and women included in the study had a DEXA T score of -1.0 or lower. Data was analyzed using descriptive statistics, correlations, and t-tests.

Of the 100 women, 16 had osteoporosis and 84 had osteopenia. Average age was 57 years, time since menopause, 11.4 years, and women had been diagnosed with stage 0, I, or II breast cancer an average of 6 years. The 66 women who had received cytotoxic chemotherapy had lower BMD of the spine (p = .05) compared to women who had not received chemotherapy. There was a trend for the 9 women who had received aromatase inhibitors to have lower BMD of the hip (p = .09) compared to women who had not received aromatase inhibitors. Fifty of the 100 women had been prescribed tamoxifen. Compared to women who had not been prescribed tamoxifen, these women had higher BMD of the hip (p = .02), and there was a trend for higher BMD of the spine (p = .10). Implications are that risk for osteoporosis in BCS may be related to the treatment they received for their cancer. Funding Sources: R01 from NINR/ NIH

Poster Abstract
DYNAMICS OF INFORMAL BMT CAREGIVING. Loretta Williams RN, MSN, OCN, AOCN, UT-Houston School of Nursing and M. D. Anderson Cancer Center, Houston TX US

Blood and marrow transplant (BMT) is a potentially curative therapy for patients with life-threatening illnesses. Over the last 10 years, family caregiving has become an essential aspect of BMT to support the patient emotionally and to provide assistance with physical care and symptom management.

An important goal for nurses is to assist family caregivers to improve outcomes for BMT patients and caregivers. Achieving this goal requires understanding of the dynamics of caregiving. The purpose of this study is to refine a conceptual Informal Caregiving Dynamics model by describing caregiver commitment, expectations, and role negotiation and identifying other important caregiver energy sources not currently included in the model. The refined and validated model will give guidance in the development of supportive caregiver interventions.

A conceptual Informal Caregiving Dynamics model derived from the literature identifies commitment, expectations, and role negotiation as moving the caregiving relationship along an illness trajectory.

This is a qualitative, cross-sectional study. The study sample is projected to be 40 informal caregivers of patients undergoing BMT at a comprehensive cancer center in the south central United States. The caregivers’ are describing their experiences approximately one month after the BMT in audiotaped dialogues, focusing on their present experience of caregiving with attention to past experiences and future hopes that affect the present experience. Analysis and identification of categories and themes by the researcher will be reviewed and confirmed by other researchers experienced in qualitative analysis, oncology nursing, informal caregiving, and BMT.

An exploratory descriptive method is being used to identify categories and themes from the transcribed dialogues. Descriptive statistics are being used to describe the sociodemographic characteristics of the caregivers and the sociodemographic and disease/treatment characteristics of the patients.

Evidence of commitment, expectations, and role negotiation has been found after initial analysis of the first five interviews. Commitment is the most consistently present category and includes recurring themes of becoming one with the patient and caregiving never ending. Analysis of dialogues will enable refinement of the Informal Caregiving Dynamics model to guide care of caregivers and ongoing research to test supportive interventions. Funding Sources: American Cancer Society Grant DSCN-02-211-01-SCN

Poster Abstract
THE ROLE OF INFLAMMATORY CYTOKINES IN CANCER-TREATMENT-RELATED FATIGUE. Lisa Wood PhD RN, Oregon Health & Science University, Portland OR US; Lillian Nail PhD RN FAAN, Oregon Health & Science University, Portland OR US; April Fischer BSN, Oregon Health & Science University, Portland OR US; Collin Elsea BS, Oregon Health & Science University, Portland OR US

Cancer patients undergoing treatment often experience a debilitating fatigue, that has a profound impact on quality of life and on the ability of patients to adhere to treatment schedules. Thus, the management of fatigue is an essential component of cancer management.

Increased levels of cytokines appear to play a role in cancer treatment related fatigue (CTRF) which has lead to the hypothesis that CTRF may be homologous to the fatigue associated with "sickness behavior," a response to infection in higher vertebrates caused by increased levels of inflammatory cytokines in the peripheral circulation and the brain. To date, the molecular mechanisms underlying the initiation and perpetuation of CTRF are not established, and thus, the precise role of cytokine expression, if any, in CTRF remains unclear. We hypothesize that cancer, cancer plus chemotherapy, and chemotherapy drugs alone can induce inflammatory cytokine expression leading to fatigue.

We have initiated a study to evaluate the relationships among cancer and its treatment, fatigue, and inflammatory cytokine levels using an innovative approach that takes advantage of an established mouse model of sickness behavior. In this model, voluntary wheel running activity (VWRA) is used as a proxy for fatigue.

Mice bearing Lewis Lung Carcinoma will either be given cancer chemotherapy or left untreated. The effect of cancer and cancer plus chemotherapy on fatigue will be determined by assessing decline in VWRA from baseline.

Using a repeated measure ANCOVA we will determine the effect of cancer or cancer and chemotherapy on daily VWRA. To determine the relationship between fatigue and cytokine deregulation in mice receiving cancer and chemotherapy, we will 1) assess for changes in inflammatory cytokine gene expression in the brains of fatigued mice and 2) assess for changes in the levels of inflammatory cytokines in the peripheral circulation of fatigued mice.

As an outcome of these studies, we expect to determine whether cytokines play a role in CTFR and, if so, the mechanisms by which cytokine levels are increased in cancer treatment. This research is significant because it holds potential for the development of nursing interventions to reduce CTRF and improve the QOL of cancer patients. Funding Sources: Pilot study funding (10,000) to determine the role of cancer chemotherapy in fatigue in mice (9/1/03 - 8/31/04). Awarded from Center Grant: NINR 1P20NR07807-03 (PI: Nail, Lillian)

Poster Abstract
ASSESSING CONSTIPATION IN THE PEDIATRIC ONCOLOGY POPULATION: A PILOT STUDY. Myra Woolery-Antill MN, RN, National Institutes of Health, Bethesda MD US; Ellen Carroll BSN, RN, National Institutes of Health, Bethesda MD US; Gwenyth Wallen PhD, RN, National Institutes of Health, Bethesda MD US Paul Jarosinski Pharm D, National Institutes of Health, Bethesda MD US; Barbara Corey MSN, RN, National Institutes of Heatlh, Bethesda MD US; Holly Wieland MPH, RN, Food and Drug Administration, Rockville MD US; Ramzi Dagher MD, Food and Drug Administraiton, Rockville MD US

Constipation is prevalent in oncology patients and children are particularly susceptible. Sequelae of constipation include anorexia, nausea, vomiting, abdominal pain, ER visits, and a decrease in quality of life. However cancer treatment plans often overlook the presence of constipation and reflect a lack of consensus for effective assessment and management. Treatment with vinca alkaloids and/or narcotics and lifestyle changes secondary to the disease process can have a negative impact on the child’s bowel elimination status. There is no consistent definition for constipation. There are no reliable and valid instruments to measure the presence and severity of constipation in children, and only one validated adult measure, the Constipation Assessment Scale (CAS).

This study defined constipation from the medical position statement made by the North American Society of Pediatric Gastroenterology and Nutrition.

Sample: A pilot study (N=34) evaluating the presence and severity of constipation, and the reliability and validity of a modified CAS in children with cancer is being conducted. Patients 2 times per day for pain are included in the study.

Methods: Initial and subsequent assessments for bowel function are conducted for 7 days after the last dose of vinca alkaloid and/or narcotic for a maximum of 6 weeks. A standardized nursing and nutrition assessment, a review of the history and physical, and a baseline CAS repeated at one-hour to assess test-retest reliability are obtained. Subsequent assessments include CAS administered 3 times per week and daily bowel diaries completed by the patient and/or their family.

T-tests will determine if the mean CAS scores in non-constipated group differs from the constipated group. The Wilcoxon rank sum test will be used if data are not normally distributed. Cronbach’s alpha will be utilized to assess internal consistency of the CAS instrument.

Seventeen patients have enrolled in this study, with a mean age 15.7 years (range 9 to 21 years). Completing the bowel diaries has been difficult for adolescents, whereas children are more compliant possibly because their parents are more involved.

Poster Abstract
COMPARISON OF EMOTIONAL/BEHAVIORAL PROBLEMS IN ADOLESCENT WITH CANCER AND EPILEPSY. Chao Hsing Yeh Associate Professor, Chang Gung University, Tao Yuen TW

To understand and compare the emotional/behavioral problems in adolescents with cancer and epilepsy

Pediatric patients with chronic illness (e.g., cancer and epilepsy) endure prolonged and endless treatment protocols, severe side effects caused by treatment, and uncertainty regarding progression of their disease. Few studies have investigated adolescents’ behavioral problems that begin after the diagnosis of a chronic illness has been confirmed. To date, there is no official documentation regarding the epidemiological prevalence of mental health conditions of adolescents with chronic illness (e.g., cancer and epilepsy) in Taiwan. This study was aimed to compare the emotional/behavioral problems of adolescents with different chronic illnesses (for example, cancer and epilepsy.

The child behavior checklist/4-18 (CBCL/4-18), reported by their parents was used. In this current study, eight narrow-band syndromes of CBCL/4-18 (Withdrawn, Somatic complaints, Anxious/Depressed, Social problems, Thought problems, Attention problems, Delinquent behavior, and Aggressive behavior) was used to compare the emotional/behavioral problems. The subjects of this descriptive and cross sectional study include parents with adolescents with cancer (n=63) and epilepsy (n=113), who were convenience selected from the outpatients clinics in a children’s hospital.

Descriptive analysis is used to describe the demographic information of the subjects and patients. Student t test with Bonferroni correction is used to compare the raw scores of syndrome problems.

The findings shown that adolescents with epilepsy reported higher problems in Withdrawn, Anxious, Social, Thought, Attention, Delinquent and Aggressive than adolescents with cancer. Among those higher scores of syndromes, Anxious and Attention in epilepsy group are statistically significant higher than adolescents in cancer group. Adolescents in cancer group report higher score in Somatic problems, but this difference does not reach statistically significances. The finding suggested that adolescents with different chronic illness bear different level of emotional/behavioral problems. It is necessary to understand the emotional/behavioral problems of children in Taiwan who have cancer so that healthcare professionals can provide better mental health services for these children.

Poster Abstract
SATISFACTION WITH COMMUNICATION AND QUALITY OF LIFE FOR ADVANCED CANCER PATIENTS. Kim Wagler Ziner RN, MSN, Indiana University School of Nursing, Indianapolis IN US; Victoria L. Champion DNS, RN, FAAN, Indiana University School of Nursing, Indianapolis IN US; Barbara Given Ph. D., RN, FAAN, Michigan State University, East Lansing MI US; Charles W. Given Ph.D., Michigan State University, East Lansing MI US

The National Cancer Institute SEER cancer statistics review 1975-2000, estimates 556,500 deaths in the USA due to cancer in the year 2004. Health professionals may play a pivotal role in helping patients to adjust to recurrent disease. Patients look to their healthcare providers, sometimes unconsciously; for social and decisional support, interpersonal skills, and information regarding how to manage treatment which can influence their subsequent Quality of Life (QOL) outcomes.

The purpose of this paper is to examine the relationship of patient satisfaction with the communication of their health care provider and subsequent Quality of Life outcomes of depression, anxiety, general health, mental health, social functioning and vitality.

Haase’s theoretical framework of Resilience was used to support the relationship of provider support and subsequent QOL outcomes.

Instruments: Reliable and valid instruments were used for all variables. Depression was measured by CESD. Anxiety was measured by STAI. The SF-36 was used to measure; general health, mental health, social functioning and vitality. Satisfaction with communication was measured with Satisfaction with Care (Loblaw, Bezjak & Bunston, 1999). Sample: Data were obtained from a larger intervention trial that included a convenience sample of 124 patients with advanced cancer shortly after the recurrence or progression of their disease was diagnosed. Data from interviews at baseline and 20 weeks following the intervention are used to answer the research question.

First, descriptive statistics with one-way ANOVA were run to determine if type of cancer influenced satisfaction with communication. There was no significant difference in satisfaction with communication related to cancer type (F (8,101) = 1.796; p= .086). Also, there was no significant difference in satisfaction with communication scores related to gender (t (.05, 110) = -1.944; p = .054). Pearson Correlations were used to test the relationship of satisfaction with communication and Quality of Life (QOL) outcomes at baseline and 20 weeks. Baseline QOL outcomes were not correlated with satisfaction with communication but at 20 weeks, satisfaction with communication was significantly correlated with depression(r= -.287, p=.02), anxiety(r= .286, p=.02), general health(r=.386, p=.001), mental health(r=.286, p=.019), social function( r=.369, p=.002) and vitality(r=.417, p<.001).

Results indicate that cancer survivor’s QOL outcomes may be influenced by their satisfaction with provider communication. Funding Sources: Walther Cancer Institute