Congress Abstracts 2005

12

PREDICTIVE VARIABLES IN MEN’S TREATMENT CHOICES FOR LOCALIZED PROSTATE CANCER. Donna Berry, PhD, RN, AOCN^®, and William Ellis, MD, University of Washington, Seattle, WA; and John Blasko, MD, Seattle Prostate Institute, Seattle, WA.

Men diagnosed with localized prostate cancer (LPC) often have the opportunity to participate in the treatment choice. Nurses have important roles in assessment of personal factors that influence this decision.

The purpose of this study was to evaluate relationships between influential personal factors and medical factors on treatment choice and decision-related outcomes.

The Decision Support Framework (DSF) is organized by determinants of decisions, decision support interventions and evaluation process and outcomes of the decision support. This study represents exploration and analyses of decision determinants that are required prior to intervention development.

Patients in urology and radiation oncology practices in western Washington were invited to participate in this descriptive study by their clinicians during 2001–2003. Men completed questionnaires at home within 2 weeks of the informational clinic visit with the cancer clinician, but prior to treatment. This report presents data from 260 men diagnosed with LPC. Personal factors (influential people and outcomes), treatment choice (TC) and decision-related outcomes were queried along with demographic data.

Relationships between all variables and the outcomes, Satisfaction with Decision (SWD) and TC, were explored using exhaustive CHAID (Chi-squared Automatic Interaction Detector) in AnswerTree 3.0 (SPSS, Chicago, IL). CHAID enables exploration of relationships through successive partitioning of variables.

The typical respondent was a white, married man in his 60’s, educated and currently employed. The strongest predictor partition variable for SWD was the subscale “factors contributing to uncertainty” (adj. p < 0.0001) followed by the Trait Anxiety score (adj. p = 0.038). The strongest predictive partition for the actual TC was age group (adj. p < 0.0001), followed by interacting marital status (adj. p = 0.0003), influence of the urologist (adj. p = 0.0008) and use of the Internet (adj. p = 0.0479). Men with LPC were more satisfied with their TC when they reported fewer uncertainty factors; these are factors mainly relevant to information needed to understand the pros and cons and to make a decision. Consistent with this finding for TC, is the use of the Internet, though this factor interacted with age, the influence of their surgeon and marital status. This study suggests that personally meaningful information communicated by patients and to patients is paramount.