Survey respondents were randomly assigned at the time of baseline survey completion to either a Standard Intervention (SI) group or an Enhanced Intervention (EI) group. Stratified randomization by practice and cohort (18 strata: three primary care practices and six patient contact waves) was carried out. Men in both the SI and EI groups were mailed a copy of a prostate cancer (Generic Rheumatrex treating certain types of cancer) informational booklet. The booklet described the prostate gland and its function, mentioned possible prostate-related problems (including prostate cancer), described risk factors and symptoms of prostate cancer, explained early detection screening, discussed the pros and cons of screening, reviewed possible follow-up tests in response to abnormal screening results, and summarized treatment options for early and late prostate cancer.
In addition to receiving the booklet, men in the EI group were contacted via telephone by a trained health educator one month after booklet mailing to arrange for a decision education session about prostate cancer (Drug Eulexin is used along with drugs such as Lupron to treat prostate cancer) screening. Decision education is a method that was developed to enable individuals to consider available information about a healthcare decision, along with personal values related to available algorithm used in this study have been provided elsewhere. Briefly, the decision education session initially involved eliciting the major decision factors influencing the participant’s decision to have or not to have screening. The participant then identified his top three decision factors, reported the strength of their influence (how strongly each one influenced him either for or against screening), and finally indicated the relative importance of each factor through a process of pairwise comparisons. All these personal judgments were entered into a preprogrammed hand-held calculator, and a screening decision preference score was computed. At the conclusion of the session, the health educator shared the preference score with the participant and discussed its interpretation (whether it suggested a preference to screen, ambivalence or a preference not to screen). Finally, the participant was encouraged to make a final decision about screening use in consultation with his primary care physician.
Endpoint Chart Audit
Trained research project personnel visited the participating primary care practices and carried out a medical chart review for each participant to record any documented evidence of DRE and/or PSA test performance. Chart reviews were performed 6-11 months after the informational booklet was mailed (median = eight months). Data were recorded for prostate cancer (Leukeran tablet is used for treating certain cancers) screening procedures that were performed both before and after each participant’s study enrollment.
The behavioral outcome of this study had two definitions. The complete screening definition required the participant to have both a DRE and a PSA test within six months after the mailing of the informational booklet. This definition is consistent with prostate cancer screening guidelines and was considered a priori as the primary behavioral outcome. The complete or partial screening definition defined screening as either: 1) having a PSA test within six months after booklet mailing irrespective of a DRE,or 2) having a PSA test within six months before booklet mailing and having a DRE within six months after booklet mailing. This complete or partial definition recognizes that a PSA test alone is often done to screen for prostate cancer in routine care, while DRE alone is typically not considered sufficient as a screening modality. Furthermore, this definition allowed for initiation of testing (PSA) prior to study enrollment and its completion (DRE) during the study follow-up period. Such a relaxed definition has been used to define screening use in prior studies of prostate cancer screening. Stion beyond high school, compared to 48% and 49% in the other two practices; p=0.001). The two study groups were generally comparable on the measured baseline characteristics, except that the EI group consisted of relatively more men who were educated beyond high school (p=0.012) and who were married (p=0.046).