Results
There were 3592 invitation letters mailed to potential subjects, of which 572 men responded. Of the respondents, 257 (44.9%) were enrolled in the study Figure 2. Reasons for exclusion were history of previous cancer (50), history of prostate or genitourinary disease (102), poor mental status (12), and being an active employee at the medical center (23). Reasons for not participating among eligible patients included: not interested in participating (52), no phone (13), distance or transportation problems (7), the patient felt that he was too ill (26), and miscellaneous reasons (30). Experimental and comparison groups were similar in age, racial distribution, comorbidity, and education Table 1.
Prostate Cancer Screening Knowledge
The knowledge questionnaire listed 18 items. The range of total correct responses on postintervention scores was 5 to 18. There was no difference at baseline in total knowledge scores between the experimental (mean=11.7, standard deviation [SD] =2.4) and comparison (mean=11.4, SD=2.4) groups (P=.32). On postintervention assessments, the experimental group had a higher total knowledge score (mean=15.0, SD=2.3) than the comparison group (mean=14.1, SD=2.7; P <.01). On the postintervention survey, the experimental group was more likely than the comparison group to be aware of the possibility of false-negative and false-positive screening test results and had better knowledge of the natural history of prostate cancer Table 2. When asked to identify the risk of a false-negative test result, the experimental group was more likely than the comparison group (70% vs 49%, P <.05), to correctly identify 1/100 as the frequency of false negative results.
Prostate Cancer Screening Beliefs
Beliefs regarding the performance of prostate cancer screening tests differed between the groups. Specifically, fewer men in the experimental group than in the comparison group believed that screening tests were infallible Table 3. At baseline 79% of the subjects felt that “most men can be cured” if prostate cancer is caught in the early stages and treatment is received. Fifty-six percent of the subjects believed that of those men who have prostate cancer, most died of something else; 35% believed that approximately half die of prostate cancer; and only 9% believed that most men die of their prostate cancer. After the intervention, subjects in the experimental group were more likely than those in the comparison group (67% vs 46%) to respond that most men with prostate cancer die of something else (P <.01).
At baseline, 84% and 87% of the total study cohort stated that they were very likely to have a PSA and DRE, respectively. Ninety-eight percent of the subjects stated that they would have screening for prostate cancer if their physician recommended it. Finally, at baseline 77% of the subjects felt well informed enough to make a decision about prostate cancer screening. Perceptions of being well informed increased to 93% after the intervention but with no difference between groups.
Prostate Cancer Screening Decisions
Eighty-two percent of the experimental group, compared with 84% of the comparison group underwent prostate cancer screening (P=.60). Subjects who chose not to be screened did not differ from screened subjects in age, race, comorbidity level, education, or postintervention prostate cancer screening knowledge. Of the 214 subjects who chose to be screened, 32 had abnormal test results: 15 subjects had a PSA greater than 4.0, and 18 subjects had an abnormal DRE (one subject had both an abnormal DRE and a high PSA). Of the 32 abnormal exams, 21 had a prostate biopsy, and 7 prostate cancers were diagnosed. Of the 11 subjects with a positive screen who did not proceed to biopsy, 1 subject with an elevated PSA deferred a prostate biopsy and subsequently developed metastatic colon cancer. A second subject with elevated PSA refused TRUS and biopsy and continues to be followed up clinically. Of the remaining 9 patients who did not have further testing, one subject refused biopsy and elected to be followed clinically. Eight subjects were evaluated by urology tests, and the recommendation was for clinical follow-up without TRUS or rectal biopsy.
Discussion
We report that a prostate cancer screening aid consisting of an illustrated pamphlet was effective in improving knowledge and changing beliefs about prostate cancer screening when tested in a randomized controlled trial. The visual display of quantitative information improved knowledge about screening outcomes, but this knowledge alone did not change prostate cancer screening test use.
Prostate cancer screening is a clinical decision for which the risks are difficult to balance, a type of decision referred to as a “tossup” dilemma.21,22 The Health Belief Model posits that a change in perceived risks and benefits of screening may affect the likelihood of the patient’s taking preventive action (undergoing prostate cancer screening).16,17 Decision-aids have improved knowledge regarding decision outcomes, reduced decision conflict, and encouraged patients to be more active in the decision-making process.11,13,23 A recent meta-analysis24 shows that although decision-aids have a consistent effect on improving knowledge, they are less likely to alter decisions about a health care intervention. Previous studies of prostate cancer screening decision-aids have provided conflicting results. In one clinical trial, 12% of a primary care practice group exposed to a shared decision-making videotape intervention had a PSA test at their next scheduled clinic visit, compared with 23% of a control group (P=.04). However, a different arm of the same study found no effect of the intervention on the high rates of prostate cancer screening tests in a free PSA screening clinic. In a second clinical trial, men exposed to a scripted informational intervention were significantly less interested in PSA screening than those in a control group,12,25 but the subsequent use of screening tests was not evaluated. A third clinical trial in Canadian men found that a prostate cancer screening informational intervention in a discussion format increased participation in the decision-making process and decreased decisional conflict but did not alter the subsequent use of prostate cancer screening tests.13