Does the manner in which information about prostate-specific antigen testing is presented affect screening rates? - PSA

Journal of Family Practice, Oct, 2001 by Andrew R. Lockman

Frosch DL, Kaplan RM, Felitti V. Evaluation of two methods to facilitate shared decision making for men considering the prostate-specific antigen test. J Gen Intern Med 2001; 16:391-98.

* BACKGROUND Prostate cancer is the second leading cause of death due to cancer in men, but substantial controversy surrounds the role of PSA in screening asymptomatic patients. Although the test can help detect prostate cancer earlier, we currently lack evidence that this early detection will increase length or quality of life. The issues involved in the controversy include frequent false-positive results, potential for complications from treatment, and the frequency of slow-growing tumors that may never become clinically significant. In recognition of this complexity, the American Cancer Society and the American Urological Association revised their recommendations for men older than 50 years, calling for shared decision making in which the patient and physician discuss options and together make a decision that agrees with the patient's individual health preferences. Of note, most other major organizations do not advocate routine PSA tests for screening. Previous research has shown that the way information about PSA testing is presented can influence the outcome of shared decision making. This study compared the effect of either a scripted discussion or videotape on men's opinions about taking the PSA test.

* POPULATION STUDIED The investigators approached all men older than 50 years who were presenting for an annual preventive care evaluation at a large health maintenance organization (HMO). The participants (n=176) were sequentially assigned to 1 of 4 interventions: usual care, a discussion about risks and benefits of PSA, a shared decision-making video, or the video plus discussion. Baseline characteristics including age, education, marital status, ethnicity (more than 70% white), history of friend or family member with prostate cancer, and previous PSA testing did not differ among groups. Approximately 40% of those contacted to participate in 1 of the 3 interventions refused, but recruitment rates did not differ among the intervention groups.

* STUDY DESIGN AND VALIDITY The investigators used a nonrandomized unblinded 2x2 factorial comparison of the discussion and video formats that yielded the 4 groups described above. A previous study evaluated and described the 25-minute video, and the lecture-format discussion closely followed the content of the video.

The authors chose not to randomize the subjects because of logistic concerns. Sequential accrual allowed the interventions to be presented in small groups, perhaps minimizing potential differences in conditions, and allowed the study to be integrated more easily into a busy practice setting. However, the investigators did approach all consecutive patients, and the intervention groups all had similar participation rates. The usual caveats concerning the generalizability of results from a predominantly white HMO population apply. Also, the 40% who refused to participate in an intervention may have left the intervention groups populated with those who would be more open to forgoing the PSA test.

* OUTCOMES MEASURED Measured responses in all groups included whether they wanted PSA testing, their level of confidence in their decision, and levels of knowledge and concern about prostate cancer. Those in the intervention groups also rated the amount, clarity, and perceived balance and fairness of the presentations. The study did not measure the actual incidence of subsequent PSA testing.

* RESULTS Almost all (97%) in the usual care group opted for testing. Discussion decreased the testing rate to 82% (P [is less than] .05), and members of the video (63%) and video/discussion (500,4) groups chose testing even less frequently (P [is less than] .05 for difference between the discussion-only and either video group). All of the interventions (other than usual care) significantly increased knowledge about prostate cancer (3.4-3.9 correct responses to 5 questions vs 1.6, P [is less than] .001) and decreased confidence in the decision regarding PSA. Subjects in the usual care group expressed more concern about prostate cancer than those in the intervention groups. Less than 1% of subjects felt negative about participating in the interventions. Eighty-two percent considered the presentation balanced; 8% felt it was slanted in favor of screening; and 11% felt it was slanted against having PSA testing.

RECOMMENDATIONS FOR CLINICAL PRACTICE

Incorporating a detailed, balanced presentation on PSA testing, using either discussion or a videotape, into a health maintenance visit can give men better knowledge about prostate cancer and result in lower rates of PSA screening, and the method of presenting the information may further affect the patient's decision. Forty to 50% of men in this study decided against PSA testing after watching an informational video.

Andrew R. Lockman, MD
University of Virginia
Charlottesville
E-mail: alockman@virginia.edu