Screening With the Prostate-Specific Antigen Test-Texas, 1997

Prostate cancer is the second leadingcause of cancer-related deaths among men in Texas. From 1990 to 1997, theaverage annual number of prostate cancer-related deaths in Texas was 1900, andthe average annual death rate was 20.9 per 100,000 population. An estimated10,186 new prostate cancer cases will occur in Texas in 2000. Several screeningmethods are available for early detection of prostate cancer, including digitalrectal examination, transrectal ultrasound, and prostate-specific antigen (PSA)testing, which involves a simple phlebotomy.

To assess the proportion of men in Texas receiving PSA testingand to identify factors associated with receipt of this testing, the TexasDepartment of Health added three questions to its 1997 Behavioral Risk FactorSurveillance System (BRFSS) survey relating to PSA testing. This reportsummarizes this analysis and indicates that approximately 37% of men aged ³40 years had received PSA testing and that receipt of PSA testing was associatedwith a doctor’s recommendation.

The Behavioral Risk Factor Surveillance System

The BRFSS is a state-based, random-digit-dialed telephonesurvey of the civilian, noninstitutionalized US population aged ³18 years. In 1997, men aged ³ 40 years who respondedto the Texas BRFSS were asked, "Have you heard about the PSA bloodtest?", "Have you ever been told by a doctor that you should have aPSA blood test?", and "Have you ever had a PSA blood test?"Responses were weighted to provide statewide estimates; standard errors and 95%confidence intervals (CIs) were calculated, and univariate and multivariateanalyses were performed using survey data analysis (SUDAAN) software.

Among respondents, 60% (95% CI = 55%-65%) said they had heardof the PSA test. Of those who had heard of the test, 52% (95% CI = 45%-59%)were told by their doctor that they should receive the test. Of those who hadheard of the test and whose doctor recommended it, 91% (95% CI = 85%-96%)reported having received the test. Overall, 37% (95% CI = 32%-42%) of menreceived the PSA test, representing approximately 62% of men who had heard ofthe test. Of those who were not told by a doctor to have the test, 24% (95% CI =14%-30%) received the test.

Results of Analyses

Univariate analysis indicated that receiving the PSA test wasassociated with a doctor’s recommendation (odds ratio [OR]=28.2; 95% CI = 13.3-59.8)(Table 1). Other factors associated with receiving the test were having hada physical examination during the preceding 2 years (OR = 5.4; 95% CI = 2.6-11.0),being aged ³ 50 years (OR = 5.2; 95% CI = 5.1-5.2),being covered by a health plan (OR = 3.8; 95% CI = 2.0-7.1), having everreceived a proctoscopic examination (OR = 3.4; 95% CI = 2.0-5.7), beingnon-Hispanic (OR = 2.2; 95 CI = 2.2-2.3), and having ³16 years of education (OR = 1.9; 95% CI = 1.2-3.0).

Logistic regression analysis indicated that receiving a PSA testwas associated with a doctor’s recommendation (adjusted OR = 80.4; 95% CI=21.4-301.9), and being aged > 50 years (adjusted OR = 4.0; 95% CI = 1.3-12.3).

Screening Recommendations

Prostate cancer screening recommendations differ among nationalorganizations. The American Cancer Society recommends that men aged ³50 years who receive an annual examination be offered the digital rectalexamination and the PSA test. The American Cancer Society also recommends thatmen aged ³ 40 years be informed about the risk forprostate cancer.

In comparison, because no evidence exists that early detectionand treatment influences the overall death rate from this disease and about halfof the men who undergo surgical treatment of localized lesions experience sideeffects (eg, incontinence and impotence), the US Preventive Services Task Force,the American College of Physicians, the American Society of Internal Medicine,the National Cancer Institute, the American Association of Family Practitioners,and the American College of Preventive Medicine do not advocate routinescreening. Despite the conflicting recommendations, PSA testing has increasedrapidly among asymptomatic men in the United States.

Factors Qualifying This Report

The findings in this report are subject to at least fourlimitations. First, BRFSS questions did not distinguish prostate cancerscreening from diagnostic testing. Some respondents may have received PSAtesting as part of a diagnostic evaluation for symptoms or to monitor treatmentfor existing prostate cancer. Second, respondents may have had the PSA testperformed, but did not know that it had been done. Third, among men for whom PSAtesting was not recommended, it is not possible to distinguish men whosephysicians discouraged screening from those whose physicians did not mentionscreening. Finally, because the survey did not ask men how they heard about thetest, the proportion of men hearing about the test from sources other than theirdoctor is not known.

The findings of this report suggest that interest in prostatecancer and awareness about the available screening tests for this disease issubstantial. These data are consistent with information from other studies thatindicate a substantial proportion of men aged > 40 years have received PSAtesting. Because PSA testing potentially can have an impact on statistics aboutprostate cancer incidence and outcomes, ongoing surveillance of the trends anddeterminants of the use of this procedure are warranted. As a result, theCenters for Disease Control (CDC) is incorporating questions about PSA testingin the 2001 BRFSS for every state. This effort, and continuing surveillance instates such as Texas, will provide information on the use of prostate cancertesting and facilitate a clearer delineation of the impact of this screening onpublic health.

Although prostate cancer screening recommendations vary, oneconsistent element is that physicians should counsel patients about the risksfor and potential benefits of treatment for early prostate cancer so thatpatients can participate in making the decision about whether to be screened. Tocounsel patients effectively about the risks for and benefits of treatment ofearly prostate cancer, physicians need access to current information and toincorporate it into their practices.

Editorial Note From the CDC

The findings in this report document a strong associationbetween a doctor’s recommendation and receipt of a PSA test, indicating thatphysician advice is a key determinant of whether men are tested. In Texas,approximately 60% of men who had heard of PSA testing reported receiving thetest. Nearly all of the men whose doctor recommended the PSA test took thatadvice.

This Commentary was reported by K. Condon, Behavioral RiskFactor Surveillance System Program; L. Suarez, phd, Office of the AssociateCommissioner of Disease Control and Prevention; D. Perrotta, phd, StateEpidemiologist, Texas Department of Health; Epidemiology, and Health ServicesResearch Bureau, Division of Cancer Prevention and Control, National Center forChronic Disease Prevention and Health Promotion; State Bureau, Division ofApplied Public Health Training, Epidemiology Program Office; and an EIS Officer,CDC. Adapted from Morbidity and Mortality Weekly Report 49(36):818-822, 2000.