The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #56926

Evelyn Chan, MD, Dept. of Medicine, University of Texas, 6431 Fannin, Suite 1.122MSB, Houston, TX 77030, 713-500-6705, evelyn.c.chan@uth.tmc.edu, Sally W Vernon, PhD, School of Public Health/Center for Health Promotion and Prevention Research, University of Texas - Houston, 7000 Fannin, Suite 2560, Houston, TX 77030, Chul Ahn, PhD, Clinical epidemiology and Biostatistics, Department of Medicine, University of Texas, 6431 Fannin, MSB 1.110, Houston, TX 77030, and Anthony Greisinger, PhD, Kelsey-Seybold Clinic, Kelsey Research Foundation, 7800 Fannin Street, Suite 209, Houston, TX 77054.

Background: Because prostate cancer screening with prostate specific antigen (PSA) is controversial, informed consent is recommended. To target educational efforts and to assess mass screenings, we determined the accuracy of patient self-reports of PSA testing and identified variables that would predict accuracy. Methods: Men 50 years and older (n=402) attending the medicine Kelsey-Seybold Clinic (KSC) and the University of Texas-Houston Clinic (UT-H) completed a survey assessing whether they had taken a PSA test with that visit and their knowledge and experience with PSA testing. A physician reviewed the medical record (n= 402) to determine whether a screening PSA was done. We conducted chi-square tests for accuracy of self-reports across variables and stepwise logistic regression to identify predictors of accuracy. Results: The prevalence of PSA testing was 90% at KSC and 28% at UT-H. The concordance between self-reports and the medical record at KSC was 65%, and at UT-H, it was 88%. At KSC, the sensitivity, specificity, false positive and false negative values to “Did you have a PSA today?” were 67%, 43%, 57%, and 33%, respectively. At UT-H, the sensitivity, specificity, false positive, false negative values were 64%, 98%, 2%, and 36%, respectively. At KSC, only “ever had a PSA” and “doctor recommended a PSA test today” predicted accuracy; at UT-H, no variables were significantly associated with accuracy. Conclusion: The accuracy of patient self-reports of PSA testing may differ when the prevalence of PSA testing is different. Caution may be required when interpreting the accuracy of self-reports of PSA testing.

Learning Objectives:

At the conclusion of this session, the participant will be able to:
• Recognize the potential relationship between prevalence of prostate specific antigen testing and accuracy of patient self-reports of having completed the test.
• Identify predictors of accuracy of patient self-reports of prostate specific antigen testing.

Keywords: Screening, Primary Care

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.