Differential prostate-specific antigen (PSA) testing practices according to obesity-related comorbid conditions may contribute to inconsistent results in studies of obesity and prostate cancer. We investigated the relationship between obesity and PSA testing, and evaluated the role of prior diagnoses and disease screening on PSA testing patterns.
Men, 40 and 79 years old and without prior prostate cancer were recruited from 25 health centers in the Southern US (n = 11,558, 85% African-American). An extensive in-person interview measured medical and other characteristics of study participants, including PSA test histories, weight, height, demographics, and disease history. Odds ratios (OR) and (95% confidence intervals) from logistic regression summarized the body mass index (BMI) and PSA test association while adjusting for socio-economic status (SES).
BMI between 25 and 40 was significantly associated with recent PSA testing (past 12 months) (OR25.0–29.9 = 1.23 (1.09, 1.39); OR30–34.9 = 1.36 (1.18, 1.57); OR35.0–39.9 = 1.44 (1.18, 1.76); OR≥40 = 1.15 (0.87, 1.51)). Prior severe disease diagnoses, such as heart disease, did not influence the obesity and PSA test association. However, adjustment for prior high blood pressure or high cholesterol diagnoses reduced the BMI-PSA testing associations. Physician PSA test recommendations were not associated with BMI, and results did not appreciably vary by race.
Overweight and obese men were preferentially PSA tested within the past 12 months. BMI was not associated with physician screening recommendations. Data suggest that clinical diagnoses related to obesity increase clinical encounters that lead to preferential selection of obese men for prostate cancer diagnosis. This detection effect may bias epidemiologic investigations of obesity and prostate cancer incidence. Prostate 66: 1371–1380, 2006. © 2006 Wiley-Liss, Inc.