State-by-State Variation in Prostate-Specific Antigen Screening Trends Following the 2011 United States Preventive Services Task Force Panel Update

OBJECTIVE: To evaluate state-by-state trends in prostate-specific antigen (PSA) screening prevalence after the 2011 United States Preventive Services Task Force (USPSTF) recommendation against this practice.

METHODS: We included 222,475 men who responded to the Behavioral Risk Factor Surveillance System 2012 and 2014 surveys, corresponding to early and late post-USPSTF populations. Logistic regression was used to identify predictors of PSA screening and to calculate the adjusted and weighted state-by-state PSA screening prevalence and respective relative percent changes between 2012 and 2014. To account for unmeasured factors, the correlation between changes in PSA screening over time and changes in screening for colorectal and breast cancer were assessed. All analyses were conducted in 2016.

RESULTS: Overall, 38.9% (95% confidence interval [CI] = 38.6%-39.2%) reported receiving PSA screening in 2012 vs 35.8% (95% CI = 35.1%-36.2%) in 2014. State of residence, age, race, education, income, insurance, access to care, marital status, and smoking status were independent predictors of PSA screening in both years (all P <.001). In adjusted analyses, the nationwide PSA screening prevalence decreased by a relative 8.5% (95% CI = 6.4%-10.5%; P <.001) between 2012 and 2014. There was a vast state-by-state heterogeneity, ranging from a relative 26.6% decrease in Vermont to 10.2% increase in Hawaii. Overall, 81.5% and 84.0% of the observed changes were not accompanied by matching changes in respective colorectal and breast cancer screening utilization, for which there were no updates in USPSTF recommendations.

CONCLUSION: There is a significant state-by-state variation in PSA screening trends following the 2011 USPSTF recommendation. Further research is needed to elucidate the reasons for this heterogeneity in screening behavior among the states.