Use of the prostate‐specific antigen test in the U.S. for men age 30 to 64 in 2011 to 2017 using a large commercial claims database: Implications for practice interventions

Abstract Background Given the public health relevance of PSA‐based screening, various professional organizations have issued recommendations on the use of the PSA test to screen for prostate cancer in different age groups. Aim Using a large commercial claims database, we aimed to determine the most recent rates of PSA testing for privately insured men age 30 to 64 in the context of screening recommendations. Methods and Results Data from employer plans were from MarketScan commercial claims database. Annual PSA testing rate was the proportion of men with ≥1 paid test(s) per 12 months of continuous enrollment. Men with diagnosis of any prostate‐related condition were excluded. Annual percent change (APC) in PSA test use was estimated using joinpoint regression analysis. In 2011 to 2017, annual testing rate encompassing 5.02 to 5.53 million men was approximately 1.4%, age 30 to 34; 3.4% to 4.1%, age 35 to 39; 11% to 13%, age 40 to 44; 18% to 21%, age 45 to 49; 31% to 33%, age 50 to 54; 35% to 37%, age 55 to 59; and 38% to 41%, age 60 to 64. APC for 2011 to 2017 was −0.5% (P = .11), age 30 to 34; −3.0% (P = .001), age 35‐39; −3.1% (P < .001), age 40 to 44; −2.4% (P = .001), age 45 to 49; −0.2% (P = .66), age 50 to 54; 0.0% (P = .997), age 55 to 59; and −3.3% (P = .054) from 2011 to 2013 and 1.2% (P = .045) from 2013 to 2017, age 60 to 64. PSA testing rate decreased from 2011 to 2017 for age groups between 35 and 49 by 13.4% to 16.9%. Conclusions Based on these data, PSA testing rate has modestly decreased from 2011 to 2017. These results, however, should be considered in view of the limitation that MarketScan claims data may not be equated to actual PSA testing practices in the entire U.S. population age 30 to 64. Future research should be directed to understand why clinicians continue ordering PSA test for men younger than 50.


| INTRODUCTION
The test for prostate-specific antigen (PSA) has been used effectively in the monitoring of diagnosed prostate cancer; however, there have been differing guidelines in using this test to screen for prostate cancer, and the recommendations for screening practices have evolved over the years. In 2008, the U.S. Preventive Services Task Force (USPSTF) noted that there was insufficient evidence to screen men younger than 75. 1 In 2012, the USPSTF recommended against PSA screening of all men for prostate cancer. 2 Subsequent to the 2012 recommendation, the USPSTF released a draft recommendation in April 2017 noting that the potential benefits and harms of PSA-based screening were closely balanced in men age 55 to 69 years, and that the decision whether to be screened should be an individual one. 3 The USPSTF issued its most recent recommendation in May 2018, 4 that was identical to the 2017 draft recommendation. 3 The recommendations by the American Urological Association (AUA) 5 and American College of Physicians (ACP) 6 in 2013, and American Cancer Society (ACS) in 2016 7 although consistent with current USPSTF recommendations, 4 did not agree with the 2012 USPSTF recommendation against routine screening of all men. 2 The AUA and USPSTF recommend offering to screen men age 55 to 69 for prostate cancer after a process of shared decision-making. 4,5 The ACP recommends limiting routine screening to men age 50 to 69 with at least 10 years of life expectancy if they agree to be tested after discussions with their clinicians, 6 while the ACS recommends PSA-based prostate cancer screening of men age ≥ 50 with at least 10 years of life expectancy. 7 In the context of PSA-based screening recommendations for prostate cancer and given its public health relevance, the objective of this study was to determine the more recent rate and trend for annual PSA testing in a sample of privately insured U.S. men age 30 to 64 in 2011 through 2017 using a large insurance claims database. This study is unique in that it is, to our knowledge, the only report of PSA testing rates and trends through 2017 that also includes men in younger age groups between 30

| Annual percent change and PSA testing trends
Annual percent change (APC) in PSA test use was estimated using joinpoint regression analysis, fitting trend data to identify the loglinear model with the fewest number of inflection points. 9 The APC was the log-linear slope of each trend line, and the 2-sided P values related to statistical significance (ie, P < .05) for each APC estimate being different from zero.

| Hemoglobin testing trends
To ascertain that the trends observed for PSA testing rate were not due to changes in clinical laboratory testing in general, we determined testing trends for hemoglobin (CPT Codes 83026, 83051, and 85018), a test not impacted by any changes in practice recommendations, as a measure of trend in laboratory testing.

| Included populations
The number of men included in each year from 2011 through 2017 ranged from 4.55 to 5.14 million after excluding claim encounters with diagnostic or non-laboratory procedure codes relating to prostate related diseases or conditions, including codes related to elevated PSA and history of prostate cancer. Consequently, 92.6% to 93.1% of men remained for analysis after this exclusion. Table 1 shows the age and regional distribution of men included in the analyses from 2011 to 2017. No regional assignment could be made for 0.2% to 0.3% of men due to missing data.  Table 2 shows the range of annual PSA testing rate and annual percent change (APC) for men in each of these seven groups.

| Annual PSA testing trends
Except for PSA testing trend in men age 60 to 64, there was no inflection point in annual testing rate from 2011 to 2017. Annual testing showed significantly downward trends in men age 35 to 39 (P = .001), 40 to 44 (P < .001) and 45 to 49 (P = .001). There were no significant trends for PSA testing rate in men age 30    Our findings should be considered in view of several limitations of this study. The sample of U.S. men studied is restricted to those with private health insurance coverage. Also, we captured only paid claims data using non-random MarketScan samples of the U.S. population. 8  This is a vital component of laboratory quality and diagnostic excellence, and it has been promulgated by initiatives, such as Choosing Wisely. 30 Explorations of future PSA-based testing rates and trends are avenues for future investigations.

ACKNOWLEDGMENTS
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of CDC.

CONFLICT OF INTEREST
The authors have stated explicitly that there are no conflicts of interest in connection with this article.

AUTHOR CONTRIBUTION
All authors had full access to the data in the study and take responsi-

ETHICAL STATEMENT
Instituitional clearance and approval have been obtained prior to submission. Requirement for patient consent is not applicable for this study.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.