Abstract
Background: In 2018, prostate cancer screening with prostate specific antigen (PSA) received a “C” recommendation from the US Preventive Services Task Force for men aged 55 to 69 years. In January 2023, our health system implemented a point-of-care electronic reminder for primary care clinicians to discuss PSA screening for men aged 55 to 69.
Methods: We assessed the impact of reminder implementation on monthly rates of PSA ordering from January 1, 2022 to July 31, 2024 by performing interrupted time series analyses for men aged 55 to 69 years (for whom the reminder was implemented) and men aged 50 to 54, 70 to 74, and ≥75 years.
Results: Before reminder implementation, PSA was ordered in a median of 6.4% of visits for men aged 55 to 69 years versus a median of 10.2% of visits after reminder [adjusted incidence rate ratio (aIRR) 1.57 (95% CI: 1.43-1.73)]. The postreminder period was associated with smaller but significantly increased rates of PSA ordering in men aged 50 to 54 [aIRR 1.44 (1.25–1.65)], 70 to 74 [aIRR 1.26 (1.14–1.39)], and ≥75 years [aIRR 1.11 (1.02–1.11)].
Conclusions: Implementation of an electonic medical record (EMR) reminder to discuss PSA screening was associated with a large increase in PSA ordering among men in the targeted age-group, but also smaller increases in age groups for whom the balance of benefits and harms of PSA screening may be less favorable.
- Cancer Screening
- Electronic Medical Records
- Preventive Medicine
- Primary Health Care
- Prostate Cancer
- Prostate-Specific Antigen
- Reminder Systems
Introduction
Rates of prostate specific antigen (PSA) screening declined after a 2012 US Preventive Services Task Force (USPSTF) “D” recommendation.1 In 2018, the USPSTF issued an updated “C” recommendation for men aged 55 to 69 years, recommending that “men should have the opportunity to discuss the benefits and risks of PSA screening” before deciding whether to be screened.2 The USPSTF maintained the “D” recommendation for men ≥70 years.2 The American Cancer Society recommends that healthy men over age 50 years should have the chance to make an informed decision about prostate cancer screening.3 In 2023, 38% of US men aged 55 to 69 years reported having a PSA test in the past year.4
Our health system has historically maintained electronic medical record (EMR) reminders for preventive services with evidence demonstrating net benefits relative to harms (eg, USPSTF “A” or “B” recommendations). In 2022, a committee designed a reminder related to discussing PSA screening for men aged 55 to 69 years, which received the USPSTF “C” recommendation because a patients’ perceptions of whether the benefits outweigh the harms of screening is likely to vary across individuals.2
In light of the proliferation of preventive care reminders within many systems (including ours),5 we were concerned that most primary care visits lacked time for nuanced informed decision making discussions about PSA screening and that clinicians may order screening after cursory discussion.6 We also believed the reminder may have a relatively large impact compared with most on-screen reminders7 because PSA screening entails a simple blood test that may often be added to other blood tests. We conducted time series analyses of the association between reminder implementation and monthly rates of PSA ordering among men receiving primary care within the target age-group and among younger and older age groups.
Methods
Data derive from the Epic EMR of University of California, Davis (UCD) Health – an integrated system with 17 primary care clinics in the Sacramento region. On January 1, 2023, the PSA discussion reminder was implemented for men aged 55 to 69 years. For eligible patients, clinicians see “Discuss Risks and Benefits of PSA” listed as a “Current Care Gap” that is listed as “overdue” until a PSA is ordered, or clinicians document a PSA discussion. If PSA is declined by patients after discussion, clinicians can document PSA discussion by checking a box in the Health Maintenance section of the Epic chart or by using a system-designed phrase in the visit progress note. For younger and older men, no reminder occurs. Clinicians received no training in discussing PSA, and the health system did not promote or endorse use of PSA screening decision aides.
We used Epic SlicerDicer to identify the number of primary care visits by men aged ≥50 years in each month from January 1, 2022 to July 31, 2024, and the number of visits with PSA ordered. We constructed separate time series for men aged 55 to 69 years (for whom the reminder was implemented) and men aged 50 to 54, 70 to 74, and over 75 years (for whom no reminder occurred). The UC Davis IRB determined the study to be exempt.
In descriptive analyses, we identified the median number of primary care visits in each month in the pre-and postreminder periods and the median percentage of visits with PSA ordered in the 4 age groups. To estimate associations between reminder implementation and PSA ordering, we performed interrupted time series analysis, using negative binomial regression to model the monthly count of PSA tests ordered as a function of an indicator signifying the pre-reminder vs. post-reminder period, age-group, and interaction terms between the pre-reminder vs. post-reminder indicator and each age-group. We included the natural logarithm of the monthly number of visits within each group as an offset, so that the regression analyses modeled monthly rates of PSA ordering per visit and estimated incident rate ratios comparing post- and prereminder periods. We controlled for seasonal effects using cosinor analysis terms,8 and used robust standard errors to account for nesting of visits within patients, clinicians, or clinics. Analyses were performed using Stata S.E., Version 18.0 (College Station, TX).
Results
Monthly rates of PSA ordering by age-group are shown in Figure 1. Before reminder implementation, men aged 55 to 69 years attended a median of 4,309 monthly visits during which PSA was ordered in a median of 6.4% (Table 1). After reminder implementation, the median percentage of visits with PSA ordered was 10.2% – a 57% relative increase in time series analysis [adjusted incidence rate ratio 1.57 (95% CI: 1.43-1.73)]. Although the reminder was only triggered for men aged 55 to 69 years, the postreminder period was associated with smaller but significantly increased rates of PSA ordering in men aged 50 to 54 years, 70 to 74 years, and ≥75 years.
Percent of primary care visits with prostate specific antigen (PSA) ordered by age-group and month among men (January 2022 through July 2024).
Prostate Specific Antigen (PSA) Ordering Before and After Implementation of an Electronic Medical Reminder to Discuss Prostate Cancer Screening by Age Group
Discussion
Implementation of an EMR reminder to primary care clinicians to discuss PSA screening was associated with a large and statistically significant increase in PSA ordering among men in the targeted age-group of 55 to 69 years, but also smaller increases in age groups that were not targeted.
Time-series analyses may be confounded by unmeasured variables or secular trends. However, we believe a causal association between reminder implementation and the observed rise in PSA ordering is likely as the magnitude of increase was large and greatest in the targeted group. While little is known about recent trends in PSA screening, an uptick in PSA screening occurred among men aged 55 to 69 from 2020 to 2022 in a national US sample.9 Hence, the increases in PSA screening observed in men aged 50 to 54, 70 to 74, and >=75 years may be partly or wholly attributable to underlying secular trends. It is also possible that the increases in nontargeted groups may reflect unintended spillover effects. While our data do not inform about mechanisms of potential spillover effects, exposure to the reminder in the targeted 55 to 69 year age-group may have normalized PSA screening among our primary care workforce, leading to greater PSA screening across the entire population, including age groups where the balance of harms and benefits of PSA screening are less favorable.
One might consider that reminder implementation may have led to a large increase in shared decision making conversations, followed by many men choosing to pursue screening. However, engaging men in informed decision making regarding PSA screening is complex,10,11 and some have argued that such discussions are infeasible in busy primary care visits with many competing demands.6 We lacked data on the depth of PSA screening discussions but suspect most clinicians responded to the reminder by ordering PSA after cursory discussion. Survey or qualitative data from health system clinicians would be needed to confirm these suspicions, although a qualitative study found that Canadian family physicians with high rates of PSA screening often ordered PSA after limited patient discussion.12
In our analysis, a primary care-directed EMR reminder targeting PSA screening discussion was associated with a substantial increase in PSA ordering. We believe the magnitude of these associations are concerning considering the likelihood that many PSA tests were ordered without carefully informed decision making. Health systems should proceed cautiously in the implementation of preventive service reminders when the benefits of targeted services do not clearly outweigh associated harms.
Notes
This article was externally peer reviewed.
Conflict of interest: Dr. Fenton led the Evidence Synthesis for the 2018 US Preventive Services Task Force recommendation on prostate cancer screening. The authors have no other conflicts of interest to disclose.
Funding: None.
- Received for publication June 13, 2025.
- Revision received July 24, 2025.
- Accepted for publication August 4, 2025.
Jump to section
Related Articles
Cited By...
- No citing articles found.