Prostate cancer is one of the most common cancers diagnosed among men in the U.S. The prostate-specific antigen (PSA) test is the primary diagnostic tool used in primary care for individuals with symptoms suggestive of prostate cancer. The NICE NG12 guideline recommends PSA testing for men presenting with urinary symptoms, visible hematuria, or erectile dysfunction, and referral to secondary care when PSA levels exceed age-specific thresholds. However, these guidelines do not specify retesting intervals or age limits. The Prostate Cancer Risk Management Program (PCRMP) permits PSA testing for asymptomatic men aged over 50 years through shared decision making but provides no guidance on retesting frequency. The lack of clear international or national recommendations has led to variability in testing practices, increased risk of overtreatment and overdiagnosis, and uncertainty about the optimal retesting interval.
This study aimed to investigate PSA testing patterns in primary care in England from 2000 to 2018, focusing on the population-based trends and associations with regional, demographic, and clinical factors, as well as individual variation in testing and retesting frequency.
This study employed a population-based cohort analysis using electronic health records from the Clinical Practice Research Datalink (CPRD) Aurum. It was linked with data from the National Cancer Registration and Analysis Service (NCRAS), Hospital Episode Statistics (HES), and the Office for National Statistics (ONS). This study included men aged over 18 years who were registered at general practices in England with at least one year of follow-up from 2000 to 2018. Individuals with a prior prostate cancer diagnosis were excluded. PSA test results were classified based on age-specific thresholds defined by NICE NG12 guidelines, ranging from > 2.5 ng/mL in men aged 18-49 years to > 6.5 ng/mL in men aged 70 years or older.
Demographic variables included ethnicity, age, region, family history of prostate cancer, symptom presentation, and deprivation index. Clinical symptoms like urinary complaints, hematuria, erectile dysfunction, back pain, and fatigue were identified from SNOMED-CT data within 90 days before PSA testing.
Statistical analyses included age-standardized testing rates, mixed-effects negative binomial regression to identify the determinants of PSA testing, and linear mixed-effects models to assess retesting intervals. Analyses were adjusted for demographic and clinical covariates, incorporating random effects for both patients and general practices.
This large population-based study included 10,235,805 men from 1442 general practices in England, contributing over 81.7 million person-years of follow-up from 2000 to 2018. About 1.52 million men (14.9%) underwent at least one PSA test, yielding 3.84 million test results, with nearly half of the participants receiving multiple tests. PSA testing increased substantially over time from 11.8 per 1000 person-years in 2000 to 69.7 in 2018. Testing was more common in men aged 70-89 years, those of white ethnicity, residents of southern England, and individuals from less deprived areas. Men with a family history of prostate cancer were almost five times more likely to undergo testing. Asymptomatic men accounted for the majority of tests, and low PSA values became more prevalent. Multivariable analysis showed that age, ethnicity, deprivation, region, symptoms, and prior raised PSA levels were significant determinants of testing frequency.
Among 735,750 men with two or more PSA tests, 2.31 million retesting intervals were recorded with a median interval of 12.6 months and a geometric mean of 19.3 months. Retesting occurred more frequently among older men, those of non-White ethnicity, those with previously elevated PSA levels, or those with a family history of prostate cancer or symptoms like hematuria or weight loss. Notably, 73% of men never exceeded PSA thresholds, indicating possible over-testing, with retesting intervals shorter than the internationally recommended 2-4 years.
Overall, PSA testing in England has increased markedly since 2000, accompanied by significant demographic and regional disparities. Despite the absence of a national screening program, many men underwent repeated PSA testing at intervals shorter than the evidence-based recommendations. These findings indicate potential over-testing in low-risk populations and under-testing in more deprived areas. These results suggest an urgent need for evidence-based and clear national guidelines on PSA testing frequency to balance the benefits of early detection against the risks of overdiagnosis and unnecessary treatment.
Reference: Collins KK, Oke JL, Virdee PS, Perera R, Nicholson BD. Prostate specific antigen retesting intervals and trends in England: population based cohort study. BMJ. 2025;391:e083800. doi:10.1136/bmj-2024-083800
