We aimed to investigate whether diabetes cases detected through screening have better health outcomes than clinically detected cases in a population-based cohort of adults who were eligible to be screened for diabetes at 10 year intervals.

The Västerbotten Intervention Programme is a community- and individual-based public health programme in Västerbotten County, Sweden. Residents are invited to clinical examinations that include screening for diabetes by OGTTs at age 30, 40, 50 and 60 years (individuals eligible for screening, n = 142,037). Between 1992 and 2013, we identified 1024 screen-detected cases and 8642 clinically detected cases of diabetes using registry data. Clinically detected individuals were either prior screening participants (n = 4506) or people who did not participate in screening (non-participants, n = 4136). Study individuals with diabetes were followed from date of detection until end of follow-up, emigration, death or incident cardiovascular disease (CVD), renal disease or retinopathy event, and compared using Cox proportional hazard regression adjusted for calendar time, age at detection, year of detection, sex and socioeconomic status.

The average age at diabetes diagnosis was 4.6 years lower for screen-detected individuals compared with clinically detected individuals. Overall, those who were clinically detected had worse health outcomes than those who were screen-detected (HR for all-cause mortality 2.07 [95% CI 1.63, 2.62]). Compared with screen-detected study individuals, all-cause mortality was higher for clinically detected individuals who were screening non-participants (HR 2.31 [95% CI 1.82, 2.94]) than for those clinically detected who were prior screening participants (HR 1.70 [95% CI 1.32, 2.18]). Estimates followed a similar pattern for CVD, renal disease and retinopathy.

Individuals with screen-detected diabetes were diagnosed earlier and appeared to fare better than those who were clinically detected with regard to all-cause mortality, CVD, renal disease and retinopathy. How much of these associations can be explained by earlier treatment because of screening rather than healthy user bias, lead time bias and length time bias warrants further investigation.