OBJECTIVES: Most of the current evidence based for screening is based on T2DM populations.This study explored the impact of increasing screening intervals for diabetic retinopathy (DR) for people with either type 1 diabetes (T1DM) or type 2 diabetes (T2DM) without DR.

METHODS: Cost utility analysis (CUA) was undertaken comparing either annual or biennial screening for DR in people with either T1DM or T2DM and no DR, attending the national Diabetic Eye Screening Wales (DESW) program. Data from DESW and Primary Care were linked using secure anonymised information linkage. Regression analysis estimated risks of DR progression requiring referral to hospital eye services (RDR). The CUA was facilitated by a time varying Markov model and DR risk factors (HbA1c, blood pressure, duration of diabetes etc.) enabled estimates of relative costs and quality adjusted life years (QALY) gained associated with an increase to biennial screening for people with T1DM and T2DM.

RESULTS: Data from 2,286 and 36,202 people with T1DM and T2DM respectively were analysed. Increasing screening intervals to two years for people with T1DM and HbAof 6.5, 8.0 and 9.5 at the time of screening were estimated to lose one QALY for a cost saving to national health services (NHS) of £118,612, £48,449 and £15,818 respectively. For people with T2DM at the same HbA level suggested NHS saving of £109,320, £60,482 and £32,510 for each QALY lost. The incremental cost effectiveness ratio for biennial screening for people with T1DM was £30,995 for and £73,475 for people with T2DM.

CONCLUSIONS: Our findings, primarily driven by the difference in the risk of progression to RDR between people with T1DM and T2DM, suggest that biennial screening intervals in people with T2DM makes best use of NHS resources. For people with T1DM however annual screening remains justified and should not be increased to biennial.