BackgroundThe FLASH‐UK trial showed lower HbA1c with intermittently‐scanned continuous glucose monitoring (isCGM), as compared with self‐monitoring of blood glucose (SMBG), in adults with type 1 diabetes and HbA1c ≥58 mmol/mol. (≥ 7.5%). Here we present results from the pre‐specified subgroup analysis for the 24‐week HbA1c (primary outcome) and selected sensor‐based secondary outcomes.MethodsThis was a multicentre, parallel‐design, randomised controlled trial. The difference in treatment effect between subgroups (baseline HbA1c (≤75 vs >75 mmol/mol (≤9.0 vs > 9.0%), treatment modality (pump vs injections), prior participation in structured education, age, educational level, impaired awareness of hypoglycaemia, deprivation index quintile sex, ethnic group, and Patient Health Questionnaire‐9 (PHQ‐9) detected depression category) were evaluated.ResultsOne hundred fifty‐six participants [females 44%, mean (SD) baseline HbA1c 71 (9) mmol/mol 8.6(0.8%), age 44 (15)] were randomly assigned, in a 1:1 ratio to isCGM (n=78) or SMBG (n=78). The mean (SD) baseline HbA1c (%) was 8.7 (0.9) in the isCGM group and 8.5 (0.8) in the SMBG group, lowering to 7.9 (0.8) vs 8.3 (0.9) respectively at 24 weeks (adjusted mean difference ‐0.5, 95% confidence interval[CI] ‐0.7 to ‐0.3; p<0.001]. For HbA1c, there was no impact of treatment modality, prior participation in structured education, deprivation index quintile, sex or baseline depression category. The between‐group difference in HbA1c was larger for younger people [a reduction of 2.7 (95%CI 0.3 to 5.0; p=0.028) mmol/mol for every additional 15 years of age]. Those with HbA1c 76‐97 mmol/mol (>9.0‐11.0%) had a marginally non‐significant higher reduction in HbA1c of 8.4 mmol/mol (3.3 to 13.5) compared to 3.1 (0.3 to 6.0) in those with HbA1c 58‐75mmol/mol (p=0.08). For "Time in range" (% 3.9 to 10 mmol/l), the difference was larger for those with at least a bachelor's degree. For "Time below range" (% <3.9 mmol/l), the difference was larger for those using injections, older people, and those with less than bachelor's degree.ConclusionsIntermittently‐scanned continuous glucose monitoring is generally effective across a range of baseline characteristics.