Aims/hypothesis: In a population-based setting, we investigated whether diabetes-related morbidity and all-cause mortality within 2 years of HbA(1c) measurement were associated with that HbA(1c) level in individuals with type 2 diabetes. The main objective was to compare outcomes in those with HbA(1c) ≥ and <7% (53 mmol/mol).
Methods: Individuals with type 2 diabetes from Aarhus County, Denmark, were identified from public data files in a 3 year period (2001-2003). Stratifying the 17,760 individuals by HbA(1c), we estimated HRs for diabetes-related morbidities and all-cause mortality using Cox regression. Results were also stratified by treatment modality.
Results: In total, 1,805 individuals experienced at least one diabetes-related morbidity and 1,859 individuals died. In general, the HRs in adjusted analyses of diabetes-related morbidity and mortality were increased for HbA(1c) ≥ 7% (53 mmol/mol): morbidity, HR 1.48 (95% CI 1.34, 1.63); and mortality, HR 1.26 (95% CI 1.15, 1.39). On grouping individuals according to HbA(1c) <5% (31 mmol/mol), 5.0-5.9% (31-41 mmol/mol), 6.0-6.9% (42-52 mmol/mol), 7.0-7.9% (53-63 mmol/mol), 8.0-8.9% (64-74 mmol/mol) and ≥ 9% (75 mmol/mol), the HRs for mortality formed a U shape, with HbA(1c) 6.0-6.9% (42-52 mmol/mol) at the lowest point. For diabetes-related morbidity, a dose-response pattern appeared (lowest for HbA(1c) < 5% [31 mmol/mol]). Patterns of HR differed with treatment modality.
Conclusions/interpretation: An HbA(1c) level ≥ 7% (53 mmol/mol) was associated with increased morbidity and mortality. Both high and very low levels of HbA(1c) were associated with increased mortality. A dose-response pattern appeared for morbidity. The impact of HbA(1c) level on morbidity and mortality depended on treatment modality.