Age at Diagnosis of Type 2 Diabetes Mellitus and Associations With Cardiovascular and Mortality Risks

Abstract

Word count: Background: Risk of cardiovascular disease (CVD) and mortality for patients with versus without type 2 diabetes mellitus (T2DM) appears to vary by age of T2DM diagnosis, but few population studies have analyzed mortality and CVD outcomes associations across the full age range. Methods: Using the Swedish National Diabetes Registry (NDR), everyone with T2DM registered in the NDR between 1998, and, 2012 were included. Controls were randomly selected from the general population matched for age, sex, and county. The analysis cohort comprised 318,083 patients with T2DM matched with just under 1•6 million controls. Participants were followed from 1998 to 2013 for CVD outcomes and to 2014 for mortality. Outcomes of interest were total mortality, CV mortality, non-CV mortality, coronary heart disease, acute myocardial infarction, stroke, heart failure, and atrial fibrillation. We also examined life expectancy by age of diagnosis. We conducted the primary analyses using Cox proportional hazards models in those with no prior cardiovascular disease and repeated the work in the entire cohort. Results: Over a median follow-up period of 2.52 years, T2DM patients diagnosed under 40 years or less had the highest excess risk for most outcomes relative to controls with adjusted HR [95% CI] of 2•05 [1•81 to 2•83] for total mortality, 2•72 [2•13 to 3•48] for CV related mortality, 1•95 [1•68-2•25] for non-CV mortality, 4•77 [3•86-5•89] for HF and 4•33 [3•82-4•91] for CHD. All risks attenuated progressively with each increasing decade at diagnostic age; by the time T2DM was diagnosed above 80 years, the adjusted HRs for CVD and non-CVD mortality were below one, with excess risks for other CVD outcomes substantially attenuated. Moreover, survival in those diagnosed beyond 80 was the same as controls, whereas it was beyond a decade less when T2DM was diagnosed in adolescence. Finally, HRs for most outcomes were numerically greater in younger women with T2DM. 4 Conclusions: Age at diagnosis of T2DM is prognostically important for survival and cardiovascular risks, with implications for determining timing and intensity of risk factor interventions for clinical decision making and for guideline-directed care. These observations amplify support for preventing / delaying T2DM onset in younger individuals.

Increased Mortality Burden in Young Asian Subjects with Dysglycemia and Comorbidities

“…In a cohort study using Clinical Practice Research Datalink data from 383 general practices in England, it was found that in 187,968 patients with incident type 2 diabetes, the life expectancy differences between subjects with and without diabetes declined with age attained [23]. In the Swedish National Diabetes Registry, those who were diagnosed with diabetes at ≤40 years had 2-fold excess mortality risk relative to the controls, and this risk attenuated progressively with each increasing decade of age, similar to our study results [14]. Our results as well as previous studies suggest that the mortality risk associated with diabetes differs markedly with age at diagnosis and presumably by ethnicity, with the highest mortality risk observed in those with early diagnosed diabetes.…”

“…Although the presence of diabetes affects life expectancy, the outcome differs according to age at the onset of diabetes. In a study comprising 318,083 patients with diabetes, who were matched with under 1·6 million controls and followed up for 5.6 years, patients who were diagnosed at younger ages showed significantly increased risks for most serious outcomes compared with the controls [14]. Interestingly, if the patient was diagnosed with diabetes at >80 years of age, the hazard ratio for the outcomes was <1, suggesting that by this age, survival in those with diabetes was the same as that in the controls.…”

Cardiovascular outcomes trials in type 2 diabetes: Time to include young adults

“…1,2 Adults with YOD lose up to 12 years of life, compared to 6 years of life lost among adults with usual-onset T2D (defined here as age of T2D onset ≥40 years). 3 Treatment targets are seldom achieved in YOD. 1 By the time they reach 60 years of age, adults with YOD are hospitalized at a rate of 2600 bed-days per 1000 patient-years, which is double that of those with usual-onset T2D.…”

“…YOD is a heterogeneous and aggressive phenotype with sharply rising incidence rates worldwide, especially in Asian populations (defined here as people of South‐, East‐ and Southeast‐Asian ethnicity), where up to one in five adults with T2D has YOD . Adults with YOD lose up to 12 years of life, compared to 6 years of life lost among adults with usual‐onset T2D (defined here as age of T2D onset ≥40 years) . Treatment targets are seldom achieved in YOD .…”

Diabetes management gains: teaching old dogs new tricks

“…Over the past three decades, excess mortality rates in people with diabetes relative to the general population have declined substantially in high-income countries such that those with diabetes are living longer than ever. That said, the gap in life expectancy remains at an average of 6 years in those diagnosed with diabetes in middle age, with considerably more life years lost when diabetes presents much earlier in life [1]. This is an issue because the numbers presenting earlier are rising, leading to greater challenges in care.…”