The HbA1c and all-cause mortality relationship in patients with type 2 diabetes is J-shaped: a meta-analysis of observational studies

Rev Diabet Stud. 2014 Summer;11(2):138-52. doi: 10.1900/RDS.2014.11.138. Epub 2014 Aug 10.


Background: Low blood glucose and HbA1c levels are recommended in the literature on management of diabetes. However, data have shown that low blood glucose is associated with serious adverse effects for the patients and the recommendation has been criticized. Therefore, this article revisits the relationship between HbA1c and all-cause mortality by a meta-analysis of observational studies.

Aim: The aim of this study is to determine whether there is a J- or U-shaped non-linear relationship between HbA1c and all-cause mortality in type 2 diabetes patients, implying an increased risk to premature all-cause mortality at high and low levels of HbA1c.

Methods: A comprehensive literature search was conducted using PubMed, Medline, and Cochrane Library databases with strict inclusion/exclusion criteria. The published adjusted hazard ratios (HR) with 95% confidence intervals of all-cause mortality for each HbA1c category and per study were analyzed. Fractional polynomial regression was used with random effect modeling to assess the non-linear relationship of the HR trends between studies. Seven eligible observational studies with a total of 147,424 participants were included in the study.

Results: A significant J-shaped relationship was observed between HbA1c and all-cause mortality. Crude relative risk for all-cause mortality identified a decreased risk per 1% increase in HbA1c below 7.5% (58 mmol/mol) (0.90, CI 0.86-0.94) and an increased risk per 1% increase in HbA1c above 7.5% (58 mmol/mol) (1.04, CI 1.01-1.06). Observational studies revealed a J-shaped relationship between HbA1c and all-cause mortality, equivalent to an increased risk of mortality at high and low HbA1c levels.

Conclusions: This increased mortality at high and low HbA1c levels has significant implications on investigating optimum clinical HbA1c targets as it suggests that there are upper and lower limits for creating a 'security zone' for diabetes management.

Publication types

  • Meta-Analysis
  • Research Support, Non-U.S. Gov't
  • Review

MeSH terms

  • Blood Glucose / analysis
  • Cardiovascular Diseases / blood
  • Cause of Death
  • Diabetes Complications / blood
  • Diabetes Complications / mortality
  • Diabetes Mellitus, Type 2 / blood*
  • Diabetes Mellitus, Type 2 / mortality*
  • Glycated Hemoglobin / analysis*
  • Humans
  • Risk Factors


  • Blood Glucose
  • Glycated Hemoglobin A