brand logo

Am Fam Physician. 2001;64(5):841-844

Mortality rates in men with diabetes are more than double the rates in other men. For some causes of death, such as cardiovascular and ischemic heart disease, the rates are three to four times higher, independent of age and other risk factors. Based on studies of microvascular complications such as retinopathy, a “threshold” of glycosylated hemoglobin (HbA1c) of 5 percent has been thought to determine the risk of mortality. Studies indicate that the relationship between glycemic control and macrovascular mortality is direct and continuous; therefore, control of blood sugar is important in all men, not just those with diabetes. Khaw and associates conducted a prospective population study to examine the value of HbA1c concentration as a predictor of death from cardiovascular disease and all causes in men.

The authors used data from a large study of general practices in eastern England. Beginning in 1993, data from more than 25,600 men and women 45 to 79 years of age were collected from questionnaires, physical examinations and laboratory tests. Regular monitoring of this population included reporting of all deaths by cause. By 1999, follow-up data were available on 4,662 men whose early profiles included HbA1c testing.

Glycosylated hemoglobin (%)
Cause of death< 5 (n = 1,204)5 to 5.4 (n = 1,606)5.5 to 6.9 (n = 1,611)≥ 7 (n = 81)Self-reported diabetes (n = 160)
All causes (n = 135):
Age adjusted rate/100 (no. of events)1.65 (18)2.33 (35)3.43 (61)4.35 (5)5.92 (16)
Relative risk1.001.412.072.643.59
Cardiovascular disease (n = 60):
Age-adjusted rate/100 (no. of events)0.50 (5)1.27 (19)1.24 (22)2.54 (3)4.11 (11)
Relative risk1.002.532.465.048.16
Ischemic heart disease (n = 42)
Age-adjusted rate/100 (no. of events)0.31 (3)0.86 (13)0.87 (15)1.63 (2)3.43 (9)
Relative risk1.002.742.775.2010.91
Noncardiovascular disease (n = 75):
Age-adjusted rate/100 (no. of events)1.15 (13)1.06 (16)2.19 (39)1.81 (2)1.82 (5)
Relative risk1.000.921.911.581.58

Although only 5 percent of the population is diabetic, about 70 percent have HbA1c levels between 5 and 6.9 percent. The age-adjusted death rate for all causes was more than 2.5 times higher in men with HbA1c levels of at least 7 percent than in men with HbA1c levels of less than 5 percent (see accompanying table). Death rates from cardiovascular disease and ischemic heart disease were increased 8.16 and 10.91 times, respectively. The risk of death from these causes was directly related to HbA1c concentration. The authors calculate that each 1 percent increase in the HbA1c level was associated with an increase of 29 percent in all-cause mortality, 38 percent in cardiovascular mortality and 44 percent in ischemic heart-disease mortality, after adjustment for age, blood pressure, cholesterol level, body mass index, smoking status and history of myocardial infarction or stroke. About 37 percent of the total deaths in the entire study population were attributed to excess mortality in men with HbA1c concentrations of 5 percent or more.

The authors conclude that HbA1c concentration significantly predicts mortality and that there is no threshold effect. Independent of other risk factors, lowering HbA1c levels has great potential to reduce mortality in men regardless of diabetic status. The authors calculate that reducing the average HbA1c level by just 0.2 percent could lower mortality by 10 percent. They point out that very small changes in HbA1c levels in very large numbers of men would have great impact, but that achieving these changes requires reversing the current population trends of obesity and an increasing rate of diabetes.

Continue Reading

More in AFP

Copyright © 2001 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See for copyright questions and/or permission requests.