Current issues of ACP Journal Club are published in Annals of Internal Medicine


Men with type 2 diabetes mellitus or impaired glucose tolerance had an elevated risk for early death

ACP J Club. 1994 Jan-Feb;120:22. doi:10.7326/ACPJC-1994-120-1-022

Related Content in the Archives
Nondiabetic men with high blood glucose levels were at increased risk for death

Source Citation

Balkau B, Eschwège E, Papoz L, et al. Risk factors for early death in non-insulin dependent diabetes and men with known glucose tolerance status. BMJ. 1993 Aug 1;307:297-301.



To identify the causes of death and the risk factors for all-cause mortality in men according to glucose tolerance status.


A cohort study of men in France followed for a mean of 16 years.


Paris, France.


Male civil servants born between 1917 and 1928 were enrolled starting in 1967. 7038 of 7166 men without diabetes had a 2-hour glucose tolerance test at the first annual examination; 128 men with known diabetes who were not treated with insulin were also included. Follow-up was complete. The cause of 72% of deaths was ascertained.

Assessment of risk factors

Smoking habits; body mass index; central obesity; blood pressure; lipid, glucose, and insulin levels; nonesterified fatty acids; and mean corpuscular volume (MCV).

Main outcome measures

Treating physicians, hospital records, and interviews with family members provided data on date and cause of death.

Main results

6161 men had normal glucose tolerance, 697 had impaired glucose tolerance (IGT), 180 had newly diagnosed diabetes, and 128 had known diabetes. 975 men died (42% from cancer, 20% from circulatory diseases, 7% from digestive causes, 10% from injury and poisoning, and 22% from poorly defined causes). Compared with men who had normal glucose tolerance, the risk for death from all causes was higher for men with IGT (relative risk [RR] 1.6, 95% CI 1.3 to 2.0), known diabetes (RR 2.0, CI 1.4 to 3.0), and newly diagnosed diabetes (RR 2.7, CI 2.0 to 3.6). Multivariate analysis identified smoking, systolic hypertension, high nonesterified fatty acid levels, high MCV, central obesity, hyperinsulinemia, IGT, and newly diagnosed diabetes as independent risk factors for death. In men with known diabetes, central obesity, a high MCV, and hyperinsulinemia were associated with a higher risk for death than for men with normal glucose tolerance.


Men with diabetes and impaired glucose tolerance had similar elevated risk for early mortality compared with men with normal glucose tolerance. Men with diabetes had mortality risk factors that were similar to but more prevalent than those in men with normal glucose tolerance.

Sources of funding: Institut National de la Santé et de la Recherche Médicale and Direction de l'Action Sociale.

For article reprint: Dr. B. Balkau, Institut National de la Santé et de la Recherche Médicale, Unit 21, Hôpital Paul Brousse, 94807 Villejuif, France. FAX 33-1-4726-9454.


This study by Balkau and colleagues has good follow-up and confirms what has been shown in several studies—persons with type 2 diabetes mellitus or IGT have increased mortality when compared with persons who have normal glucose tolerance. Although this study showed that 42% of all deaths in the cohort were caused by cancer and only 20% were caused by cardiovascular diseases, other studies have shown that most of the excess deaths in patients with type 2 diabetes result from cardiovascular disease (1). This difference probably results from the high levels of alcohol and tobacco use among this cohort and perhaps results from the methods used to ascertain causes of death.

The risk factors that predict mortality in this cohort are similar to those that have been identified in earlier studies. The investigators also found that hyperinsulinemia predicts subsequent death among all participants, especially those with type 2 diabetes. It is unfortunate that neither proteinuria nor microalbuminuria were evaluated in this study, because recent studies among patients with type 2 diabetes have shown that these are important predictors of death (2).

Patients with type 2 diabetes and IGT have an increased risk for death. More than half of the patients with type 2 diabetes were not diagnosed before the study. The other factors posing independent risks in these patients with IGT or NIDDM predicted death in men with normal glucose tolerance. Many of these risk factors can be modified and, therefore, should be attended to in all older men. Finally, the independent role of hyperinsulinemia should be considered and may influence clinicians' choice of therapy for patients with type 2 diabetes, who frequently also have elevated insulin levels.

Linda Humphrey, MD, MPH
Department of Veterans Affairs Medical CenterPortland, Oregon, USA


1. Panzram G. Mortality and survival in type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia. 1987;30:123-31.

2. Damsgaard EM, Froland A, Jorgensen OD, Mogensen CE. Eight to nine year mortality in known non-insulin dependent diabetics and controls. Kidney Int. 1992;41:731-5.