Different Mortality Rate Trends in Diabetic Men and Women

Summarized by Robert W. Griffith, MD
September 11, 2007

Summary

Analysis of NHANES data shows reductions in mortality rates in diabetic men, but not in women, over the 40-year period between 1971 and 2000. This is most probably related to less satisfactory prevention, diagnosis, and management of cardiovascular disease in women, compared with that in men.

Introduction

People in the USA are living longer, as shown by the overall decline in mortality rate over the last 40 years or so. This is largely due to a decrease in cardiovascular deaths, brought about by improvements in treatment as well as in lifestyle. Most deaths in diabetics are due to cardiovascular disease, so one might expect that changes in the mortality rate of diabetics would parallel those of the population in general.

US Centers for Disease Control (CDC) scientists have studied this question, and reported their findings in the Annals of Internal Medicine. They analyzed data from the National Health and Nutritional Examination Surveys (NHANES I, II, and III). This is a summary of their principle findings.

What was done

NHANES I, II, and III are three cohorts or groups of people aged 35 to 74 at enrollment; NHANES I ran from 1971 to 1974, NHANES II from 1976 to 1980, and NHANES III from 1988 to 1994. There were 26,000 people in the analyses conducted after an average of 12.2 years. About 8% of them had diabetes (types I and II combined).

The information collected and analyzed included demographics, diabetes status (self-reported), duration of the disease, insulin use, and history of cardiovascular disease (heart attack, hear failure, or stroke). Height and weight were measured and used to calculate the body mass index (BMI). Deaths and their causes were listed.

Mortality rates were determined through 1986, 1992, and 2000 for the three cohorts, respectively. Separate rates were calculated for each gender, and for diabetics and non-diabetics.

What was found

The percentage of nonwhite persons doubled across the survey years, in both men and women, and the level of education increased. BMI scores increased in the later years, more so in diabetic women than in diabetic men. The average age at diagnosis of diabetes decreased and the average age of diabetic people decreased, by roughly 2½ years.

The mortality rates for different groups are shown in the table (given as the number of deaths per 1000):

* statistically significant decrease
@ statistically significant increase

it can be seen that the findings in the overall population obscured important gender-related differences. In men, all-cause and cardiovascular mortality decreased in a similar fashion for those with and without diabetes. However, in women there were less pronounced falls in those without diabetes, and a significant increase in mortality in all-cause deaths in those with diabetes.

What the results show

It's clear that diabetic women have been left behind in the otherwise general reduction in mortality (and especially cardiovascular mortality) over the last 35-40 years. The difference in mortality between diabetic and non-diabetic women has actually doubled, a disturbing fact.

It's also clear that the decline in overall mortality reflects an accompanying, or underlying, decline in cardiovascular mortality. The results for men-only parallel those in the men-plus-women totals. However, in women, there was no improvement in cardiovascular mortality over the years (10.5 vs. 9.4 per 1000).

The reason for this difference should probably therefore be sought in the cardiovascular care women receive. There have been reports of smaller increases in aspirin use and antihypertensive medication in women, compared with men. Women are also less likely to receive aggressive medical management, including revascularization procedures, after hospitalization for coronary artery narrowing. There are also sex differences in inflammatory and hormonal responses to cardiovascular risk factors, while the symptoms of heart attack in women are less readily recognized. All these factors might influence the outcome in women unfavorably.

The study has certain limitations. Diabetes was assessed by self-reporting, and there were insufficient numbers of subjects to provide all the statistical information required to explain the gender difference detected. However, the findings should lead to greater efforts to improve public health information about diabetic and cardiovascular risk factors, especially in women.

Source

• Mortality trends in men and women with diabetes, 1971-2000. EW. Gregg , Q. Gu, YJ. Cheng,  et al. , Annals Intern Med , 2007, vol. 147, pp. 60520--59

Related Links
Diabetic Women At Risk
When Should Diabetics Start a Heart Protection Program?
Women's New Cardiac Health Guidelines