Current Trends Premature Mortality from Diabetes Mellitus - - Use of Sentinel Health Event Surveillance to Assess Causes

According to national death certificate data (Table V), diabetes was the 12th leading cause of years of potential life lost in the United States in 1984. Actually diabetes contributes to a much larger proportion of mortality, since it is reported on only about half of the death certificates for persons who die with the disease and is listed as the underlying cause on only one-quarter of the certificates on which it appears (1). The most frequent causes of death among persons with diabetes are ischemic and other forms of heart disease, cerebrovascular disease, and other forms of atherosclerosis; renal disease, including nephritis/nephrosis and uremia; respiratory disease; and infection.

Optimal diabetes control depends on access to health care and the application of modern health care practices by both providers and patients. A method recently suggested for ascertaining the causes of premature mortality among persons with diabetes is the enumeration and investigation of deaths among young persons with diabetes (2). For the purpose of this investigation, each death is considered a "sentinel health event" worthy of detailed evaluation (3). The frequency of and factors contributing to these events may serve as a measure of the quality of health care in the community.

During 1985, Diabetes Control Programs (DCP) in Colorado, Illinois, Kentucky, Louisiana, Michigan, and Washington participated in a pilot project to test a surveillance system for sentinel health events for persons with diabetes. An event was defined as the death of a person who was 45 years of age and whose death certificate made any reference to diabetes. A DCP investigator gathered basic demographic data from the death certificate and interviewed a family member and a physician associated with the case. The information collected included demographic, clinical, and health care related data.

The cause of death for each event was coded into one of the following three categories: 1) acute complications of diabetes (diabetic ketoacidosis or coma, infection, and hypoglycemia), 2) chronic complications (heart disease, end-stage renal disease, and cerebrovascular disease), and 3) other (diabetes only, non-diabetes-related, and unknown). Coding was based on the most contributory cause as indicated on the death certificate and by the DCP investigation.

Two hundred and thirty-three events were identified. The mean observed mortality rate (adjusting for variable observation times among sites) was 32.7/10,000 persons who were 45 years of age and had diabetes. Based on 1983 national death certificate data, the expected rate is 36.3/10,000 persons 45 years of age with diabetes.

The cohort of events included 146 males and 87 females. The median age at death was 38 years for males and 36 years for females. The distribution of causes of death did not differ between males and females. However, the cause of death differed significantly by age (Chi-square=14.5, p0.01); younger persons tended to die from acute complications, and older persons, from chronic complications (Table 1).

Several conditions were incidentally reported with notable frequency. They include alcohol abuse (13 events), suicide (5 events), and cardiomyopathy or congestive heart failure in the absence of ischemic heart disease (7 events). Eight of the persons with a history of alcohol abuse were male. Five of the 13 persons with a history of alcohol abuse died from acute complications; three, from chronic complications; and five, from other causes. All of the suicide events involved males.

The care practices of the cohort are shown by sex in Table 2. Such practices as blood pressure measurement and urinalysis were reported to have been conducted within the last year of life for virtually every person in the cohort. Other care practices, including glycohemoglobin measurement and funduscopic examination in the last year of life, were reported less commonly. Only half (48%) of the funduscopic examinations were performed by ophthalmologists. Among those persons using insulin (84%), 46% had had a glycohemoglobin test in the last year of life, and 61% had used self-monitoring of blood glucose. Although virtually all persons with hypertension had been under treatment for it, the condition had not been controlled for 57% of them ( greater than or equal to 140/90 mm Hg at the last examination). Forty-one percent were smokers at the time of death.

Individual differences in care practices between those who died from acute complications and those who died from chronic complications were not statistically significant. However, those who died from acute complications were reported to have had fewer average yearly physician visits than those who died from chronic complications (Wilcoxon rank sum test, p0.05).

There were more males than expected in the cohort. The ratio of males to females was 1.6, whereas the ratio of males to females in the living diabetic population 45 years of age is 0.7. Differences in individual care practices between males and females were not statistically significant. However, within the last year of life, females had consistently higher rates of beneficial health care practices such as blood pressure check, urinalysis and glycohemoglobin testing, self-monitoring of blood glucose, and funduscopic examination. Fewer females had been current smokers, and females had seen physicians more frequently than had males, though the difference was not significant (median frequency 9 visits per year for females as compared with 6.5 visits per year for males). Reported by B Gabella, S Michael, Colorado Dept of Health; B Hudspeth, R Shubert, Illinois Dept of Public Health; C Gollmar, Kentucky Dept of Health Svcs; D Kaplan, Louisiana Dept of Health and Human Resources; J Beasley, M Halpern, S Longabaugh, Michigan Dept of Public Health; J Will, Washington Dept of Social and Health Svcs; Div of Diabetes Control, Center for Prevention Svcs, CDC.

Editorial Note

Editorial Note: Despite the heavy toll of chronic diseases on the health of persons in the United States, there are relatively few surveillance systems for them. The model surveillance system employed for infectious diseases, which includes reporting and compiling cases at a central source and rapidly disseminating results, is not currently a feasible approach for chronic disease surveillance. Infectious disease control--the investigation of causes and development of control strategies--is relatively straightforward. Chronic disease surveillance and control are comparatively more difficult because 1) chronic diseases are more complicated to diagnose, 2) there are often multiple and poorly defined etiologic factors for disease, and 3) there are long latency periods between these factors and the onset of disease. In addition, the factors responsible may be behaviors or practices (or neglect thereof) of affected persons or health care providers.

Between 50% and 85% of the acute and chronic complications that are associated with diabetes and contribute to mortality are preventable or treatable (4). Previous studies on premature mortality among persons with diabetes have found that a significant proportion of deaths were due to preventable factors. The mortality rate for persons with diabetes who are 45 years of age has been reported to be 8 times that of the same age group in the general population (2). A detailed review of Washington State death certificates revealed that almost one-third of the deaths of persons 45 years of age with diabetes were due to acute complications for which there is definitive therapy (2). In a study in Great Britain, "neglect of diabetes" was considered a contributing factor in 27% of 447 deaths involving persons with diabetes who died at 50 years of age (5).

The rates of routine care practices of blood pressure check and urinalysis testing in this reported cohort were high. However, all persons with diabetes would be expected to have such basic examinations, especially in the last year of life when manifest complications of diabetes cause many to seek more health care. The rates of utilization of state-of-the-art care technologies, such as glycohemoglobin measurement or self-monitoring of blood glucose, were considerably lower and might be a more sensitive reflection of the quality of care. Finally, considering that this group is already at high risk for vascular disease, the rate of smoking was high.

With advancing age, the frequency of complications among persons with diabetes increases, and the potential to prevent complications and attendant mortality decreases. The greatest potential to prevent mortality and years of potential life lost from diabetes exists among the youngest age groups. Problems that can be remedied by improved health care for this age group could also be expected to be remediable for the rest of the diabetic population. Thus, recommendations for preventing future deaths among persons 45 years of age should benefit the entire diabetic population.

The combination of alcohol and diabetes appears to be of particular concern. Alcohol has well-observed adverse effects on the health of persons with diabetes. Metabolic disturbances and compromise

Disclaimer   All MMWR HTML documents published before January 1993 are electronic conversions from ASCII text into HTML. This conversion may have resulted in character translation or format errors in the HTML version. Users should not rely on this HTML document, but are referred to the original MMWR paper copy for the official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.

Page converted: 08/05/98

HOME  |  ABOUT MMWR  |  MMWR SEARCH  |  DOWNLOADS  |  RSS |  CONTACT POLICY  |  DISCLAIMER  |  ACCESSIBILITY

Morbidity and Mortality Weekly Report Centers for Disease Control and Prevention 1600 Clifton Rd, MailStop E-90, Atlanta, GA 30333, U.S.A Department of Health and Human Services

This page last reviewed 5/2/01