Mortality trends in men and women with diabetes, 1971 to 2000.

@article{citeulike:2066153, abstract = {BACKGROUND: Whether mortality rates among diabetic adults or excess mortality associated with diabetes in the United States has declined in recent decades is not known. OBJECTIVE: To examine whether all-cause and cardiovascular disease mortality rates have declined among the U.S. population with and without self-reported diabetes. DESIGN: Comparison of 3 consecutive, nationally representative cohorts. SETTING: Population-based health surveys (National Health and Nutrition Examination Surveys I, II, and III) with mortality follow-up assessment. PATIENTS: Survey participants age 35 to 74 years with and without diabetes. MEASUREMENTS: Diabetes was determined by self-report for each survey (1971-1975, 1976-1980, and 1988-1994), and mortality rates were determined through 1986, 1992, and 2000 for the 3 surveys, respectively. RESULTS: Among diabetic men, the all-cause mortality rate decreased by 18.2 annual deaths per 1000 persons (from 42.6 to 24.4 annual deaths per 1000 persons; P = 0.03) between 1971 to 1986 and 1988 to 2000, accompanying decreases in the nondiabetic population. Trends for cardiovascular disease mortality paralleled those of all-cause mortality, with 26.4 annual deaths per 1000 persons in 1971 to 1986 and 12.8 annual deaths per 1000 persons in 1988 to 2000 (P = 0.06). Among women with diabetes, however, neither all-cause nor cardiovascular disease mortality declined between 1971 to 1986 and 1988 to 2000, and the all-cause mortality rate difference between diabetic and nondiabetic women more than doubled (from a difference of 8.3 to 18.2 annual deaths per 1000 persons). The difference in all-cause mortality rates by sex among people with diabetes in 1971 to 1986 were essentially eliminated in 1988 to 2000. LIMITATIONS: Diabetes was assessed by self-report, and statistical power to examine the factors explaining mortality trends was limited. CONCLUSIONS: Progress in reducing mortality rates among persons with diabetes has been limited to men. Diabetes continues to greatly increase the risk for death, particularly among women.}, address = {National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia 30341-3727, USA.}, author = {Gregg, E. W. and Gu, Q. and Cheng, Y. J. and Narayan, K. M. and Cowie, C. C. }, citeulike-article-id = {2066153}, comment = {Most physicians and patients are interested in mortality and cardiovascular disease trends. NHANES has released new mortality information on both men and women in subjects with and without diabetes. One of the limitations of NHANES is self-reporting and the statistical power to examine the factors explaining mortality trends. Gregg et al reported on all cause and CV mortality in men and women with and without diabetes from NHANES I (1971–1986), II (1976–1992), and III (1988–2000) cohorts. The trends in all cause mortality for men without diabetes 19 annual deaths per 1000 persons, 14.6/1000, 11.6/1000 respectively (1986, 1992 and 2000). Men with diabetes annual deaths were 42.6/1000, 32.3/1000 and 24.4 per 1000. Note that men had a progressive decline in all cause mortality with or without diabetes. Women had an annual all cause mortality without diabetes of 10.1/1000 (NHANES I), 7.9/1000 (NHANES II), 7.7/1000 (NHANES III). Women with diabetes had an annual death rate of 18.4 per 1000 (NHANES I), 15.1 per 1000 (NHANES II) and 25.9 per 1000 (NHANES III) respectively. This revealed an increase that was not seen in men. Reasons for the difference are complex, possibly as pointed at from the analysis of 2005 Health Effectiveness Data and Information Set (HEDIS) cardiovascular disease and diabetes measures showed sex disparities for ambulatory preventive care. Many other studies also have found similar results as HEDIS. In summary the advantage seen in longevity in women vs men may be lost with the development of diabetes. The development of diabetes for both men and women, there is a double or tripling of all cause mortality. --Robert J Chilton, MD This report continues the lessons that repeated National Health and Nutrition Examination Surveys (NHANES) have taught us about diabetes in the United States. Although each of the three surveys conducted respectively in 1971-1975, 1976-1980 and 1988-1994 is a separate cross-sectional study of different cohorts, each cohort is population based, though omitting institutionalized persons. Thus the results are considered representative, although data which was self-reported, like a previous diagnosis of diabetes, is always subject to participant bias. With these limitations in mind, this report is astonishing and concerning. 12 year mortality rates were markedly decreased in diabetic men from the first to the third cohorts: all cause mortality from 42.6 to 24.4 deaths per year per thousand persons and cardiovascular mortality from 26.4 to 12.8. The results for diabetic women were dramatically different. The first to third cohort values were: all cause mortality from18.4 to 25.9 and cardiovascular mortality from 10.4 to 9.5 deaths per year per 1000 persons. Thus the difference in mortality between diabetic men and diabetic women from 1974 to 1986 was largely obliterated in the period from 1988 to 2000. Men’s mortality rates improved and women’s did not, though in the non-diabetic population, mortality decreased in both genders. The former diabetic women’s advantage in mortality over diabetic men was lost. The results remained the same after adjustment for age, ethnicity, and duration of diabetes or BMI. The reasons for this gender disparity in the trends for improvement in mortality rates are not clear. Nonetheless it behooves all primary care physicians, diabetologists and cardiologists to exert every effort to improve modifiable risk factors for CVD and provide the most scrupulous updated proven care when CVD events do occur, in women with diabetes. If there are biological reasons why resistance to improvement in mortality account for these differences, they must be found and dealt with as much as possible.--Saul Genuth, MD }, issn = {1539-3704}, journal = {Ann Intern Med}, keywords = {diabetes, epidemiology}, month = {August}, number = {3}, pages = {149--155}, posted-at = {2008-01-08 00:41:44}, priority = {2}, title = {Mortality trends in men and women with diabetes, 1971 to 2000.}, url = {http://view.ncbi.nlm.nih.gov/pubmed/17576993}, volume = {147}, year = {2007} }

TY - JOUR ID - citeulike:2066153 L3 - citeulike-article-id:2066153 TI - Mortality trends in men and women with diabetes, 1971 to 2000. JF - Ann Intern Med VL - 147 IS - 3 SP - 149 EP - 155 AD - National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia 30341-3727, USA. SN - 1539-3704 N2 - BACKGROUND: Whether mortality rates among diabetic adults or excess mortality associated with diabetes in the United States has declined in recent decades is not known. OBJECTIVE: To examine whether all-cause and cardiovascular disease mortality rates have declined among the U.S. population with and without self-reported diabetes. DESIGN: Comparison of 3 consecutive, nationally representative cohorts. SETTING: Population-based health surveys (National Health and Nutrition Examination Surveys I, II, and III) with mortality follow-up assessment. PATIENTS: Survey participants age 35 to 74 years with and without diabetes. MEASUREMENTS: Diabetes was determined by self-report for each survey (1971-1975, 1976-1980, and 1988-1994), and mortality rates were determined through 1986, 1992, and 2000 for the 3 surveys, respectively. RESULTS: Among diabetic men, the all-cause mortality rate decreased by 18.2 annual deaths per 1000 persons (from 42.6 to 24.4 annual deaths per 1000 persons; P = 0.03) between 1971 to 1986 and 1988 to 2000, accompanying decreases in the nondiabetic population. Trends for cardiovascular disease mortality paralleled those of all-cause mortality, with 26.4 annual deaths per 1000 persons in 1971 to 1986 and 12.8 annual deaths per 1000 persons in 1988 to 2000 (P = 0.06). Among women with diabetes, however, neither all-cause nor cardiovascular disease mortality declined between 1971 to 1986 and 1988 to 2000, and the all-cause mortality rate difference between diabetic and nondiabetic women more than doubled (from a difference of 8.3 to 18.2 annual deaths per 1000 persons). The difference in all-cause mortality rates by sex among people with diabetes in 1971 to 1986 were essentially eliminated in 1988 to 2000. LIMITATIONS: Diabetes was assessed by self-report, and statistical power to examine the factors explaining mortality trends was limited. CONCLUSIONS: Progress in reducing mortality rates among persons with diabetes has been limited to men. Diabetes continues to greatly increase the risk for death, particularly among women. KW - diabetes KW - epidemiology N1 - Most physicians and patients are interested in mortality and cardiovascular disease trends. NHANES has released new mortality information on both men and women in subjects with and without diabetes. One of the limitations of NHANES is self-reporting and the statistical power to examine the factors explaining mortality trends. Gregg et al reported on all cause and CV mortality in men and women with and without diabetes from NHANES I (1971–1986), II (1976–1992), and III (1988–2000) cohorts. The trends in all cause mortality for men without diabetes 19 annual deaths per 1000 persons, 14.6/1000, 11.6/1000 respectively (1986, 1992 and 2000). Men with diabetes annual deaths were 42.6/1000, 32.3/1000 and 24.4 per 1000. Note that men had a progressive decline in all cause mortality with or without diabetes. Women had an annual all cause mortality without diabetes of 10.1/1000 (NHANES I), 7.9/1000 (NHANES II), 7.7/1000 (NHANES III). Women with diabetes had an annual death rate of 18.4 per 1000 (NHANES I), 15.1 per 1000 (NHANES II) and 25.9 per 1000 (NHANES III) respectively. This revealed an increase that was not seen in men. Reasons for the difference are complex, possibly as pointed at from the analysis of 2005 Health Effectiveness Data and Information Set (HEDIS) cardiovascular disease and diabetes measures showed sex disparities for ambulatory preventive care. Many other studies also have found similar results as HEDIS. In summary the advantage seen in longevity in women vs men may be lost with the development of diabetes. The development of diabetes for both men and women, there is a double or tripling of all cause mortality. --Robert J Chilton, MD This report continues the lessons that repeated National Health and Nutrition Examination Surveys (NHANES) have taught us about diabetes in the United States. Although each of the three surveys conducted respectively in 1971-1975, 1976-1980 and 1988-1994 is a separate cross-sectional study of different cohorts, each cohort is population based, though omitting institutionalized persons. Thus the results are considered representative, although data which was self-reported, like a previous diagnosis of diabetes, is always subject to participant bias. With these limitations in mind, this report is astonishing and concerning. 12 year mortality rates were markedly decreased in diabetic men from the first to the third cohorts: all cause mortality from 42.6 to 24.4 deaths per year per thousand persons and cardiovascular mortality from 26.4 to 12.8. The results for diabetic women were dramatically different. The first to third cohort values were: all cause mortality from18.4 to 25.9 and cardiovascular mortality from 10.4 to 9.5 deaths per year per 1000 persons. Thus the difference in mortality between diabetic men and diabetic women from 1974 to 1986 was largely obliterated in the period from 1988 to 2000. Men’s mortality rates improved and women’s did not, though in the non-diabetic population, mortality decreased in both genders. The former diabetic women’s advantage in mortality over diabetic men was lost. The results remained the same after adjustment for age, ethnicity, and duration of diabetes or BMI. The reasons for this gender disparity in the trends for improvement in mortality rates are not clear. Nonetheless it behooves all primary care physicians, diabetologists and cardiologists to exert every effort to improve modifiable risk factors for CVD and provide the most scrupulous updated proven care when CVD events do occur, in women with diabetes. If there are biological reasons why resistance to improvement in mortality account for these differences, they must be found and dealt with as much as possible.--Saul Genuth, MD AU - Gregg, EW AU - Gu, Q AU - Cheng, YJ AU - Narayan, KM AU - Cowie, CC PY - 2007/08/07/ UR - http://view.ncbi.nlm.nih.gov/pubmed/17576993 ER -