Diabetes Treatments

and 95 women had diabetes, while 222 men and of established CHD vs diabetes on the risk of CHD mor the CHD and the diabetes log hazards was +0.73 (95 …

ORIGINAL INVESTIGATION

Sex Differences in Risk for Coronary Heart Disease Mortality Associated With Diabetes and Established Coronary Heart Disease
Sundar Natarajan, MD, MSc; Youlian Liao, MD; Guichan Cao, MS; Stuart R Lipsitz, ScD; Daniel L McGee, PhD

Background: The sex-specific independent effect of dia-

betes mellitus and established coronary heart disease CHD on subsequent CHD mortality is not known
Methods: This is an analysis of pooled data n5243 from the Framingham Heart Study and the Framingham Offspring Study with follow-up of 20 years At baseline 1971-1975, 134 men and 95 women had diabetes, while 222 men and 129 women had CHD Risk for CHD death was analyzed by proportional hazards models, adjusting for age, hypertension, serum cholesterol levels, smoking, and body mass index The comparative effect of established CHD vs diabetes on the risk of CHD mortality was tested by testing the difference in log hazards Results: The adjusted hazard ratios HRs with 95 confidence intervals CIs for death from CHD were 21 95

CI, 13-33 in men with diabetes only, and 42 95 CI, 32-56 in men with CHD only compared with men without diabetes or CHD The HR for CHD death was 38 95 CI, 22-66
in women with diabetes, and 19 95 CI, 11-34 in women with CHD The difference between the CHD and the diabetes log hazards was 073 95 CI, 072-075 in men and -065 95 CI, -068 to -063 in women
Conclusions: In men, established CHD signifies a higher

risk for CHD mortality than diabetes This is reversed in women, with diabetes being associated with greater risk for CHD mortality Current treatment recommendations for women with diabetes may need to be more aggressive to match CHD mortality risk Arch Intern Med 2003;163:1735-1740 threshold and goals of treatment equivalent to the goals for patients with established CHD Though diabetes has a greater effect on CHD mortality in women compared with men,7 the magnitude of sex differences in CHD mortality in individuals with diabetes, particularly in comparison to an accepted marker of increased risk like established CHD, have not been elucidated To determine this, evaluation of a population sample with both men and women in the study is needed The specific aims of this investigation were 1 to evaluate the independent effect of diabetes and established CHD on subsequent CHD mortality and 2 to determine the differential sex-specific effects of
diabetes on CHD mortality compared with established CHD
METHODS STUDY DESIGN AND STUDY SAMPLE This analysis used public use cohort data involving participants from the Framingham Heart Study8 or the Framingham Offspring

From the Ralph H Johnson Veterans Affairs Medical Center, Charleston, SC Dr Natarajan; the Center for Health Care Research Dr Natarajan, Department of Medicine Dr Natarajan, and Department of Biometry and Epidemiology Drs Liao and Lipsitz and Ms Cao, Medical University of South Carolina, Charleston; and Department of Statistics, Florida State University, Tallahassee Dr McGee The authors have no relevant financial interest in this article

ORONARY HEART disease CHD remains the leading cause of mortality and morbidity in developed countries,1 with approximately 30 dying of their first CHD event,2 emphasizing the need for aggressive preventive strategies Recent data indicate that individuals with diabetes but without established CHD have as high a risk for fatal CHD as persons with established CHD but without diabetes3 Prior studies have shown that diabetes has greater impact on womens risk for CHD than on men4 National data indicate that men with diabetes have seen a
slower decline in heart disease mortality than men without diabetes while women with diabetes have noted an increase in heart disease mortality,5 emphasizing the need to further understand the sex-specific magnitude of risk associated with diabetes Guidelines from national organizations such as the American Diabetes Association6 recommend aggressive management of other CHD risk factors in patients with diabetes, with the initiation

C

REPRINTED ARCH INTERN MED/ VOL 163, JULY 28, 2003 1735

WWWARCHINTERNMEDCOM

Downloaded from wwwarchinternmedcom on November 16, 2008 2003 American Medical Association All rights reserved

Study9 who were 35 to 74 years old during 1970-1975 The examination at which the participants had their complete lipid profile determined was defined as the baseline This corresponds to the first examination of the Framingham Offspring Study and the 11th examination for most in the Framingham Heart Study others had it in the 10th or 12th examination MEASUREMENTS Coronary heart disease was defined as myocardial infarction, coronary insufficiency, or angina pectoris The outcome measure was CHD mortality, which was ascertained by a panel of clinical investigators by
reviewing records that included detailed history, clinical findings, electrocardiograms, autopsy reports, and death certificates10 Briefly, CHD death was categorized as either sudden or nonsudden death Sudden death was defined as death within 1 hour from onset of symptoms where the death could not reasonably be attributed to some other nonCHD cause Nonsudden death was diagnosed if the terminal episode lasted longer than 1 hour, the available documentation suggested CHD as the cause, and no other cause could be ascribed Participants were considered to have probable diabetes based on 2 casual plasma glucose levels greater than 150 mg/dL 83 mmol/L or the use of hypoglycemic medications insulin or oral hypoglycemic agents in the Framingham Heart Study These individuals then had their records reviewed including glucose tolerance tests by the investigators and a final diagnosis of diabetes was made based on corroborating evidence A fasting plasma glucose level greater than 140 mg/dL 78 mmol/L or the use of hypoglycemic agents defined diabetes in the Framingham Offspring Study Smoking status was obtained by self-report and participants were classified as current smokers regular smoking in
the year prior to the visit and nonsmokers Hypertension was defined as systolic blood pressure of 140 mm Hg or higher, diastolic blood pressure of 90 mm Hg or higher, or taking antihypertensive medications11 Lipid measures included total cholesterol, high-density lipoprotein cholesterol HDL-C, and low-density lipoprotein cholesterol LDL-C Height and weight were measured during each visit and body mass index BMI was calculated as weight in kilograms divided by the square of the height in meters STATISTICAL ANALYSIS The analyses were performed separately by sex Baseline characteristics were compared for the 4 CHD and diabetes groups: neither CHD nor diabetes, CHD only, diabetes only, and both CHD and diabetes Life-table analysis was used to determine the cumulative CHD mortality rate and to produce CHD mortality curves for the 4 groups Coronary heart disease mortality was adjusted for baseline age using the direct method and a log-rank test was used to test the differences in survival The independent effect of diabetes or established CHD on CHD mortality was determined using proportional hazards models10 All multivariate analyses were adjusted for age, hypertension, smoking, serum
cholesterol either total and HDL-C or LDL-C and HDL-C, and BMI The risk of CHD death for the 3 CHD and diabetes groups CHD alone, diabetes alone, and both CHD and diabetes was evaluated using persons without diabetes or CHD as the reference To determine the effect of CHD severity on subsequent CHD mortality, patients with CHD were classified into more severe myocardial infarction and less severe coronary insufficiency or angina pectoris categories and evaluated as independent variables, along with diabetes and other covariates, in a multivariate proportional hazards model

Bootstrap resampling was used to compare the CHD and diabetes proportional hazards regression coefficients on the risk of death from CHD12 Two separate models were fit for each bootstrap sample The first model contained CHD and all of the covariates while the second model contained diabetes and all of the covariates The coefficients for CHD and diabetes were calculated for each bootstrap sample One thousand bootstrap samples were drawn and we used the empirical distribution of these samples to calculate a 95 confidence interval CI for the difference in regression coefficients To further evaluate sex differences
in the effect of diabetes and established CHD on CHD mortality, sex-diabetes, sexCHD, diabetes-CHD interactions as well as the sex-diabetesCHD interaction were tested in a hierarchical Cox model combining men and women All analyses were performed using the Statistical Analysis System13 RESULTS
Table 1 presents the baseline characteristics of the

pooled sample Of 5336 participants aged 35 to 74 years, 93 were not included in the analysis because of missing data Among the 2494 men and 2749 women, 134 men and 95 women had diabetes, while 222 men and 129 women had CHD Compared with women with CHD only, women with diabetes only had a higher proportion of smokers, higher BMI, were younger, and had similar blood pressures and lipid levels Men with diabetes only were younger, had slightly higher BMI, similar proportion of smokers, lower total cholesterol and LDL-C levels, and similar blood pressure compared with men with CHD only The distribution of the type of CHD differs in men and women, with fewer than half the prevalent CHD cases in men being classified as angina and about two thirds of CHD cases in women classified as angina The age-adjusted CHD mortality curves for men and women
are displayed in the Figure The CHD mortality for participants with both diabetes and CHD was substantially greater than other groups The age-adjusted 20year CHD mortality was 58 in men and 34 in women Men with CHD alone had a significantly greater CHD mortality than men with diabetes alone log-rank test, P 001, with CHD mortality of 44 and 23, respectively, at 20 years This was reversed in women, with diabetic women being at greater risk for CHD mortality than women with only CHD log-rank test, P 05, with CHD mortality of 19 and 6, respectively, at 20 years Though men had higher overall cumulative CHD mortality than women, the age-adjusted CHD mortality rate for women with diabetes was higher than men without diabetes and approached the rate for men with diabetes To evaluate the risk of CHD mortality attributable to diabetes and/or established CHD, proportional hazards regression analyses adjusted for age and other covariates were undertaken Table 2 After multivariate adjustment, the hazard ratio HR for CHD death in men with diabetes only was 21 95 CI, 13-33 while it was 42 95 CI, 32-56 in men with CHD only In women, the corresponding HRs were 38 95 CI, 22- 66 and 19 95 CI, 11-34,
respectively The results were very similar when LDL-C instead of total cholesterol was included in the model
WWWARCHINTERNMEDCOM

REPRINTED ARCH INTERN MED/ VOL 163, JULY 28, 2003 1736

Downloaded from wwwarchinternmedcom on November 16, 2008 2003 American Medical Association All rights reserved

Table 1 Baseline Characteristics of the Study Sample
Men Diabetes Variable Participants, No Age, y BMI Current smoking, Total cholesterol, mg/dL HDL-C, mg/dL LDL-C, mg/dL Systolic BP, mm Hg Diastolic BP, mm Hg CHD categories, No MI CI AP CHD 31 12 626 68 27 27 48 226 43 39 10 145 38 140 23 82 13 19 613 5 161 7 226 No CHD 103 41 55 102 284 43 47 210 47 41 11 134 33 143 23 86 12 No Diabetes CHD 191 77 594 89 268 37 46 223 42 41 12 147 42 141 20 84 11 96 503 12 63 83 435 No CHD 2169 870 507 101 27 35 47 216 39 45 13 141 35 133 19 84 11 CHD 20 07 627 88 29 9 45 234 48 48 16 140 45 165 31 86 17 5 25 1 5 14 70 Diabetes No CHD 75 27 589 101 291 63 37 247 48 49 13 162 44 149 24 85 12 Women No Diabetes CHD 109 40 644 69 269 52 16 252 49 54 17 166 45 151 25 84 11 27 248 10 92 72 661 No CHD 2545 926 519 103 254 44 38 222 44 58 16 141 40 130 21
80 11

Abbreviations: AP, angina pectoris; BMI, body mass index calculated as weight in kilograms divided by the square of height in meters; BP, blood pressure; CHD, coronary heart disease; CI, coronary insufficiency; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; MI, myocardial infarction SI conversion factor: To convert cholesterol to millimoles per liter, multiply by 002586 Data are given as mean SD unless otherwise specified Some of the percentages may not sum to 100 because of rounding

To determine if the differential risk for CHD mortality in men and women is due to differences in severity of CHD, in the multivariate analysis, we separated CHD into 2 groups: myocardial infarction and angina pectoris/ coronary insufficiency In men, diabetes had an HR for CHD mortality of 17 95 CI, 12-25, angina pectoris/ coronary insufficiency had an HR of 32 95 CI, 2245, and myocardial infarction had an HR of 50 95 CI, 36-69 In women, the corresponding HRs were 36 95 CI, 22-59 for diabetes, 15 95 CI, 09-27 for angina pectoris/coronary insufficiency, and 31 95 CI, 12-76 for myocardial infarction Thus, men with prior myocardial infarction or other
forms of CHD were at a higher risk for CHD death than men with diabetes In women, diabetes still conferred a higher risk than the 2 CHD groups To compare the magnitude of risk for CHD mortality in individuals with CHD with the magnitude of risk in individuals with diabetes, the difference in regression coefficients equivalent to log hazard ratios between CHD and diabetes was determined The difference between the CHD coefficient and the diabetes coefficient in men was 073 95 CI, 072-075 This indicates that, in men, established CHD has a greater magnitude of risk for CHD mortality than diabetes HR, 208; 95 CI, 205-212 In contrast, the difference was -066 95 CI, -068 to -063 in women, implying lower risk of CHD death from prior CHD than diabetes HR, 052; 95 CI, 051-053 To further evaluate sex differences in the relationship between diabetes and CHD on CHD mortality, sex, diabetes, and CHD interactions were tested with men and women combined using hierarchical modeling principles Because the sex-diabetes-CHD interaction P96 and the CHD-diabetes interaction P 31 were not significant, they were not included in the final model The sex-CHD interaction was associated with an HR of 050 95
CI, 029-086, which indicates that the relative risk

for fatal CHD among women with CHD is significantly lower than the relative risk for men with CHD In contrast, the sex-diabetes interaction was associated with an HR of 231 95 CI, 126-423, indicating that the relative risk for CHD death in women with diabetes is higher than the relative risk for CHD death among men with diabetes
COMMENT

The findings from this prospective, community-based study emphasize the magnitude of diabetes as a major risk factor for CHD mortality in men and women These findings quantify sex differences in the risk for CHD mortality in individuals with diabetes by comparing it with established CHD In men, while diabetes is an important risk factor for fatal CHD, established CHD is associated with a larger magnitude of risk In women, the magnitude of the association is reversed and diabetes is a larger risk for fatal CHD than established CHD In both men and women, individuals with both diabetes and CHD were at dramatically higher risk Though it is well known that the CHD mortality rate in general is lower in women than in men of the same age, the age-adjusted CHD mortality rate in diabetic women is higher
than in men without diabetes and approaches the mortality rate seen in men with diabetes The sex difference in the relative magnitude of risk for CHD mortality may be explained by several biological mechanisms In our analysis, diabetic women without CHD were more likely to smoke, have lower HDL-C and lower LDL-C levels compared with nondiabetic women with CHD However, even after adjusting for these and other risk factors, diabetes was associated with a significant increased risk for CHD mortality Data from the Nurses Health Study14 indicate that at any level of other risk factors, women with diabetes are more likely to have
WWWARCHINTERNMEDCOM

REPRINTED ARCH INTERN MED/ VOL 163, JULY 28, 2003 1737

Downloaded from wwwarchinternmedcom on November 16, 2008 2003 American Medical Association All rights reserved

A
60

Table 2 Rate of Fatal CHD and Its Relationship to Diabetes and Established CHD in Men and Women
CHD Hazard Ratio 95 CI Deaths, Rate/1000 No Person-Years Age Adjusted Multivariate Neither CHD only Diabetes only Both diabetes and CHD Neither CHD only Diabetes only Both diabetes and CHD 172 78 19 13 Men 45 329 121 478 10 Reference 48 36-63 22 13-35 69 39-123 10 Reference 42
32-56 21 13-33 61 34-109

50

CHD Mortality,

40

30

20

10

0 0 5 10 15 20

76 14 16 7

Women 16 10 Reference 83 23 13-41 138 52 30-90 317 92 42-204

10 Reference 19 11-34 38 22-66 54 24-123

B
60 CHD DM CHD DM Neither

Abbreviations: CHD, coronary heart disease; CI, confidence interval Adjusted for age, smoking, hypertension, total serum cholesterol, high-density lipoprotein cholesterol, and body mass index

50

CHD Mortality,

40

30

20

10

0 0 5 10 15 20

Years of Follow-up

Age-adjusted cumulative coronary heart disease CHD mortality by CHD and diabetes mellitus DM status for men A and women B

cardiovascular events than women without diabetes Women with diabetes have been shown to have lower HDL-C and higher triglyceride levels than men with diabetes15 Diabetes has been demonstrated to have greater adverse effects in women with regard to waist-to-hip ratio, LDL-C, HDL-C, LDL particle size, apolipoprotein B, apolipoprotein A1, and fibrinogen16 Compared with diabetic men, diabetic women may have greater levels of lipid peroxidation, independent of glycemic control17 In addition to the other CHD risk factors, excess circulating glucose may adversely affect the
estrogen-related cardiovascular protection by decreasing vascular and platelet nitric oxide production,18 thereby increasing vascular tone, platelet aggregation, and enhance vascular proliferation While premenopausal nondiabetic women have greater endothelium-dependent vasodilation than nondiabetic men, premenopausal diabetic women have significant impairment of endothelial function, leading to endothelial dysfunction similar to diabetic men19 In addition to these markers of increased risk, since women have less severe coronary atherosclerosis and less collateral vessels than men,20 they tend to sustain greater myocardial damage with coronary occlusion and thus diabe-

tes may impact women more than men, both for CHD morbidity and mortality21,22 For example, in the Framingham Study, 66 of CHD deaths in women occurred in those without prior angina23 Because the weight of evidence indicates that diabetes and CHD have marked sex differences in subsequent CHD rates, it is crucial to analyze the data by sex24 This analytic approach is probably responsible for the differences between this study and the previous study,3 which did not formally test for sex differences Haffner et al,3
combining Finnish men and women, compared the risk for fatal CHD in 890 diabetic individuals without prior myocardial infarction 48 female with 69 nondiabetic individuals with prior myocardial infarction 26 female They found an HR for fatal CHD of 12 95 CI, 06-24 and inferred that the risk associated with diabetes and that associated with previous CHD were similar Though prior studies have shown a greater impact of diabetes in women compared with men, they have not determined the relative strength of the relationship compared with established CHD by sex7,25,26 Hu et al27 evaluated the impact of diabetes and myocardial infarction on CHD mortality using self-reported data from the Nurses Health Study only women with 20 years follow-up They reported a relative risk RR of 87 95 CI, 74-103 associated with diabetes and an RR of 106 95 CI, 81138 with myocardial infarction However, they did not directly compare the risks associated with diabetes and CHD using formal statistical procedures Also, in a validation study of this cohort, only 68 of self-reported myocardial infarction cases were actually confirmed to have myocardial infarction28 In addition, all women did not have uniform
assessment for diabetes and thus there may have been contamination of the nondiabetic reference group with undiagnosed diabetics,29 which could have led to underestimation of relative risk among women with diabetes In a report from the Physicians Health Study only US male physicians using self-reported information on diabetes, CHD and risk factors, Lotufo et al30 found
WWWARCHINTERNMEDCOM

REPRINTED ARCH INTERN MED/ VOL 163, JULY 28, 2003 1738

Downloaded from wwwarchinternmedcom on November 16, 2008 2003 American Medical Association All rights reserved

that a history of prevalent CHD was associated with greater relative risk of fatal CHD RR, 54; 95 CI, 47-62 than prevalent diabetes RR, 29; 95 CI, 23-37 Because the Nurses Health Study and the Physicians Health Study are 2 separate studies with very different study designs and populations, it is impossible to evaluate sex differences from them directly Our analysis found that men and women with both diabetes and CHD were at greatest risk for CHD death, which is consistent with other studies Compared with individuals without CHD or diabetes at baseline, prior research has reported that women with both diabetes and CHD27 had an RR
for CHD mortality of 176 95 CI, 105-294 while for men the RR was 12 95 CI, 99-14630 Malmberg et al31 found that prior diabetes in a patient recently hospitalized for unstable angina or nonQ-wave myocardial infarction was associated with a 2-year cardiovascular mortality rate of 93, with greater adverse impact of diabetes in women compared with men Miettinen et al32 reported a high mortality rate in diabetic patients after their first myocardial infarction, with the difference being particularly high in women The results from this investigation should be interpreted while taking into account certain potential limitations First, this community-based study comprised almost totally white participants and thus this same effect may not be seen in nonwhite persons Second, information regarding family history of CHD, renal function, severity of diabetes, abdominal obesity, physical activity, homeostatic factors, inflammatory markers, other vascular risk factors, and socioeconomic status was not available Therefore, we were unable to adjust for these potential confounders Third, because angina is a less sensitive and specific symptom of coronary disease in women, a certain proportion of
women reporting angina may be misclassified as having CHD Even when severity of CHD was considered in the analysis, men with prior myocardial infarction or other forms of CHD were at a higher risk for CHD death than men with diabetes, while in women diabetes conferred a higher risk than the CHD groups Finally, this study followed up participants over a 20-year period and these analyses have not accounted for differences in diagnostic criteria and treatment for diabetes and CHD over this period Despite these potential limitations, this analysis adds to the body of knowledge regarding the effect of diabetes on CHD mortality by quantifying the dramatic impact of diabetes in women after accounting for other known CHD risk factors The findings from this study support aggressive management of diabetes to prevent CHD, particularly in women While there may be a decrease in CHD events such as myocardial infarction with intensive glycemic control,33 the benefits from aggressive treatment of hypertension,34 dyslipidemia,35 and platelet responsiveness36 are unambiguous Of public health concern, estimates indicate that the number of persons with diabetes is likely to double in the first quarter
of the 21st century with a corresponding increase in social and financial burden37 A recent costeffectiveness analysis38 found that treating dyslipidemia in diabetic patients without cardiovascular disease 506323792 per year of life saved was as cost-effective as treat-

ing nondiabetic patients with cardiovascular disease 8799-21628 per year of life saved Based on our data, since women are at higher risk, it is likely that treatment of women with diabetes will be even more costeffective Since the intensity of management of diabetic patients is based on their risk for CHD, and because women with diabetes may be at higher risk for CHD than women with established CHD, current guidelines for treatment of women with diabetes may need to be more aggressive In conclusion, this community-based prospective study identifies diabetes as worse than prior established CHD in risk for subsequent CHD mortality in women In men, prior CHD has greater risk for subsequent fatal CHD than diabetes This analysis should provide the impetus to further refine treatment guidelines to match the intensity of treatment to patients risk for future CHD events Accepted for publication October 31, 2002 We thank
John A Colwell, MD, PhD, and Robert Fletcher, MD, for reviewing earlier versions of the manuscript Public use Framingham Heart Study and Framingham Offspring Study data were obtained from the National Heart, Lung, and Blood Institute, Bethesda, Md The views expressed in this article are those of the authors and do not necessarily reflect those of this agency This study was supported in part by the Department of Veterans Affairs Health Services Research and Development Career Development Award RCD 000211 and Public Health Service grants DK52329 National Institute of Diabetes, Digestive and Kidney Diseases, National Institutes of Health HL67460, HL68900, HL52329 National Heart Lung and Blood Institute, National Institutes of Health, and AHRQ 10871 Agency for Healthcare Research and Quality Corresponding author: Sundar Natarajan, MD, MSc, 423 E 23rd St, Room 11101-S, New York, NY 10010 e-mail: sundarnatarajan@mednyuedu
REFERENCES
1 Labarthe DR Epidemiology and Prevention of Cardiovascular Diseases: A Global Challenge Gaithersburg, Md: Aspen Publishers; 1998 2 Chambless L, Keil U, Dobson A, et al Population versus clinical view of case fatality from acute coronary heart disease:
results from the WHO MONICA Project 1985-1990: Multinational Monitoring of Trends and Determinants in Cardiovascular Disease Circulation 1997;96:3849-3859 3 Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction N Engl J Med 1998;339:229234 4 Mosca L, Grundy SM, Judelson D, et al Guide to Preventive Cardiology for Women: AHA/ACC Scientific Statement Consensus panel statement Circulation 1999; 99:2480-2484 5 Gu K, Cowie CC, Harris MI Diabetes and decline in heart disease mortality in US adults JAMA 1999;281:1291-1297 6 American Diabetes Association: clinical practice recommendations 2001 Diabetes Care 2001;24suppl 1:S1-121 7 Barrett-Connor EL, Cohn BA, Wingard DL, Edelstein SL Why is diabetes mellitus a stronger risk factor for fatal ischemic heart disease in women than in men? the Rancho Bernardo Study JAMA 1991;265:627-631 8 Dawber TR, Meadors GF, Moore FE Epidemiologic approaches to heart disease: the Framingham Study Am J Public Health 1951;41:279-286 9 Kannel WB, Feinleib M, McNamara PM, Garrison RJ, Castelli WP An investiga-

REPRINTED
ARCH INTERN MED/ VOL 163, JULY 28, 2003 1739

WWWARCHINTERNMEDCOM

Downloaded from wwwarchinternmedcom on November 16, 2008 2003 American Medical Association All rights reserved

10

11

12 13 14

15

16

17 18 19

20

21

22

23

24

tion of coronary heart disease in families: the Framingham Offspring Study Am J Epidemiol 1979;110:281-290 Shurtleff D Some characteristics related to the incidence of cardiovascular disease and death: Framingham Heart Study 18-year follow-up In: Kannel WB GT, ed The Framingham Study: An Epidemiological Investigation of Cardiovascular Disease Section 30 Washington, DC: US Government Printing Office; 1974:1725 The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure Arch Intern Med 1997;157:24132446 Efron B An Introduction to the Bootstrap New York, NY: Chapman Hall; 1993 SAS/STAT Users Guide Cary, NC: SAS Institute Inc; 1999 Manson JE, Colditz GA, Stampfer MJ, et al A prospective study of maturityonset diabetes mellitus and risk of coronary heart disease and stroke in women Arch Intern Med 1991;151:1141-1147 Walden CE, Knopp RH, Wahl PW, Beach KW, Strandness E Jr Sex differences in
the effect of diabetes mellitus on lipoprotein triglyceride and cholesterol concentrations N Engl J Med 1984;311:953-959 Howard BV, Cowan LD, Go O, Welty TK, Robbins DC, Lee ET Adverse effects of diabetes on multiple cardiovascular disease risk factors in women: the Strong Heart Study Diabetes Care 1998;21:1258-1265 Evans RW, Orchard TJ Oxidized lipids in insulin-dependent diabetes mellitus: a sex-diabetes interaction? Metabolism 1994;43:1196-1200 Sowers JR Insulin and insulin-like growth factor in normal and pathological cardiovascular physiology Hypertension 1997;29:691-9 Steinberg HO, Paradisi G, Cronin J, et al Type II diabetes abrogates sex differences in endothelial function in premenopausal women Circulation 2000;101: 2040-2046 Chaitman BR, Bourassa MG, Davis K, et al Angiographic prevalence of highrisk coronary artery disease in patient subsets CASS Circulation 1981;64: 360-367 Rentrop KP, Thornton JC, Feit F, Van Buskirk M Determinants and protective potential of coronary arterial collaterals as assessed by an angioplasty model Am J Cardiol 1988;61:677-684 Williams DO, Amsterdam EA, Miller RR, Mason DT Functional significance of coronary collateral vessels in patients with
acute myocardial infarction: relation to pump performance, cardiogenic shock and survival Am J Cardiol 1976;37: 345-351 Kannel WB, Abbott RD Incidence and prognosis of myocardial infarction in women: the Framingham Study In: Eaker ED, Wenger NK, Clarkson TB, Tyroler HA, eds Coronary Heart Disease in Women Bethesda, Md: National Heart, Lung, and Blood Institute, National Institutes of Health; 1987 Committee on Understanding the Biology of Sex and Gender Differences, Institute of Medicine Sex affects health In: Wizemann TM, Pardue M, eds Explor-

25

26

27

28

29

30

31

32

33

34

35

36 37 38

ing the Biological Contributions to Human Health: Does Sex Matter? Washington, DC: National Academy Press; 2001:93-136 Gu K, Cowie CC, Harris MI Mortality in adults with and without diabetes in a national cohort of the US population, 1971-1993 Diabetes Care 1998;21:11381145 Abbott RD, Donahue RP, Kannel WB, Wilson PW The impact of diabetes on survival following myocardial infarction in men vs women: the Framingham Study JAMA 1988;260:3456-3460 Hu FB, Stampfer MJ, Solomon CG, et al The impact of diabetes mellitus on mortality from all causes and coronary heart disease in women: 20 years of
followup Arch Intern Med 2001;161:1717-1723 Colditz GA, Martin P, Stampfer MJ, et al Validation of questionnaire information on risk factors and disease outcomes in a prospective cohort study of women Am J Epidemiol 1986;123:894-900 Harris MI, Flegal KM, Cowie CC, et al Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in US adults: the Third National Health and Nutrition Examination Survey, 1988-1994 Diabetes Care 1998;21:518-24 Lotufo PA, Gaziano JM, Chae CU, et al Diabetes and all-cause and coronary heart disease mortality among US male physicians Arch Intern Med 2001;161:242247 Malmberg K, Yusuf S, Gerstein HC, et al Impact of diabetes on long-term prognosis in patients with unstable angina and non-Q-wave myocardial infarction: results of the OASIS Organization to Assess Strategies for Ischemic Syndromes Registry Circulation 2000;102:1014-1019 Miettinen H, Lehto S, Salomaa V, et al Impact of diabetes on mortality after the first myocardial infarction: the FINMONICA Myocardial Infarction Register Study Group Diabetes Care 1998;21:69-75 Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of
complications in patients with type 2 diabetes UKPDS 33: UK Prospective Diabetes Study UKPDS Group Lancet 1998; 352:837-53 Staessen JA, Thijs L, Gasowski J, Cells H, Fagard RH Treatment of isolated systolic hypertension in the elderly: further evidence from the Systolic Hypertension in Europe Syst-Eur trial Am J Cardiol 1998;82:20R-22R Haffner SM, Alexander CM, Cook TJ, et al Reduced coronary events in simvastatintreated patients with coronary heart disease and diabetes or impaired fasting glucose levels: subgroup analyses in the Scandinavian Simvastatin Survival Study Arch Intern Med 1999;159:2661-2667 Colwell JA Aspirin therapy in diabetes Diabetes Care 1997;20:1767-1771 King H, Aubert RE, Herman WH Global burden of diabetes, 1995-2025: prevalence, numerical estimates, and projections Diabetes Care 1998;21:1414-1431 Grover SA, Coupal L, Zowall H, Alexander CM, Weiss TW, Gomes DR How costeffective is the treatment of dyslipidemia in patients with diabetes but without cardivascular disease? Diabetes Care 2001;24:45-50

REPRINTED ARCH INTERN MED/ VOL 163, JULY 28, 2003 1740

WWWARCHINTERNMEDCOM

Downloaded from wwwarchinternmedcom on November 16, 2008 2003 American Medical
Association All rights reserved