adjustment for established diabetes risk factors. Depressive symptoms had no impact on diabetes diabetes was heterogeneous according to these characteris …
American Journal of Epidemiology Copyright 2003 by the Johns Hopkins Bloomberg School of Public Health All rights reserved
Vol 158, No 5 Printed in USA DOI: 101093/aje/kwg172
Symptoms of Depression as a Risk Factor for Incident Diabetes: Findings from the National Health and Nutrition Examination Epidemiologic Follow-up Study, 19711992
Mercedes R Carnethon1,2, Leslie S Kinder1, Joan M Fair1, Randall S Stafford1, and Stephen P Fortmann1
1 Stanford Center for Research in Disease Prevention and the Department of Medicine, Stanford University School of Medicine, Palo Alto, CA 2 Current affiliation: Department of Preventive Medicine, The Feinberg School of Medicine, Northwestern University, Chicago, IL
Received for publication August 12, 2002; accepted for publication February 28, 2003
Symptoms of depression may predict incident diabetes independently or through established risk factors for diabetes US men and women aged 2574 years who were free of diabetes at baseline n 6,190 were followed from 1971 to 1992 mean, 156 years; standard deviation, 6 for incident diabetes Depressive symptoms were measured by using the General Well-Being Depression subscale and were categorized to
compare persons with high 9, intermediate 32, and low 59 numbers of symptoms The incidence of diabetes was highest among participants reporting high numbers of depressive symptoms 73 per 1,000 person-years and did not differ between persons reporting intermediate and low numbers of symptoms 34 and 36 per 1,000 personyears, respectively p 001 for high vs low In the subset of participants with less than a high school education a marker of low socioeconomic status, the risk of developing diabetes was three times higher 95 confidence interval: 20, 47 for persons reporting high versus low numbers of depressive symptoms These results persisted following adjustment for established diabetes risk factors Depressive symptoms had no impact on diabetes incidence among persons with at least a high school education Results suggest an independent role for depressive symptoms in the development of diabetes in populations with low educational attainment depression; diabetes mellitus; educational status; longitudinal studies; social class
Abbreviations: CI, confidence interval; NHANES I, First National Health and Nutrition Examination Survey; NHEFS, National Health and Nutrition Examination
Epidemiologic Follow-up Survey; RR, relative risk
The elevated prevalence of comorbid depression among persons with type 1 and type 2 diabetes mellitus is well established Depression affects 1520 percent of patients with diabetes, three times the rate in the general population 1 However, the mechanism linking these disorders and the causal direction of the association is unclear There are a number of plausible explanations for the relation between depression and diabetes The most commonly proposed is that depression and hopelessness arise from having a chronic disease such as diabetes 24 Alternatively, one study suggested that clinical depression may precede incident diabetes 5 It is not known whether
depression independently increases the risk of developing diabetes or acts through established risk factors such as obesity, sedentary lifestyle, poor diet, and cigarette smoking 6 The objective of the current study was to test whether symptoms of depression preceded the onset of diabetes and whether this association is mediated through established risk factors for diabetes Depression is known to differ by demographic characteristics including socioeconomic status, health behaviors,
and comorbid illness 7, so we evaluated whether the association between depressive symptoms and diabetes was heterogeneous according to these characteris-
Correspondence to Dr Mercedes R Carnethon, Department of Preventive Medicine, The Feinberg School of Medicine, Northwestern University, 680 North Lake Shore Drive, Suite 1102, Chicago, IL 60611-4402 e-mail: carnethon@northwesternedu
416
Am J Epidemiol 2003;158:416423
Depressive Symptoms and Incident Diabetes 417
tics To our knowledge, this is one of the first investigations of this question in a nationally representative sample
MATERIALS AND METHODS Study design and population
Participants from the First National Health and Nutrition Examination Survey NHANES I who were followed as part of the National Health and Nutrition Examination Epidemiologic Follow-up Survey NHEFS were included in this study NHANES I, conducted from 1971 to 1975, was a cross-sectional survey of health conditions and healthrelated behaviors in a probability sample of the noninstitutionalized civilian population of the United States aged 174 years After the baseline examination, participants were contacted and medical and health care records were
abstracted in four cycles, 19821984, 1986, 19871989, and 19901992 A detailed description of the study design and sampling methods is available elsewhere 8, 9 The NHEFS traced 93 percent of the original sample of 14,407 adults aged 2574 years n 13,383 through 1992 10, 11 This study includes the subset of NHEFS participants who received a more detailed health examination at baseline n 6,910 From those participants, we excluded those with prevalent diabetes based on self-reported physician diagnosis, use of diabetes control medication, urine glucose test, or a diagnosis date prior to the first examination n 656 or those whose race was reported as other, because of small numbers n 64 This paper includes information on 6,190 participants: 2,858 men and 3,332 women
Measurements
The General Well-Being survey, administered during NHANES I at mobile examination centers by trained personnel, includes six independent but related components of general psychological well-being 12 This paper focuses on the cheerfulness versus depressed mood subscale, a series of four questions highly correlated to depressive symptomatology Previous research indicates that the General WellBeing Depression
subscale predicts clinically trained interviewers ratings of depression and is correlated with scores from other instruments designed to assess depression 12 Participants were asked the following questions: Have you felt down-hearted or blue?; How have you been feeling in general?; Have you felt so sad, discouraged, hopeless or had so many problems that you wondered if anything was worthwhile?; and, How DEPRESSED or CHEERFUL have you been [in the past month]? Responses are provided on a Likert scale, with lower values indicating a more depressed mood In this study, the association between depression and diabetes was examined both continuously 025 and in categories Categories for analysis were based on previous research indicating that scores of 012 correspond with high numbers of depressive symptoms 13 and scores of 1925 indicate few or no depressive symptoms 12 Secondary analyses included the similarly rated anxious versus relaxed subscale 12
Am J Epidemiol 2003;158:416423
Incident diabetes was defined by the report of this condition on any of the following: death certificates International Classification of Diseases, Ninth Revision, codes 2500 2509, health care facility records
nursing home or hospital, or self-report 11 Biochemical testing to detect diabetes was not available Previous research in this cohort indicates that at least half of all cases of diabetes were confirmed by two or more sources 14, 15 We could not identify whether incident diabetes was type 1 or type 2, but type 2 diabetes represents over 90 percent of diabetes diagnoses and is much more likely to develop after age 30 years 16 The diagnosis date for diabetes was recorded as the date recorded on death certificates or facility discharge records or at first report by the respondent In the case of incomplete data, the date was imputed on the basis of the amount of information available 17 Covariates associated with diabetes and depressive symptoms were ascertained from NHANES I surveys on medical history, health care needs, and cardiovascular conditions as well as from medical examinations that included laboratory determinants and anthropometric measurements 8, 9 Briefly, age at interview, gender, race Black or White, marital status, and educational attainment were ascertained via self-report NHANES I interviewers determined whether a participant resided in a low-poverty area by using
the US Census 9 On the basis of information collected from health interview surveys, we categorized level of recreational physical activity low, moderate, or high, cigarette smoking current, former, or never, and number of drinks of alcohol per day 0, 12, or 3 At the baseline medical examination, blood pressure was measured while participants were seated We identified a participant as hypertensive if he or she met one of the following conditions: systolic blood pressure of 140 mmHg, diastolic blood pressure of 90 mmHg, antihypertensive medication use, or a previous physician diagnosis of hypertension We calculated baseline body mass index as the ratio of standing height in meters, squared, to measured weight in kilograms Self-reported weight in pounds; 1 pound 0454 kg was ascertained during follow-up interviews for 5,038 participants; estimated weight change was calculated as the difference between self-reported weight during follow-up and measured weight at baseline converted from kilograms to pounds
Statistical methods
We compared the distribution of baseline characteristics by depressive symptom categories and evaluated statistical significance by using t tests for means and 2
tests for proportions Poisson regression was used to estimate diabetes incidence rates and corresponding rate ratios by depressive symptoms Follow-up time was calculated as the difference between the date last seen alive and baseline for persons without diabetes and as the difference between the diabetes diagnosis date and baseline for persons with diabetes Multivariable Cox proportional hazards regression was used to calculate relative risks and 95 percent confidence intervals of incident diabetes by depressive symptom categories and by a standard deviation increase in continuous depressive symptom scores We tested and confirmed the
418 Carnethon et al
validity of the proportional hazards assumption by using lognegative log survival plots All covariates were evaluated as potential effect modifiers heterogeneity by using firstorder interaction terms between each covariate and depressive symptom categories A significant p 005 change in the maximum likelihood 2 value following removal of the interaction term from the model indicated statistical interaction When there was evidence of effect modification, we retained the interaction term in the model To investigate the extent to
which established risk factors explain the association between depressive symptoms and incident diabetes, we calculated the excess risk attributable to depressive symptoms when groups of covariates were entered into the model We used the following formula: percentage excess risk RR1 RR2/RR1 1, where RR1 is the relative risk of diabetes for persons with high versus low numbers of depressive symptoms in the minimally adjusted model, RR2 is the relative risk after adjustment for a group of covariates, and RR1 1 is the excess risk of diabetes among persons with high versus low numbers of depressive symptoms 18 The resulting percentage explains how much of the association between depressive symptoms and diabetes can be explained by the covariates; the relative risk following adjustment represents the residual association between depressive symptoms and diabetes Older adults, low-income persons, and women of childbearing age were oversampled in NHANES I Corresponding sample weights are provided so investigators can make national prevalence estimates However, the objective of the current study was to examine the association between a specific risk factor, depressive symptoms, and the
risk of developing diabetes, not to provide national estimates Furthermore, there has been controversy surrounding the use of the sampling weights in NHEFS 19 Previous authors reported conducting both weighted and unweighted analyses, but, because the results were generally comparable, authors presented only the unweighted results Similarly, we report only unweighted results in this paper All analyses were conducted by using SAS software, version 81 SAS Institute, Inc, Cary, North Carolina Statistical significance is denoted at p 005
RESULTS Baseline characteristics
FIGURE 1 Incidence rates of diabetes per 1,000 person-years, by depressive symptoms category, in the full study sample A n 6,190 and stratified by education B high school HS education Educ n 2,600; high school education n 3,590, the National Health and Nutrition Examination Epidemiologic Follow-up Study, United States, 19711992
alence of hypertension by depressive symptoms On average, participants gained 47 pounds during follow-up standard deviation, 226 Weight change was similar between persons with high and low numbers of depressive symptoms, but persons with intermediate numbers of symptoms gained significantly
more weight than persons with low numbers of symptoms
Incident diabetes
At baseline, 3,618 585 percent participants reported low numbers of depressive symptoms, 2,006 324 percent reported intermediate numbers of symptoms, and 566 91 percent reported high numbers of symptoms There was no age difference across depressive symptom categories, but participants with high or intermediate numbers of symptoms were more likely to be women, to be Black, to be unmarried, to have less than a high school education, and to live in a low-poverty area table 1 Furthermore, participants with higher numbers of depressive symptoms were less physically active, were less likely to drink alcohol, had a higher baseline body mass index, were more often obese, and were more likely to be current cigarette smokers We found no differences in systolic or diastolic blood pressure or the prev-
Over an average of 156 years of follow-up maximum, 219; standard deviation, 55 years, 15 incident cases of diabetes were identified by death certificate, 362 were selfreported, and 49 were identified by a health care facility stay In total, 369 participants 6 percent developed diabetes The distribution of diabetes
incidence by age at follow-up was 43 percent, 173 percent, 274 percent, 274 percent, 203 percent, and 33 percent for participants aged 3544, 4554, 5564, 6574, 7584, and 85 years, respectively The incidence of diabetes was highest among participants reporting high numbers of depressive symptoms 73 per 1,000 personyears and did not differ between persons reporting intermediate and low numbers of symptoms 34 and 36 per 1,000 person-years, respectively figure 1A
Am J Epidemiol 2003;158:416423
Depressive Symptoms and Incident Diabetes 419
TABLE 1 Baseline characteristics of the study population, by depressive symptoms category, in the National Health and Nutrition Examination Epidemiologic Follow-up Study, United States, 19711975
Depressive symptoms no Covariate Low n 3,618 484 02 1,761 482 370 101 2,932 803 1,360 372 840 230 274 75 836 231 255 01 521 143 1,300 356 1336 04 844 02 1,443 395 41 04 Intermediate n 2,006 475 03 1,217 607 269 133 1,511 746 905 447 559 276 231 114 511 255 253 01 309 153 802 396 1325 05 842 03 784 387 59 06 High n 566 478 06 384 667 121 210 381 662 356 618 181 314 129 224 162 286 267 02 138 240 270 469 1321 09 856 05 251 436 37 11 High vs low 034 00001
00001 00001 00001 00001 00001 0004 00001 00001 00001 016 004 006 071
p value
Intermediate vs low 004 00001 00003 00001 00001 00001 00001 005 030 031 0003 008 054 055 0014
Age years mean SE Female gender no Black race no Married no Less than high school education no Residence in low-poverty area no Low level of physical activity no No alcohol intake no Body mass index mean SD Obesity body mass index of 30 no Current cigarette smoker no Systolic blood pressure mmHg mean SE Diastolic blood pressure mmHg mean SE Hypertension no Change in weight over follow-up pounds
p values from a chi-square test of proportions or t tests for least-squares means SE, standard error Zero alcohol drinks per day Ratio of standing height in meters, squared, to measured weight in kilograms Difference between measured baseline weight in pounds; 1 pound 0454 kg and self-reported weight at follow-up among 5,038 participants for whom information was available
A statistically significant difference was found in the association between depressive symptoms and diabetes for participants with less than a high school education and those with at least a high school education 2 2 df 625, p 004
Among persons with less than a high school education, the incidence rate of diabetes was 219 times higher 95 percent confidence interval CI: 150, 313 for those with high versus low numbers of depressive symptoms figure 1B In contrast, the same relation between high numbers of depressive symptoms and diabetes was not significant for persons with at least a high school education incidence rate ratio 117, 95 percent CI: 065, 212 No association was found between intermediate or low numbers of depressive symptoms and incident diabetes in either group There was some evidence of heterogeneity of effect by gender 2 2 df 446, p 011; the association between depressive symptoms and incident diabetes was strongest among women However, the difference in association was primarily in magnitude and was most likely attributable to the increased prevalence of women with high numbers of depressive symptoms table 1 Among persons with low levels of education, the effect was identical by gender Because heterogeneity by gender became less important in the presence of strong effect modification by education, this paper does not present a stratified analysis, but we included an interaction term between
gender and depressive symptom categories in all further modeling
Am J Epidemiol 2003;158:416423
Multivariate analyses
In the entire cohort, the age-, race-, and gender-adjusted relative risk of developing diabetes was 252 95 percent CI: 173, 367 for persons with high versus low numbers of depressive symptoms, and it was nonsignificantly elevated for persons with intermediate numbers of symptoms RR 124, 95 percent CI: 091, 170 vs low numbers of symptoms With adjustment for health behaviors current vs never or former smoking, current or former alcohol drinking vs never drinking, and low vs moderate or high levels of recreational physical activity and baseline body mass index, the risk of developing diabetes remained 86 percent higher 95 percent CI: 127, 271 for persons with high versus low numbers of depressive symptoms In fully adjusted models, the risk of developing diabetes increased 4 percent per standard deviation 45-unit increase in depressive symptoms 95 percent CI: 102, 107 Among persons with low levels of education, those with high depressive symptom scores were three times more likely to develop diabetes than those with low scores table 2 In this population, the risk
increased 8 percent per 45-unit increase in depressive symptoms The relative risk for diabetes attenuated only slightly after further adjustment for behavioral and anthropometric risk factors A small part 6 percent of the association between high numbers of depressive symptoms and diabetes was explained by health behaviors cigarette smoking, alcohol consumption, and physical activity With the addition of baseline body mass index into the multivari-
420 Carnethon et al
TABLE 2 Relative risk 95 
confidence interval of developing diabetes, by depressive symptoms and education, National Health and Nutrition Examination Epidemiologic Follow-up Study, United States, 19711992
Depressive symptoms category nos of symptoms; referent: low Education category High vs low Relative risk High school Model 1 2 3 High school Model 1 2 3 305 293 230 197, 470 190, 453 149, 354 117 114 105 078, 175 076, 170 071, 155 108 107 106 104, 111 104, 111 102,109 High vs low Reference 59 366 Continuous Reference 125 250 143 139 104 076, 268 074, 260 055, 197 126 123 135 086, 185 084, 181 091, 201 103 102 102 098, 107 098, 106 098, 106 95 CI Intermediate vs low Relative risk 95 CI Continuous depressive
symptoms score per 45 standard deviation-unit increase Relative risk 95 CI
excess risk explained
excess risk explained RR1 RR2/RR1 1, where RR1 is the relative risk of diabetes for high vs low levels of depression or per 45-unit decrease, RR2 is the relative risk of diabetes for high vs low levels of depression per 45 units adjusted for terms specified in the model, and RR1 1 represents the excess risk of diabetes for persons with high vs low levels of depression CI, confidence interval Model 1 was adjusted for age, race Black vs White, and gender; model 2 was adjusted for model 1 smoking status current vs former or never, no of alcohol drinks/day 0 vs 12 and 3 vs 12, and level of recreational physical activity low vs medium or high; and model 3 was adjusted for model 2 body mass index
able model, these factors accounted for 37 percent of the association between high numbers of depressive symptoms and diabetes In the subset of participants for whom data were available on self-reported weight change over follow-up, the effect estimates did not change when weight change was substituted for baseline body mass index or was added to a model that included baseline body mass
index
Secondary analyses
Symptoms of depression and anxiety are highly correlated r 075, p 00001 in this sample, so we conducted additional analyses to evaluate the specific role of depressive symptomatology, independent of anxiety, on diabetes incidence High numbers of depressive symptoms remained a strong, significant predictor of incident diabetes following adjustment for symptoms of anxiety RR 177, 95 percent CI: 109, 287 vs low numbers of symptoms and all other covariates in the total population In the sample of persons with lower levels of education, the risk of developing diabetes remained 219 times higher 95 percent CI: 131, 368 for those with high versus low numbers of depressive symptoms after adjustment for anxiety The same association was not significant for persons with at least a high school education Symptoms of anxiety explained an additional 8 percent excess risk 230 219/230 1 of the association between depressive symptoms and diabetes risk beyond behavioral and anthropometric risk factors in lesseducated persons
DISCUSSION
In this population-based sample, men and women with less than a high school education who reported the highest
numbers of symptoms of
depression were at increased risk of developing diabetes This relation appears to act independently of established risk factors for the development of diabetes and a related psychological construct, anxiety Biologic mechanisms that may explain the association between depressive symptoms and diabetes include inflammation, activation of the hypothalamic-pituitary-adrenal axis, or an interaction between genetic predisposition and depression or stress In cross-sectional studies, inflammatory markers including the cytokines interleukin-1, and tumor necrosis factor- 20 and C-reactive protein 21 were found to be elevated in depressed persons Two population studies 17, 22 reported that inflammatory markers are associated with the development of diabetes One suggested mechanism is that obesity or atherosclerosis is associated with the expression of low-grade inflammation that can be detected prior to the development of diabetes 17 Alternatively, inflammation may be associated with oxidative damage and the release of free radicals 23 that damage pancreatic cells 24, thus limiting the release of insulin The inflammatory process also may inhibit insulin uptake 25, a critical process in
glucose regulation Further research is needed into a possible role for inflammation regarding the relation between depressive symptoms and incident diabetes A dysregulation of the hypothalamic-pituitary-adrenal axis in depressed persons may result in elevated cortisol levels 26 Cortisol may antagonize the actions of insulinmediated glucose disposal or cause preferential deposition of fat in the abdomen visceral adiposity, which is a risk factor for developing diabetes It has also been shown that insulin sensitivity, an important mechanism in the development of type 2 diabetes, can be manipulated by treatment for depression In an experimental study, depressed patients had lower
Am J Epidemiol 2003;158:416423
Depressive Symptoms and Incident Diabetes 421
insulin sensitivity, but, when treated with heterocyclic antidepressants, depressive symptoms improved concurrently with insulin sensitivity, independent of changes in body weight 27 Depression is highly correlated with physiologic and psychological stress; therefore, it is possible that the reported relation between stress and hyperglycemia may also mediate the relation between depression and diabetes Stressful situations have
been shown to induce hyperglycemia in euglycemic animals 28 and in humans with a genetic predisposition toward developing diabetes 29 After Pima Indians, thought to have a genetic predisposition to developing diabetes, completed a mental arithmetic task, their fasting serum glucose levels were significantly higher and remained higher longer than those in Whites 29 This finding supports the proposal by Wales 30 that stress only precipitates clinical diabetes in persons predisposed to developing diabetes Depressed mood is highly correlated with health behaviors associated with the development of diabetes Previous research indicates that health behaviors such as sedentary lifestyle and poor diet that lead to weight gain are more prevalent among depressed persons 31 Thus, it was unexpected that self-reported weight change over time did not differ between persons with high and low numbers of depressive symptoms in this sample and that only intermediate numbers of symptoms were associated with significantly more weight gain However, we did confirm that elevated numbers of depressive symptoms were related to a sedentary lifestyle and that baseline body mass index based on measured weight
was higher and the proportion of obese persons body mass index of 30 at baseline was greater Our analysis did not include dietary components, so it is possible that control for lifestyle characteristics was incomplete Additionally, body weight during follow-up was selfreported, and its accuracy is unknown Our results may have been biased differentially if the likelihood of accurate selfreport was related to depressive symptoms or development of diabetes Despite these limitations, we estimated that 37 percent of the association between depressive symptoms and diabetes could be explained by health behaviors and baseline body mass index as measured in this sample The strong residual association between depressive symptoms and diabetes incidence following adjustment for baseline body mass index, which we hypothesize is related to diet unmeasured in this study and physical activity measured, suggests that depressive symptoms play an independent role in the development of diabetes We found that the relation between depressive symptoms and diabetes was characterized by strong heterogeneity by educational attainment, a proxy for socioeconomic status While persons with lower levels of
education were more likely to report high numbers of depressive symptoms than were their more highly educated counterparts in this study 13 percent vs 6 percent and others 7, the absolute number in each group was still substantial 350 vs 216 Furthermore, the proportion of persons who developed diabetes, although higher among persons with less education 72 percent vs 51 percent again, lower socioeconomic status has been reported as a strong risk factor for the development
Am J Epidemiol 2003;158:416423
of diabetes 32, 33, was not subject to sparse numbers in each strata Thus, statistical power to detect a relation was good in both educational strata Rather, we think that our findings represent a real not statistical artifact difference in the association between depressive symptoms and the risk of diabetes by educational attainment Socioeconomic status reflects material and financial resources, social prestige, and knowledge, all of which act independently or in combination to influence health-related behaviors Persons of lower socioeconomic status may be more likely to consume diets high in fats, carbohydrates, and alcohol; to smoke cigarettes more frequently; and to engage in a
sedentary lifestyle, all of which are predisposing factors for developing diabetes Persons of depressed mood and fewer economic and education resources may be especially vulnerable to these maladaptive behaviors; thus, depressed mood and low socioeconomic status may act synergistically to increase the risk of diabetes A second study in the NHEFS cohort that used the Center for Epidemiologic Studies Depression Scale measured in 19821984 second examination found no effect of depressive symptoms on diabetes incidence 34 The authors did not report investigating the association by education, and participants were followed for a relatively shorter time average, 42 years Another earlier study 5 tested the temporal relation between major depressive disorder and incident diabetes in a Baltimore, Maryland, population n 3,481 by using the National Institute of Mental Healths Diagnostic Interview Schedule to identify major depressive disorder In this study, Eaton et al 5 found a strong twofold increase in the risk of diabetes, but, probably because of the smaller sample size, were unable to confirm the statistical significance of this relation However, a previous study by Kawakami et al 35 of
2,764 Japanese men reported findings similar to our own Men reporting higher numbers of depressive symptoms according to the Zung Self-Rating Depression scale were more likely to develop diabetes Our large sample and length of follow-up afforded sufficient power to test our primary and secondary hypotheses Although we might have observed a stronger association by using a measure of clinical depression, such as that of Eaton et al 5, depression and diabetes share common symptoms, such as fatigue and weakness 2 The internal consistency of the General Well-Being Depression scale was 077 in previously reported studies in this sample 36, 37, and, in a 9year longitudinal study, the level of negative affect was relatively stable over time 38 Thus, we are confident about the validity, reliability, and robustness of our measure of depressive symptoms and about the related psychological components it may encompass In addition, our measure of depressive symptoms assessed affect rather than somatic symptoms of depression, so we may have avoided the possibility that persons reporting depressive symptoms were actually experiencing symptoms of diabetes at baseline Conclusions from this study must
be interpreted in light of some limitations Incident and prevalent diabetes were based on self-report, medical records, and death certificates These sources are less rigorous than definitions based on results from oral glucose tolerance testing or fasting serum glucose concentrations Both misclassification and an underdiag-
422 Carnethon et al
nosis of diabetes are possible since a clinical diagnosis is usually made 47 years after the beginning of the disease 39; an estimated 27 percent of the US population has undiagnosed diabetes 40 Consequently, a diagnosis of diabetes during follow-up may represent a progression from undiagnosed, nonsymptomatic disease at baseline to a clinical diagnosis In secondary analyses, we excluded 26 persons 7 percent of diabetes diagnoses who reported a diagnosis of diabetes during the first 10 years of follow-up; our results did not change data not shown However, if depressed persons are more likely to have undiagnosed diabetes, then our results may have been confounded We were also unable to confirm whether diabetes was type 1 or type 2 Because type 2 is more likely to be diagnosed after age 30 years 16, we conducted secondary analyses that
excluded 920 participants 19 diabetes events who were younger than age 30 years at baseline, and our results were similar data not shown The impact of depressive symptoms on diabetes, as reported in this study, suggests that practitioners may have an additional reason to institute targeted screening and treatment programs for depression in low-socioeconomic-status populations Strategies for diabetes prevention in this population could be more effective if this relation is addressed
ACKNOWLEDGMENTS
This work was conducted while Drs Carnethon and Kinder were National Research Service Award National Heart, Lung, and Blood Institute postdoctoral fellows at the Stanford Center for Research and Disease Prevention supported by Public Health Service Training Grant 5 T32 HL 07034
REFERENCES 1 Peyrot M Levels and risks of depression and anxiety symptomatology among diabetic adults Diabetes Care 1997;20:58590 2 Stevens DE, Merikangas KR, Merikangas JR Comorbidity of depression and other medical conditions In: Beckham EE, Leber WR, eds Handbook of depression New York, NY: The Guilford Press, 1995:147202 3 Lustman PJ, Griffith LS, Gavard JA, et al Depression in adults with diabetes Diabetes
Care 1992;15:16319 4 Lustman PJ The course of major depression in diabetes Psychosom Med 1997;59:2431 5 Eaton WW, Armenian H, Gallo J, et al Depression and risk for onset of type II diabetes: a prospective population-based study Diabetes Care 1996;19:1097102 6 Rewers M, Hamman RF Risk factors for non-insulin dependent diabetes In: National Diabetes Data Group, ed Diabetes in America Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disorders, 1995:179220 7 Kaelber CT, Moul DE, Farmer ME Epidemiology of depression Handbook of depression New York, NY: The Guilford Press, 1995:335 8 Plan and operation of the health and nutrition examination survey: United States–19711973 Vital Health Stat 1 1973;1:1
77 9 Miller HW Plan and operation of the health and nutrition examination survey United States–19711973 Vital Health Stat 1 1973;1:146 10 Engel A, Murphy RS, Maurer K, et al Plan and operation of the HANES I augmentation survey of adults 2574 years United States, 19741975 Vital Health Stat 1 1978;Jun:1110 11 Cox CS, Mussolino ME, Rothwell ST, et al Plan and operation of the NHANES I Epidemiologic Followup Study, 1992 Vital Health Stat 1
1997;Dec:1231 12 Fazio AF A concurrent validational study of the NCHS General Well-Being Schedule Vital Health Stat 2 1977;Sep:153 13 Zonderman AB Depression as a risk for cancer morbidity and mortality in a nationally representative sample Cardiovasc Res 1999;44:31524 14 Resnick HE Differential effects of BMI on diabetes risk among black and white Americans Diabetes Care 1998;21:182835 15 Lipton RB, Liao Y, Cao G Determinants of incident noninsulin-dependent diabetes mellitus among blacks and whites in a national sample The NHANES I Epidemiologic Follow-up Study Am J Epidemiol 1993;138:82639 16 Harris MI, Robbins DC Prevalence of adult onset IDDM in the US population Diabetes Care 1994;17:133740 17 Ford ES Leukocyte count, erythrocyte sedimentation rate, and diabetes incidence in a national sample of US adults Am J Epidemiol 2002;155:5764 18 Breslow NE, Day NE, eds Statistical methods in cancer research Vol 1 The analysis of case-control studies Lyon, France: International Agency for Research on Cancer, 1980: 768 IARC scientific publication no 32 19 Ingram DD, Makuc DM Statistical issues in analyzing the NHANES I Epidemiologic Followup Study Series 2: data evaluation and methods
research Vital Health Stat 2 1994;May:1 30 20 Appels A, Bar FW, Bar J, et al Inflammation, depressive symptomatology, and coronary artery disease Psychosom Med 2000;62:6015 21 Dotevall A, Hulthe J, Rosengren A, et al Autoantibodies against oxidized low-density lipoprotein and C-reactive protein are associated with diabetes and myocardial infarction in women Clin Sci 2001;101:13644 22 Schmidt MI, Duncan BB, Sharrett AR, et al Markers of inflammation and prediction of diabetes mellitus in adults Atherosclerosis Risk in Communities study: a cohort study Lancet 1999;353:164952 23 Paolisso G, DAmore A, Di Maro G, et al Evidence for a relationship between free radicals and insulin action in the elderly Metabolism 1993;42:65963 24 Rabinovitch A, Suarez WL, Thomas PD, et al Cytotoxic effects of cytokines on rat islets: evidence for involvement of free radicals and lipid peroxidation Diabetologia 1992;35:409 13 25 Paolisso G, DAmore A, Giugliano D, et al Pharmacologic doses of vitamin E improve insulin action in healthy subjects and non-insulin dependent diabetic patients Am J Clin Nutr 1993;57:6506 26 Bjorntorp P Body fat distribution, insulin resistance, and metabolic diseases Nutrition
1997;13:795803 27 Okamura F, Tashiro A, Utumi A, et al Insulin resistance in patients with depression and its changes during the clinical course of depression: minimal model analysis Metabolism 2000;49:125560 28 Surwit RS, Schneider MS, Feinglos MN Stress and diabetes mellitus Diabetes Care 1992;15:141322 29 Esposito-Del Puente A, Lillioja S, Bogardus C, et al Glycemic response to stress is altered in euglycemic Pima Indians Int J
Am J Epidemiol 2003;158:416423
Depressive Symptoms and Incident Diabetes 423
Obes 1994;18:76670 30 Wales JK Does psychological stress cause diabetes? Diabet Med 1995;12:10912 31 Wurtman RJ, Wurtman JJ Brain serotonin, carbohydratecraving, obesity and depression Obes Res 1995;3suppl 4: 477S480S 32 Socioeconomic status of women with diabetes–United States, 2000 MMWR Morb Mortal Wkly Rep 2002;51:1478,159 33 Robbins JM, Vaccarino V, Zhang H, et al Socioeconomic status and type 2 diabetes in African American and non-Hispanic white women and men: evidence from the Third National Health and Nutrition Examination Survey Am J Public Health 2001;91:7683 34 Saydah SH, Hill-Golden S, Brancati FL, et al Depression and risk of type 2 diabetes mellitus in the US
Abstract Diabetes 2002;51:A65 35 Kawakami N, Takatsuka N, Shimizu H, et al Depressive symp-
36 37 38 39 40
toms and occurrence of type 2 diabetes among Japanese men Diabetes Care 1999;22:10716 Cronbach L Essentials of psychological testing New York, NY: Harper-Collins, 1970 Jonas BS, Mussolino ME Symptoms of depression as a prospective risk factor for stroke Psychosom Med 2000;62:463 71 Costa PT Jr Longitudinal analyses of psychological well-being in a national sample: stability of mean levels J Gerontol 1987; 42:505 Harris MI, Klein R, Welborn TA, et al Onset of NIDDM occurs at least 47 years before clinical diagnosis Diabetes Care 1992;15:81519 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, 19881994 Diabetes Care 1998;21:51824
Am J Epidemiol 2003;158:416423
Source:gda-online.org
