Diabetes, Major Depression, and Functional Disability Among U.S. Adults

Data from the sample adult core of 1999 NHIS (19) were analyzed. The NHIS is an ongoing national household survey of nonmilitary and noninstitutionalized individuals in the U.S. sponsored by the National Center for Health Statistics of the Centers for Disease Control and Prevention. Adults aged ≥18 years were randomly selected to respond to a computer-assisted personal interview questionnaire. The sample was selected by a complex sampling design involving stratification, clustering, and multistage sampling with a nonzero probability of selection for each person. Final weights were constructed for the NHIS to reflect the unequal probability of selection and to adjust for nonresponse and poststratification so that estimates from the NHIS can be generalized to the adult civilian population of the U.S. Details about the methodology of the 1999 NHIS are available (19,20).

In 1999, the NHIS introduced the World Health Organization Composite International Diagnostic Interview Short Form (CIDI-SF) as part of the survey questionnaire. The CIDI-SF is a diagnostic interview designed for use by trained interviewers who are not clinicians. The CIDI-SF was developed from the longer and more complex CIDI (21), and it was revised to screen for disorders defined in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (22). The CIDI-SF is a valid and reliable diagnostic interview and has classification accuracy of 93% for major depressive disorder (23). The classification accuracy of the CIDI-SF is derived from comparing the CIDI-SF cases and noncases to those derived from structured clinical interviews performed as part of the National Comorbidity Survey (24).

A complete copy of the CIDI-SF questions and scoring instructions is available from the World Health Organization website (www.who.int/msa/cidi/index.htm). The CIDI-SF uses a stem-branch logic in which a small number of initial diagnostic stem questions are used in each section to skip-out people who are least likely to be considered case subjects before they are asked further symptom questions (25). There are two ways to meet the diagnostic stem requirement for major depressive disorder: either by endorsing all questions about having 2 weeks of dysphoric mood or by endorsing all questions about having 2 weeks of anhedonia. In addition, the symptoms of dysphoric mood and anhedonia should last at least most of the day almost every day. Respondents who deny either the existence of symptoms or the persistence of symptoms are defined as not having major depressive disorder. If the respondent endorses dysphoric mood, seven additional questions are asked about losing interest, feeling tired, change in weight, difficulty sleeping, trouble concentrating, feeling down, and thoughts about death, and then a summary major depressive disorder score is calculated based on positive responses to these additional seven questions (range 0–7).

Similarly, respondents who endorse anhedonia are asked additional symptom questions, including questions about losing interest, feeling tired, change in weight, difficulty sleeping, trouble concentrating, feeling down, and thoughts about death. A summary major depressive disorder score is also calculated based on number of positive responses (range 0–7). Based on the recommendations for scoring (25), an individual was classified as having major depression if they endorsed the stem questions and had positive responses to three or more of the symptom questions. Individuals who endorsed the stem questions but had fewer than three positive responses to the symptoms questions were defined as not having major depression.

In 1999, the NHIS assessed functional status by asking respondents about their ability to perform 12 routine tasks without special equipment. These questions, derived from the work of Nagi (26), Nagi and Marsh (27), and Rosow and Breslau (28), were first incorporated into the NHIS in 1997 (11). Respondents were asked the following questions: “By yourself and without using any special equipment, how difficult is it for you to: 1) walk a quarter of a mile (approximately three city blocks); 2) walk up 10 steps without resting; 3) stand or be on your feet for approximately 2 h; 4) sit for approximately 2 h; 5) stoop, bend, or kneel; 6) reach up over your head; 7) use your fingers to grasp or handle small objects; 8) lift or carry something as heavy as 10 pounds such as a full bag of groceries; 9) push or pull large objects like a living room chair; 10) go out to things like shopping, movies, or sporting events; 11) participate in social activities such as visiting friends, attending clubs and meetings, or going to parties; 12) do things to relax at home or for leisure (reading, watching TV, sewing, listening to music)?”

There were five response options: not at all difficult, only a little difficult, somewhat difficult, very difficult, or cannot do at all. For each of the 12 tasks, an individual was deemed to have a functional limitation if they reported difficulty performing the task without special equipment. Overall functional disability was defined by NHIS as difficulty performing one or more of the 12 tasks listed above without special equipment. The definition of functional limitations and functional disability are consistent with prior work (11,29).

Four racial/ethnic groups defined by NHIS were used: non-Hispanic white, Hispanic, non-Hispanic black, non-Hispanic other. Two age categories were created: 18–64 and ≥65 years. Education was classified as less than high school graduate and at least high school graduate or greater. Household income was categorized as <$20,000 and ≥$20,000. Geographic location was based on census region: West, South, Midwest, and Northeast.

A multilevel variable that defined four disease categories was created. The four categories were: no diabetes and no major depression, major depression alone, diabetes alone, and diabetes and comorbid major depression. The diagnosis of diabetes was based on self-report of being diagnosed with diabetes by a physician excluding a diagnosis made during pregnancy. Major depression was defined as described above. To account for the potential confounding effect of comorbid conditions other than diabetes or major depression on odds of functional disability, a variable called disability-associated comorbidity was created. An individual was deemed to have a disability-associated comorbid condition if they had any of the following conditions: chronic arthritis (rheumatoid arthritis, osteoarthritis, or arthritis due to gout), chronic obstructive pulmonary disease, coronary artery disease, stroke, or heart failure. These conditions were selected because of their high prevalence and independent association with functional disability (29).

Statistical software, STATA (30), which accounts for the multistage sampling, clustering, and stratification design of national surveys such as the NHIS (31), was used for statistical analyses and to generate population estimates. Three sets of analyses were performed. First, demographic characteristics and the distribution of disability-associated comorbid conditions were compared across the four disease categories (no diabetes and no major depression, major depression only, diabetes only, and diabetes and comorbid major depression) using χ² statistics. Second, proportions of respondents with functional limitations were compared across the four disease categories. Similarly, the proportions of respondents with overall functional disability (defined as difficulty performing one or more of the 12 previously mentioned tasks without special equipment) were compared across the four disease categories using χ² statistics. Third, multiple logistic regression was used to determine the independent odds of functional limitations and overall functional disability across the four disease categories controlling for age, sex, race/ethnicity, education, household income, census region, and disability-associated comorbidity. For each of the 12 tasks, a multiple logistic regression model was run with inability to perform the task without special equipment (functional limitation) as the dependent variable (yes versus no) and disease category (four-level variable) as the independent variable, controlling for age, sex, race/ethnicity, education, household income, census region, and disability-associated comorbidity. Similarly, for overall functional disability, the dependent variable was overall functional disability (yes versus no), and disease category, age, sex, race/ethnicity, education, household income, census region, and disability-associated comorbidity were entered as independent variables. Additional analyses were performed to determine whether there was an interaction between the effects of diabetes and major depression on functional disability. Separate logistic regression models were run with functional disability as the dependent variable and diabetes, major depression, and an interaction term (diabetes and major depression) as independent variables along with other covariates.

In 1999, 30,801 individuals aged ≥18 years were interviewed, and the overall response rate was 70%. Of this number, 30,022 with complete data on relevant variables were included in the analysis. Approximately 170 individuals had both diabetes and major depression, 1,624 had only diabetes, 1,852 had only major depression, and 26,376 individuals had neither diabetes nor major depression. Extrapolating to the U.S. population in 1999, ∼1 million individuals had both diabetes and major depression, 9.4 million had diabetes alone, 11 million had major depression alone, and 174 million had neither diabetes nor major depression.

Table 1 compares demographic characteristics and distribution of disability-associated comorbid conditions across the four disease categories. There were significant differences in demographic characteristics across the four disease categories. In addition, there were significant differences in the prevalence of disability-associated comorbidity (chronic arthritis, chronic obstructive pulmonary disease, coronary artery disease, stroke, and heart failure) across the four disease categories; however, individuals with diabetes and major depression had significantly higher prevalence of these conditions compared with the other categories.

Table 2 shows the proportion of respondents with functional limitations and overall functional disability by disease category, and Table 3 shows the adjusted odds of functional limitations by disease category. There seemed to be a progressive increase in the proportion of individuals with functional limitations and overall functional disability across the four disease categories. A significantly higher proportion of individuals with diabetes and major depression reported having overall functional disability compared with the other three groups. Specifically, ∼78% of individuals with diabetes and comorbid major depression reported having overall functional disability in contrast with 58% of those with diabetes, 51% of those with major depression, and 25% of individuals with neither diabetes nor major depression.

Table 4 shows factors that are independently associated with overall functional disability. All variables entered into the multiple logistic regression model were independently associated with odds of functional disability, but the magnitude of effect differed across the variables (Table 4). Independent of disease category, individuals with one or more disability-associated comorbid condition, the elderly, women, people with lower household income, those with less than high school education, and residents of the Southern and Northeastern regions of the U.S. had significantly higher odds of functional disability. On the contrary, non-Hispanic blacks and Hispanic/others had significantly lower odds of functional disability compared with non-Hispanic whites.

Table 5 shows the unadjusted and adjusted odds of functional disability by disease category. In the unadjusted model, individuals with diabetes and major depression (odds ratio [OR] 9.18, 95% CI 6.29–13.41) had significantly higher odds of overall functional disability compared with individuals with either major depression (3.21, 2.89–3.57) or diabetes (4.18, 3.74–4.67). This pattern remained even after adjusting for relevant covariates: diabetes and major depression (6.15, 3.86–9.80), major depression (3.02, 2.66–3.44), and diabetes (2.46, 2.15–2.82). In addition, in both unadjusted and adjusted models, the odds of overall functional disability in individuals with diabetes and comorbid major depression (OR 6.15) was significantly higher than the sum of the independent odds of functional disability in individuals with diabetes and major depression (OR 5.48), which suggest a synergistic effect rather than an additive effect.

This study, which assessed the relationship among diabetes, major depression, and functional disability in a nationally representative sample of U.S. adults, has two important findings. First, it provides evidence that the odds of functional disability are significantly higher in individuals with diabetes and comorbid major depression than in individuals with either diabetes or major depression alone. Second, it indicates that there is a synergistic relationship between coexistent diabetes and major depression and functional disability even after controlling for relevant confounding factors.

There is a growing body of literature on the relationship between diabetes and depression and the effect of diabetes and comorbid depression on health outcomes and costs. Specifically, there is evidence that ∼10% of individuals with diabetes have comorbid major depression (16,17) and that the coexistence of diabetes and major depression is associated with increased health care use (18), increased health care costs (18,32), and adverse health outcomes for diabetes (33,34). The finding of this study, which has not been previously reported, adds to that body of literature. It shows that individuals with diabetes and comorbid major depression have higher odds of functional limitations and higher odds of functional disability compared with individuals with either major depression or diabetes alone.

The results of this study, which show that major depression and diabetes each independently increase the odds of functional disability, are consistent with the work of others (9–11,35–38). The mechanism by which major depression and diabetes each affects physical and psychosocial functioning and thus mediate functional disability is not clearly understood. Depression is hypothesized to decrease physical health by a combination of biological and psychological mechanisms (15). Psychological distress and subsequent neurohormonal and immunological changes are thought to increase susceptibility to disease, persistent somatic symptoms of depression is thought to worsen physical health over time, and depressed mood is thought to interfere with physical recovery by impeding treatment seeking, adherence to treatment, and adoption of healthy lifestyles (15). Diabetes, on the other hand, is assumed to mediate functional disability via the development of multiple complications (39). In this study, functional limitations in individuals with diabetes and comorbid major depression encompassed both physical and psychosocial function, which suggest that multiple mechanisms are likely to mediate the relationship between diabetes and comorbid major depression and functional disability.

Previous studies have shown that certain factors including older age, female sex, minority race/ethnicity, and multiple chronic diseases increase the risk of functional disability (9,29). In this study, similar factors were independently associated with odds of functional disability except race/ethnicity. Specifically, disability-associated comorbid conditions, older age, female sex, lower income, lower educational attainment, and living in the Southern and Northeastern regions of the U.S. were independently associated with higher odds of functional disability. However, individuals of minority ethnicity had lower odds of functional disability in contrast to previous reports. The reason for this finding is unclear and will need to be clarified in future studies.

The results of this study have two major public health implications. First, the burden of functional disability is tremendous. Extrapolating to the adult population of the U.S., ∼54.5 million of an estimated 195 million adult population in the U.S. have a functional disability. Similarly, 5.6 million of an estimated 11 million adults with major depression, 5.5 million of an estimated 9.4 million adults with diabetes, and 0.7 million of an estimated 1 million adults with diabetes and comorbid major depression have a functional disability. The public health implication of these estimates are further magnified if put in the context of available evidence, which indicate that functional disability predicts further functional decline, functional dependence, increased health services use, increased health care costs, and increased risk of death (3–7).

Second, evidence from the literature indicate that recognition and treatment of major depression is less than ideal (40–43), that control of diabetes is suboptimal (44,45), and that the coexistence of diabetes and comorbid depression has adverse effects on health outcomes (18,33,34,46). These findings suggest that individuals with coexisting diabetes and major depression represent a subgroup at high risk of diminished quality of life, increased health care use and costs, and increased mortality. Unlike the situation with diabetes, where functional disability probably occurs after years of living with the disease (39), functional disability resulting from depression starts early and spans the course of the illness if untreated (15,47). However, there is good evidence that effective treatment for depression decreases functional disability (48,49,50); therefore, strategies to improve detection and treatment of depression, particularly in individuals with diabetes and comorbid depression are critical to improving functional outcomes and decreasing disability.

The findings of this study are subject to several limitations. First, as with all cross-sectional studies, this study cannot speak to causality or temporality. Functional disability may have preceded diabetes or major depression; however, this is unlikely to be the case because currently available data indicate that diabetes and major depression most often precede functional disability (10,15,47). Second, this study may have underestimated the prevalence of functional limitations and functional disability because the sample excluded institutionalized individuals such as people in nursing homes and rehabilitation facilities, who may have higher prevalence of disability. In addition, the definition of disability excluded those individuals that used assisting devices to compensate for functional limitations, which may have further underestimated the overall prevalence of functional disability. Similarly, the complex statistical adjustment procedures incorporated into the NHIS may not have adequately accounted for lower response rates for individuals with functional disability, which may have further underestimated the prevalence of functional disability.

Third, due to unavailability of reliable data in the NHIS on duration of diabetes, diabetes severity, or comorbid psychiatric conditions, the effect of these factors could not be ascertained, and future studies will need to address their effect on functional disability. Fourth, the questions used to assess functional disability in this study did not include all domains of disability such as body function, body structures, activities and participation, and environmental factors as recommended by the International Classification of Functioning, Disability, and Health (51). In addition, the disability questions did not assess other perspectives such as institutional or societal perspectives as detailed in the International Classification of Functioning, Disability, and Health (51). Finally, the diagnosis of diabetes was based on self-report. This is unlikely to have biased the estimates because previous studies have established the reliability of self-reported diabetes as a measure of diagnosed diabetes (52,53).

Despite these limitations, this study provides evidence that the odds of functional disability are significantly higher in individuals with diabetes and comorbid major depression than in individuals with either diabetes or major depression alone and that there is a synergistic relationship between coexistent diabetes and major depression and functional disability even after controlling for relevant confounding factors.

This work was supported by the Agency for Healthcare Research and Quality (Rockville, MD) Grant 5K08HS11418.

This study was presented in part at the American Diabetes Association 63rd Scientific Sessions, New Orleans, Louisiana, 13–17 June 2003.