Abstract
Background:
The population of older adults living with multiple chronic conditions (MCC) continues to grow. MCC is independently associated with functional limitation and obesity. The aim of our study was to evaluate the association between obesity and MCC, and secondarily, the combined presence of obesity and functional limitations with MCC.
Methods:
We analyzed cross-sectional survey data from the National Health and Aging Trends Survey (NHATS) 2011 baseline data, a nationally representative Medicare beneficiary cohort of adults in the United States. We evaluated the coexistent prevalence of obesity and MCC overall, and by standard body mass index (BMI) categories. We then evaluated the prevalence of functional limitations (mobility, self-care, and household activities) and Fried-defined frailty status in persons with a BMI ≥ 30 kg/m2. Logistic regression was used to measure the association between MCC and BMI, and functional limitations and MCC among those with obesity.
Results:
In the 6,600 participants, the prevalence of concurrent obesity and MCC was 30.4%. Of those with obesity, the prevalence of MCC was 84.0%, and were more likely to have MCC (adjusted OR: 2.17, 95% CI 1.86, 2.54) compared to a normal BMI. Obesity and functional limitations or frailty were more likely have MCC than individuals with obesity alone.
Conclusions:
We found that individuals with obesity is strongly associated with MCC and that functional limitations and frailty status have a greater association with having MCC than individuals with obesity without MCC. Future longitudinal analyses are needed to ascertain this relationship.
Keywords: older adults, multimorbidity, functional limitation
INTRODUCTION
The population of older adults in the United States living with multiple chronic conditions (MCC) continues to grow, with recent estimates exceeding 70% [1]. MCC, often termed multimorbidity, is defined as the occurrence of two or more chronic diseases and is associated with reduced quality of life [2] increased mortality [3], and accounts for 84% of total health care expenditures [4, 5]. In older adults, MCC has been shown to be associated with a decline in physical function [6]. A strong association has been seen between the number of chronic diseases and disability in both cross-sectional and longitudinal cohort studies. In fact, for each additional chronic disease, an older adult with MCC experiences a 50% increased risk of further functional decline [7, 8]. Hence, understanding the relationship between MCC with physical function or frailty in persons with obesity is also public health priority.
The rise in the prevalence of MCC can be attributed, at least partially, to the rapid rise in the epidemic of obesity [9]. Recent population-based estimates demonstrate that adults over age 60 have obesity prevalence exceeding 42.2% in males and 43.3% in females [10]. Obesity is an established risk factor for many age-related chronic diseases including hypertension, diabetes, cardiovascular disease, stroke, cancer, arthritis, and pulmonary abnormalities [11]. Studies in China have shown that the prevalence of MCC in people classified as overweight or having obesity over age 60, is 2.5–3.0 times greater than that of the general population [12]. Cross-sectional data from a Canadian cohort noted the prevalence rates of MCC have significantly increased over time, with older adults with obesity making up the largest proportion of individuals across the life span [13].
Additionally, obesity has shown to be predictive of functional decline in older adults. Excess body fat mass and a BMI ≥ 30 kg/m2 in older subjects is associated with physical and mobility impairments and is predictive of a decline in functional status and future disability [14–16]. Using the Health, Aging and Body Composition data, adults classified as overweight or with obesity using BMI over the life course had high rates of incident disability over a seven-year period [17].
Both MCC and obesity in older adults are public health concerns that are attributable to worse health outcomes and increased mortality. Therefore, an understanding of the associations between MCC and obesity are important for understanding this phenomenon. There is clear evidence of the independent association of MCC with either functional limitations or obesity, themselves. This study will use cross-sectional data from an existing nationally representative cohort of older adults in the US to further understand this association. Our primary evaluation will be to ascertain the association between MCC and obesity, defined using both BMI and waist circumference (WC). Our secondary findings will allow us to ascertain the relationship between markers of functional limitation (including frailty status) and MCC in individuals with obesity. As the proportion of individuals living with MCC continues to increase in the US, there is a greater need to understand the patterns associated with this geriatric syndrome. Gaining an understanding of this relationship will enable the development of interventions that could improve mortality, quality of life, and prevent the progression of MCC, frailty and functional limitation.
METHODS
Study Design & Participants
Data were obtained from the National Health and Aging Trends Study (NHATS), a nationally representative survey of Medicare beneficiaries aged ≥65 years funded by the National Institute on Aging whose aim is to provide information and an understanding of trends in late-life functioning. Respondents are interviewed annually in person, and asked to report health, social, economic, and other characteristics. Objective measures of physical function are collected at each annual interview in participants’ home using trained staff. For the purposes of this analysis, the 2011 baseline data was used (n=8,245). Participants were excluded if they lacked information on height or weight preventing the calculation of body mass index (BMI) or lacked a measurement for waist circumference. Participants with insufficient data on key study variables (n=1,645) were excluded from analyses, producing an analytic sample of 6,600 participants. Data from participants who were excluded in our analysis (n=1645) due to missing BMI or waist circumference (WC) data are presented in eTable 1 in the supplement. Compared to our included cohort, excluded participants had higher rates of older age, Black race, physical inactivity, and smoking. This study was deemed exempt from the local institutional review board at University of North Carolina at Chapel Hill (#20-2777) as this data was de-identified in nature. All participant data collected via NHATS was obtained via documented informed consent processes during the observational study cohort.
Outcome measures - Multiple chronic conditions (MCC)
Primary analyses assessed the prevalence of MCC. Participants fulfilled criteria for MCC if they had two or more chronic conditions; they were not considered to have MCC if they had either one or no chronic conditions. Participants reported whether they had any of ten Medicare reported chronic medical conditions or events in the previous year: stroke, ischemic heart disease, hypertension, diabetes, chronic obstructive pulmonary/lung disease, Alzheimer’s disease or dementia, arthritis, osteoporosis, or cancer (other than skin cancer). Information regarding health conditions was asked using a self-reported questionnaire to participants as to whether a doctor had ever told them if they had any of several conditions. Previous research demonstrates good concordance between self-reported and medical record extracted health conditions [18, 19] using variables that were based on a validated algorithm from the Medicare Chronic Condition Data Warehouse [20, 21].
Other measures
Weight Status:
Obesity status was defined in two manners: body mass index (BMI) or waist circumference (WC). Current height and weight were assessed using a self-reported questionnaire. Standard BMI categories were used: underweight (<18.5 kg/m2), normal (18.5-24.9 kg/m2), overweight (25-29.9 kg/m2) or obesity (≥30 kg/m2). Waist circumference was measured using a flexible tape measure on the participant’s abdomen aligned with his or her navel in a snug, but not tight manner, holding their breath at the end of exhalation, recording to the nearest one-quarter inch. These values were then converted to metric system. Waist circumference was defined to be high if higher than 88 cm for women or 102 cm for men.
Functional Limitations:
All limitations described were assessed using self-reported questionnaires. Self-care activities (activities of daily living) were defined as eating, bathing, toileting, and dressing; household activities (instrumental activities of daily living) included laundry, grocery shopping, meal preparation, banking or paying bills, and medication tracking; and mobility activities were defined as going outside, moving around inside, and getting out of bed. For each activity, participants were assigned to one of three categories: having no difficulty, having some difficulty but able to perform activity without assistance, or unable to do and requiring assistance. For each category (self-care, household, and mobility), the presence of functional limitation was formulated as a dichotomous variable, with no functional limitation (e.g., no difficulty) coded as ‘no limitation’ and any functional limitation coded as having a ‘limitation.’ An overall value for functional limitation was additionally formulated as a dichotomous variable, with an overall limitation present if the participant had a limitation in any of three categories (ADL, IADL, or mobility).
Frailty:
Frailty was assessed according five criteria outlined in the phenotypic frailty definition: weakness, slowness, exhaustion, low physical activity and shrinking [22]. We adapted these criteria based on available data as noted below in parentheses: weakness, defined according to grip strength (grip strength using maximum dominant hand grip strength over 2 trials as ≤ 20th percentile within 8 sex-by–body mass index (BMI) categories); slow walking speed (gait speed using the first of 2 usual-pace walking trials as being ≤ 20th percentile of the weighted population distribution within 4 sex-by-height categories); self-reported exhaustion (easily exhausted, limiting activities); low physical activity (ever go walking or do vigorous activities) and shrinking defined as unintentional weight loss of more than 10 pounds in a year (weight loss of 10 pounds in the last year without trying) [23]. Participants were categorized in three frailty categories on the number of criteria fulfilled: robust (no criteria), moderate frailty (one or two criteria) and high frailty (three or more criteria).
Covariates:
We included age range at baseline, sex, level of education (less than high school, high school to some college, college, and more than college), race (White, Black, Hispanic, other), smoking status (current, former, or never smoker), level of physical activity, and place of residence. Age was considered a restricted variable, hence, standard NHATS five-year increment categories from 65 – 85 years and 85+ years were used. Smoking status was defined as current, past, or never smoked using the question “ever smoked cigarettes at least 1 cigarette a day regularly?” Level of physical activity was dichotomized to ‘ever walk’ or ‘never walk’ using the answer to the question “in the last month, did you (respondent) ever go walking for exercise?”
Statistical Analysis
All data were aggregated according to NHATS guidelines (http://www.nhats.org). Demographic and baseline characteristics were evaluated using descriptive statistics. Continuous variables are presented as mean ± standard deviation (SD) and categorical variables as frequencies (percent), where associations between categorical variables and MCC status were evaluated using chi-squared test while associations between continuous variables and MCC status were evaluated using unpaired t-test. Our primary analysis focused on the relationship between BMI-based obesity categories and the odds of having MCC. We used logistic regression to estimate the adjusted odds ratios (aOR) and 95% confidence intervals (95% CI) of the relationship between obesity (standard BMI categories – referent = normal) and MCC, after adjusting for covariates. Unadjusted (Model 1) and adjusted models are presented. Model 2 adjusted for demographic variables including age group, sex, and race. Model 3 additionally adjusted for social and behavioral confounders that were available in the dataset that could impact obesity and MCC, including smoking status, education status, physical activity, and place of residence.
Within those classified as having obesity, logistic regression estimated aOR and 95% CI for relationship between frailty status or functional limitation (limitation in mobility, self-care, or household activities) and the presence of MCC. Separate analyses were conducted for BMI and waist circumference defined obesity. Study data were analyzed using R version 3.5.2 (www.R-project.org). pandas 1.0.3, and statsmodels 0.11.1. A P value of <.05 was considered statistically significant [24].
RESULTS
A total of 6,600 participants were included in the analyses. Table 1 displays baseline characteristics of participants as a function of presence of MCC, stratified by body mass index (BMI) categories. Of those included, 27.7% of participants had BMI defined obesity, 43.1% were male, 25.6% reported less than high school education, 7.9% were smokers, and 39.4% reported physical inactivity. In this analytic sample, 75.2% of participants were classified as having MCC. eTable 2 in the supplement displays baseline characteristics of participants as a function of presence of MCC, stratified by waist circumference (WC) categories.
Table 1:
Baseline Characteristics of Cohort (n = 6,600)
| Presence of MCC (≥2 co-morbidities) | Absence of MCC (0 or 1 co-morbidities) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Overall | BMI (kg/m2) | Overall | BMI (kg/m2) | |||||||
| <18.5 | 18.5-24.9 | 25-29.9 | ≥30 | <18.5 | 18.5-24.9 | 25-29.9 | ≥30 | |||
| N, % | N=4967 (75.2) | 102 (2.1) | 1572 (31.6) | 1782 (35.9) | 1511 (30.4) | N=1633 (32.9) | 42 (2.6) | 648 (39.7) | 656 (40.2) | 287 (17.6) |
| Age group, years | ||||||||||
| 65-70 | 852 (17.2) | 2 (2.0) | 182 (11.6) | 291 (16.3) | 377 (25.0) | 445 (27.3) | 8 (19.0) | 153 (23.6) | 191 (29.1) | 93 (32.4) |
| 70-75 | 1053 (21.2) | 17 (16.7) | 258 (16.4) | 358 (20.1) | 420 (27.8) | 382 (23.4) | 9 (21.4) | 136 (21.0) | 165 (25.2) | 72 (25.1) |
| 75-80 | 1053 (21.2) | 15 (14.7) | 289 (18.4) | 422 (23.7) | 327 (21.6) | 289 (17.7) | 4 (9.5) | 101 (15.6) | 124 (18.9) | 60 (20.9) |
| 80-85 | 1032 (20.8) | 32 (31.4) | 361 (23.0) | 400 (22.4) | 239 (15.8) | 264 (16.2) | 9 (21.4) | 122 (18.8) | 94 (14.3) | 39 (13.6) |
| 85+ | 977 (19.7) | 36 (35.3) | 482 (30.7) | 311 (17.5) | 148 (9.8) | 253 (15.5) | 12 (28.6) | 136 (21.0) | 82 (12.5) | 23 (8.0) |
| Male Sex | 2002 (40.3) | 22 (21.6) | 587 (37.3) | 836 (46.9) | 557 (36.9) | 843 (51.6) | 12 (28.6) | 294 (45.4) | 389 (59.3) | 148 (51.6) |
| Race a | ||||||||||
| White | 3440 (69.3) | 77 (75.5) | 1126 (71.6) | 1283 (72.0) | 954 (63.1) | 1213 (74.3) | 32 (76.2) | 483 (74.5) | 506 (77.1) | 192 (66.9) |
| Black | 1109 (22.3) | 19 (18.6) | 284 (18.1) | 370 (20.8) | 436 (28.9) | 270 (16.5) | 7 (16.7) | 105 (16.2) | 94 (14.3) | 64 (22.3) |
| Hispanic | 283 (5.7) | 2 (2.0) | 96 (6.1) | 94 (5.3) | 91 (6.0) | 94 (5.8) | 0 (0.0) | 28 (4.3) | 43 (6.6) | 23 (8.0) |
| Other | 130 (2.6) | 3 (2.9) | 65 (4.0) | 34 (1.9) | 28 (1.9) | 54 (3.3) | 3 (7.1) | 31 (4.8) | 13 (2.0) | 7 (2.4) |
| Missing | 5 | 1 | 1 | 1 | 2 | 2 | -- | 1 | -- | 1 |
| Education b | ||||||||||
| < High School | 1365 (27.5) | 34 (33.3) | 395 (25.1) | 491 (27.6) | 445 (29.5) | 326 (20.0) | 11 (26.2) | 138 (21.3) | 124 (18.9) | 53 (18.5) |
| High School to Some College | 2392 (48.2) | 47 (46.1) | 777 (49.4) | 816 (45.8) | 752 (49.8) | 769 (47.1) | 19 (45.2) | 306 (47.2) | 303 (46.2) | 141 (49.1) |
| College | 752 (15.1) | 12 (11.8) | 248 (15.8) | 283 (15.9) | 209 (13.8) | 321 (19.7) | 8 (19.0) | 126 (19.4) | 136 (20.7) | 51 (17.8) |
| Post College | 451 (9.1) | 7 (6.9) | 149 (9.5) | 190 (10.7) | 105 (6.9) | 216 (13.2) | 4 (9.5) | 77 (11.9) | 93 (14.2) | 42 (14.6) |
| Missing | 7 (0.1) | 2 (2.0) | 3 (0.2) | 2 (0.1) | -- | 1 (0.1) | -- | 1 (0.1) | -- | -- |
| Physical Activity | ||||||||||
| Ever Walk | 2839 (57.2) | 48 (47.1) | 927 (59.0) | 1085 (60.9) | 779 (51.6) | 1158 (70.9) | 25 (59.5) | 475 (73.3) | 477 (72.7) | 181 (63.1) |
| Smoking Status c | ||||||||||
| Current | 380 (7.7) | 20 (19.6) | 154 (9.8) | 126 (7.1) | 80 (5.3) | 144 (8.8) | 10 (23.8) | 74 (11.4) | 46 (7.0) | 14 (4.9) |
| Former | 2215 (44.6) | 43 (42.2) | 665 (42.3) | 811 (45.5) | 696 (46.1) | 640 (39.2) | 7 (16.7) | 237 (36.6) | 277 (42.2) | 119 (41.5) |
| Never | 2369 (47.7) | 39 (38.2) | 751 (47.8) | 845 (47.4) | 734 (48.6) | 849 (52.0) | 25 (59.5) | 337 (52.0) | 333 (50.8) | 154 (53.7) |
| Missing | 3 (0.1) | -- | 2 (0.2) | -- | 1 (0.1) | -- | -- | -- | -- | -- |
| Residence Type d | ||||||||||
| Private Residence | 3926 (79.0) | 75 (73.5) | 1219 (77.5) | 1435 (80.5) | 1197 (79.2) | 1364 (83.5) | 35 (83.3) | 527 (81.3) | 567 (86.4) | 235 (81.9) |
| Group Home e | 7 (0.1) | 0 (0.0) | 0 (0.0) | 6 (0.3) | 1 (0.1) | 1 (0.1) | 0 (0.0) | 0 (0.0) | 1 (0.2) | 0 (0.0) |
| ALF, CCRC | 14 (0.3) | 0 (0.0) | 4 (0.3) | 4 (0.2) | 6 (0.4) | 2 (0.1) | 0 (0.0) | 1 (0.2) | 1 (0.2) | 0 (0.0) |
| Other | 15 (0.3) | -- | 6 (0.4) | 7 (0.4) | 2 (0.1) | 261 (16.0) | 7 (16.7) | 120 (18.5) | 84 (12.8) | 50 (17.4) |
| Religious | 1 (0.1) | -- | -- | 1 (0.1) | -- | 1 (0.1) | -- | -- | 1 (0.2) | -- |
| Not Listed | 1004 (20.2) | 27 (26.5) | 343 (21.8) | 329 (18.5) | 305 (20.2) | 4 (0.2) | -- | -- | 2 (0.3) | 2 (0.3) |
Note. Abbreviations: BMI: body mass index, BMI <18.5kg/m2 was defined as underweight, 18.5-24.9kg/m2 was defined as normal, 24.9 to 30.0 kg/m2 was defined as overweight, ≥ 30kg/m2 was defined as obese. ALF: assisted living facility. CCRC: continuing care retirement community.
All values represented are counts (%). An analysis of variance was performed for all continuous variables, and chi-square test used for all category p-values. Physical activity is defined as an affirmative response to ‘In the last month, did you ever go walking for exercise?’ Counts may be different due to missing values.
Race – other and unknown race are not included in this table.
Education level – those with education levels not reported were not included in the table.
Smoking status – those with smoking status not reported were not included in the table.
Residence type – those who lived in religious group quarter or other type of housing were not included in the table.
Group home includes: Board/Care, Supervised Housing
Table 2 shows the association between BMI/WC categories and MCC categories using univariate and multivariate logistic regression analyses. Compared to older adults with a normal BMI, older adults with BMI-defined obesity were more likely to have MCC (OR: 2.17, 95% CI 1.86, 2.54). Adjusting for covariates demonstrated an even greater association of BMI-defined obesity with MCC (adjusted OR: 2.59, 95% CI 2.15, 3.11). Older adults with a high waist circumference were also more likely to have MCC when compared to individuals with a low waist circumference (adjusted OR: 1.84, 95% CI 1.61, 2.09).
Table 2:
Cross-Sectional Associations of Obesity Status and Multiple Chronic Conditions
| Univariate OR [95% CI] | Multivariate [95% CI] | ||
|---|---|---|---|
| Model 1 | Model 2 | Model 3 | |
| Body Mass Index | |||
| Underweight | 1.0 (0.69, 1.45) | 0.86 (0.59, 1.25) | 0.71 (0.46, 1.09) |
| Normal | Referent | Referent | Referent |
| Overweight | 1.12 (0.99, 1.27) | 1.26 (1.11, 1.44) | 1.25 (1.08, 1.45) |
| Obesity | 2.17 (1.86, 2.54) | 2.53 (2.15, 2.98) | 2.59 (2.15, 3.11) |
| Waist Circumference | |||
| Low | Referent | Referent | |
| High | 1.95 (1.74, 2.19) | 1.86 (1.65, 2.09) | 1.74 (1.52, 1.99) |
Note. Values presented as odds ratios [95% confidence intervals] as determined by logistic regression models.
Odds of MCC as a function of weight status using either BMI or WC calculations. A normal BMI and a low WC were used as referents for BMI and WC status, respectively. Unadjusted (Model 1) and adjusted models are presented. Model 2 adjusted for demographic variables including age group, sex, and race. Model 3 additionally adjusted for social and behavioral confounders that were available in the dataset that could impact obesity and MCC, including smoking status, education status, physical activity, and place of residence
Baseline characteristics of individuals with BMI-defined obesity and high WC are presented in eTable 3. In participants with BMI-defined obesity, the rate of MCC was 84.0%, with 59.5% having at least three chronic conditions. The univariate associations of covariates with MCC were analyzed, with strong associations of MCC with age, Black race, and current/former smoking status (eTable 4 in the supplement). Individuals who were 85 years or older were more likely have MCC compared to individuals who were 65-70 (adjusted OR: 2.83, 95% CI 2.27, 3.53; eTable 4 in the supplement). Black individuals were more likely to have MCC compared to White individuals (adjusted OR: 1.41, 95% CI 1.18, 1.68; eTable 4 in the supplement).
Table 3 displays prevalence of limitation (overall, mobility, self-care, household activities) and frailty status in individuals with obesity stratified by MCC status. Persons with MCC and obesity had high rates of frailty and functional limitation. Among participants with obesity and MCC, 14.8% had a high frailty status, while 3.8% of individuals with obesity but without MCC had a high frailty status. 39.7% of individuals with MCC and obesity had a mobility limitation specifically, and 52.3% of individuals with MCC and obesity had at least one type of limitation.
Table 3:
Prevalence of Functional Limitations and Frailty Status in Participants with Obesity
| Body Mass Index ≥ 30kg/m2 | High Waist Circumference | |||||
|---|---|---|---|---|---|---|
| Presence of MCC (≥2 co-morbidities) | Absence of MCC (0 or 1 co-morbidities) | p-value | Presence of MCC (≥2 co-morbidities) | Absence of MCC (0 or 1 co-morbidities) | p-value | |
| N, % | N=1508 | N=286 | N=3354 | N=841 | ||
| Frailty Status | 0.92 | 0.93 | ||||
| Robust | 426 (28.2) | 153 (53.5) | 992 (29.6) | 456 (54.2) | ||
| Moderate | 859 (57.0) | 122 (42.7) | 1853 (55.2) | 345 (41.0) | ||
| High | 223 (14.8) | 11 (3.8) | 509 (15.2) | 40 (4.8) | ||
| Functional Limitation Type | ||||||
| Overall | 788 (52.3) | 57 (19.9) | 0.32 | 1735 (51.7) | 177 (21.0) | 0.31 |
| Mobility | 598 (39.7) | 36 (12.6) | 0.24 | 1307 (39.0) | 116 (13.8) | 0.23 |
| Self-Care activities | 247 (16.4) | 13 (4.5) | 0.04 | 549 (16.4) | 45 (5.4) | 0.04 |
| Household activities | 423 (28.0) | 28 (9.8) | 0.14 | 939 (28.0) | 82 (9.8) | 0.14 |
Note. All values represented are counts (%). Associations between categorical variables and MCC status were evaluated using chi-squared test while associations between continuous variables and MCC status were evaluated using unpaired t-test.
Frailty was assessed according five criteria outlined in the phenotypic frailty definition: weakness, slowness, exhaustion, low physical activity and shrinking. Criteria were adapted based on available data as reported in the methods section. Participants were categorized in three frailty categories based on the number of frailty criteria fulfilled: robust (none), moderate (one or two criteria) and high (three or more criteria).
Participants were defined as having a limitation in mobility, self-care, or household activities if were unable to perform an activity at all or only able to perform with assistance. Mobility activities included getting around inside, going outside, getting out of bed. Self-care activities included eating, bathing, toileting, and dressing. Household activities included laundry, grocery shopping, meal preparation, banking or paying bills, and medication tracking. An overall value for functional limitation was formulated if the participant had a limitation in any of three categories (mobility, self-care, or household). For each limitation category (overall, mobility, self-care and household), the referent is the non-impaired state for that category.
Table 4 outlines the unadjusted and adjusted multivariate regression models to determine the association of MCC with frailty and functional limitations in patients with obesity. When adjusting for all covariates, individuals living with obesity who had a high frailty status were strongly associated with having MCC as compared to individuals who were obese but had a robust frailty status (adjusted OR: 8.35, 95% CI 3.87, 13.87). Impairments in mobility, self-care, and household activities were also associated with MCC in individuals with obesity, with the strongest association seen for mobility limitation (adjusted OR: 4.56, 95% CI 3.17, 6.57).
Table 4:
Association of Presence of MCC with Frailty and Disability Impairment in Patients with Obesity
| Body Mass Index ≥ 30kg/m2 | High Waist Circumference | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | |
| Frailty Status | ||||||
| Robust | Referent | Referent | Referent | Referent | Referent | Referent |
| Moderate | 2.51 (1.93, 3.27) | 2.47 (1.88, 3.25) | 2.52 (1.77, 3.59) | 2.47 (2.11, 2.9) | 2.4 (2.03, 2.83) | 2.38 (1.94, 2.91) |
| High | 7.26 (3.86, 13.68) | 7.17 (3.74, 13.72) | 8.35 (3.7, 18.85) | 5.87 (4.18, 8.25) | 5.68 (4.01, 8.06) | 5.89 (3.83, 9.06) |
| Functional Limitation Type | ||||||
| Overall | 4.40 (3.23, 5.98) | 4.20 (3.07, 5.76) | 3.76 (2.63, 5.37) | 4.02 (3.36, 4.81) | 3.85 (3.20, 4.62) | 3.76 (3.05, 4.64) |
| Mobility Activities | 4.56 (3.17, 6.57) | 4.32 (2.98, 6.27) | 3.32 (2.21, 4.98) | 3.99 (3.24, 4.91) | 3.82 (3.09, 4.72) | 3.58 (2.81, 4.57) |
| Self-Care Activities | 4.11 (2.32, 7.29) | 3.99 (2.23, 7.12) | 3.48 (1.83, 6.61) | 3.46 (2.53, 4.74) | 3.26 (2.37, 4.49) | 3.15 (2.17, 4.57) |
| Household Activities | 3.59 (2.39, 5.59) | 3.27 (2.17, 4.93) | 3.42 (2.1, 5.59) | 3.6 (2.83, 4.58) | 3.38 (2.65, 4.3) | 3.36 (2.55, 4.43) |
Note. Values presented as odds ratios [95% confidence intervals] as determined by logistic regression models
Odds of MCC+ as a function of functional limitation or frailty status, within an obese (BMI ≥ 30) or high WC population. Frailty was assessed according five criteria outlined in the phenotypic frailty definition: weakness, slowness, exhaustion, low physical activity and shrinking. Criteria were adapted based on available data as reported in the methods section. Participants were categorized in three frailty categories based on the number of frailty criteria fulfilled: robust (none), moderate (one or two criteria) and high (three or more criteria). Odds of MCC as a function of moderate or high frailty status is compared to robust frailty status (referent).
Participants were defined as having a limitation in mobility, self-care, or household activities if were unable to perform an activity at all or only able to perform with assistance. Mobility activities included getting around inside, going outside, getting out of bed. Self-care activities included eating, bathing, toileting, and dressing. Household activities included laundry, grocery shopping, meal preparation, banking or paying bills, and medication tracking. An overall value for functional limitation was formulated if the participant had a limitation in any of three categories (mobility, self-care, or household). For each limitation category (overall, mobility, self-care, and household), the referent is the non-impaired state for that category.
Model 1 is unadjusted; model 2 includes demographic adjustment (age category, sex, race); model 3 is fully adjusted (age category, sex, race, education, ever walk, smoking status, residence)
DISCUSSION
This nationally representative study of older adults demonstrates that approximately three in four community-dwelling participants fulfill criteria for having multiple chronic conditions. In addition, our findings suggest that obesity and functional limitations and frailty are strongly related to MCC. These findings are critically important from a public health standpoint as MCC is associated with poor health outcomes, increased mortality, and increased healthcare utilization [2–4], while functional limitation and frailty are strong predictors of nursing home placement [25] and increased healthcare utilization [26, 27].
Previous research has indicated independent relationships between obesity and MCC, obesity and functional limitation, or MCC and functional limitations. To our knowledge, our study is the first to examine the combined presence of functional limitation and obesity and their association with MCC in a nationally representative cohort of older adults. The co-existence counts are high and suggest that important preventive health measures should be considered across the lifespan to prevent their development. Previous data from weight loss interventions suggest that even early life course interventions may reduce the impact of long-term disability [28]. This is critically important as the strong associations with functional limitations and frailty will ultimately become a challenge to the healthcare system. Our results demonstrate the strong relationship and impact that our society is currently facing.
We found that individuals with obesity who also had functional limitations had a much higher association with having MCC compared to those without functional imitations after adjusting for covariates. These findings demonstrate a potential synergistic effect between obesity and functional limitations on the association with MCC. The finding that obesity, functional limitations, and MCC may be strongly associated with one another may provide insight into the pathogenesis of this relationship. Our results provide an early foundation that demonstrates the need to understand common pathways that may underlie functional limitation and MCC.
Identifying risk factors and pathways that are common to multimorbidity and functional limitation could aid in the design of interventions to delay, prevent, or alleviate the progression of this health decline in older adults. Old age, sedentarism, inappropriate medication use, and obesity have all been implicated as underlying pathways for the development of MCC and frailty [29] While all potential pathways merit exploration, our findings may provide additional credence to obesity’s role in this scheme. Even after adjusting for age, our analysis demonstrates the notable association between obesity, functional limitations, and MCC.
Our analysis deliberately included two clinical and low-cost measures to define obesity, BMI and WC. While we acknowledge that gold-standard methods (e.g., computer tomography, magnetic resonance imaging) or dual energy x-ray absorptiometry provide a more complete assessment of body composition, using both BMI and WC provide indirect measures of overall adiposity and central adiposity, respectively. Including WC overcomes some of the limited diagnostic accuracy of BMI [30]. The strong associations observed between obesity and MCC were consistent when using either definition. Future research should evaluate segmental body composition and its relationship in MCC with functional limitations and frailty.
We emphasize that this cross-sectional analysis does not evaluate causality. For instance, functional limitations in individuals with obesity may be attributed to the increased presence of obesity-related chronic conditions such as diabetes, cardiovascular disease, osteoarthritis, heart failure, hypertension, and some cancers [31]. Additionally, it has been suggested that some chronic disease combinations may be more associated with functional limitation than others [32]. Future studies should be conducted to analyze specific disease combinations and their relationship with functional limitation, frailty, and obesity. The presence of inflammatory markers induced by central obesity may also mediate the relationship between obesity, multimorbidity, and frailty. Studies have hypothesized that inflammation induced by central obesity induces multimorbidity and frailty by inhibiting growth factors, increasing catabolism, and interfering with homeostatic signalling [33]. Yet, while obesity-related chronic conditions may precede the development of functional limitations in older adults, it is also possible that obesity and functional limitations are independent of one another or that one precedes the other. Future longitudinal research is needed to ascertain and confirm the findings in our cross-sectional analysis. Specifically, this includes confirming whether indeed there is a causal or bidirectional relationship between obesity, functional limitation, and MCC. Importantly, it is highly likely that the progression of both MCC and obesity may lead to further functional impairment that can only be captured using a longitudinal study.
Our study has the known limitations of any cross-sectional analysis, including above all, the lack of causality. We also excluded 1,645 (20%) participants from our analytic sample who had higher counts of smoking, physically inactivity, categorized as being part of an underrepresented minority group, having lower education, and being over the age of 85+. The exclusion of these participants may have led to an underestimation of the prevalence of MCC or functional limitations in our population. Furthermore, even in our included cohort, we relied on complete case ascertainment as a result of missing data basing our procedures on previous guidelines. [34] The current sample is also composed of Medicare beneficiaries and therefore does not represent the uninsured or undocumented who might be at greater risk for health disparities and for being classified as having obesity, having functional limitations, or having MCC. Additionally, some medical conditions were excluded in the analysis because they were not part of the NHATS administered questionnaires, and hence our study may under capture the extent of comorbidity in our population. Our analysis was deliberately based on 2011 data from NHATS Wave 1 to maximize the cohort included. We acknowledge that this data may under capture the extent of obesity, multimorbidity, and functional limitation in the older adult population given that the prevalenec of these trends are known to be higher at presents [35–37].
Our results convey that a significant proportion of older adults who experience MCC are classified as having obesity with co-existent functional limitation. The magnitude of the obesity epidemic in older adults should be prioritized in future research and interventional efforts. The current findings provide important insights and foundational knowledge needed to evaluate causality. For clinicians, as the number of chronic diseases increases with age, particularly in patients with obesity, the risk of unrecognized/diagnosed functional impairment or frailty should not be overstated. Future efforts should elucidate steps to counteract or prevent the development of MCC in older adults. Given the potential to modify risk for the development of additional chronic conditions through exercise and dietary interventions, our study provides insight on the potential benefits of employing weight loss interventions that minimize loss of muscle mass in a population that faces concomitant obesity and frailty.
Supplementary Material
Highlights.
What is already known on this topic:
Both obesity and multiple chronic conditions are public health concerns with rates rising in older adult populations leading to increased risk of adverse events. Few studies have evaluated the implications of both on physical function in persons with obesity
What this study adds:
This epidemiological study highlights the importance of co-existing obesity and multiple chronic conditions in older adults. Rates of concurrent obesity and multiple chronic conditions are high, as are the rates of functional limitations and risk of frailty, compared to those without obesity.
How this study might affect research, practice or policy:
The relationship of obesity and multiple chronic conditions, with frailty and physical limitations signifies the importance of developing interventions in this subgroup.
Acknowledgements
We thank Dr. Joshua Niznik PhD and Ms. Kristen Ruck for their comments on the manuscript.
Funding
Dr. Batsis’ research reported in this publication was supported in part by the National Institute on Aging of the National Institutes of Health under Award Number K23AG051681. Ms. Shirazi’s research reported in this publication was supported by the National Institute of Aging of the National Institute of Health under award number NIA 2T35AG038047-11- UNC-CH Summer Research Training in Aging for Medical Students. Dr. Kahkoska is supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant KL2TR002490. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest
The authors declare no conflict of interest.
REFERENCES
- 1.Jessie Gerteis M, et al. Multiple Chronic Conditions Chartbook 2010 MEDICAL EXPENDITURE PANEL SURVEY DATA 2014; Available from: https://www.ahrq.gov/sites/default/files/wysiwyg/professionals/prevention-chronic-care/decision/mcc/mccchartbook.pdf.
- 2.Newman AB, et al. , A Physiologic Index of Comorbidity: Relationship to Mortality and Disability. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 2008. 63(6): p. 603–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bayliss EA, Ellis JL, and Steiner JF, Subjective assessments of comorbidity correlate with quality of life health outcomes: initial validation of a comorbidity assessment instrument. Health Qual Life Outcomes, 2005. 3: p. 51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.G A Chronic Care: making the case for ongoing care. 2010. [cited 2021; Available from: http://www.rwjf.org/content/dam/farm/reports/reports/2010/rwjf54583.
- 5.Thorpe KE, Ogden LL, and Galactionova K, Chronic conditions account for rise in Medicare spending from 1987 to 2006. Health Affairs, 2010. 29(4): p. 718–24. [DOI] [PubMed] [Google Scholar]
- 6.Marengoni A, et al. , Aging with multimorbidity: a systematic review of the literature. Ageing Res Rev, 2011. 10(4): p. 430–9. [DOI] [PubMed] [Google Scholar]
- 7.Guralnik JM, et al. , Maintaining mobility in late life. I. Demographic characteristics and chronic conditions. Am J Epidemiol, 1993. 137(8): p. 845–57. [DOI] [PubMed] [Google Scholar]
- 8.Fuchs Z, et al. , Morbidity, comorbidity, and their association with disability among community-dwelling oldest-old in Israel. J Gerontol A Biol Sci Med Sci, 1998. 53(6): p. M447–55. [DOI] [PubMed] [Google Scholar]
- 9.Canizares M, et al. , Increasing Trajectories of Multimorbidity Over Time: Birth Cohort Differences and the Role of Changes in Obesity and Income. Journals of Gerontology. Series B: Psychological Sciences and Social Sciences, 2018. 73(7): p. 1303–1314. [DOI] [PubMed] [Google Scholar]
- 10.Hales CM, et al. , Prevalence of Obesity and Severe Obesity Among Adults: United States, 2017-2018. NCHS Data Brief, 2020(360): p. 1–8. [PubMed] [Google Scholar]
- 11.Guh DP, et al. , The incidence of co-morbidities related to obesity and overweight: a systematic review and meta-analysis. BMC Public Health, 2009. 9: p. 88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu X, et al. , Prevalence and influencing factors of overweight and obesity in a Chinese rural population: the Henan Rural Cohort Study. Sci Rep, 2018. 8(1): p. 13101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lebenbaum M, et al. , Trends in obesity and multimorbidity in Canada. Prev Med, 2018. 116: p. 173–179. [DOI] [PubMed] [Google Scholar]
- 14.Allan JL, Johnston M, and Campbell N, Missed by an inch or a mile? Predicting the size of intention-behaviour gap from measures of executive control. Psychol Health, 2011. 26(6): p. 635–50. [DOI] [PubMed] [Google Scholar]
- 15.Nigg CR and Long CR, A systematic review of single health behavior change interventions vs. multiple health behavior change interventions among older adults. Transl Behav Med, 2012. 2(2): p. 163–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.French DP, et al. , Which behaviour change techniques are most effective at increasing older adults’ self-efficacy and physical activity behaviour? A systematic review. Ann Behav Med, 2014. 48(2): p. 225–34. [DOI] [PubMed] [Google Scholar]
- 17.Houston DK, et al. , Overweight and obesity over the adult life course and incident mobility limitation in older adults: the health, aging and body composition study. Am J Epidemiol, 2009. 169(8): p. 927–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Haapanen N, et al. , Agreement between questionnaire data and medical records of chronic diseases in middle-aged and elderly Finnish men and women. Am J Epidemiol, 1997. 145(8): p. 762–9. [DOI] [PubMed] [Google Scholar]
- 19.Zhu K, et al. , Comparison of self-report data and medical records data: results from a case-control study on prostate cancer. Int J Epidemiol, 1999. 28(3): p. 409–17. [DOI] [PubMed] [Google Scholar]
- 20.Pope GC, et al. , Principal inpatient diagnostic cost group model for Medicare risk adjustment. Health Care Financ Rev, 2000. 21(3): p. 93–118. [PMC free article] [PubMed] [Google Scholar]
- 21.Chronic Conditions Data Warehouse. 2020; Available from: https://www2.ccwdata.org/web/guest/condition-categories.
- 22.Fried LP, et al. , Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci, 2001. 56(3): p. M146–56. [DOI] [PubMed] [Google Scholar]
- 23.Crow RS, et al. , Frailty Versus Stopping Elderly Accidents, Deaths and Injuries Initiative Fall Risk Score: Ability to Predict Future Falls. Journal of the American Geriatrics Society, 2018. 66(3): p. 577–583. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Seabold S, and Perktold Josef, statsmodels: Econometric and statistical modeling with python, in Proceedings of the 9th Python in Science Conference. 2010. [Google Scholar]
- 25.Kojima G, Frailty as a Predictor of Nursing Home Placement Among Community-Dwelling Older Adidts: A Systematic Review and Meta-anclysis. J Geriatr Phys Ther, 2018. 41(1): p. 42–48. [DOI] [PubMed] [Google Scholar]
- 26.Roe L, et al. , The impact of frailty on healthcare utilisation in Ireland: evidence from the Irish longitudinal study on ageing. BMC Geriatr, 2017. 17(1): p. 203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cegri F, et al. , The impact of frailty on admission to home care services and nursing homes: eight-year follow-up of a community-dwelling, older adult, Spanish cohort. BMC Geriatrics, 2020. 20(1): p. 281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Villareal DT, et al. , Weight loss, exercise, or both and physical function in obese older adidts. New England Journal of Medicine, 2011. 364(13): p. 1218–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Calderón-Larrañaga A, et al. , Multimorbidity and functional impairment-bidirectional interplay, synergistic effects and common pathways. Journal of Internal Medicine, 2019. 285(3): p. 255–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Batsis JA, et al. , Diagnostic accuracy of body mass index to identify obesity in older adults: NHANES 1999-2004. International Journal of Obesity (2005), 2016. 40(5): p. 761–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Pi-Sunyer X, The Medical Risks of Obesity. Postgraduate Medicine, 2009. 121(6): p. 21–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Vogeli C SA, Lee TA, et al. , Multiple chronic conditions: prevalence, health consequences, and implications for quality, care management, and costs. . J Gen Intern Med, 2007. 22: p. 391–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ferrucci L and Fabbri E, Inflammageing: chronic inflammation in ageing, cardiovascular disease, and frailty. Nat Rev Cardiol, 2018. 15(9): p. 505–522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Jakobsen JC, et al. , When and how should multiple imputation be used for handling missing data in randomised clinical trials – a practical guide with flowcharts. BMC Medical Research Methodology, 2017. 17(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Flegal KM, et al. , Prevalence and trends in obesity among US adidts, 1999-2008. Jama, 2010. 303(3): p. 235–41. [DOI] [PubMed] [Google Scholar]
- 36.Gale CR, Cooper C, and Sayer AA, Prevalence of frailty and disability: findings from the English Longitudinal Study of Ageing. Age Ageing, 2015. 44(1): p. 162–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Pefoyo AJ, et al. , The increasing burden and complexity of multimorbidity. BMC Public Health, 2015. 15: p. 415. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.