Weil E, Wachterman M, McCarthy EP, Davis RB, O'Day B, Iezzoni LI, Wee CC. Obesity Among Adults With Disabling Conditions. JAMA. 2002;288(10):1265-1268. doi:10.1001/jama.288.10.1265
Author Affiliations: Division of General Medicine and Primary Care, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Mass (Drs McCarthy, Davis, Iezzoni, and Wee and Mss Weil and Wachterman) and Cherry Engineering Support Services, Inc, McLean, Va (Dr O'Day).
Context Obesity, a leading cause of preventable death and chronic disease, is
associated with disability. Little is known about obesity among adults with
Objectives To determine the prevalence of obesity in adults with physical and sensory
limitations and serious mental illness.
Design, Setting, and Participants The 1994-1995 National Health Interview Survey of 145 007 US community-dwelling
respondents, 25 626 of whom had 1 or more disabilities.
Main Outcome Measures Likelihood of being obese, attempting weight loss, and receiving exercise
counseling among adults with and without disabilities.
Results Among adults with disabilities, 24.9% were obese vs 15.1% of those without
disabilities. After adjusting for sociodemographic factors, adults with a
disability were more likely to be obese, with an adjusted odds ratio (AOR)
of 1.9 (95% confidence interval [CI], 1.8-2.0). The highest risk occurred
among adults with some (AOR, 2.4; 95% CI, 2.3-2.5) or severe (AOR, 2.5; 95%
CI, 2.3-2.7) lower extremity mobility difficulties. After further adjustment
for comorbid conditions, adults with disabilities were as likely to attempt
weight loss as those without disabilities, except for adults with severe lower
extremity mobility difficulties, who were less likely (AOR, 0.7; 95% CI, 0.5-0.9]),
and adults with mental illness, who were more likely (AOR, 1.4; 95% CI, 1.2-1.8).
Physician exercise counseling was reported less often among adults with severe
lower extremity (AOR, 0.5; 95% CI, 0.4-0.7) and upper extremity (AOR, 0.7;
95% CI, 0.5-1.0) mobility difficulties.
Conclusion Obesity appears to be more prevalent in adults with sensory, physical,
and mental health conditions. Health care practitioners should address weight
control and exercise among adults with disabilities.
Obesity, a leading cause of preventable deaths, is more prevalent among
adults with disabilities.1- 4
One contributing factor is physical inactivity since adults with disabling
conditions or disabilities are more likely to face barriers to regular exercise.
Nevertheless, healthy weight and exercise are essential goals for the entire
population, and adults with disabilities should derive benefits for health
and overall functioning.1,5- 8
We examined the prevalence of obesity, weight loss attempts, and physician
exercise counseling among adults with mobility and sensory disabilities and
We pooled data from the 1994-1995 National Health Interview Survey (NHIS),
the 1994-1995 Disability Supplement (NHIS-D), and the 1995 Healthy People
2000 Supplement.9 The NHIS is a continuing
household survey of noninstitutionalized US adults (including adults in group
homes) conducted by the National Center for Health Statistics. In 1994 and
1995, all respondents were asked about sociodemographic factors (core survey)
and about specific sensory and physical limitations and psychiatric conditions
(disability supplement). One adult from half of the households in 1995 was
then asked about 6 chronic medical conditions (diabetes; chronic lung, kidney,
liver, or cardiac disease; and cancer), tobacco use, attempts to lose weight
and exercise counseling (Healthy People 2000 Supplement). The overall combined
response rate was 87% to the core survey and NHIS-D in both years and 94%
to the Healthy People 2000 Supplement. Proxies (35%) responded for adults
who were unable or unavailable to answer. The National Center for Health Statistics
weighted figures to account for nonresponse.9
We classified respondents into 6 categories, as described in detail
previously: blind/low vision (blind or serious difficulty seeing); deaf/hard
of hearing (difficulty hearing normal conversations or uses hearing aid);
lower extremity mobility difficulty (trouble walking, climbing stairs, standing,
or uses mobility aid); upper extremity mobility difficulty (difficulty reaching);
hand dexterity difficulty (difficulty grasping or holding a pen); and serious
mental illness (schizophrenia, major depression or paranoid, bipolar or severe
personality disorder).10 We divided the physical
impairment groups into 2 severity levels based on self-reports of serious
difficulties or use of a wheelchair or scooter.10
Although we call these "disabilities," many people with these conditions may
not view themselves as disabled.
We used SUDAAN's direct standardization method to adjust for age with
the entire NHIS sample as the standard population.11,12
We calculated age-sex standardized rates of obesity (body mass index [BMI]
>30 kg/m2), attempted weight loss, and physician exercise counseling
in the previous year. We conducted logistic regression analyses of disability
and obesity and simultaneously adjusted for relevant demographic factors including
age, sex, race/ethnicity, education, living alone, and family income. Among
the respondents to the Healthy People 2000 Supplement, we examined weight
loss and exercise counseling, adjusting additionally for BMI, current smoking,
and available comorbidities. We used SUDAAN to estimate SEs with Taylor series
linearization. We assessed for collinearity by comparing SEs from adjusted
and unadjusted models.11,12
Of 145 007 respondents, 25 626 reported at least 1 disabling
condition. Lower extremity mobility difficulties were most common. Poverty,
living alone, low education, inability to work, and smoking were more frequent
among adults with than without disabilities (Table 1).
Among adults with disabling conditions, 24.9% were obese compared to
15.1% among those without disabilities. Mild, moderate, and severe obesity
were more prevalent in adults with than without disabilities (Table 2). Rates of overweight were slightly lower among adults with
disabilities, except for those who were deaf or hard of hearing. After full
adjustment, adults with disabilities remained significantly more likely to
be in the obese category; adults with lower extremity mobility difficulties
were the most likely to be obese (Table
3). Although adults with upper extremity mobility difficulties were
also more likely to be obese, adjusting for other disabilities attenuated
this effect (adjusted odds ratios [AORs], 1.2 [95% confidence interval [CI],
1.1-1.4] and 1.1 [95% CI 1.0-1.3] for some and severe difficulty, respectively).
Results were similar when we limited analyses to self-responders.
Respondents with most disabilities were as likely to attempt weight
loss as adults without disabilities (Table
4). However, adults with severe lower extremity mobility difficulties
were less likely and adults with mental illness were more likely to attempt
weight loss. When we limited analyses to obese adults, the results were similar
for those with mental illness (AOR, 1.5 [95% CI, 0.9-2.3]); but adults with
severe lower extremity mobility difficulty were now as likely to attempt weight
loss (AOR, 1.0 [95% CI, 0.6-1.5]). However, overweight adults (BMI, 25-29.9
kg/m2) with severe mobility difficulty were still significantly
less likely to attempt weight loss (AOR, 0.4 [95% CI, 0.2-0.8]) than overweight
adults without mobility difficulties.
Compared with those without disabilities, most respondents with disabilities
were as likely to report exercise counseling. Exceptions included adults with
severe upper or lower extremity mobility difficulties (Table 4). When we limited analyses to obese adults, counseling differences
for adults with severe upper extremity mobility difficulty were no longer
statistically significant (AOR, 0.6 [95% CI, 0.3-1.2]).
We found that obesity was more prevalent in adults with disabling sensory,
physical, and mental health conditions than in the general US population.
Most adults with disabilities were as likely to attempt weight loss or report
exercise counseling as adults without disabilities. However, attempted weight
loss was less common for nonobese adults with lower extremity mobility difficulties
and more common for those with mental illness. Adults with severe upper or
lower extremity difficulties reported lower rates of physician exercise counseling.
Our results are consistent with previous studies showing an association
between obesity and disability. Numerous studies suggest that obesity increases
the risk of developing physical disability.2,3
However, few studies have examined obesity prevalence in populations with
specific types of disabilities and mental illness.13,14
Unlike many other studies, which use a broad definition of disability based
on limited functioning, we assessed obesity among adults with physical and
sensory impairments, and mental health conditions using a national sample.
Obesity is a serious public health concern that not only increases mortality
and morbidity, but also diminishes quality of life.1,4,15
Our estimates of higher obesity rates among those with disabilities are likely
underestimates because the prevalence of obesity has increased since 1994.16 Furthermore, we did not consider weight loss associated
with drug addiction or eating disorders such as anorexia nervosa or bulimia,
particularly among those with some types of mental illness who are likely
to have higher rates of these conditions. Interestingly, most adults with
disabilities were as likely as those without disabilities to attempt weight
loss, which suggests that adults with disabilities share concerns about their
weight. The finding that most adults with and without disabilities reported
exercise counseling at similar rates is encouraging, but must be interpreted
in light of overall low levels of physician counseling about exercise and
Particularly troubling is the prevalence of obesity in those with mobility
difficulties, especially since population-standardized BMI categories likely
underestimate health risks for adults with reduced lean muscle mass. Fortunately,
obese adults with severe mobility difficulties were as likely to attempt weight
loss as obese adults without these difficulties. However, overweight adults
with these impairments were substantially less likely to attempt weight loss
compared with overweight adults without mobility impairments, even though
being overweight confers substantial health risk and predisposes one to developing
obesity.15 Furthermore, regardless of BMI,
adults with mobility difficulties were less likely to report exercise counseling.
Physicians may hesitate to encourage exercise because they perceive greater
impediments to physical activity.19 Although
exercise contributes only moderately to weight loss, it is important for weight
maintenance since inactivity contributes to a "vicious cycle" where inactivity
contributes to obesity, obesity exacerbates disability, and disability impedes
Adults with mental illness were also distinct as the only group to attempt
weight loss more frequently than the general population. This may be partly
due to weight gain induced by psychotropic medications, a common reason for
medication nonadherence. Higher weight loss attempts may be provoked by this
Interestingly, adults who were deaf or hard of hearing were also more
likely to be obese. This group may have fewer "apparent" barriers to exercise
than adults with other disabilities, but clearly more research is needed to
understand the reasons for this excess risk of obesity.
Guidelines that recommend obesity screening presumably apply to adults
with disabilities, but few offer guidance on how to counsel this population.1,5,6 Effective counseling
must address obstacles to weight control and exercise, such as time constraints,
limited availability of exercise places or equipment, and inadequate reimbursement
for weight control treatment.21 Furthermore,
adults with disabilities must confront environmental and disability-specific
barriers, such as availability of accessible facilities and transportation.
Additionally, physical impairments, including pain and weakness, may hinder
or preclude certain physical activities.
Our study has limitations. Weight and height figures were self-reported
and may be inaccurate: overweight respondents are more likely to underestimate
weight and overestimate height than thinner respondents, and those who cannot
stand on a scale or see scale readings may be more prone to reporting imprecise
weights.22 Proxies may provide incorrect information,
although analyses restricted to self-responders were similar to the overall
sample. Unfortunately, the NHIS did not assess potential contributors to obesity,
such as energy intake or physical activity among adults with disabilities,
and did not address nutritional counseling, which is central to weight management.
Finally, because we did not have information on the temporal relationship
of disability and obesity, we could not draw inferences on causality.
Nonetheless, our study demonstrates that obesity is more prevalent in
adults with disabling sensory, mobility, and mental health conditions. Physicians
should recognize that patients with disabilities face increased risks for
obesity and address their weight concerns. Additional studies and more detailed
clinical guidelines are needed to help physicians promote weight control and
exercise among adults with disabilities.