Disability and Health Journal 7 (2014) 124e130 www.disabilityandhealthjnl.com
Brief Report
Obesity and symptoms and quality of life indicators of individuals with disabilities Rana Salem, M.A.*, Alyssa M. Bamer, M.P.H., Kevin N. Alschuler, Ph.D., Kurt L. Johnson, Ph.D., and Dagmar Amtmann, Ph.D. Department of Rehabilitation Medicine, University of Washington, Seattle, WA, USA
Abstract Background: Health risks of obesity are well known, but effects of obesity on health-related quality of life (HRQOL) have not been well-studied in people with physical disabilities. Objective/hypothesis: We examined the association between obesity and HRQOL in people with disabilities relative to the general US population. We hypothesized (a) overall, individuals with disabilities will report worse HRQOL than the general US population and (b) obese individuals with disabilities will report worse HRQOL than non-obese individuals. Methods: Individuals with muscular dystrophy, multiple sclerosis, post-polio syndrome, and spinal cord injury (N 5 1849) completed Patient Reported Outcomes Measurement Information System (PROMIS) measures of fatigue, pain interference, physical and social function, depression, sleep disturbance, and sleep-related impairment. Participants were classified as obese or non-obese based on self-reported weight and height (BMI) and/or waist circumference (WC). PROMIS T-scores were compared to norms and between obesity groups. Results: Mean BMI was 26.4 kg/m2 with 23.4% classified as obese. Mean WC was 37.5 inches (males) and 34.0 inches (females); 26.4% reported abdominal obesity. Based on BMI and/or WC, 33.3% (n 5 616) were classified obese. Compared to PROMIS norms, obese individuals reported worse HRQOL on all domains ( p ! 0.0001). Compared to non-obese individuals, obese individuals reported worse functioning on all domains except depression ( p ! 0.01). Obese individuals with MS and MD reported worse outcomes than non-obese counterparts. Conclusions: Obesity in people with physical disabilities is associated with poorer HRQOL. More research is needed to inform clinicians in identifying obese patients and helping them achieve healthy weight, reduce symptom burden, and improve QOL. Ó 2014 Elsevier Inc. All rights reserved. Keywords: Quality of life; Body mass index; Waist circumference; Obesity; Physical disabilities
Individuals living with a long-term physical disability are often susceptible to obesity due to difficulties maintaining a healthy diet, medication side-effects, mobility limitations, and disability-related sequelae (e.g., gait impairment, spasticity, pain, fatigue) that may impede physical activity.1e7 The serious health risks of obesity (e.g., type 2 diabetes, coronary heart disease, cancer) are well known.8 However, its association with various secondary conditions such as depression,9e13 sleep disturbance,14,15 and pain,16,17 The authors have no conflicts of interest. An abstract based on this data was previously presented at the American Public Health Association 140th Annual Meeting, October 29, 2012. The contents of this manuscript were developed under a grant from the Department of Education, National Institute on Disability and Rehabilitation Research (NIDRR) grant number H133B080024. However, those contents do not necessarily represent the policy of the Department of Education, and you should not assume endorsement by the Federal Government. * Corresponding author. Tel.: þ1 206 462 5070; fax: þ1 206 685 9224. E-mail address: [email protected] (R. Salem). 1936-6574/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.dhjo.2013.10.003
has been less well studied, especially in persons with disabilities. People with disabilities tend to experience certain secondary conditions, such as fatigue, depression, and sleep disturbance, at a higher rate than the general population.18e23 These conditions can further limit mobility and activity, resulting in negative health consequences, including obesity. However, previous research in persons with spinal cord injury (SCI) have found inconsistent associations between obesity and health-related quality of life (HRQOL) indicators. Hetz and colleagues24 found that individuals with SCI who self-reported being overweight (by responding to a yes or no question) had greater pain and depressive symptoms, and lower satisfaction with life than individuals who did not report being overweight. Chen and colleagues1 found an association between obesity and pain, but no relationship between body mass index (BMI) and depression and satisfaction with life in persons with SCI. BMI may not be the best way to investigate obesity in some individuals with disabilities25e27 and may contribute
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
to the mixed findings between HRQOL and BMI. Limited mobility, spinal deformities and spasticity may make it difficult to obtain accurate height and weight measurements, resulting in inaccurate BMI.26,28 Furthermore, individuals with a physical disability may have different body composition than people without physical disabilities. For example, persons with SCI may have atrophied muscle mass25 and a greater amount of fatty tissue29 than individuals without SCI, underestimating BMI and obesity in individuals with SCI. More accurate measures of obesity in persons with physical disabilities include bioelectrical impedance analysis26,30 and anthropometric index.31 However, these methods can be costly, require in-person measurement, and are not feasible in research studies where data are collected from a large geographically diverse sample. Abdominal obesity has recently been implicated as a significant risk factor for diseases such as type 2 diabetes and cardiovascular disease.32 Waist circumference (WC) provides a good estimate of abdominal visceral fat33 and has been suggested as a potentially low-cost and valid measure of obesity in individuals with physical disabilities.26,28 Past studies on the association between obesity and secondary conditions in persons with physical disabilities have primarily focused on the SCI population. Therefore, the purpose of this study was to examine the association between obesity and HRQOL indicators in a large group of individuals living with a physical disability relative to the general US population. We hypothesized that (a) individuals with disabilities, regardless of their obesity status, will report worse HRQOL indicators than the general US population and (b) obese individuals with disabilities will report on average worse HRQOL indicators than non-obese individuals with disabilities.
Methods Participants Individuals participating in a longitudinal study of aging with a physical disability completed self-report surveys every year for four years to collect information about their disease progression, symptom ratings (pain, fatigue, depression, sleep, cognition), and other factors such as coping skills, resilience, and well-being. Cross-sectional data from the baseline survey collected in 2009e2010 were used for this study; additional information about this study sample and procedures have been described in previously published articles.34,35 Briefly, individuals were recruited from multiple sources, including: (1) participants from previous research studies who indicated interest in future research, (2) University of Washington’s disability-specific registry of individuals interested in being contacted about future research, (3) referral by family and friends, and (4) print and web advertisements. Eligible participants were age 18 years or older with a self-reported diagnosis of muscular dystrophy (MD), multiple sclerosis (MS), post-polio syndrome (PPS), or
125
spinal cord injury (SCI). A total of 2041 surveys were mailed to eligible and interested individuals. Of those, 1862 completed the survey (91.2% response rate) and 1849 participants provided complete BMI or WC data (MD: 339; MS: 579: PPS: 443; SCI: 488). Participants from all 50 states were represented, with the top three states being Washington (28.6%), Georgia (7.1%), and California (6.2%). All participants provided informed consent and received $25 upon survey completion; procedures were approved by University of Washington’s Human Subjects division. Measures Demographics Basic demographic information (age at survey, disability, race/ethnicity, education, household income, and marital status) was collected from all participants. Urban or rural status was based on participant zip code at survey completion.36 Obesity Obesity status was measured by two indexes. First, body mass index (BMI) was calculated using self-reported weight and height. Calculated BMI (kg/m2) was then classified into four categories: underweight (!18.5), normalweight (18.5e24.9), overweight (25.0e29.9), or obese (>30).37 Second, waist circumference (WC) was used: Participants provided self-reported WC in inches and indicated if it was based on an actual recent measure, pants size, or similar measure. Independent of BMI, abdominal obesity as defined by a WC O 40 inches (males) or O35 inches (females) is predictive of disease.38 The National Heart, Lung, and Blood Institute has recommended that these WC cut-offs be used in conjunction with a BMI of >25 kg/m2 to identify individuals who are at increased risk of disease because of their WC.38 For this study, obese status was categorized as having an obese BMI and/or high-risk abdominal obesity. All others were categorized as non-obese. Adding the abdominal obesity component may identify study respondents with underestimated BMI who have a greater amount of visceral fat. HRQOL indicators Seven constructs (i.e., domains) were measured using Patient Reported Outcomes Measurement Information System (PROMIS) v1.0 short forms: physical function, fatigue, pain interference, depression, satisfaction with participation in social roles, sleep disturbance, and sleep-related impairment. PROMIS measures were developed using item response theory, have excellent psychometric properties and have been validated in various clinical populations.39e41 PROMIS norming sample data were primarily collected in 2006e2007 by YougovPolimetrix.com, an internet polling company. Polimetrix selected participants from a database of more than 1 million individuals who provided contact information and regularly participated in
126
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
online surveys; 13250 general population respondents provided data. A subset of this PROMIS general population sample (n 5 5239) matching the 2000 US Census on gender, age, and race40 was created to establish US population norms: 57% female; 15% 18e29, 22% 30e44, 28% 45e59, 22% 60e74 and 13% >75 years old; 74% White, 10% Black, 11% Hispanic/Latino, and 4% other; 51% had more than a high school education. Each PROMIS domain raw score is converted to a standardized T-score metric with a US general population mean of 50 and standard deviation (SD) of 10. The estimated US general population mean is the mean of the PROMIS norming sample. A higher score on any measure means more of the domain being measured (see www.nihpromis.org). Analyses Demographic and QOL means and frequencies were first calculated by obesity status to describe the sample. Demographics were compared between obesity groups using Student t-tests for continuous variables and chi-square tests for categorical variables. To test the hypothesis that individuals with disabilities report worse HRQOL than the PROMIS norm sample, onesample t-tests were used to compare each study sample domain mean to the norm mean of 50, overall and by obesity status. To account for multiple comparisons, a Bonferroni correction adjusted the alpha level to 0.0167 (0.05/3). To test the hypothesis that obese individuals with disabilities report worse HRQOL than non-obese individuals, PROMIS T-scores were compared between obesity status groups using Student t-tests. To provide an at-a-glance graph that shows symptom patterns, we plotted all mean domain T-scores by obesity status. We also compared Tscores by obesity status for each disability group; alpha level adjusted to 0.0125 (0.05/4). A multivariate regression model was used to examine the effects of obesity on HRQOL after adjusting for demographic characteristics. The model included the following independent variables: age, gender, race/ethnicity (non-Hispanic white vs. other), marital status (married/living with partner vs. other), education (college degree or greater vs. other), household income, years since diagnosis, state of residency at survey completion (WA vs. other), rural/urban designation, and obesity status group. Analyses were conducted using SAS software v9.3.42 Results Demographics The mean BMI of all respondents was 26.4 kg/m2 with 5.3% classified as underweight, 41.6% normal-weight, 29.7% overweight, and 23.4% obese. The mean WC was 37.5 inches for men and 34.0 inches for women, with 26.4% of respondents who provided a WC categorized as having abdominal obesity. A total of 33.3% (n 5 616) were
classified as obese BMI and/or having abdominal obesity; 291 of these individuals were classified as obese using both indexes. Obese respondents were older and less educated than non-obese individuals ( p ! 0.001) (Table 1). The obese group also tended to be female ( p ! 0.01) and reported less household income than non-obese individuals ( p ! 0.001). Comparisons between sample and normative PROMIS means Except for sleep-related disturbance for the non-obese group, both non-obese and obese individuals with Table 1 Participant characteristics by obesity status Non-obese (n 5 1233)
Age Years diagnosed Height (inches) Weight (pounds) Waist circumference (inches)
Obese (n 5 616)
Mean 6 SD
Mean 6 SD
pa
55.2 6 14.1 15.6 6 10.8 66.8 6 4.3 148.1 6 29.7 32.6 6 4.3
58.0 6 11.3 15.1 6 10.4 66.4 6 4.2 206.5 6 39.3 40.7 6 5.5
!0.0001 0.3648 0.0241 !0.0001 !0.0001
n (%)
n (%)
b
Diagnosis Muscular dystrophy 240 (70.8) 99 (29.2) Multiple sclerosis 379 (65.5) 200 (34.5) Post-polio syndrome 263 (59.4) 180 (40.6) Spinal cord injury 351 (71.9) 137 (28.1) Sexc Male 480 (38.9) 200 (32.5) Female 753 (61.1) 416 (67.5) Race/ethnicityc Non-Hispanic white 1131 (92.5) 561 (91.2) Other 92 (7.5) 54 (8.8) Educationc Some college or less 517 (41.9) 314 (51.0) College or graduate degree 716 (58.1) 302 (49.0) Marital statusc Married/living with partner 774 (62.9) 391 (63.5) Other 457 (37.1) 225 (36.5) Household incomec !$25,000 260 (22.3) 152 (25.9) $25,000e$40,000 177 (15.2) 123 (21.0) $41,000e$55,000 157 (13.5) 84 (14.3) $56,000e$70,000 132 (11.3) 63 (10.8) $71,000e$85,000 117 (10.0) 48 (8.2) $86,000e$100,000 121 (10.4) 54 (9.2) O$100,000 203 (17.4) 62 (10.6) Urban/rural designation of zip code at survey completionc Urban 986 (80.0) 480 (77.9) Rural 246 (20.0) 136 (22.1) State of residence at survey completionc Washington 350 (28.4) 179 (29.1) Other 883 (71.6) 437 (70.9)
0.0002
0.0064
0.3502
0.0002
0.8016
0.0004
0.2928
0.7632
BMI 5 body mass index. Demographics were compared between obesity groups using Student ttests for continuous variables or chi-square tests for categorical variables. b Distribution of each disability across obesity status categories. c Distribution of each characteristic per obesity status category. a
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
disabilities reported worse HRQOL scores on all domains relative to the PROMIS normative sample (all p ! 0.0001; Fig. 1). Compared to PROMIS norms, the difference in sample T-scores ranged from 0.5 points (nonobese sleep-related disturbance) to 15.1 points (obese physical function). Comparisons between sample obesity status means Compared to the non-obese group, obese individuals reported statistically significantly worse functioning on all domains except depression (all p ! 0.01; Fig. 1). The largest T-score difference between obesity groups was 2.6 points for physical function. The smallest difference in Tscores was 1.0 points (depression). Disability-specific results The proportion of obese individuals was highest for PPS (40.6%) followed by MS (34.5%), MD (29.2%), and SCI (28.1%). Obese individuals with MS reported significantly worse outcomes on all domains except depression when compared to the non-obese MS group (all p ! 0.0125). The largest T-score difference was 4.6 points for sleeprelated impairment. Obese individuals with MD reported worse physical function, satisfaction with participation in social roles, and fatigue relative to non-obese individuals with MD. Only physical function scores were significantly worse for obese individuals with PPS compared to
127
non-obese individuals with PPS. There were no statistically significant differences for the SCI group. Multivariate regression model results After controlling for demographic variables, obese status was statistically significantly associated with all PROMIS domains except depression ( p ! 0.05, Table 2 contains standardized Beta coefficients). Discussion The purpose of this study was to examine the association between obesity and levels of symptoms and QOL indicators within disability groups and also relative to a normative sample representative of the US population in age, gender, and race/ethnicity. As previously reported for these populations27 and for others,1 higher BMI levels were associated with older age and less education. Consistent with our hypothesis, results suggest that people with disabilities report worse HRQOL indicators when compared to the PROMIS normative sample. Among persons with disabilities, important differences emerged between persons classified as obese relative to those classified as non-obese including poorer physical and social functioning, and higher fatigue, pain interference and sleep and sleep-related disturbance. The association between obesity and these HRQOL domains remained statistically significant after we controlled for
Fig. 1. PROMIS domain T-scores for individuals with physical disabilities, by obesity status. The figure shows the mean PROMIS T-scores of the obese and non-obese groups. The T-score of 50 represents the PROMIS norm sample mean. A higher score on any measure means more of the domain being measured, i.e., higher physical function score indicates better physical function while higher depression score indicates more depression. Left-facing bars indicate scores that are worse than the PROMIS normative sample mean for domains where a higher score is better (physical function and social role satisfaction) and right-facing bars indicate scores that are worse than the PROMIS normative sample mean for the domains where lower scores are better (sleep disturbance, sleep-related impairment depression, pain interference, and fatigue). Asterisk indicates statistically significant difference in T-score between the obese and non-obese groups ( p ! 0.05).
128
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
Table 2 Multiple linear regression results (standardized Beta coefficients) examining factors associated with PROMIS domains in individuals with a physical disability PROMIS domain Predictors in model
Physical functiona
Social role satisfactiona
Fatigueb
Pain interferenceb
Depressionb
Sleep disturbanceb
Sleep-related impairmentb
Age (years) Female sex Non-Hispanic white Married/living with partner College or graduate degree Household income Years diagnosed Washington state residence Rural zip code Obese statusc
0.16* 0.09* 0.05* 0.04 0.01 0.20* 0.20* 0.12* 0.01 0.10*
0.13* 0.03 0.02 0.05 0.04 0.23* 0.05 0.06* 0.02 0.09*
0.09* 0.21* 0.03 0.09* 0.03 0.16* 0.02 0.05* 0.03 0.10*
0.08* 0.00 0.07 0.03 0.09* 0.19* 0.06 0.04 0.04 0.10*
0.02 0.03 0.00 0.07 0.03 0.16* 0.02 0.07* 0.00 0.06
0.12* 0.05* 0.00 0.00 0.10* 0.07* 0.00 0.03 0.03 0.05*
0.12* 0.07 0.06 0.03 0.02 0.10* 0.07 0.06 0.02 0.09*
PROMIS 5 Patient Reported Outcomes Measurement Information System. *Significantly associated with PROMIS domain ( p ! 0.05). a Higher score indicates better functioning. b Higher score indicates worse functioning. c Association of obesity status with health-related quality of life scores after adjusting for age, gender, race/ethnicity, marital status, education, household income, years since diagnosis, state of residency at survey completion, and rural/urban designation.
several demographic characteristics. Results from this study also support disability-specific differences in how obesity may affect symptoms and QOL, extending the findings made by Alschuler and colleagues.27 Obese individuals with MS and, to a lesser degree MD, reported worse functioning than normal weight individuals. We did not see this effect for the PPS and SCI groups. It is possible that disability-specific characteristics (e.g., disability severity) may moderate the effects of obesity on HRQOL. Future studies of obesity in persons with disabilities are needed given the paucity of research in this area. The differences between the obese and non-obese groups, while consistent, were relatively small and may not reach clinical significance. Clinical significance is often evaluated by using minimally clinically important differences (MCID). MCIDs have been developed for some PROMIS domains in advanced-stage cancer patients,43 but such guidelines have not been developed for this study’s population. An alternative is to use 0.5 SD, which is generally considered to be a good approximation for clinically important difference.44 For PROMIS instruments, 5 points is equivalent to 0.5 SD. With this criterion in mind, differences in average physical function, fatigue, pain interference and satisfaction with participation in social roles reached clinical significance when comparing T-scores between the PROMIS normative sample and obese group. However, we observed clinically significant differences for these domains between the non-obese group and the PROMIS normative sample as well. This suggests that simply living with a physical disability is associated with worse HRQOL when compared with a general population. The obese group was generally about 2.5 points worse than the non-obese group for each of these domains,
indicating that obesity adds a clinically small, but statistically significant burden on individuals living with a disability. BMI may not capture obesity in persons with disabilities and may even vary by disability.27 This may partially account for differences in study sample (23.4%) and national (35.7%)45 obese BMI levels. By including individuals with abdominal obesity, this study identified an additional 183 individuals with disabilities who are at high-risk for disease and produced a study sample obesity level (33.3%) that is much closer to the national rate. Future research is needed to examine whether BMI or WC or a combination is the best indicator of obesity in individuals with physical disabilities and to establish acceptable cut-off levels for this population. In this study we grouped individuals with under-, normal- and normal WC/over-weight BMI into a nonobese group to focus on the association between obesity and HRQOL. We recognize that there are health issues specific to being under-46 or over-weight.47 Published studies have reported that unlike those who are obese, being overweight does not result in an increased risk of mortality relative to normal-weight individuals,48 suggesting that the adverse effects of increased BMI are especially true for those classified as obese. Indeed, in results not shown, study sample T-scores were similar between the normalweight and the under- and over-weight groups except for physical function and pain interference, respectively. Finally, self-reported BMI at the lower and higher levels tend to be over- and under-reported, respectively.49 However, as direct measurement of BMI is sometimes not practicable in large studies like this, self-report has been found to be an acceptable alternative.50
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
Conclusion The study findings make an important contribution to the literature by being the first study to examine the relationship between obesity and symptoms and QOL indicators in people with MD, MS, and PPS. We conclude that there is much more to learn about obesity in persons with physical disabilities. BMI and WC cut-off levels should be established in persons with disabilities so we can better study obesity in large epidemiological-type studies in which physical measurements are impractical. The difference in outcomes between the obese and non-obese groups suggests an association between obesity and HRQOL, especially among people with MS or MD. However, prospective studies must be conducted to establish causality and to examine disability-specific factors associated with obesity. With this information, health care professionals will be better able to identify and counsel patients about effective ways to improve their health and well-being. References 1. Chen Y, Cao Y, Allen V, Richards J. Weight matters: physical and psychosocial well being of persons with spinal cord injury in relation to body mass index. Arch Phys Med Rehabil. 2011;92(3):391e398. 2. Marrie R, Horwitz R, Cutter G, Tyry T, Campagnolo D, Vollmer T. High frequency of adverse health behaviors in multiple sclerosis. Mult Scler. 2009;15(1):105e113. 3. Timmerman G, Stuifbergin A. Eating patterns in women with multiple sclerosis. J Neurosci Nurs. 1999;31(3):152e158. 4. Ahlstr€ om G, Karlsson U. Disability and quality of life in individuals with postpolio syndrome. Disabil Rehabil. 2000;22(9):416e422. 5. van Nimwegen M, Speelman AD, Hofman-van Rossum EJ, et al. Physical inactivity in Parkinson’s disease. J Neurol. 2011;258(12): 2214e2221. 6. Sosnoff JJ, Gappmaier E, Frame A, Motl R. Influence of spasticity on mobility and balance in persons with multiple sclerosis. J Neurol Phys Ther. 2011;35(3):129e132. 7. Motl RW, Snook EM, Schapiro R. Symptoms and physical activity behavior in individuals with multiple sclerosis. Res Nurs Health. 2008;31(5):466e475. 8. National Task Force on the Prevention and Treatment of Obesity. Overweight, obesity, and health risk. Arch Intern Med. 2000;160(7):898e904. 9. Luppino FS, de Wit LM, Bouvy PF, et al. Overweight, obesity, and depression: a systematic review and meta-analysis of longitudinal studies. Arch Gen Psychiatry. 2010;67(3):220e229. 10. Patterson RE, Frank LL, Kristal AR, White E. A comprehensive examination of health conditions associated with obesity in older adults. Am J Prev Med. 2004;27(5). 11. Zhao G, Ford ES, Li C, Tsai J, Dhingra S, Balluz L. Waist circumference, abdominal obesity, and depression among overweight and obese U.S. adults: National Health and Nutrition Examination Survey 2005-2006. BMC Psychiatry. 2011;11(130). 12. Ma J, Xiao L. Obesity and depression in US women: results from the 2005-2006 National Health and Nutritional Examination Survey. Obesity (Silver Spring). 2010;18(2):347e353. 13. Herva A, Laitinen J, Miettunen J, et al. Obesity and depression: results from the longitudinal Northern Finland 1966 Birth Cohort Study. Int J Obes. 2006;30(3):520e527. 14. Resta O, Foschino Barbaro MP, Bonfitto P, et al. Low sleep quality and daytime sleepiness in obese patients without obstructive sleep apnoea syndrome. J Intern Med. 2003;253(5):536e543.
129
15. Vorona RD, Winn MP, Babineau TW, Eng BP, Feldman HR, Ware J. Overweight and obese patients in a primary care population report less sleep than patients with a normal body mass index. Arch Intern Med. 2005;165:25e30. 16. Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Viikari-Juntura E. The association between obesity and low back pain: a meta-analysis. Am J Epidemiol. 2010;171(2):135e154. 17. Stone AA, Broderick J. Obesity and pain are associated in the United States. Obesity (Silver Spring). 2012;20(7):1491e1495. 18. Cook K, Molton I, Jensen M. Fatigue and aging with a disability. Arch Phys Med Rehabil. 2011;92(7):1126e1133. 19. Bamer A, Johnson K, Amtmann D, Kraft G. Prevalence of sleep problems in individuals with multiple sclerosis. Mult Scler. 2008;14(8): 1127e1130. 20. Wu N, Minden S, Hoaglin D, Hadden L, Frankel D. Quality of life in people with multiple sclerosis: data from the Sonya Slifka Longitudinal Multiple Sclerosis Study. J Health Hum Serv Adm. 2007;30(3): 233e267. 21. Kinne S, Patrick D, Doyle D. Prevalence of secondary conditions among people with disabilities. Am J Public Health. 2004;94(3): 443e445. 22. Robinson-Whelen S, Taylor H, Hughes R, Nosek M. Depressive symptoms in women with physical disabilities: identifying correlates to inform practice. Arch Phys Med Rehabil; 2013 [in press], http://dx. doi.org/10.1016/j.apmr.2013.07.013. 23. Okoro C, Strine T, Balluz L, et al. Serious psychological distress among adults with and without disabilities. Int J Public Health. 2009;54(suppl 1):52e60. 24. Hetz SP, Latimer AE, Arbour-Nicitopoulos KP, Martin Ginis K. Secondary complications and subjective well-being in individuals with chronic spinal cord injury: associations with self-reported adiposity. Spinal Cord. 2011;49(2):266e272. 25. Giangregorio L, McCartney N. Bone loss and muscle atrophy in spinal cord injury: epidemiology, fracture prediction, and rehabilitation strategies. J Spinal Cord Med. 2006;29(5):489e500. 26. Eriks-Hoogland I, Hilfiker R, Baumberger M, Balk S, Stucki G, Perret C. Clinical assessment of obesity in persons with spinal cord injury: validity of waist circumference, body mass index, and anthropometric index. J Spinal Cord Med. 2011;34(4):416e422. 27. Alschuler KN, Gibbons LE, Rosenberg DE, et al. Body mass index and waist circumference in persons aging with muscular dystrophy, multiple sclerosis, post-polio syndrome, and spinal cord injury. Disabil Health J. 2012;5(3):177e184. 28. Buchholz AC, Bugaresti J. A review of body mass index and waist circumference as markers of obesity and coronary heart disease risk in persons with chronic spinal cord injury. Spinal Cord. 2005;43(9): 513e518. 29. Jones LM, Legge M, Goulding A. Healthy body mass index values often underestimate body fat in men with spinal cord injury. Arch Phys Med Rehabil. 2003;84:1068e1071. 30. Spungen AM, Bauman WA, Wang J, Pierson RN Jr. Measurement of body fat in individuals with tetraplegia: a comparison of eight clinical methods. Paraplegia. 1995;33(7):402e408. 31. Bulbulian R, Johnson RE, Gruber JJ, Darabos B. Body composition in paraplegic male athletes. Med Sci Sports Exerc. 1987;19(3): 195e201. 32. Siren R, Eriksson JG, Vanhanen H. Waist circumference a good indicator of future risk for type 2 diabetes and cardiovascular disease. BMC Public Health. 2012;12:631. 33. Lemieux S, Prud’homme D, Bouchard C, Tremblay A, Despres J. A single threshold value of waist girth identifies normal-weight and overweight subjects with excess visceral adipose tissue. Am J Clin Nutr. 1996;64:685e693. 34. Alschuler KN, Jensen MP, Goetz MC, Smith AE, Verrall A, Molton IR. Effects of pain and fatigue on physical functioning and depression in persons with muscular dystrophy. Disabil Health J. 2012;5(4):277e283.
130
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
35. Cook KF, Bamer AM, Amtmann D, Molton IR, Jensen M. Six patient-reported outcome measurement information system short form measures have negligible age- or diagnosis-related differential item functioning in individuals with disabilities. Arch Phys Med Rehabil. 2012;93(7):1289e1291. 36. Centers for Medicare and Medicaid Services. Ambulance Fee Schedule: National Breakout of Geographic Area Definitions by Zip Code; 2013. 37. World Health Organization. Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation. WHO Technical Report Series 894. Geneva: World Health Organization; 2000. 38. National Institutes of Health NH, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Obes Res. 1998;6(suppl 2):S51eS209. 39. Cella D, Riley W, Sone A, et al. Initial item banks and first wave testing of the Patient-Reported Outcomes Measurement Information System (PROMIS) network: 2005-2008. J Clin Epidemiol. 2010;63(11):1179e1194. 40. Liu H, Cella D, Gershon R, et al. Representativeness of the PROMIS Internet panel. J Clin Epidemiol. 2010;63(11):1169e1178. 41. Rothrock N, Hays R, Spritzer K, Yount S, Riley W, Cella D. Relative to the general US population, chronic diseases are associated with poorer health-related quality of life as measured by the PatientReported Outcomes Measurement Information System (PROMIS). J Clin Epidemiol. 2010;63(11):1195e1204.
42. SAS Software [program]. Cary, NC: SAS Institute Inc.; 2011. 43. Yost KJ, Eton DT, Garcia SF, Cella D. Minimally important differences were estimated for six Patient-Reported Outcomes Measurement Information System-Cancer scales in advanced-stage cancer patients. J Clin Epidemiol. 2011;64(5):507e516. 44. Farivar SS, Liu H, Hays R. Half standard deviation estimate of the minimally important difference in HRQOL scores? Expert Rev Pharmacoecon Outcomes Res. 2004;4(5):515e523. 45. Flegal K, Carroll M, Kit B, Ogden C. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. J Am Med Assoc. 2012;307(5):491e497. 46. Flegal KM, Graubard BI, Williamson DF, Gail M. Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA. 2007;298(17):2028e2037. 47. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003;289(1):76e79. 48. Flegal KM, Graubard BI, Williamson DF, Gail M. Excess deaths associated with underweight, overweight, and obesity. JAMA. 2005;293(15):1861e1867. 49. Stommel M, Schoenborn C. Accuracy and usefulness of BMI measures based on self-reported weight and height: findings from the NHANES & NHIS 2001-2006. BMC Public Health. 2009;9:421. 50. Bes-Rastrollo M, Sabate J, Jaceldo-Siegl K, Fraser G. Validation of self-reported anthropometrics in the Adventist Health Study 2. BMC Public Health. 2011;5(11):213.
Brief Report
Obesity and symptoms and quality of life indicators of individuals with disabilities Rana Salem, M.A.*, Alyssa M. Bamer, M.P.H., Kevin N. Alschuler, Ph.D., Kurt L. Johnson, Ph.D., and Dagmar Amtmann, Ph.D. Department of Rehabilitation Medicine, University of Washington, Seattle, WA, USA
Abstract Background: Health risks of obesity are well known, but effects of obesity on health-related quality of life (HRQOL) have not been well-studied in people with physical disabilities. Objective/hypothesis: We examined the association between obesity and HRQOL in people with disabilities relative to the general US population. We hypothesized (a) overall, individuals with disabilities will report worse HRQOL than the general US population and (b) obese individuals with disabilities will report worse HRQOL than non-obese individuals. Methods: Individuals with muscular dystrophy, multiple sclerosis, post-polio syndrome, and spinal cord injury (N 5 1849) completed Patient Reported Outcomes Measurement Information System (PROMIS) measures of fatigue, pain interference, physical and social function, depression, sleep disturbance, and sleep-related impairment. Participants were classified as obese or non-obese based on self-reported weight and height (BMI) and/or waist circumference (WC). PROMIS T-scores were compared to norms and between obesity groups. Results: Mean BMI was 26.4 kg/m2 with 23.4% classified as obese. Mean WC was 37.5 inches (males) and 34.0 inches (females); 26.4% reported abdominal obesity. Based on BMI and/or WC, 33.3% (n 5 616) were classified obese. Compared to PROMIS norms, obese individuals reported worse HRQOL on all domains ( p ! 0.0001). Compared to non-obese individuals, obese individuals reported worse functioning on all domains except depression ( p ! 0.01). Obese individuals with MS and MD reported worse outcomes than non-obese counterparts. Conclusions: Obesity in people with physical disabilities is associated with poorer HRQOL. More research is needed to inform clinicians in identifying obese patients and helping them achieve healthy weight, reduce symptom burden, and improve QOL. Ó 2014 Elsevier Inc. All rights reserved. Keywords: Quality of life; Body mass index; Waist circumference; Obesity; Physical disabilities
Individuals living with a long-term physical disability are often susceptible to obesity due to difficulties maintaining a healthy diet, medication side-effects, mobility limitations, and disability-related sequelae (e.g., gait impairment, spasticity, pain, fatigue) that may impede physical activity.1e7 The serious health risks of obesity (e.g., type 2 diabetes, coronary heart disease, cancer) are well known.8 However, its association with various secondary conditions such as depression,9e13 sleep disturbance,14,15 and pain,16,17 The authors have no conflicts of interest. An abstract based on this data was previously presented at the American Public Health Association 140th Annual Meeting, October 29, 2012. The contents of this manuscript were developed under a grant from the Department of Education, National Institute on Disability and Rehabilitation Research (NIDRR) grant number H133B080024. However, those contents do not necessarily represent the policy of the Department of Education, and you should not assume endorsement by the Federal Government. * Corresponding author. Tel.: þ1 206 462 5070; fax: þ1 206 685 9224. E-mail address: [email protected] (R. Salem). 1936-6574/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.dhjo.2013.10.003
has been less well studied, especially in persons with disabilities. People with disabilities tend to experience certain secondary conditions, such as fatigue, depression, and sleep disturbance, at a higher rate than the general population.18e23 These conditions can further limit mobility and activity, resulting in negative health consequences, including obesity. However, previous research in persons with spinal cord injury (SCI) have found inconsistent associations between obesity and health-related quality of life (HRQOL) indicators. Hetz and colleagues24 found that individuals with SCI who self-reported being overweight (by responding to a yes or no question) had greater pain and depressive symptoms, and lower satisfaction with life than individuals who did not report being overweight. Chen and colleagues1 found an association between obesity and pain, but no relationship between body mass index (BMI) and depression and satisfaction with life in persons with SCI. BMI may not be the best way to investigate obesity in some individuals with disabilities25e27 and may contribute
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
to the mixed findings between HRQOL and BMI. Limited mobility, spinal deformities and spasticity may make it difficult to obtain accurate height and weight measurements, resulting in inaccurate BMI.26,28 Furthermore, individuals with a physical disability may have different body composition than people without physical disabilities. For example, persons with SCI may have atrophied muscle mass25 and a greater amount of fatty tissue29 than individuals without SCI, underestimating BMI and obesity in individuals with SCI. More accurate measures of obesity in persons with physical disabilities include bioelectrical impedance analysis26,30 and anthropometric index.31 However, these methods can be costly, require in-person measurement, and are not feasible in research studies where data are collected from a large geographically diverse sample. Abdominal obesity has recently been implicated as a significant risk factor for diseases such as type 2 diabetes and cardiovascular disease.32 Waist circumference (WC) provides a good estimate of abdominal visceral fat33 and has been suggested as a potentially low-cost and valid measure of obesity in individuals with physical disabilities.26,28 Past studies on the association between obesity and secondary conditions in persons with physical disabilities have primarily focused on the SCI population. Therefore, the purpose of this study was to examine the association between obesity and HRQOL indicators in a large group of individuals living with a physical disability relative to the general US population. We hypothesized that (a) individuals with disabilities, regardless of their obesity status, will report worse HRQOL indicators than the general US population and (b) obese individuals with disabilities will report on average worse HRQOL indicators than non-obese individuals with disabilities.
Methods Participants Individuals participating in a longitudinal study of aging with a physical disability completed self-report surveys every year for four years to collect information about their disease progression, symptom ratings (pain, fatigue, depression, sleep, cognition), and other factors such as coping skills, resilience, and well-being. Cross-sectional data from the baseline survey collected in 2009e2010 were used for this study; additional information about this study sample and procedures have been described in previously published articles.34,35 Briefly, individuals were recruited from multiple sources, including: (1) participants from previous research studies who indicated interest in future research, (2) University of Washington’s disability-specific registry of individuals interested in being contacted about future research, (3) referral by family and friends, and (4) print and web advertisements. Eligible participants were age 18 years or older with a self-reported diagnosis of muscular dystrophy (MD), multiple sclerosis (MS), post-polio syndrome (PPS), or
125
spinal cord injury (SCI). A total of 2041 surveys were mailed to eligible and interested individuals. Of those, 1862 completed the survey (91.2% response rate) and 1849 participants provided complete BMI or WC data (MD: 339; MS: 579: PPS: 443; SCI: 488). Participants from all 50 states were represented, with the top three states being Washington (28.6%), Georgia (7.1%), and California (6.2%). All participants provided informed consent and received $25 upon survey completion; procedures were approved by University of Washington’s Human Subjects division. Measures Demographics Basic demographic information (age at survey, disability, race/ethnicity, education, household income, and marital status) was collected from all participants. Urban or rural status was based on participant zip code at survey completion.36 Obesity Obesity status was measured by two indexes. First, body mass index (BMI) was calculated using self-reported weight and height. Calculated BMI (kg/m2) was then classified into four categories: underweight (!18.5), normalweight (18.5e24.9), overweight (25.0e29.9), or obese (>30).37 Second, waist circumference (WC) was used: Participants provided self-reported WC in inches and indicated if it was based on an actual recent measure, pants size, or similar measure. Independent of BMI, abdominal obesity as defined by a WC O 40 inches (males) or O35 inches (females) is predictive of disease.38 The National Heart, Lung, and Blood Institute has recommended that these WC cut-offs be used in conjunction with a BMI of >25 kg/m2 to identify individuals who are at increased risk of disease because of their WC.38 For this study, obese status was categorized as having an obese BMI and/or high-risk abdominal obesity. All others were categorized as non-obese. Adding the abdominal obesity component may identify study respondents with underestimated BMI who have a greater amount of visceral fat. HRQOL indicators Seven constructs (i.e., domains) were measured using Patient Reported Outcomes Measurement Information System (PROMIS) v1.0 short forms: physical function, fatigue, pain interference, depression, satisfaction with participation in social roles, sleep disturbance, and sleep-related impairment. PROMIS measures were developed using item response theory, have excellent psychometric properties and have been validated in various clinical populations.39e41 PROMIS norming sample data were primarily collected in 2006e2007 by YougovPolimetrix.com, an internet polling company. Polimetrix selected participants from a database of more than 1 million individuals who provided contact information and regularly participated in
126
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
online surveys; 13250 general population respondents provided data. A subset of this PROMIS general population sample (n 5 5239) matching the 2000 US Census on gender, age, and race40 was created to establish US population norms: 57% female; 15% 18e29, 22% 30e44, 28% 45e59, 22% 60e74 and 13% >75 years old; 74% White, 10% Black, 11% Hispanic/Latino, and 4% other; 51% had more than a high school education. Each PROMIS domain raw score is converted to a standardized T-score metric with a US general population mean of 50 and standard deviation (SD) of 10. The estimated US general population mean is the mean of the PROMIS norming sample. A higher score on any measure means more of the domain being measured (see www.nihpromis.org). Analyses Demographic and QOL means and frequencies were first calculated by obesity status to describe the sample. Demographics were compared between obesity groups using Student t-tests for continuous variables and chi-square tests for categorical variables. To test the hypothesis that individuals with disabilities report worse HRQOL than the PROMIS norm sample, onesample t-tests were used to compare each study sample domain mean to the norm mean of 50, overall and by obesity status. To account for multiple comparisons, a Bonferroni correction adjusted the alpha level to 0.0167 (0.05/3). To test the hypothesis that obese individuals with disabilities report worse HRQOL than non-obese individuals, PROMIS T-scores were compared between obesity status groups using Student t-tests. To provide an at-a-glance graph that shows symptom patterns, we plotted all mean domain T-scores by obesity status. We also compared Tscores by obesity status for each disability group; alpha level adjusted to 0.0125 (0.05/4). A multivariate regression model was used to examine the effects of obesity on HRQOL after adjusting for demographic characteristics. The model included the following independent variables: age, gender, race/ethnicity (non-Hispanic white vs. other), marital status (married/living with partner vs. other), education (college degree or greater vs. other), household income, years since diagnosis, state of residency at survey completion (WA vs. other), rural/urban designation, and obesity status group. Analyses were conducted using SAS software v9.3.42 Results Demographics The mean BMI of all respondents was 26.4 kg/m2 with 5.3% classified as underweight, 41.6% normal-weight, 29.7% overweight, and 23.4% obese. The mean WC was 37.5 inches for men and 34.0 inches for women, with 26.4% of respondents who provided a WC categorized as having abdominal obesity. A total of 33.3% (n 5 616) were
classified as obese BMI and/or having abdominal obesity; 291 of these individuals were classified as obese using both indexes. Obese respondents were older and less educated than non-obese individuals ( p ! 0.001) (Table 1). The obese group also tended to be female ( p ! 0.01) and reported less household income than non-obese individuals ( p ! 0.001). Comparisons between sample and normative PROMIS means Except for sleep-related disturbance for the non-obese group, both non-obese and obese individuals with Table 1 Participant characteristics by obesity status Non-obese (n 5 1233)
Age Years diagnosed Height (inches) Weight (pounds) Waist circumference (inches)
Obese (n 5 616)
Mean 6 SD
Mean 6 SD
pa
55.2 6 14.1 15.6 6 10.8 66.8 6 4.3 148.1 6 29.7 32.6 6 4.3
58.0 6 11.3 15.1 6 10.4 66.4 6 4.2 206.5 6 39.3 40.7 6 5.5
!0.0001 0.3648 0.0241 !0.0001 !0.0001
n (%)
n (%)
b
Diagnosis Muscular dystrophy 240 (70.8) 99 (29.2) Multiple sclerosis 379 (65.5) 200 (34.5) Post-polio syndrome 263 (59.4) 180 (40.6) Spinal cord injury 351 (71.9) 137 (28.1) Sexc Male 480 (38.9) 200 (32.5) Female 753 (61.1) 416 (67.5) Race/ethnicityc Non-Hispanic white 1131 (92.5) 561 (91.2) Other 92 (7.5) 54 (8.8) Educationc Some college or less 517 (41.9) 314 (51.0) College or graduate degree 716 (58.1) 302 (49.0) Marital statusc Married/living with partner 774 (62.9) 391 (63.5) Other 457 (37.1) 225 (36.5) Household incomec !$25,000 260 (22.3) 152 (25.9) $25,000e$40,000 177 (15.2) 123 (21.0) $41,000e$55,000 157 (13.5) 84 (14.3) $56,000e$70,000 132 (11.3) 63 (10.8) $71,000e$85,000 117 (10.0) 48 (8.2) $86,000e$100,000 121 (10.4) 54 (9.2) O$100,000 203 (17.4) 62 (10.6) Urban/rural designation of zip code at survey completionc Urban 986 (80.0) 480 (77.9) Rural 246 (20.0) 136 (22.1) State of residence at survey completionc Washington 350 (28.4) 179 (29.1) Other 883 (71.6) 437 (70.9)
0.0002
0.0064
0.3502
0.0002
0.8016
0.0004
0.2928
0.7632
BMI 5 body mass index. Demographics were compared between obesity groups using Student ttests for continuous variables or chi-square tests for categorical variables. b Distribution of each disability across obesity status categories. c Distribution of each characteristic per obesity status category. a
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
disabilities reported worse HRQOL scores on all domains relative to the PROMIS normative sample (all p ! 0.0001; Fig. 1). Compared to PROMIS norms, the difference in sample T-scores ranged from 0.5 points (nonobese sleep-related disturbance) to 15.1 points (obese physical function). Comparisons between sample obesity status means Compared to the non-obese group, obese individuals reported statistically significantly worse functioning on all domains except depression (all p ! 0.01; Fig. 1). The largest T-score difference between obesity groups was 2.6 points for physical function. The smallest difference in Tscores was 1.0 points (depression). Disability-specific results The proportion of obese individuals was highest for PPS (40.6%) followed by MS (34.5%), MD (29.2%), and SCI (28.1%). Obese individuals with MS reported significantly worse outcomes on all domains except depression when compared to the non-obese MS group (all p ! 0.0125). The largest T-score difference was 4.6 points for sleeprelated impairment. Obese individuals with MD reported worse physical function, satisfaction with participation in social roles, and fatigue relative to non-obese individuals with MD. Only physical function scores were significantly worse for obese individuals with PPS compared to
127
non-obese individuals with PPS. There were no statistically significant differences for the SCI group. Multivariate regression model results After controlling for demographic variables, obese status was statistically significantly associated with all PROMIS domains except depression ( p ! 0.05, Table 2 contains standardized Beta coefficients). Discussion The purpose of this study was to examine the association between obesity and levels of symptoms and QOL indicators within disability groups and also relative to a normative sample representative of the US population in age, gender, and race/ethnicity. As previously reported for these populations27 and for others,1 higher BMI levels were associated with older age and less education. Consistent with our hypothesis, results suggest that people with disabilities report worse HRQOL indicators when compared to the PROMIS normative sample. Among persons with disabilities, important differences emerged between persons classified as obese relative to those classified as non-obese including poorer physical and social functioning, and higher fatigue, pain interference and sleep and sleep-related disturbance. The association between obesity and these HRQOL domains remained statistically significant after we controlled for
Fig. 1. PROMIS domain T-scores for individuals with physical disabilities, by obesity status. The figure shows the mean PROMIS T-scores of the obese and non-obese groups. The T-score of 50 represents the PROMIS norm sample mean. A higher score on any measure means more of the domain being measured, i.e., higher physical function score indicates better physical function while higher depression score indicates more depression. Left-facing bars indicate scores that are worse than the PROMIS normative sample mean for domains where a higher score is better (physical function and social role satisfaction) and right-facing bars indicate scores that are worse than the PROMIS normative sample mean for the domains where lower scores are better (sleep disturbance, sleep-related impairment depression, pain interference, and fatigue). Asterisk indicates statistically significant difference in T-score between the obese and non-obese groups ( p ! 0.05).
128
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
Table 2 Multiple linear regression results (standardized Beta coefficients) examining factors associated with PROMIS domains in individuals with a physical disability PROMIS domain Predictors in model
Physical functiona
Social role satisfactiona
Fatigueb
Pain interferenceb
Depressionb
Sleep disturbanceb
Sleep-related impairmentb
Age (years) Female sex Non-Hispanic white Married/living with partner College or graduate degree Household income Years diagnosed Washington state residence Rural zip code Obese statusc
0.16* 0.09* 0.05* 0.04 0.01 0.20* 0.20* 0.12* 0.01 0.10*
0.13* 0.03 0.02 0.05 0.04 0.23* 0.05 0.06* 0.02 0.09*
0.09* 0.21* 0.03 0.09* 0.03 0.16* 0.02 0.05* 0.03 0.10*
0.08* 0.00 0.07 0.03 0.09* 0.19* 0.06 0.04 0.04 0.10*
0.02 0.03 0.00 0.07 0.03 0.16* 0.02 0.07* 0.00 0.06
0.12* 0.05* 0.00 0.00 0.10* 0.07* 0.00 0.03 0.03 0.05*
0.12* 0.07 0.06 0.03 0.02 0.10* 0.07 0.06 0.02 0.09*
PROMIS 5 Patient Reported Outcomes Measurement Information System. *Significantly associated with PROMIS domain ( p ! 0.05). a Higher score indicates better functioning. b Higher score indicates worse functioning. c Association of obesity status with health-related quality of life scores after adjusting for age, gender, race/ethnicity, marital status, education, household income, years since diagnosis, state of residency at survey completion, and rural/urban designation.
several demographic characteristics. Results from this study also support disability-specific differences in how obesity may affect symptoms and QOL, extending the findings made by Alschuler and colleagues.27 Obese individuals with MS and, to a lesser degree MD, reported worse functioning than normal weight individuals. We did not see this effect for the PPS and SCI groups. It is possible that disability-specific characteristics (e.g., disability severity) may moderate the effects of obesity on HRQOL. Future studies of obesity in persons with disabilities are needed given the paucity of research in this area. The differences between the obese and non-obese groups, while consistent, were relatively small and may not reach clinical significance. Clinical significance is often evaluated by using minimally clinically important differences (MCID). MCIDs have been developed for some PROMIS domains in advanced-stage cancer patients,43 but such guidelines have not been developed for this study’s population. An alternative is to use 0.5 SD, which is generally considered to be a good approximation for clinically important difference.44 For PROMIS instruments, 5 points is equivalent to 0.5 SD. With this criterion in mind, differences in average physical function, fatigue, pain interference and satisfaction with participation in social roles reached clinical significance when comparing T-scores between the PROMIS normative sample and obese group. However, we observed clinically significant differences for these domains between the non-obese group and the PROMIS normative sample as well. This suggests that simply living with a physical disability is associated with worse HRQOL when compared with a general population. The obese group was generally about 2.5 points worse than the non-obese group for each of these domains,
indicating that obesity adds a clinically small, but statistically significant burden on individuals living with a disability. BMI may not capture obesity in persons with disabilities and may even vary by disability.27 This may partially account for differences in study sample (23.4%) and national (35.7%)45 obese BMI levels. By including individuals with abdominal obesity, this study identified an additional 183 individuals with disabilities who are at high-risk for disease and produced a study sample obesity level (33.3%) that is much closer to the national rate. Future research is needed to examine whether BMI or WC or a combination is the best indicator of obesity in individuals with physical disabilities and to establish acceptable cut-off levels for this population. In this study we grouped individuals with under-, normal- and normal WC/over-weight BMI into a nonobese group to focus on the association between obesity and HRQOL. We recognize that there are health issues specific to being under-46 or over-weight.47 Published studies have reported that unlike those who are obese, being overweight does not result in an increased risk of mortality relative to normal-weight individuals,48 suggesting that the adverse effects of increased BMI are especially true for those classified as obese. Indeed, in results not shown, study sample T-scores were similar between the normalweight and the under- and over-weight groups except for physical function and pain interference, respectively. Finally, self-reported BMI at the lower and higher levels tend to be over- and under-reported, respectively.49 However, as direct measurement of BMI is sometimes not practicable in large studies like this, self-report has been found to be an acceptable alternative.50
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
Conclusion The study findings make an important contribution to the literature by being the first study to examine the relationship between obesity and symptoms and QOL indicators in people with MD, MS, and PPS. We conclude that there is much more to learn about obesity in persons with physical disabilities. BMI and WC cut-off levels should be established in persons with disabilities so we can better study obesity in large epidemiological-type studies in which physical measurements are impractical. The difference in outcomes between the obese and non-obese groups suggests an association between obesity and HRQOL, especially among people with MS or MD. However, prospective studies must be conducted to establish causality and to examine disability-specific factors associated with obesity. With this information, health care professionals will be better able to identify and counsel patients about effective ways to improve their health and well-being. References 1. Chen Y, Cao Y, Allen V, Richards J. Weight matters: physical and psychosocial well being of persons with spinal cord injury in relation to body mass index. Arch Phys Med Rehabil. 2011;92(3):391e398. 2. Marrie R, Horwitz R, Cutter G, Tyry T, Campagnolo D, Vollmer T. High frequency of adverse health behaviors in multiple sclerosis. Mult Scler. 2009;15(1):105e113. 3. Timmerman G, Stuifbergin A. Eating patterns in women with multiple sclerosis. J Neurosci Nurs. 1999;31(3):152e158. 4. Ahlstr€ om G, Karlsson U. Disability and quality of life in individuals with postpolio syndrome. Disabil Rehabil. 2000;22(9):416e422. 5. van Nimwegen M, Speelman AD, Hofman-van Rossum EJ, et al. Physical inactivity in Parkinson’s disease. J Neurol. 2011;258(12): 2214e2221. 6. Sosnoff JJ, Gappmaier E, Frame A, Motl R. Influence of spasticity on mobility and balance in persons with multiple sclerosis. J Neurol Phys Ther. 2011;35(3):129e132. 7. Motl RW, Snook EM, Schapiro R. Symptoms and physical activity behavior in individuals with multiple sclerosis. Res Nurs Health. 2008;31(5):466e475. 8. National Task Force on the Prevention and Treatment of Obesity. Overweight, obesity, and health risk. Arch Intern Med. 2000;160(7):898e904. 9. Luppino FS, de Wit LM, Bouvy PF, et al. Overweight, obesity, and depression: a systematic review and meta-analysis of longitudinal studies. Arch Gen Psychiatry. 2010;67(3):220e229. 10. Patterson RE, Frank LL, Kristal AR, White E. A comprehensive examination of health conditions associated with obesity in older adults. Am J Prev Med. 2004;27(5). 11. Zhao G, Ford ES, Li C, Tsai J, Dhingra S, Balluz L. Waist circumference, abdominal obesity, and depression among overweight and obese U.S. adults: National Health and Nutrition Examination Survey 2005-2006. BMC Psychiatry. 2011;11(130). 12. Ma J, Xiao L. Obesity and depression in US women: results from the 2005-2006 National Health and Nutritional Examination Survey. Obesity (Silver Spring). 2010;18(2):347e353. 13. Herva A, Laitinen J, Miettunen J, et al. Obesity and depression: results from the longitudinal Northern Finland 1966 Birth Cohort Study. Int J Obes. 2006;30(3):520e527. 14. Resta O, Foschino Barbaro MP, Bonfitto P, et al. Low sleep quality and daytime sleepiness in obese patients without obstructive sleep apnoea syndrome. J Intern Med. 2003;253(5):536e543.
129
15. Vorona RD, Winn MP, Babineau TW, Eng BP, Feldman HR, Ware J. Overweight and obese patients in a primary care population report less sleep than patients with a normal body mass index. Arch Intern Med. 2005;165:25e30. 16. Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Viikari-Juntura E. The association between obesity and low back pain: a meta-analysis. Am J Epidemiol. 2010;171(2):135e154. 17. Stone AA, Broderick J. Obesity and pain are associated in the United States. Obesity (Silver Spring). 2012;20(7):1491e1495. 18. Cook K, Molton I, Jensen M. Fatigue and aging with a disability. Arch Phys Med Rehabil. 2011;92(7):1126e1133. 19. Bamer A, Johnson K, Amtmann D, Kraft G. Prevalence of sleep problems in individuals with multiple sclerosis. Mult Scler. 2008;14(8): 1127e1130. 20. Wu N, Minden S, Hoaglin D, Hadden L, Frankel D. Quality of life in people with multiple sclerosis: data from the Sonya Slifka Longitudinal Multiple Sclerosis Study. J Health Hum Serv Adm. 2007;30(3): 233e267. 21. Kinne S, Patrick D, Doyle D. Prevalence of secondary conditions among people with disabilities. Am J Public Health. 2004;94(3): 443e445. 22. Robinson-Whelen S, Taylor H, Hughes R, Nosek M. Depressive symptoms in women with physical disabilities: identifying correlates to inform practice. Arch Phys Med Rehabil; 2013 [in press], http://dx. doi.org/10.1016/j.apmr.2013.07.013. 23. Okoro C, Strine T, Balluz L, et al. Serious psychological distress among adults with and without disabilities. Int J Public Health. 2009;54(suppl 1):52e60. 24. Hetz SP, Latimer AE, Arbour-Nicitopoulos KP, Martin Ginis K. Secondary complications and subjective well-being in individuals with chronic spinal cord injury: associations with self-reported adiposity. Spinal Cord. 2011;49(2):266e272. 25. Giangregorio L, McCartney N. Bone loss and muscle atrophy in spinal cord injury: epidemiology, fracture prediction, and rehabilitation strategies. J Spinal Cord Med. 2006;29(5):489e500. 26. Eriks-Hoogland I, Hilfiker R, Baumberger M, Balk S, Stucki G, Perret C. Clinical assessment of obesity in persons with spinal cord injury: validity of waist circumference, body mass index, and anthropometric index. J Spinal Cord Med. 2011;34(4):416e422. 27. Alschuler KN, Gibbons LE, Rosenberg DE, et al. Body mass index and waist circumference in persons aging with muscular dystrophy, multiple sclerosis, post-polio syndrome, and spinal cord injury. Disabil Health J. 2012;5(3):177e184. 28. Buchholz AC, Bugaresti J. A review of body mass index and waist circumference as markers of obesity and coronary heart disease risk in persons with chronic spinal cord injury. Spinal Cord. 2005;43(9): 513e518. 29. Jones LM, Legge M, Goulding A. Healthy body mass index values often underestimate body fat in men with spinal cord injury. Arch Phys Med Rehabil. 2003;84:1068e1071. 30. Spungen AM, Bauman WA, Wang J, Pierson RN Jr. Measurement of body fat in individuals with tetraplegia: a comparison of eight clinical methods. Paraplegia. 1995;33(7):402e408. 31. Bulbulian R, Johnson RE, Gruber JJ, Darabos B. Body composition in paraplegic male athletes. Med Sci Sports Exerc. 1987;19(3): 195e201. 32. Siren R, Eriksson JG, Vanhanen H. Waist circumference a good indicator of future risk for type 2 diabetes and cardiovascular disease. BMC Public Health. 2012;12:631. 33. Lemieux S, Prud’homme D, Bouchard C, Tremblay A, Despres J. A single threshold value of waist girth identifies normal-weight and overweight subjects with excess visceral adipose tissue. Am J Clin Nutr. 1996;64:685e693. 34. Alschuler KN, Jensen MP, Goetz MC, Smith AE, Verrall A, Molton IR. Effects of pain and fatigue on physical functioning and depression in persons with muscular dystrophy. Disabil Health J. 2012;5(4):277e283.
130
R. Salem et al. / Disability and Health Journal 7 (2014) 124e130
35. Cook KF, Bamer AM, Amtmann D, Molton IR, Jensen M. Six patient-reported outcome measurement information system short form measures have negligible age- or diagnosis-related differential item functioning in individuals with disabilities. Arch Phys Med Rehabil. 2012;93(7):1289e1291. 36. Centers for Medicare and Medicaid Services. Ambulance Fee Schedule: National Breakout of Geographic Area Definitions by Zip Code; 2013. 37. World Health Organization. Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation. WHO Technical Report Series 894. Geneva: World Health Organization; 2000. 38. National Institutes of Health NH, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. Obes Res. 1998;6(suppl 2):S51eS209. 39. Cella D, Riley W, Sone A, et al. Initial item banks and first wave testing of the Patient-Reported Outcomes Measurement Information System (PROMIS) network: 2005-2008. J Clin Epidemiol. 2010;63(11):1179e1194. 40. Liu H, Cella D, Gershon R, et al. Representativeness of the PROMIS Internet panel. J Clin Epidemiol. 2010;63(11):1169e1178. 41. Rothrock N, Hays R, Spritzer K, Yount S, Riley W, Cella D. Relative to the general US population, chronic diseases are associated with poorer health-related quality of life as measured by the PatientReported Outcomes Measurement Information System (PROMIS). J Clin Epidemiol. 2010;63(11):1195e1204.
42. SAS Software [program]. Cary, NC: SAS Institute Inc.; 2011. 43. Yost KJ, Eton DT, Garcia SF, Cella D. Minimally important differences were estimated for six Patient-Reported Outcomes Measurement Information System-Cancer scales in advanced-stage cancer patients. J Clin Epidemiol. 2011;64(5):507e516. 44. Farivar SS, Liu H, Hays R. Half standard deviation estimate of the minimally important difference in HRQOL scores? Expert Rev Pharmacoecon Outcomes Res. 2004;4(5):515e523. 45. Flegal K, Carroll M, Kit B, Ogden C. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. J Am Med Assoc. 2012;307(5):491e497. 46. Flegal KM, Graubard BI, Williamson DF, Gail M. Cause-specific excess deaths associated with underweight, overweight, and obesity. JAMA. 2007;298(17):2028e2037. 47. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003;289(1):76e79. 48. Flegal KM, Graubard BI, Williamson DF, Gail M. Excess deaths associated with underweight, overweight, and obesity. JAMA. 2005;293(15):1861e1867. 49. Stommel M, Schoenborn C. Accuracy and usefulness of BMI measures based on self-reported weight and height: findings from the NHANES & NHIS 2001-2006. BMC Public Health. 2009;9:421. 50. Bes-Rastrollo M, Sabate J, Jaceldo-Siegl K, Fraser G. Validation of self-reported anthropometrics in the Adventist Health Study 2. BMC Public Health. 2011;5(11):213.
