Archives of Physical Medicine and Rehabilitation
Volume 89, Issue 2 , Pages 210-218, February 2008

Health of Community-Dwelling Adults With Mobility Limitations in the United States: Prevalent Health Conditions. Part I

Presented as an abstract to the American Public Health Association, December 14, 2005, Philadelphia, PA.

  • Elizabeth K. Rasch, PT, PhD

      Affiliations

    • Rehabilitation Medicine Department, Clinical Research Center, National Institutes of Health, Bethesda, MD
    • Rasch was employed at the National Center for Health Statistics, Center for Disease Control and Prevention when this work was completed.
    • Corresponding Author InformationCorrespondence to Elizabeth K. Rasch, PT, PhD, National Institutes of Health, Bldg 10, CRC, Room 1469, 10 Center Dr, MSC 1604, Bethesda, MD 20892-1604. Reprints are not available from the author.
    • ,
  • Marc C. Hochberg, MD, MPH

      Affiliations

    • University of Maryland School of Medicine, Baltimore, MD
    • ,
  • Larry Magder, PhD, MPH

      Affiliations

    • University of Maryland School of Medicine, Baltimore, MD
    • ,
  • Jay Magaziner, PhD, MSHyg

      Affiliations

    • University of Maryland School of Medicine, Baltimore, MD
    • ,
  • Barbara M. Altman, PhD

      Affiliations

    • National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, MD

Article Outline

Abstract 

Rasch EK, Hochberg MC, Magder L, Magaziner J, Altman BM. Health of community-dwelling adults with mobility limitations in the United States: prevalent health conditions. Part I.

Objective

To characterize the extent and types of prevalent health conditions among nationally representative groups of adults with mobility, nonmobility, and no limitations.

Design

Data were collected during 5 rounds of household interviews from a probability subsample of households that represent the civilian, noninstitutionalized U.S. population. With some exceptions, round 1 variables were used for this analysis.

Setting

Community.

Participants

Data were analyzed on the same respondents from the 1996 to 1997 Medical Expenditure Panel Survey (MEPS) and the 1995 National Health Interview Survey Disability Supplement. Respondents were categorized into 3 groups for analysis: those with mobility limitations, nonmobility limitations; and no limitations. The analytic sample included 13,897 MEPS adults (≥18y).

Interventions

Not applicable.

Main Outcome Measures

Number, types, and prevalence of self-reported health conditions compared across groups.

Results

On average, adults with mobility limitations had significantly more prevalent conditions (3.6) than those with nonmobility limitations (2.4), or no limitations (1.3). Greater comorbidity existed in the context of fewer personal resources and more than half of adults with mobility limitations were working age.

Conclusions

Determining factors that influence the health of adults with mobility limitations is a critical public health issue.

Key Words: Comorbidity, Persons with disabilities, Public health, Rehabilitation

 

RECENT NATIONAL DATA have indicated that approximately 1 of every 5 civilian, noninstitutionalized adults in the U.S. report the presence of some type of disability as defined by the presence of sensory, communication, or cognitive impairment, functional or activity limitations, the use of assistive devices, or receipt of federal work disability benefits.1, 2 This finding is of particular significance because the health of people with disabilities appears to be, in aggregate, worse than the general population as reflected by poorer self-reported physical and mental health status as well as greater comorbidity.3, 4, 5 Furthermore, many new health conditions occur after the onset of disability.4, 6, 7, 8, 9, 10, 11, 12 These conditions contribute to the prevalent comorbid disease burden which includes conditions that lead to disability and those that precede disability but are unrelated, as well as those that follow. Comorbidity among those with disability is consequential because it may increase the likelihood of subsequent disability,13, 14, 15 the frequency of hospitalization,16, 17 and the number of days that routine activities cannot be performed.12, 17 Medical expenditures are disproportionately high for people with activity limitations,18, 19, 20 especially when more than 1 comorbid health condition is present.

Despite inclusion in the Healthy People 2010 objectives,21 there is insufficient information at the national level about health conditions among Americans with disabilities. Survey-based queries of adults with disabilities about new and existing health conditions have not been uniform or comprehensive,4, 6, 7, 8, 9, 10, 11, 12 comparisons with reference groups without disability are generally lacking,6, 7, 9, 10, 11, 12 and the composition of study populations has not been nationally representative nor diverse with regard to the types of disabilities that are represented.4, 7, 10, 11, 16, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 This lack of information, coincident with the recent Surgeon General’s Call to Action to Improve the Health and Wellness of Persons with Disabilities,35 underscores the need for and timeliness of the current study.

This work is the first part of a 2-part study with an overall objective of characterizing prevalent and incident health conditions among nationally representative groups of adults with mobility limitations, nonmobility limitations, and without limitations. This study necessitated use of longitudinal data, although the first part is limited to examining prevalent conditions that were present at the start of the study. The objective of this study was to compare the extent and types of prevalent health conditions in adults with mobility limitations, nonmobility limitations, and without limitations, using a data source that provided comprehensive, open-ended enumeration of conditions by respondents.

Back to Article Outline

Methods 

Data Source 

We derived the data from the 1996 to 1997 Medical Expenditure Panel Survey (MEPS)36 and the 1995 National Health Interview Survey (NHIS) Disability Supplement (NHIS-D),37 using the same respondents from both surveys. The 1995 NHIS used a multistage probability sampling design and served as the sampling frame for the 1996 MEPS, which included a subsample of NHIS households that were nationally representative of the noninstitutionalized, civilian, U.S. population. In 1995, a disability supplement to the NHIS provided additional data on self-reported limitations and use of assistive equipment that were not available from the MEPS. Data from the NHIS-D were used for limitation group classification in conjunction with data from the MEPS. The MEPS core survey (household component) was the primary source of data for this analysis in addition to selected questions from the supplement on long-term care that collected detailed information on people with limitations (round 4, 1997, panel 1). The MEPS used an overlapping panel design whereby data were collected longitudinally through 5 rounds of household interviews over a 2-year period. Although prevalent conditions were the primary focus of the current study (part 1) which necessitated use of data from round 1 (start of study), longitudinal data were required in order to distinguish prevalent conditions from subsequent incident conditions which constituted the overall aim of the project (parts 1–2). The primary advantages of linking the 1995 NHIS with the 1996 to 97 MEPS were that (1) the 1994 to 1995 NHIS included a disability supplement, not repeated since that time, that provided indicators of disability that were otherwise unavailable; and (2) respondents with disabilities were oversampled in the 1996 to 1997 MEPS (panel 1) which was a subsample of the 1995 NHIS. The oversampling of adults with disabilities provided a sufficiently large sample to address our research questions. Survey design and methods for the NHIS,38, 39 its companion disability supplement,40 and the MEPS40, 41, 42, 43 have been described.

Respondents 

Variables on self-reported limitations necessary for limitation group classification varied considerably between children and adults; therefore we limited the analyses to adults, 18 years of age and older. Of the 22,601 persons in the 1996 MEPS file, 15,386 were adults. Nearly all adults had useable (nonzero) 1996 analytic weights (n=15,029) that were necessary to produce population estimates. NHIS-D data, necessary for limitation group classification, could be obtained for 93% (n=13,979) of these adults. Finally, 65 adults were no longer part of the noninstitutionalized civilian U.S. population by round 1 of the 1996 MEPS due to death, institutionalization, or military service, and 17 had insufficient data for group classification. Thus, the analytic sample comprised 13,897 adults. The joint response rate for the 1995 NHIS core and 1996 MEPS (round 1) was 78%.36 The institutional review board (IRB) of the University of Maryland at Baltimore determined this project to be exempt from the IRB approval process.

Analytic Variables 

We used variables from the 1996 MEPS, round 1, panel 1 for analysis (exceptions noted below). Respondents were classified into 3 groups: those with mobility limitations, nonmobility limitations, and no limitations. Adults who reported any difficulty (some, a lot, or unable) climbing stairs, walking 3 blocks, walking a mile, standing for 20 minutes, or stooping to pick up an object from the floor in the 1996 MEPS were classified with mobility limitations. Indication of any use of assistive technology was obtained from the MEPS, but specification of the type of equipment was obtained from the 1995 NHIS-D. Adults who reported use of a cane, crutches, walker, medically prescribed shoes, a wheelchair, a scooter, a back, leg, foot, or knee brace, an artificial leg or foot, or those who reported use of assistive technology to get around inside the home were also classified with mobility limitations whether or not they reported difficulty with the functional mobility activities described above. Classification of respondents as having mobility limitation based on use of assistive devices in the absence of self-reported mobility difficulty has been used by other researchers.44 Adults with mobility limitations may have reported additional nonmobility limitations; however, inclusion in the mobility limitation group was based on whether or not respondents reported mobility difficulty or use of mobility equipment, regardless of the presence or absence of other types of limitations. Adults without mobility limitations who reported (1996 MEPS) any difficulty lifting, reaching overhead, grasping objects, seeing, hearing, communicating, thinking (cognition), performing usual daily activities (activities of daily living [ADLs], instrumental ADLs, major activities, or social activities), or use of assistive technology other than mobility devices (1995 NHIS-D) were classified as having nonmobility limitations. The MEPS did not include all disability indicators in each round of data collection. Because visual and hearing limitations were only identified in round 2 of the 1996 MEPS and communication limitations were only identified in the long-term care supplement of the MEPS (round 4, 1997, panel 1), responses from adults who reported these limitations were verified in the 1995 NHIS-D. It was assumed that if respondents reported visual, hearing, or communication limitations in round 2 (or the long-term care supplement) of the MEPS and in the NHIS-D, they also had the limitation in round 1 of the MEPS (start of study). This method was used to adjust for the fact that some disability indicators were not collected in the MEPS at the time point necessary for this analysis (ie, round 1). All other adults were classified with no limitations. The number of limitations per person was determined by adding dichotomous indicators of limitations in mobility functional activities, nonmobility functional activities, vision, hearing, communication, cognition, and activity limitations.

Self-reported health conditions were enumerated at the beginning of each round of MEPS data collection. The reference period for round 1 of the MEPS was from January 1, 1996, to the date of the first interview (between March–July 1996).45 Respondents were told: “We’re interested in learning about health problems that may have bothered (person) since (start date).” Health problems include physical conditions, accidents, or injuries that affect any part of the body as well as mental or emotional health conditions, such as feeling sad, blue, or anxious about something. Respondents were then asked: “Between (start date) and (end date), did (person) have any physical or mental health problems, accidents or injuries? Please include all of (person)’s conditions, accidents or injuries regardless of whether (person) saw a medical provider, received treatment, or took medications since (start date).” For affirmative replies, respondents were asked: “What did (person) have?” and were probed: “Did (person) have any other health problems, accidents, or injuries?”36 Conditions were also reported by respondents in association with health care provider visits, hospitalizations, prescription medication use, and disability days. For instance, respondents that reported a hospital stay during the reference period were asked: “Was this hospital stay related to any specific health condition or were any conditions discovered during this hospital stay?” For affirmative replies, respondents were asked: “What conditions were discovered or led (person) to enter the hospital?” and were probed “Any other conditions?”36 Interviewers recorded conditions as verbatim text, which were then coded to 1996 International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes by professional coders. To facilitate analysis, it was necessary to collapse ICD-9 codes into a manageable number of meaningful categories. We chose the only well-recognized, peer reviewed, and published system that was available when the study was conducted: the Clinical Classifications for Health Policy Research developed by the Agency for Healthcare Research and Quality (AHRQ).46, 47 Clinical Classification Software (CCS) aggregated ICD-9-CM condition codes into 259 mutually exclusive, clinically homogeneous categories and these categories were used for analysis.46, 47, 48 Prevalent conditions were identified in round 1 of the 1996 MEPS and were counted as distinct from one another if they fell into separate CCS categories.

Individual characteristics included age, sex, race and ethnicity, and body mass index (BMI). Construction of the race-ethnicity variable has been described.45 BMI was calculated from self-reported weight and height in the 1995 NHIS. Adults with a BMI of 30kg/m2 or more were classified as obese.49 Marital status, family size, education, income (1996 family income as a percentage of the national poverty threshold), health insurance coverage (summary indicator of coverage for persons in the 1996 calendar year), and access to health care (round 2) were conceptualized as potential resources because they represent assets available to respondents that could influence health. Round 1 or full year (1996) variables were used with the exception that data on access to health care were only collected in round 2. An indicator of access to health care was constructed from all combinations of responses to 2 dichotomous MEPS variables indicating (1) whether respondents had a regular source of health care and (2) whether they had difficulty obtaining health care in the past 12 months.

Statistical Analyses 

We explored potential bias due to exclusion of adults from the analytical sample because of inability to obtain NHIS-D data by comparing selected individual characteristics, availability of resources, and health status of MEPS adults who did and did not have data from the NHIS-D. Limitation groups were compared by individual characteristics and availability of resources using chi-square tests, and by the number of prevalent conditions at round 1 (mean and 95% confidence interval [CI]). Prevalences of the most frequently occurring conditions (of 259 CCS categories) were rank-ordered and compared between groups using linear contrasts of percentage estimates in the mobility limitation group and the nonmobility limitation group compared with the no limitation group. CCS categories were collapsed into system-level categories, rank-ordered, and compared between groups. Direct age-adjustment was performed on CCS and system-level estimates using the analytic sample (n=13,897) as the standard population (5-y increments). Respondents reporting more than 1 condition within a system-level category were counted only once in that category for the purpose of generating system-level prevalence estimates. The system-level categorization developed at AHRQ47 was modified such that sensory conditions (vision and hearing) were separated from diseases of the nervous system and events such as immunizations or healthy pregnancies were separated from other conditions and labeled as health services encounters. Finally, differences in the number of prevalent conditions across limitation groups were examined with linear regression models controlling for individual characteristics (sex, age, race and ethnicity, BMI), and availability of resources (marital status, household size, education, income, health insurance, access to health care). Potentially significant (P≤.05) and meaningful 2-way interactions between limitation group and all other independent variables were explored for possible inclusion in the final regression model.

MEPS 1996 person-level sample weights that adjusted for differential selection probability were used to produce population estimates. Variance estimation was accomplished through Taylor linearization taking into account the complex sampling design. Estimates were based on a minimum of 30 responses an analytic cell and relative standard errors less than 30%.40 SASa for Windows and SUDAANb software programs were used for analyses. Standard errors (SEs) are reported in parentheses unless otherwise indicated.

Back to Article Outline

Results 

Analytic Sample 

Small but statistically significant differences in sex, age, race and ethnicity, and marital status existed between adults that did (n=13,979) and did not (n=1050) have data from the NHIS-D (data not shown). Adults omitted from the analytic sample were more likely to be male, 18 to 44 years of age, Hispanic, and unmarried. There were no differences in education, poverty status, or health status between those with and without data from the NHIS-D. The mean number (95% CIs) of prevalent health conditions was slightly greater in adults with NHIS-D data (mean, 1.66; 95% CI, 1.62–1.71) compared with those without (mean, 1.45; 95% CI, 1.33–1.57).

Description of Limitation Groups 

Roughly 20% of civilian, community-dwelling adults reported some type of limitation; 12.4% (SE=0.4) reported a mobility limitation or use of mobility equipment and 7.5% (SE=0.3) reported other, nonmobility types of limitations. A single limitation was reported by 82.6% (SE=1.4) of adults with nonmobility limitations and 13.9% (SE=0.9) of adults with mobility limitation. Conversely, 61.5% (SE=1.3) of adults with mobility limitations reported 3 or more limitations while only 3.2% (SE=0.7) of those with nonmobility limitations reported multiple limitations (P<.001).

Individual characteristics and the availability of resources across the 3 limitation groups are described in table 1. A greater proportion of adults with mobility limitations were 65 years of age or older compared with the other 2 groups; however, over half of this group was working age (18–64y). The proportion of women was greater in the group with mobility limitations compared with the other groups. Also, more adults with mobility limitations were obese (28%) compared with those with nonmobility limitations (19%), or without limitations (15%). A greater proportion of adults with mobility limitations were unmarried, lived alone, had less than a high school education, had a low income or were poor, and had only public insurance⁎ in comparison with other community-dwelling adults.

Table 1. Individual Characteristics and Availability of Resources by Limitation Group
Weighted Percentage ± SE⁎
CategoriesMobility Limitation (n=1769)Nonmobility Limitation (n=1030)No Limitation (n=11,098)P (χ2)
All adults100.0100.0100.0
Age (y) <.001
18–4422.2±1.334.6±1.962.4±0.7
45–6430.3±1.329.2±1.627.2±0.5
≥6547.5±1.836.2±1.810.4±0.4
Sex <.001
Men39.0±1.151.6±1.848.3±0.4
Women61.0±1.148.5±1.851.8±0.4
Race and ethnicity <.001
NH-white79.2±1.282.0±1.574.1±0.8
NH-black11.6±1.09.9±1.211.2±0.6
Hispanic6.6±0.75.8±0.810.2±0.5
Other2.7±0.62.4±0.64.6±0.4
Obesity† <.001
BMI ≥30kg/m2‡27.7±1.118.5±1.315.1±0.4
BMI <30kg/m272.3±1.181.5±1.384.9±0.4
Marital status† <.001
Spouse in the house49.3±1.455.4±2.058.7±0.7
Not married/no spouse50.7±1.444.6±2.041.3±0.7
Family size <.001
Living alone30.9±1.424.6±1.715.7±0.5
239.7±1.640.0±2.030.9±0.7
3–423.6±1.426.5±1.538.9±0.8
≥55.9±0.68.9±1.114.5±0.6
Education† <.001
<High school40.0±1.430.6±1.817.7±0.6
High school32.7±1.236.0±1.835.0±0.7
>High school27.3±1.433.4±1.747.4±1.0
Poverty status <.001
Poor (<100% poverty line§)19.9±1.115.8±1.39.7±0.4
Low income (100%–199% poverty line)28.8±1.324.4±1.616.7±0.6
Middle income (200%–399% poverty line)30.0±1.429.3±1.833.3±0.7
High income (≥400% poverty line)21.4±1.230.5±1.640.3±0.9
Insurance coverage <.001
Any private insurance62.6±1.666.1±1.878.6±0.7
Public insurance only30.8±1.525.4±1.67.9±0.4
Uninsured6.6±0.68.6±1.013.5±0.5
Access to health care† <.001
Usual source/no difficulty obtaining health care75.9±1.373.8±1.570.6±0.7
Usual source/difficulty obtaining health care15.0±1.112.1±1.26.7±0.3
No usual source/no difficulty obtaining health care6.2±0.79.9±1.019.7±0.6
No usual source/difficulty obtaining health care2.9±0.44.3±0.83.1±0.2

Abbreviations: NH, Non-Hispanic.

⁎Proportions may not add to 100% due to rounding.

†Missing values excluded from estimates (4.1% for obesity, ≤.01% for marital status, 0.7% for education, and 1.2% for access to health care).

‡BMI calculated from self-reported weight and height.

§Poverty line: 1996 family income as percentage of 1996 national poverty threshold determined by the U.S. Census Bureau.

Prevalent Health Conditions 

Table 2 shows the 10 most frequently self-reported conditions in the general adult population (represented by the analytic sample; n=13,897) compared across the 3 limitation groups. Both unadjusted and age-adjusted results are shown because the age structures of the limitation groups differed considerably. Because many of these condition categories included other conditions and were therefore nonspecific, the table also reports the most prevalent ICD-9 codes within each of these categories to enhance clinical meaningfulness. All conditions except other upper respiratory infections and intestinal infection were much more prevalent in the mobility limitation group compared with those without limitations. Also, a gradient in the magnitude of conditions was apparent; adults with mobility limitations had the highest condition prevalences, followed by adults with nonmobility limitations and without limitations. A similar gradient in the magnitude of prevalent conditions can be observed in table 3, where conditions were collapsed into 19 system-level categories. In all system-level categories except infectious and parasitic diseases and health services encounters (age-adjusted estimates), prevalences were much greater among adults with mobility limitations compared with those without limitations. High condition prevalences were, in part, a reflection of the high comorbidity in adults with mobility limitations, where the mean number of prevalent conditions was 3.59 (95% CI, 3.46–3.73) compared with 2.39 (95% CI, 2.24–2.54) for adults with nonmobility limitations and 1.30 (95% CI, 1.26–1.34) for adults with no limitations.

Table 2. Unadjusted and Age-Adjusted Prevalence of the 10 Health Conditions (out of 259 condition categories)46 Most Frequently Reported by Adults in the General Population at Round 1 (start of study) Compared Across Limitation Groups (U.S. 1996)
Percentage ± SE
Unadjusted PrevalenceAge-Adjusted Prevalence
Most Frequent Health Conditions Reported by the General Adult Population (represented by the analytic sample, N=13,897)MLNMLNLMLNMLNL
Other upper respiratory infections46⁎11.4±1.011.9±1.112.8±0.514.1±1.512.8±1.212.4±0.5
Acute nasopharyngitis (common cold)†6.8±0.86.4±0.97.4±0.37.5±1.06.0±0.97.2±0.3
Essential hypertension29.3±1.2††17.0±1.2††8.3±0.319.2±1.0††11.2±0.810.3±0.4
Intestinal infection5.8±0.6⁎⁎7.1±1.08.1±0.37.0±0.98.2±1.27.6±0.3
Other nontraumatic joint disorders46‡22.0±1.0††9.5±1.0††3.2±0.215.0±1.0††7.1±0.8††3.9±0.2
Other and unspecified arthropathies†19.4±1.0††7.4±0.9††2.1±0.212.0±0.8††5.0±0.7⁎⁎2.8±0.2
Other upper respiratory disease46§7.1±0.7⁎⁎8.0±1.0⁎⁎5.2±0.39.2±1.2††6.8±0.95.0±0.3
Allergic rhinitis†4.3±0.64.9±0.83.7±0.26.2±1.1#4.5±0.83.5±0.2
Spondylosis, intervertebral disk disorders, other back problems14.6±0.9††5.8±0.8⁎⁎3.5±0.218.4±1.4††6.8±0.9††3.4±0.2
Other mental conditions46∥9.4±0.7††7.5±1.1††3.1±0.211.0±1.0††9.5±1.4††3.1±0.2
Depressive disorder, not elsewhere classified†8.4±0.7††6.9±1.1††2.8±0.210.2±1.0††8.5±1.3††2.7±0.2
Diabetes mellitus without complication14.4±0.9††7.4±0.9††2.2±0.210.0±0.8††5.4±0.7††2.6±0.2
Residual codes, unclassified4612.1±0.8††7.1±1.0††2.4±0.29.6±1.0††6.3±0.8††2.6±0.2
General symptoms†5.7±0.6††3.2±0.7††0.8±0.15.1±0.8††3.1±0.6††0.8±0.1
Headache, including migraine5.4±0.7⁎⁎4.8±0.7#3.3±0.27.6±1.2††7.1±1.1††3.1±0.2

NOTE. Prevalences were age-adjusted (5-y increments) using the analytic sample as the standard population (n=13,897).

Abbreviations: ML, mobility limitations; NL, no limitations; NML, nonmobility limitations.

⁎Includes conditions such as diphtheria, streptococcal sore throat, acute nasopharyngitis (common cold), acute sinusitis/pharyngitis/laryngitis/tracheitis, and chronic sinusitis.

†Most prevalent ICD-9-CM code within Clinical Classification category for all limitation groups.

‡Includes conditions such as certain arthropathies, loose bodies in joints, pathologic/recurrent joint dislocation, joint ankylosis, and other and unspecified disorders of joint (such as effusion or hemarthrosis).

§Includes conditions such as deviated septum, nasal polyps, chronic pharyngitis/nasopharyngitis, chronic laryngitis, allergic rhinitis, paralysis of vocal cords or larynx, throat pain, voice disturbances, etc.

∥Includes conditions such as other and unspecified neurotic disorders, sexual deviations and disorders, physiologic malfunctions arising from mental factors (such as psychogenic paralysis), certain adjustment reactions, etc.

Includes general symptoms such as sleep disturbances, nonspecific findings on examination of blood, other nonspecific abnormal findings, and other ill-defined and unknown causes of morbidity and mortality.

#Significantly different from no limitation group at P<.05.

⁎⁎Significantly different from no limitation group at P<.01.

††Significantly different from no limitation group at P<.001.

Table 3. Unadjusted and Age-Adjusted Prevalence of Self-Reported Health Conditions at Round 1 (start of study) Across the 3 Limitation Groups Collapsed into 19 System-Level Condition Categories47 (U.S. 1996)
Percentage ± SE
Unadjusted PrevalenceAge-Adjusted Prevalence
Conditions⁎MLNMLNLMLNMLNL
Diseases of the circulatory system47.5±1.327.9±1.611.2±0.431.9±1.818.0±1.014.0±0.4
Diseases of the musculoskeletal system and connective tissue45.3±1.421.1±1.59.4±0.344.1±1.919.1±1.410.4±0.4
Diseases of the respiratory system30.5±1.327.8±1.6∥23.1±0.633.5±1.827.4±1.6∥22.5±0.6
Endocrine, nutritional, and metabolic diseases and immunity28.6±1.218.3±1.58.2±0.321.6±1.414.1±1.39.4±0.3
Diseases of the digestive system22.2±1.118.8±1.4§15.2±0.420.2±1.318.5±1.5§15.1±0.4
Diseases of the sense organs (ear and eye)†17.1±1.013.8±1.25.5±0.311.1±0.910.4±1.06.2±0.3
Injury and poisoning15.9±1.012.2±1.3∥8.1±0.322.0±1.813.3±1.47.8±0.3
Mental disorders15.9±1.016.0±1.55.9±0.317.2±1.319.6±1.95.7±0.3
Other conditions‡15.1±0.99.6±1.14.5±0.212.9±1.38.6±1.04.7±0.2
Diseases of the nervous system†14.4±0.99.3±0.94.6±0.318.7±1.612.6±1.44.6±0.3
Diseases of the genitourinary system13.8±0.98.8±1.1§6.3±0.314.2±1.48.2±1.16.5±0.3
Health services encounters‡10.7±0.9∥9.2±1.07.5±0.310.5±1.49.8±1.37.9±0.4
Neoplasms8.2±0.86.3±0.72.5±0.25.1±0.7∥4.7±0.6∥2.9±0.2
Diseases of the skin and subcutaneous tissue7.9±0.87.2±1.0∥4.2±0.27.0±0.9∥6.1±1.04.5±0.3
Infectious and parasitic diseases4.2±0.54.1±0.73.2±0.24.3±0.74.7±0.93.1±0.2

NOTE. Prevalences were age-adjusted (5-y increments) using the analytic sample as the standard population (n=13,897).

⁎Estimates not reported due to unweighted analytic cell sizes <30 and/or relative SEs >30%. Estimates for the categories: diseases of the blood and blood forming organs, congenital anomalies, certain conditions originating in the perinatal period, and complications of pregnancy, childbirth, and the puerperium were not reported for this reason.

†Diseases of the sense organs were separated from the original category 6, diseases of the nervous system.

‡Health services encounters were separated from the original category 17, other conditions.

§Significantly different than no limitation group at P<.05.

∥Significantly different than no limitation group at P<.01.

Significantly different than no limitation group at P<.001.

In table 4, results of a linear regression model are shown that examines differences in the number of prevalent conditions across limitation groups while controlling for individual characteristics (sex, age, race and ethnicity, BMI) and availability of resources (marital status, household size, education, income, health insurance, access to health care). No interaction terms were added to the final regression model based on preliminary analyses. Adults with mobility limitations (least squares-adjusted mean ± SE, 3.15±0.06) and nonmobility limitations (2.16±0.08) had more comorbid conditions than adults without limitations (1.40±0.02) after controlling for individual characteristics and availability of resources. The only factors that did not contribute significantly to the model were poverty and marital status. The R2 value for the final model was .28.

Table 4. Results of Linear Regression Model Predicting the Number of Prevalent Conditions Across Limitation Groups
GroupMean Difference in Number of Conditions (β ± SE)P
Limitation group
No limitationRefRef
Mobility limitation1.75±0.07<.001
Other limitation0.76±0.08<.001
Sex
MenRefRef
Women0.51±0.03<.001
Age (y)
18–23RefRef
24–290.01±0.06.83
30–34−0.03±0.06.62
35–390.08±0.06.20
40–440.04±0.07.52
45–490.24±0.07.001
50–540.31±0.08<.001
55–590.55±0.09<.001
60–640.50±0.09<.001
65–690.71±0.10<.001
70–740.64±0.10<.001
75–790.52±0.11<.001
80–840.70±0.17<.001
≥850.03±0.18.86
Race and ethnicity
Non-Hispanic whiteRefRef
Non-Hispanic black−0.13±0.05.010
Hispanic−0.14±0.05.003
Other−0.21±0.07.004
BMI
<30kg/m2RefRef
≥30kg/m20.33±0.04<.001
Marital status
Spouse in the houseRefRef
Not married/no spouse−0.04±0.04.29
Family size
Living alone0.53±0.06<.001
20.28±0.05<.001
3–40.11±0.05.014
≥5RefRef
Education
<High school−0.06±0.05.24
High school−0.15±0.03<.001
>High schoolRefRef
Poverty status
Poor (<100% poverty line⁎)−0.004±0.06.94
Low income (100%–199% poverty line)−0.06±0.05.23
Middle income (200%–399% poverty line)−0.01±0.04.85
High income (≥400% poverty line)RefRef
Insurance coverage
Any private insuranceRefRef
Public insurance only−0.05±0.07.49
Uninsured−0.24±0.05<.001
Access to health care
Usual source/no difficulty obtaining health careRefRef
Usual source/difficulty obtaining health care0.46±0.08<.001
No usual source/no difficulty obtaining health care−0.60±0.04<.001
No usual source/difficulty obtaining health care−0.16±0.08.054

Abbreviation: Ref, reference group.

⁎Poverty line: 1996 family income as percentage of 1996 national poverty threshold determined by the U.S. Census Bureau.

Back to Article Outline

Discussion 

Few studies have evaluated the prevalence and pattern of comorbidity among community-dwelling adults with self-reported limitations relative to adults without limitations, despite the seriousness of adverse health consequences for this segment of the population. In previous studies, positive relationships have been found consistently between greater comorbidity, female sex, and increasing age.6, 17, 50, 51, 52 The current study supports these findings. In concurrence with previous studies,4, 6, 7, 8, 9, 10, 11, 12 we found high comorbidity among adults with limitations, exceeding that of adults without limitations. Similar to previous studies, we identified a high prevalence of obesity, mental health problems, musculoskeletal problems, hypertension, and diabetes among adults with mobility and nonmobility limitations relative to adults without limitations. In the current study, other upper respiratory diseases such as allergic rhinitis, residual codes including general symptoms such as sleep disturbances,46 and headache (including migraine) were also more prevalent in adults with mobility limitations compared with other adults. Although actual prevalence estimates may have changed in the time since these data were collected over 10 years ago, the striking differences in the relative magnitude of conditions across groups are noteworthy. Mean numbers of comorbid conditions in adults with and without limitations in Washington State (4.02 vs 1.28) were quite similar to the current study (mobility limitations, 3.6 and nonmobility limitations, 2.4 vs no limitations, 1.3).8

Several strengths contribute to the validity and generalizability of the results reported in the current study and expand on previous research in this area. The MEPS included a large analytic sample (with oversampling of respondents with disabilities) that was representative of the U.S. noninstitutionalized, civilian population, providing diversity in the types of disabilities that were represented and the underlying causal conditions. The ability to distinguish adults with and without mobility and nonmobility limitations facilitated analysis of relative disease prevalence in these subgroups. Condition enumeration was open-ended and comprehensive. Respondents were queried about all health conditions that occurred during the reference period, whether or not treatment was sought or received, and they reported conditions associated with health care encounters. Fully characterizing the range of conditions that occur in adults with mobility limitations (both chronic and acute) was important because conditions that may be uneventful to those without limitations, such as a lower respiratory infection, could lead to hospitalization for someone with respiration compromised by paralysis, for instance. The CCS codes used for the current analyses were inclusive of all possible ICD-9 codes. Thus, condition enumeration was not only comprehensive, but standardized which differs from previous studies. However, a drawback of this standardized classification system was the treatment of health care encounters that lacked a reported condition (ie, immunizations or medical evaluations). These events received ICD-9 codes, and therefore CCS codes, even though they may not directly reflect health conditions. Thus, they were included in the count of prevalent conditions. In the current study, the number of these events was small and their removal from the count of prevalent conditions had a negligible effect on the mean number of conditions across limitation groups. Furthermore, the benefits of using a comprehensive, standardized approach to the collection and analysis of data on health conditions outweighs the drawbacks because it facilitates comparisons of disease occurrence across studies, which have been limited to date because a uniform approach has been lacking.

Study Limitations 

Perhaps the most significant limitation of the current study lies in the self-report of health conditions, because reporting errors may have introduced bias.53, 54, 55, 56 Use of respondent reports has the advantage of better population coverage compared with use of medical or insurance records, but response errors can occur. Respondents may not recall all of their medical conditions, may not be aware of the presence of a condition, and may not report stigmatized conditions. Generally, household respondents underreport conditions compared with providers, especially certain types of conditions.53, 54, 56 Furthermore, respondents may not report conditions with specificity. For example, respondents with multiple heart or circulatory conditions tend to group them as a single problem.54 Consistent evidence has shown that the level of agreement between respondent and provider reports increases when conditions are collapsed into broader categories such as the CCS and system-level categories that were used in the current study.53, 54, 56 Information on the seriousness of conditions was not available in the MEPS, posing an additional study limitation. Contemporary survey-based indices of comorbidity are derived from a combination of condition enumeration and rating of condition severity.57, 58 Because the MEPS did not include indicators of condition severity, characterization of comorbidity was limited to a condition count. Therefore, the impact of conditions could not be ranked or evaluated in a meaningful way.

Back to Article Outline

Conclusions 

Adults with mobility limitations have a greater number of limitations than other adults. The comorbid disease burden of adults with mobility limitations is far in excess of other community dwelling adults, and certain disease combinations (eg, obesity, hypertension, diabetes) establish risk for acquiring additional health problems (eg, heart disease).59 In the aggregate, adults with mobility limitations experience these health problems in the context of fewer resources with which to address them. Contrary to the notion that high comorbidity is a problem of old age, more than half of adults with mobility limitations are of working age. Better understanding of the primary or secondary nature of health conditions, and the factors that drive their occurrence, is an important first-step toward informing prevention strategies.

Suppliers

Back to Article Outline

References 

  1. Centers for Disease Control and Prevention (CDC). Prevalence of disabilities and associated health conditions among adults—United States, 1999 [published erratum in: MMWR Morb Mortal Wkly Rep 2001;50(8):149]. MMWR Morb Mortal Wkly Rep. 2001;50(7):120–125
  2. Census 2000 Summary File 3 (SF 3): United States. Table QT-P21: Disability status by sex. http://factfinder.census.gov/servlet/QTTable?_bm=y&-state=qt&-qr_name=DEC_2000_SF3_U_QTP21&-ds_name=DEC_2000_SF3_U&-CONTEXT=qt&-redoLog=false&-geo_id=01000US&-format=&-_lang=en&-SubjectID=14793516Accessed October 8, 2007
  3. Centers for Disease Control and Prevention (CDC). Health-related quality of life and activity limitation—eight states, 1995. MMWR Morb Mortal Wkly Rep. 1998;47(7):134–140
  4. Campbell ML, Sheets D, Strong PS. Secondary health conditions among middle-aged individuals with chronic physical disabilities: implications for unmet needs for services. Assist Technol. 1999;11:105–122
  5. Ries P, Brown S. Disability and health: characteristics of persons by limitation of activity and assessed health status, United States, 1984-88. Adv Data. 1991;(197):1–12
  6. Chan L, Shumway-Cook A, Yorkston KM, Ciol MA, Dudgeon BJ, Hoffman JM. Design and validation of a methodology using the International Classification of Diseases, 9th Revision, to identify secondary conditions in people with disabilities. Arch Phys Med Rehabil. 2005;86:1065–1069
  7. Coyle CP, Santiago MC, Shank JW, Ma GX, Boyd R. Secondary conditions and women with physical disabilities: a descriptive study. Arch Phys Med Rehabil. 2000;81:1380–1387
  8. Kinne S, Patrick DL, Doyle DL. Prevalence of secondary conditions among people with disabilities. Am J Public Health. 2004;94:443–445
  9. Santiago M, Coyle C. Leisure-time physical activity and secondary conditions in women with physical disabilities. Disabil Rehabil. 2004;26:485–494
  10. Seekins T, Clay J, Ravesloot C. A descriptive study of secondary conditions reported by a population of adults with physical disabilities served by three independent living centers in a rural state. J Rehabil. 1994;60(April-June):47–51
  11. Traci MA, Seekins T, Szalda-Petree A, Ravesloot C. Assessing secondary conditions among adults with developmental disabilities: a preliminary study. Ment Retard. 2002;40:119–131
  12. Wilber N, Mitra M, Walker DK, Allen D, Meyers AR, Tupper P. Disability as a public health issue: findings and reflections from the Massachusetts survey of secondary conditions. Milbank Q. 2002;80:393–421
  13. Ettinger WH, Davis MA, Neuhaus JM, Mallon KP. Long-term physical functioning in persons with knee osteoarthritis from NHANES (I: Effects of comorbid medical conditions). J Clin Epidemiol. 1994;47:809–815
  14. Guralnik JM, LaCroix AZ, Abbott RD, et al. Maintaining mobility in late life (I. Demographic characteristics and chronic conditions). Am J Epidemiol. 1993;137:845–857
  15. Otiniano ME, Du XL, Ottenbacher K, Markides KS. The effect of diabetes combined with stroke on disability, self-rated health, and mortality in older Mexican Americans: results from the Hispanic EPESE. Arch Phys Med Rehabil. 2003;84:725–730
  16. Johnson RL, Gerhart KA, McCray J, Menconi JC, Whiteneck GG. Secondary conditions following spinal cord injury in a population-based sample. Spinal Cord. 1998;36:45–50
  17. Trupin L, Rice DP. Health status, medical care use, and number of disabling conditions in the United States. Disabil Stat Abstr. 1995;9:1–4
  18. Chan L, Beaver S, Maclehose RF, Jha A, Maciejewski M, Doctor JN. Disability and health care costs in the Medicare population. Arch Phys Med Rehabil. 2002;83:1196–1201
  19. Max W, Rice DP, Trupin L. Medical expenditures for people with disabilities. Disabil Stat Abstr. 1996;12:1–4
  20. Rice DP, LaPlante MP. Medical expenditures for disability and disabling comorbidity. Am J Public Health. 1992;82:739–741
  21. U.S. Department of Health and Human Services. Healthy people 2010: disability and secondary conditions (Vision for the decade). Proceedings and recommendations of a symposium, Atlanta, Georgia. December 4–5, 2000. Washington (DC): DHHS; 2001;
  22. Anson CA, Shepherd C. Incidence of secondary complications in spinal cord injury. Int J Rehabil Res. 1996;19:55–66
  23. Cifu DX, Kreutzer JS, Marwitz JH, et al. Etiology and incidence of rehospitalization after traumatic brain injury: a multicenter analysis. Arch Phys Med Rehabil. 1999;80:85–90
  24. Imai K, Kadowaki T, Aizawa Y, Fukutomi K. Morbidity rates of complications in persons with spinal cord injury according to the site of injury and with special reference to hypertension. Paraplegia. 1994;32:246–252
  25. Kapell D, Nightingale B, Rodriguez A, Lee JH, Zigman WB, Schupf N. Prevalence of chronic medical conditions in adults with mental retardation: comparison with the general population. Ment Retard. 1998;36:269–279
  26. Klotz R, Joseph PA, Ravaud JF, Wiart L, Barat M Tetrafigap Group. The Tetrafigap Survey on the long-term outcome of tetraplegic spinal cord injured persons: Part III (Medical complications and associated factors). Spinal Cord. 2002;40:457–467
  27. Marwitz JH, Cifu DX, Englander J, High WM. A multi-center analysis of rehospitalizations five years after brain injury. J Head Trauma Rehabil. 2001;16:307–317
  28. McKinley WO, Jackson AB, Cardenas DD, DeVivo MJ. Long-term medical complications after traumatic spinal cord injury: a regional model systems analysis. Arch Phys Med Rehabil. 1999;80:1402–1410
  29. Noreau L, Proulx P, Gagnon L, Drolet M, Laramee MT. Secondary impairments after spinal cord injury: a population-based study. Am J Phys Med Rehabil. 2000;79:526–535
  30. Turk MA, Geremski CA, Rosenbaum PF, Weber RJ. The health status of women with cerebral palsy. Arch Phys Med Rehabil. 1997;78(12 Suppl 5):S10–S17
  31. Vogel LC, Krajci KA, Anderson CJ. Adults with pediatric-onset spinal cord injury: part 1: prevalence of medical complications. J Spinal Cord Med. 2002;25:106–116
  32. Vogel LC, Krajci KA, Anderson CJ. Adults with pediatric-onset spinal cord injury: part 2: musculoskeletal and neurological complications. J Spinal Cord Med. 2002;25:117–123
  33. Walter JS, Sacks J, Othman R, et al. A database of self-reported secondary medical problems among VA spinal cord injury patients: its role in clinical care and management. J Rehabil Res Dev. 2002;39:53–61
  34. Whiteneck GG, Charlifue SW, Frankel HL, et al. Mortality, morbidity, and psychosocial outcomes of persons spinal cord injured more than 20 years ago. Paraplegia. 1992;30:617–630
  35. U.S. Department of Health and Human Services. The Surgeon General’s call to action to improve the health and wellness of persons with disabilities. Washington (DC): DHHS, Office of the Surgeon General; 2005;
  36. United States Department of Health and Human Services. Medical expenditure panel survey public use data files. http://www.meps.ahrq.gov/mepsweb/data_stats/download_data_files.jspAccessed October 8, 2007
  37. 1995 National Health Interview Survey on Disability, phase I and II [CD-ROM]. Hyattsville: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 1999;Series 10, No. 10B
  38. Benson V, Marano MA. Current estimates from the National Health Interview Survey, 1995. Vital Health Stat 10. 1998;(199):1–428
  39. Botman SL, Moore TF, Moriarity CL, Parsons VL. Design and estimation for the National Health Interview Survey, 1995-2004. Vital Health Stat 2. 2000;(130):1–31
  40. Hendershot GE, Larson S, Lakin KC. An overview of the National Health Interview Survey on disability. Res Soc Sci Disabil. 2003;3:9–40
  41. Cohen J. Methodology report #1: design and methods of the Medical Expenditure Panel Survey household component. Rockville: Agency for Health Care Policy and Research; 1997;AHCPR publication no. 97-0026
  42. Cohen SB. Methodology report #2: sample design of the 1996 Medical Expenditure Panel Survey household component. Rockville: Agency for Health Care Policy and Research; 1997;AHCPR publication no. 97-0027
  43. Cohen SB, DiGaetano R, Goksel H. Methodology report #5: estimation procedures in the 1996 Medical Expenditure Panel Survey household component. Rockville: Agency for Health Care Policy and Research; 1999;AHCPR publication no. 99-0027
  44. Iezzoni LI, McCarthy EP, Davis RB, Siebens H. Access to health care and insurance for working-age adults reporting mobility difficulties. Res Soc Sci Disabil. 2003;3:131–155
  45. Weigers ME, Drilea SK. Research findings #10: health status and limitations: a comparison of Hispanics, blacks, and whites, 1996. Rockville: Agency for Health Care Policy and Research; 1999;AHCPR publication no. 00-0001
  46. Elixhauser A, McCarthy E. Clinical Classifications for Health Policy Research, Version 2: hospital inpatient statistics. Rockville: Agency for Health Care Policy and Research; 1996;AHCPR publication no. 96-0017
  47. Elixhauser A, Steiner CA. Hospital inpatient statistics, 1996. HCUP-3 Research Note. Rockville: Agency for Health Care Policy and Research; 1999;http://www.hcup-us.ahrq.gov/reports/natstats/his96/clinclas.htmAccessed October 10, 2007
  48. MEPS HC-006R: 1996 medical conditions. Rockville: Agency for Health Care Policy and Research; 2007;http://www.meps.ahrq.gov/mepsweb/data_stats/download_data_files_detail.jsp?cboPufNumber=HC-006RAccessed October 10, 2007
  49. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults—The Evidence Report (National Institutes of Health [published erratum in: Obes Res 1998;6:464]). Obes Res. 1998;6(Suppl 2):51S–209S
  50. Guralnik JM, LaCroix AZ, Everett DF, Kovar MG. Aging in the eighties: the prevalence of comorbidity and its association with disability. Adv Data. 1989;(170):1–8
  51. Rice DP, LaPlante MP. Chronic illness, disability, and increasing longevity. In:  Sullivan S,  Lewin ME editor. The economics and ethics of long-term care and disability. Washington (DC): American Enterprise Institute for Public Policy Research; 1988;p. 9–55
  52. Verbrugge LM, Lepkowski JM, Imanaka Y. Comorbidity and its impact on disability. Milbank Q. 1989;67:450–484
  53. Cox BG, Iachan R. A comparison of household and provider reports of medical conditions. J Am Stat Assoc. 1987;82:1013–1018
  54. Edwards WS, Winn DM, Kurlantzick V, et al. Evaluation of National Health Interview Survey diagnostic reporting. Vital Health Stat 2. 1994;(120):1–116
  55. Guralnik JM. Assessing the impact of comorbidity in the older population. Ann Epidemiol. 1996;6:376–380
  56. Johnson AE, Sanchez ME. Household and medical provider reports on medical conditions: National Medical Expenditure Survey, 1987. J Econ Soc Meas. 1993;19:199–223
  57. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383
  58. Sangha O, Stucki G, Liang MH, Fossel AH, Katz JN. The Self-Administered Comorbidity Questionnaire: a new method to assess comorbidity for clinical and health services research. Arthritis Rheum. 2003;49:156–163
  59. Okoro CA, Denny CH, Greenlund KJ, et al. Risk factors for heart disease and stroke among diabetic persons, by disability status. J Diabetes Complications. 2005;19:201–206
  • a Release 9.1; SAS Institute Inc, 100 SAS Campus Dr, Cary, NC, 27513.
  • b Release 9.0; Research Triangle Institute, PO Box 12194, Research Triangle Park, NC 27709-2194.
  • ⁎ A higher proportion of adults with mobility limitations depend on public coverage than table 3 suggests because the any private insurance category includes Medicare beneficiaries with supplementary private coverage.

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated.

PII: S0003-9993(07)01666-8

doi:10.1016/j.apmr.2007.08.146

Archives of Physical Medicine and Rehabilitation
Volume 89, Issue 2 , Pages 210-218, February 2008

Article Tools