Determining paralysis prevalence in the United States

Disability and Health Journal 1 (2008) 172e179 www.disabilityandhealthjnl.com

Brief Report

Determining paralysis prevalence in the United States Michael H. Fox., Sc.D.a,*, Jennifer L. Rowland, Ph.D., M.P.H., P.T.b, Katherine Froehlich-Grobe, Ph.D.c, Dee Vernberg, Ph.D.d, Glen W. White, Ph.D.d, Lori Haskett, B.A.e,y a

Department of Health Policy and Management, University of Kansas Medical Center, Kansas City, KS 66160, USA b Department of Disability and Human Development, Center on Health Promotion Research for Persons with Disabilities, University of Illinois at Chicago, Chicago, IL 60608, USA c Gerontology Center, The University of Kansas, Lawrence, KS 66045, USA d Research and Training Center on Independent Living, University of Kansas, Lawrence, KS 66045, USA e Injury Prevention and Disability and Health, Office of Health Promotion, Kansas Department of Health and Environment, Topeka, KS 66612, USA

Abstract Background: Estimates of paralysis vary widely, largely owing to a lack of standard definition and nontargeted survey approaches. Like other poorly understood conditions such as fibromyalgia, chronic fatigue, or chronic pain, paralysis falls outside the scope of clearly defined medical diagnosis, further complicating surveillance efforts. This inability to identify accurate prevalence makes developing policy interventions around the needs of many persons with these disabilities problematic. The objectives were to investigate how paralysis is being measured in the United States and to examine the validity of prevalence estimates based on current approaches. Methods: We reviewed existing measurement instruments and surveyed 139 agencies and organizations to determine how they capture paralysis data. Results: There is a widespread reliance on ICD coding or broad functional capabilities for most state or federal agencies. Many organizations serving consumers depend on state registries for discrete conditions in which paralysis is not directly measured. Conclusions: Improved paralysis prevalence data will benefit from a more functional definition consistent with ICF guidelines, which can be part of future surveillance efforts at state and federal levels. Ó 2008 Elsevier Inc. All rights reserved. Keywords: Paralysis data; Disability; Prevalence; ICF

Introduction To have accurate data describing the number of persons with paralysis is an important part of developing policies and practices leading to improved health and quality of life for all persons with disabilities. As a condition defined by functionality rather than strict medical diagnosis, ‘‘paralysis’’ can demonstrate the utility of the International Classification of Disability, Functioning and Health (ICF), which views disability within the context of personenvironment interaction rather than physical insufficiency [1,2]. Understanding prevalence among conditions such as paralysis extends our ability to use the ICF model in ways consistent with a social model of disability, a goal of many in both * Corresponding author: 3901 Rainbow Boulevard, Kansas City, KS 66160; Fax: (913) 588-8236. E-mail address: [email protected] (M.H. Fox). y Funding for this project was provided to the University of Kansas through Grant # 3798 by the Christopher and Dana Reeve Foundation, which provided portions of salary support for authors Fox, Rowland, Froehlich-Grobe, Vernberg and White. 1936-6574/08/$ e see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.dhjo.2008.05.003

the public health and the disability community who have used or critiqued its use in the past [3,4]. Defining ‘‘paralysis’’ is the first step before being able to measure it. Different estimates of paralysis exist owing to the use of different criteria, varying sampling frames, or data collection methods designed for other purposes. Because of this variation in measurement, a consensus on how many persons have paralysis does not exist, even among stakeholders advocating for improved services and quality of life for persons with paralysis [5]. Persons with disabilities and certain chronic conditions characterized by functional limitations offer unique challenges to public health professionals. Conditions such as paralysis, while widely recognized and apparent to those having them, are associated with a diffuse array of underlying medical, environmental, and social factors, complicating efforts to identify prevalence. Without valid prevalence estimates, framing policy issues especially revelant to these populations at federal, state, and local levels is difficult. Current paralysis figures usually come from the 1995 National Health Interview Survey (NHIS) Disability

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179

Supplement, where respondents self-identifed as ‘‘paralyzed.’’ This led to prevalence of 9 per 1000 adults, or 2.4 million Americans [6]. Other available data yield widely varying results (Figure 1). When hospital discharge data are used, prevalence approaches 1.4 million. If estimates are calculated based on the most common conditions associated with paralysis (stroke, spinal cord injury [SCI], multiple sclerosis [MS], spina bifida, and cerebral palsy [CP]), there are closer to 5.3 million persons with paralysis in the United States [7-19]. As with the NHIS, these estimates can also be challenged, as each makes its own assumptions about how paralysis is defined or measured [20]. A number of researchers have noted the connection between valid population-based data for persons with disabilities and policy development [21-25]. Data provide a framework from which disability policy evolves. Investigating ways to collect data for ill-defined disabling conditions such as paralysis is important to develop more

173

accurate prevalence estimates needed to develop health policies that affect persons with these conditions.

Methods and Materials We began our investigation by reviewing existing measurement tools in the public domain from which many current national estimates on persons with disabilities are derived. We then surveyed by telephone government agencies and private organizations representing persons with paralysis-related disabilities. We followed up some of our surveys to make further inquiries, through either site visits, discussions with grantees of state disability agencies, or lengthy telephone interviews, focusing on agencies and organizations that appeared to show particular promise of expanding their current approaches to paralysis surveillance.

Figure 1. Estimating Paralysis Prevalence Using National Hospital Discharge Data and Most Common Conditions Associated with Paralysis.

174

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179

The purpose of our national telephone survey of government agencies and private organizations was to assess their paralysis surveillance capacities. Our funders, the Christopher and Dana Reeve Foundation (grant No. 3798; University of Kansas Human Subjects Committee Approval HSC No. 14777), provided us with a working definition of paralysis: related disorders, diseases, birth defects, and injuries that affect ambulation or paralysis of the extremities. This does not include disorders like vocal cord paralysis, gastrointestinal paralysis and facial paralysis. Copies of our survey are available from the corresponding author upon request. Our sample included 56 state and federal agencies and 83 organizations known to collect disability-related data, originally derived from a list of 334 state and federal agencies and 151 organizations we developed with project advisors at the Christopher and Dana Reeve Foundation and a national advisory group of leading authorities in the field of disability and paralysis surveillance (see Acknowledgments). The 56 state and federal agencies surveyed were selected so that each of the 10 federal regions would be represented by 5 or 6 agencies, with selection determined by the willingness of the agency to participate in our survey. Our sampling strategy randomly selected replacement agencies to contact within each region, ensuring that at least one was selected from birth defects registries, cancer registries, hospital discharge databases, trauma system registries, or ‘‘‘‘other,’’‘‘ such as Centers for Disease Control and Prevention (CDC)funded state disability and health programs or regional VA centers. Our final sample thereby provided a cross section of government agency data surveillance approaches with national geographic variation. We used snowball sampling to expand our original sample pool. States administer the federally underwritten Behavioral Risk Factor Surveillance Survey (BRFSS), which can also be a data source for obtaining prevalence for disability [7]. We contacted all 50 states via telephone to determine which used state-added questions specific to paralysis but, finding none, did not include these agencies in our larger survey. Federal agencies from our original sample pool were those managing disability programs from within the Departments of Health and Human Services, Labor, Education, Veterans Affairs, and Defense and the Social Security Administration. For our survey of private organizations, the Christopher and Dana Reeve Foundation provided us with conditions they thought were most closely associated with paralysis and mobility limitations. We subsequently contacted organizations representing people having these conditions, in addition to other organizations recommended during subsequent interviews. We did not interview people at 48 of the 151 organizations from our original sample because the organization was a regional office that referred us to the national organization. Twenty organizations contacted by

telephone did not return our calls. Examples of organizations that were included are the American Stroke Association, Easter Seals, Family of Spinal Muscular Atrophy, Miami Project to Cure Paralysis, National MS Society, The Transverse Myelitis Association, and United Cerebral Palsey Organization, as well as others. A complete listing is available from the corresponding author upon request. Results Table 1 summarizes collection methods from current surveys using a matrix approach in which approaches are organized by three types: functional, clinical, and ICF. As a functional measure, the NHIS [26] contains two questions involving paralysis: (1) ‘‘How long have you had polio (myelitis), paralysis or para/quadriplegia?’’ (2) ‘‘Enter time period for time with polio (myelitis), paralysis or para/ quadriplegia.’’ Other NHIS questions focused on conditions that may have been associated with paralysis, such as stroke or equipment use. While functional in nature (i.e, ‘‘Do you now have any health problems that require you to use special equipment, such as a cane, a wheelchair, a special bed, or a special telephone?’’ ‘‘By yourself, and without using any special equipment, how difficult is it for you to . Walk a quarter of a miledabout 3 city blocks?’’), none of them seek sufficient detail to identify a functional equivalence of paralysis, such as ability to use arms and legs and related questions. Similarly, the U.S. Census and American Community Survey [27] contains questions regarding functional levels related to disability. One example is, ‘‘Do you have a condition that substantially limits physical activities such as walking, climbing stairs, reaching, lifting, or carrying?’’ But specific factors related to paralysis are absent. SLAITS (the State and Local Area Integrated Telephone Survey [28]) collects information on functional limitations and disabling conditions. However, it is not possible to extrapolate specific paralysis information. For example, one question states: ‘‘. [are you] limited in any way because of a physical, mental or emotional problem. .’’ This defines broad functional limitations but limits more targeted conditions such as paralysis. In the Survey of Income and Program Participation (SIPP) [29], respondents complete activities of daily living (ADL) and instrumental activities of daily living (IADL) questions, in addition to those that ask about supplemental security income (SSI). This tool omits questions about more specific paralysis information. The BRFSS [7], coordinated through the CDC but administered individually in each state, contains core and module questions related to disability, while using a sample frame of only noninstitutionalized adults. Examples of broadly functional related questions include, ‘‘Are you limited in any way in any activities because of any impairment or health problem?’’ ‘‘Do you now have any health problems that require you to use special equipment such as

Table 1 Defining ParalysisdA Matrix Approach Type

Purpose(s)

Strengths

Weaknesses

Sample Questions

National Health Interview Survey (NHIS)

Functional

Collect data on the health status of the U.S. population

Ongoing since 1957

Unable to identify most, but not all, functional conditions

Address specific issues of current public health concern

Good response rate Comprehensive Links to MEPS and NDI Timeliness

Self-report bias

Provide estimates of demographic, housing, social, and economic characteristics every year for states, cities, counties, metropolitan areas, and population groups of 65,000 people or more. In-depth state and local area data to meet various program and policy needs

Ongoing since 1996

Specific conditions associated with functionality are not asked for

How long have you had polio (myelitis), paralysis, or para/ quadriplegia? Do you now have any health problems that require you to use special equipment such as a cane, a wheelchair, a special bed, or a special telephone? Do you have a condition that substantially limits physical activities such as walking, climbing stairs, reaching, lifting, or carrying?

www.cdc.gov/nchs/nhis.htm

Census and the American Community Survey (ACS) www.census.gov/acs

Functional

State and Local Area Integrated Telephone System (SLAITS) www.cdc.gov/nchs/slaits.htm

Functional

Behavioral Risk Factor Surveillance System 2(BRFSS) http://www.cdc.gov/brfss/

Functional

Survey of Income and Program Participation (SIPP)

Functional

www.bls.census.gov/sipp

National Health and Nutrition Examination Survey (NHANES) www.cdc.gov/nchs/nhanes.htm

Functional

State-based system of health surveys that collects information on health risk behaviors, preventive health practices, and health care access primarily related to chronic disease and injury.

Good response rate County- and community-level data Ongoing telephone survey that screens nearly 1 million households per year Can provide comparative data at local levels Broad range of personal health behaviors Multiyear and multistate comparisons Ongoing since 1994

Specific conditions associated with functionality are not asked for Self-report bias

Are you limited in any way because of a physical, mental, or emotional problem?

Only surveys adults

Are you limited in any way in any activities because of any impairment or health problem? Do you now have any health problems that require you to use special equipment such as a cane, wheelchair, a special bed, or a special telephone? Do you use a cane, crutches or a walker (wheelchair, electric scooter, or similar aid is related question)?

Specific conditions associated with functionality are not asked for Self-report bias

To provide accurate and comprehensive information about the income and program participation of individuals and households in the United States, and about the principal determinants of income and program participation

Good resource for family labor and income use

Functional limitation questions considered part of ‘‘topical modules’’ that are offered periodically

Ongoing panels since 1984

No connection to condition Self-report bias

To assess the health and nutritional status of adults and children in the United States through interviews and direct physical examinations

Comprehensive

Work-related disability

Ongoing panels Use of multiple domains

Self-report bias Time limited

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179

Survey

Because of a physical or mental condition, do you have difficulty . getting around inside the home .? (Activities of daily living questions) What is the main reason you did not work last week . (disabled)?

(Continued) 175

176

Table 1 Continued Survey

Type

Purpose(s)

Strengths

Weaknesses

Sample Questions

Healthcare Cost and Utilization Project (H-CUP) www.ahrq.gov/data/hcup

Clinical

Claims form abstractiondno survey questions

Clinical

Kansas Behavioral Risk Factor Surveillance System (Kansas BRFSS) http://www.kdheks.gov/brfss/

International Classification of Disability, Functioning and Health (ICF)

Scope of data: Nationwide Inpatient (NIS), Kids’ Inpatient (KID), State Inpatient (SID), State Ambulatory Surgery (SASD), State Emergency Department (SEDD) Comprehensive source of data on the cost and use of health care and health insurance coverage linking person to physician and employer Less self-report bias since given different disability domains to select from Continuous Allows comparison of behavioral risk factors between subpopulations with and without disabilities

Assumes facility use for person within a given year

Medical Expenditure Panel Survey (MEPS) www.meps.ahrq.gov

To collect data on the cost, utilization of, and access to health care using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding To collect data on a nationally representative sample of the population on the cost, use, and quality of health care

Assumes physician treatment of disability and ICD coding May exclude large numbers of persons for nonprimary conditions

‘‘I need the diagnoses for these visits. .’’ (asked of providers)

Only asked of adults in household

Question No. 8: Is your ability to move around due to paralysis? Note: If asked, ‘‘Paralysis is defined as loss of function or feeling that affects the ability to move your arms or legs but does not include amputation or missing limbs’’ 1 Yes 2 No 7 Don’t know/Not sure 9 Refused Four of 13 being tested:

Must be purchased as state added questions each year

Validation Ongoing

Christopher and Dana Reeve Foundation Consensus Conference (proposed)

ICF

To develop a personenvironment, comprehensive, and functionally based set of questions that can be used in national surveys to identify paralysis

Combines both clinical and functional definitions Consistent with domainspecific categorization of disability

Needs to be validated Will be costly to administer until in widespread use

Do you or does anyone in this household have any difficulty moving their arms or legs? Currently, how difficult is it for (you/this person) to move (your/their) arms and/or legs? What is this difficulty in movement due to? What condition or disease caused (your/their) difficulty in moving?

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179

To include nine questions that categorize individuals into broad domains of disability based on their having selfreported using the existing two national BRFSS survey questions that screen for disability (see above)

Functional disability may not be primary or secondary diagnosis

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179

a cane, wheelchair, a special bed, or a special telephone?’’ and ‘‘Are you limited in any way in any in any activities because of a physical, mental, or emotional problem?’’ In the National Health and Nutrition Examination Survey (NHANES) [30], respondents identify the main reason they did not work the previous week. Although not specifically targeted at examining paralysis, one answer choice is ‘‘disabled.’’ As an example of an information source using clinical International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding, the Healthcare Cost and Utilization Project (H-CUP) [31] includes five datasets: the Nationwide Inpatient Sample, Kid’s Inpatient Database, State Inpatient Databases, State Ambulatory Surgery Databases, and State Emergency Department Databases. Using Clinical Classification Software (CCS), ICD9-CM codes group diagnoses into one paralysis code, CCS 82. These surveys have the ability to screen for both primary and secondary identification of paralysis; they can identify paralysis by year and gender, age, and region without any conversion to statistical software. However, they are dependent on a clinical diagnosis that could be both transitory and unrelated to a person’s functional capacity, such as having facial paralysis that may not limit mobility. In addition, paralysis is excluded if it is coded as a secondary condition. A second data source using ICD-9 coding, the Medical Expenditure Panel Survey (MEPS) [32], collects ICD-9CM codes and categorizes them into clinical classification

177

software coding similar to the H-CUP data; CCS 82 is paralysis and CCS 6.3 represents a multiple-level diagnosis of paralysis. Information collected with MEPS is based on the assumption that some form of health care services is received and coded based on paralysis as a primary condition but, similar to H-CUP data, may exclude large numbers of persons for whom paralysis is a secondary condition. Recent efforts to integrate body functions, activities, and participation along the lines of the ICF model are presented last. The 2006 and 2008 Kansas BRFSS surveys use a panel of nine questions, among which (No. 8) is, ‘‘‘‘Is your ability to move around due to paralysis?’’‘‘ This follows other questions that focus on the functional nature of the disability and are available online [33]. The Christopher and Dana Reeve Foundation, through their National Paralysis Task Force, which included a national concensus conference, have also drafted a series of questions intended for use in state BRFSS or national surveys that use this ICF framework [5]. Four of the 13 questions developed are presented in our matrix, including, ‘‘ Do you or does anyone in this household have any difficulty moving their arms or legs?’’ These questions show great promise in being able to classify disability into health-related domains that integrate body function and struction with environment and performance. Our survey summaries are given in Table 2. Of 56 state and federal agencies included in our interviews, 30 (54%) collected paralysis information. Of the 83 organizations in our interviews, 14 (17%) collected some form of

Table 2 Paralysis Surveillance CapacitydSurvey Results (2005) Question Able to determine paralysis from data routinely collect? Way(s) in which paralysis is (are) identified? ICD or E coding Registry reporting Survey (including BRFSS) Enrollment or intake info Primary purpose of collecting paralysis data? Resource and funding allocation Research and evaluation Program and policy development Advocacy Conditions related to paralysis for which collect information? Spinal cord injury Traumatic brain injury Birth defects Stroke Multiple sclerosis Importance of collecting and using paralysis information? Important Not important Don’t know or no response

State or Federal agencies(n 5 56)

Disability organizations(n 5 83)

Total(n 5 139)

30/56 (54%)

14/83 (17%)

44/139 (32%)

24/30 9/30 6/30 1/30

(90%) (36%) (20%) (3%)

2/14 14/14 1/14 8/14

(14%) (100%) (7%) (57%)

26/54 23/54 7/54 9/54

(48%) (43%) (13%) (17%)

4/30 12/30 6/30 0/30

(13%) (40%) (20%) (0%)

5/14 9/14 4/14 6/14

(36%) (64%) (29%) (43%)

9/54 21/54 10/54 6/54

(17%) (39%) (19%) (11%)

21/30 20/30 12/30 8/30 8/30

(70%) (67%) (40%) (27%) (27%)

5/14 1/14 3/14 3/14 3/14

(36%) (7%) (21%) (21%) (21%)

26/54 21/54 15/54 11/54 11/54

(48%) (39%) (28%) (20%) (20%)

13/30 (43%) 6/30 (20%) 11/30 (37%)

9/14 (64%) 0/14 (0%) 5/14 (36%)

22/54 (41%) 6/54 (11%) 16/54 (30%)

Note: State and federal agencies were drawn from a sampling frame of 334 agencies identified by the Christopher and Dana Reeve Foundation as stakeholders in use of paralysis data. Organizations were drawn from a sampling frame of 151 not-for-profit organizations provided to us from the same source. Our final sample was identified in conjunction with recommendations made from our external advisory committee and foundation staff that ensured geographic and condition variation (see Methods).

178

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179

paralysis information. Within these groups, 22 of the 30 agencies (73%) screened for paralysis based on ICD codes. Among the 14 organizations, all reported using some form of state registry information, with 6 (43%) using this information to promote advocacy efforts and 9 (64%) needing it for research and evaluation. When asked to rate the importance of collecting information for their agencies or organizations among those collecting this information, 43% (13 of 30) state and federal agencies and 64% (9 of 14) organizations rated paralysis as important information to collect. Discussion Our investigation confirmed that current approaches used to gather disability data do not allow us to accurately identify paralysis prevalence. There first needs to be a uniform and more widely accepted definition that captures the breadth of possible ways in which paralysis can manifest itself. As part of initial efforts to advance this issue on the national level, the definition of ‘‘paralysis’’ needs to encompass categories inclusive of all people who may experience functional limitations and decreased social participation because of these limitations. We consider identification of a clear definition to be one of the cornerstones of future efforts to measure paralysis, and should be part of a national consensus effort directed by consumers and other key stakeholders in paralysis. The starting point should involve developing a paralysis framework around ways in which people with paralysis are limited functionally. This parallels efforts of the World Health Organization’s (WHO) International Classification of Functioning, Disability and Health [1]. One way to capture functional information uniformly across many sources would be to develop a disability module that could be added to existing state and federal surveys. We piloted questions in Kansas’s 2006 and 2008 BRFSS that could be used as a basis for this module, with statewide prevalence estimates now available online [33]. The Paralysis Task Force of the Christopher and Dana Reeve Foundation have supported this effort and advanced it further since their first national concensus conference held in Atlanta in 2006 [5]. This work could lead to adding similar questions to the national BRFSS core module, including them in national surveys such as the NHIS or NHANES, or as part of a national survey dedicated to investigating both the prevalence of paralysis and more detailed information on behavioral, environmental, and medical conditions among those identified. Conclusion In moving forward with future data collection efforts, we believe it would be worthwhile to work with researchers and program managers in states that appear closest to having functional, data-driven public policy capacities in place.

Our study revealed a small cadre of states in which demonstration programs to build a capacity for ongoing surveillance of paralysis independent of state surveys or registries could be fruitful. Approaches could include linking existing data sources or working toward other forms of surveillance that states deem most suited to broader public health needs, of which paralysis would be one. Through our many conversations in the course of this project and in our own experiences, public health agencies throughout the nation appear poised to participate in this effort if they are not already doing it. What has been principally lacking in the past has been a guiding vision of using data for surveillance in ways that integrate current collection mechanisms with innovative and consumeroriented approaches to measuring disability. This vision must come from leaders of federal or state agencies that could best capitalize from this approach, as well as leaders of advocacy organizations representing the interests of people with the most to gain. The ICF model provides the appropriate framework to help guide these new efforts. By acquiring valid, functionally based paralysis data that allow more accurate prevalence measurement, policy and existing services can improve for this segment of our population, while encouraging expansion of similar efforts in other areas of poorly defined disabling conditions.

Acknowledgments The authors appreciate the support and guidance they received from representatives of the Christopher and Dana Reeve Foundation, primarily Joe Canose, Sheila Fitzgibbon, and Anthony Cahill. In addition, the authors give thanks for the many insights and suggestions from our external advisory team of Elena Andresen, Michael DeVivo, David Gray, and Renee Johnson.

References [1] World Health Organization (WHO). International Classification of Disability, Functioning, & Health (ICF). www.who.int/icidh. Accessed June 1, 2005. [2] Centers for Disease Control and Prevention, National Committee on Vital and Health Statistics (NCVHS). Classifying and Reporting Functional Status. http://www.ncvhs.hhs.gov/010716rp.htm. Accessed December 27, 2007. [3] Khan F, Pallant JF. Use of International Classification of Functioning, Disability and Health (ICF) to describe patient-reported disability in multiple sclerosis and identification of relevant environmental factors. J Rehabil Med. 2007;39:63-70. [4] Nordenfelt L. On health, ability and activity: comments on some basic notions in the ICF. Disabil Rehabil. 2006;28:1461-1465. [5] First Report of the Working Group on Developing a Paralysis Survey Module and an Implementation Strategy to Gather National Level Data on Paralysis. Short Hills, NJ: Christopher and Dana Reeve Foundation Paralysis Resource Center; November 2006. [6] Centers for Disease Control and Prevention, National Center on Health Statistics (NCHS).http://www.cdc.gov/nchs/about/major/ nhis_dis/nhis_dis.htm. AccessedDecember 27, 2007.

M.H. Fox et al. / Disability and Health Journal 1 (2008) 172e179 [7] Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. Behavioral Risk Factor Surveillance Survey. http://www.cdc.gov/brfss/. Accessed June 14, 2007. [8] Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke. 1988;19:1497-1500. [9] Irwin ZN, Arthur M, Mullins RJ, Hart RA. Variations in injury patterns, treatment, and outcome for spinal fracture and paralysis in adult versus geriatric patients. Spine. 2004;29:796-802. [10] Jackson AB, Dijkers M, Devivo MJ, Poczatek RB. A demographic profile of new traumatic spinal cord injuries: change and stability over 30 years. Arch Phys Med Rehabil. 2004;85:1740-1748. [11] Rothwell PM, Charlton D. High incidence and prevalence of multiple sclerosis in south east Scotland: evidence of a genetic predisposition. J Neurol Neurosurg Psychiatry. 1998;64:730-735. [12] Rice-Oxley M, Williams ES, Rees JE. A prevalence survey of multiple sclerosis in Sussex. J Neurol Neurosurg Psychiatry. 1995;58:27-30. [13] Mathews TJ, Honein MA, Erickson JD. Spina bifida and anencephaly prevalencedUnited States, 1991-2001. MMWR Recomm Rep. 2002; 51:9-11. [14] Meyer RE, Siega-Riz AM. Sociodemographic patterns in spina bifida birth prevalence trendsdNorth Carolina, 1995-1999. MMWR Recomm Rep. 2002;51:12-15. [15] Winter S, Autry A, Boyle C, Yeargin-Allsopp M. Trends in the prevalence of cerebral palsy in a population-based study. Pediatrics. 2002;110:1220-1225. [16] Yeargin-Allsopp M, Boyle C. Overview: the epidemiology of neurodevelopmental disorders. Ment Retard Dev Disabil Res Rev. 2002;8: 113-116. [17] Prencipe M, Ferretti C, Casini AR, Santini M, Giubilei F, Culasso F. Stroke, disability, and dementia: results of a population survey. Stroke. 1997;28:531-536. [18] Reingold SC. Research Directions in Multiple Sclerosis. New York: National Multiple Sclerosis Society; 1995. [19] Williams LJ, Mai CT, Edmonds LD, et al. Prevalence of spina bifida and anencephaly during the transition to mandatory folic acid fortification in the United States. Teratology. 2002;66:33-39.

179

[20] Christopher and Dana Reeve Paralysis Foundation. SCI facts. Available at http://www.christopherreeve.org/site/c.geIMLPOpGjF/ b.1402847/k.EE60/SCI_Facts.htm. Accessed December 31, 2007. [21] Fox M, Kim KM, Ehrenkrantz D. Developing comprehensive statewide disability information systems. J Disabil Policy Stud. 2002;13:171179. [22] Barnard T. The Americans with Disabilities Act: nightmare for employers and dream for lawyers? St. John’s Law Rev. 1999;64:229-252. [23] Blank P. The economics of the employment provisions of the Americans with Disabilities Act: Part Idworkplace accommodations. DePaul Law Rev. 1997;46:877-914. [24] Christopher J. Title I of the Americans with Disabilities Act: disabling the disabled. Cumberland Law Rev. 1995;25:715-752. [25] Stein M. Empirical implications of Title I. Iowa Law Rev. 2000;85: 1671-1690. [26] NHIS (National Health Interview Survey). http://www.cdc.gov/nchs/ nhis.htm. Accessed July 10, 2007. [27] U.S. Census Bureau, Housing and Household Economic Statistics Division. http://www.census.gov/hhes/www/disability/disability.html. Accessed June 14, 2007. [28] National Center for Health Statistics (NCHS). SLAITS (State and Local Integrative Telephone Survey). http://www.cdc.gov/nchs/slaits. htm. Accessed January 18, 2008. [29] U.S. Census Bureau. SIPP (Survey of Income and Program Participation). http://www.bls.census.gov/sipp/. Accessed January 18, 2008. [30] National Center for Health Statistics (NCHS). NHANES (National Health and Nutrition Examination Survey). http://www.cdc.gov/nchs/ nhanes.htm. Accessed January 18, 2008. [31] HCUP Databases Healthcare Cost and Utilization Project (HCUP). Rockville, MD: Agency for Healthcare Research and Quality; 1998-2004. www.hcup-us.ahrq.gov/databases.jsp. [32] MEPS (Medical Expenditure Panel Survey). http://www.meps.ahrq. gov/mepsweb/. Accessed July 10, 2007. [33] Kansas Department of Health and Environment. Behavioral Risk Factor Surveillance Survey. http://www.kdheks.gov/brfss/Questionnaires/quest2006.html#StateDisabilityQualityLife06. Accessed December 31, 2007.