Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 11 , Pages 1513-1521, November 2007

The State-of-the-Science: Access to Postacute Care Rehabilitation Services. A Review

Presented to the State-of-the-Science Symposium on Post-Acute Rehabilitation, February 13, 2007, Washington, DC.

  • Kenneth J. Ottenbacher, PhD, OTR

      Affiliations

    • Corresponding Author InformationReprint requests to Kenneth J. Ottenbacher, PhD, OTR, Div of Rehabilitation Sciences, School of Allied Health Sciences, University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-1137
    • ,
  • James E. Graham, PhD, DC

University of Texas Medical Branch, Galveston, TX.

Article Outline

Abstract 

Ottenbacher KJ, Graham JE. The state-of-the-science: access to postacute care rehabilitation services. A review.

Objective

To identify and summarize potential factors that influence access to postacute care (PAC) rehabilitation services.

Data Sources

Ovid Medline and agency-specific (eg, RAND Corp, Research Triangle Institute, U.S. Department of Health and Human Services) searches, using combinations of the terms health services accessibility, delivery of healthcare, access to care, post-acute, postacute, and access. Names of prominent authors were also searched to retrieve their relevant articles.

Study Selection

The initial search resulted in 823 articles published between 1980 and 2006. After we reviewed titles and abstracts, we obtained and examined 49 potential sources. Our final sample of 35 studies had descriptions and/or analyses of measures that might serve as indictors of access to PAC settings.

Data Extraction

Key information abstracted from each article and document included author(s), year of publication, patient groups, sample size, health care setting, and a brief summary of the access-related issues in PAC.

Data Synthesis

Studies varied by patient group, care setting, and access categories. Indicators of access to PAC rehabilitation services were classified into 4 categories: financial (46%), personal (31%), structural (26%), and attitudinal (23%). Many studies assessed more than 1 category.

Conclusions

We identified categories of information that can be used to develop a monitoring system for PAC services. The inclusion of access indicators in existing health report cards and quality indicator systems should be further explored.

Key Words: Delivery of health care, Disabled persons, Quality indicators, Rehabilitation, Review [publication type]

 

ACCESS TO HEALTH CARE refers to the degree to which patients and groups are able to obtain needed services1; access is often equated to having a sufficient supply of health care providers and facilities (eg, hospitals) in a given geographic area. It is important to understand, however, that eligibility and availability provide increased opportunities for access to services, but do not guarantee that appropriate services will be provided.2

Beyond the frequently cited figures about how many Americans are without health insurance,3 health care providers, policy-makers, and consumers have few objective and regularly reported indicators that characterize access to health care. This is particularly true for postacute care (PAC) services that are provided in a variety of settings, including rehabilitation units, rehabilitation hospitals and centers, skilled nursing facilities (SNFs), long-term care hospitals (LTCHs), and home health agencies (HHAs). Until recently, providers, policy-makers, and researchers focused on access to primary and acute medical services. The dramatic increase in the use and cost of PAC services during the 1980s and 1990s resulted in the Balanced Budget Act of 1997 (BBA) and prospective payment systems (PPS) for all major venues of PAC.4 The history and impact of PPS on PAC settings are described in companion articles published in this series5, 6, 7 and will not be discussed here. Rather, our focus is on how to define, monitor, understand, and eventually improve access to PAC rehabilitation services.

In conducting this systematic review, we adapted a definition used by the Institute of Medicine in which “access” is defined as the timely use of individual PAC rehabilitation services to achieve the best possible health outcomes.1 This definition requires matching patients’ needs with the availability and, more importantly, the delivery of care. Identifying measures that reflect the outcomes associated with PAC services for individual patients can operationalize the definition. Thus, the first step in understanding and correcting access-related problems is identifying those problems. Access indicators should detect when and where problems occur. Indicators do not explain the cause of these problems, but they can provide a basis for generating ideas and hypotheses that can be tested about why there are differences in access. We acknowledge that even the best indicators of access will be proxies for complicated phenomenon but, over time, these indicators can provide information about changes or improvements in the amount and type of care delivered.

An important starting point in improving access to postacute rehabilitation is to develop a better understanding of how services are provided, and to identify potential barriers to receiving and monitoring services across different populations and settings. We conducted a systematic review of studies that discussed or included measures of access in 1 or more of the following settings: inpatient medical rehabilitation units, rehabilitation hospitals and centers, SNFs, LTCHs, and HHAs. Our purpose in this review was to identify potential indicators, or classes of indicators, which might contribute to a system for monitoring PAC rehabilitation services. A monitoring system would be useful for policy-makers, providers, and consumers, as well as for families of people with disabilities and/or chronic health conditions.

Back to Article Outline

Methods 

Search Strategy and Selection Criteria 

The initial search by 1 author (JEG) covered the years 1980 to 2006 and involved using default search terms from the National Library of Medicine’s Medical Subject Headings within the Ovid Medline database describing access issues in PAC. Search terms included health services accessibility, delivery of health care, access to care, skilled nursing facilities, and subacute care. A follow-up search by both authors focused on articles that contained both postacute care (postacute, postacute, or post acute) and access in the title and/or abstract. All of our subsequent searches involved identification and review of similar articles and prominent authors, as well as agency-specific searches (eg, RAND Corp, Research Triangle Institute, U.S. Department of Health and Human Services) for nonindexed reports and summaries.

Study Assessment and Data Extraction 

The primary criterion for including an article in our final sample was a description or analysis of variables or measures that might serve as indicators of access to PAC in the venues listed above. We evaluated and discussed the eligibility of potential articles. Each article was summarized based on several factors including: source, year of publication, and number and type of patients and participants (ie, medical diagnosis).8 It is important to note that the summary (table 1) may not reflect the primary aim or outcome of a given study, but rather describes information in the document that pertained to access-related issues in PAC.

Table 1. Summary of the Literature on Access to Postacute Care
StudyYearPatientsSettingAccess CategorySummary
Angelelli et al152002Medicare (N range, 222,020–213,915)SNFFinancialUsed Medicare hospital and nursing facility claims (1996−1999) to compare access to Ohio SNFs for costliest patients after the BBA. Results: Little effect of BBA (SNF PPS) on access to SNFs, even for “costliest” patients. It is not known, however, whether incentives to minimize costs are affecting care plans and outcomes.
Angelelli et al162006Older adults (N=62,601)SNFPersonalUsed 2002 MDS information to assess ethnic and education differences in SNF access. It is generally believed that choice is limited by demand, geography (proximity), availability, and hospital discharge planners (primary mediators of decision). Results: African Americans and less-educated people are more likely to be admitted to the ”worst-quality” SNFs in a given region.
Bernstein et al172003All (N≈7,800,000)AllFinancialReport by the National Center for Health Statistics based on the National Health Care Survey. Through the 1990s the home health industry was the fastest growing sector in the health care industry and Medicare was the single largest payer for home health services. Results: After reductions in payment rates (BBA 1997), the number of HHAs declined 26% by 2000. The number of elderly patients receiving home health services decreased ≈50% from 1996−2000.
Braddom142005Medicare (N=NA)IRFStructuralReviews history of government actions that have impacted IRFs: Medicare and Medicaid (1965); or hospital PPS (1982); or 75% rule (1983); or 75% rule expanded to 10 diagnoses (1984); or BBA (1997); or IRF PPS (2002); or CMS final rule: strict enforcement of 75% rule (2004). Results: The 75% rule is prohibiting needed care. In addition, local coverage determinations that define the criteria for admission and that are created by various fiscal intermediaries (11 total), are not consistent across the United States, or with standard rehabilitation practice and beliefs.
Bronskill et al182002Medicare: heart attack (N=39,837)AllStructuralData from Cooperative Cardiovascular Project, along with CMS administrative and provider files. Assessed influence of external factors (hospital and state characteristics) on PAC utilization by elderly patients after a heart attack. Results: 37% received PAC within 30 days of discharge. Home health was most common setting. Key predictors of PAC use included severity (more severe), hospital ownership (for-profit), and home health services status (provided by hospital or subsidiary). After controlling for patient- and facility-level characteristics, PAC use still varied by states.
Buczko192001Medicare (N=77,768)IRF/SNFAttitudinalUsed Medicare data to predict IRF vs SNF use after acute care (1992). Results: Patients hospitalized for the following conditions are more likely to be admitted to IRF: spinal or cranial surgery, stroke, CABG, hip fracture or replacement, and tracheostomy. Those hospitalized for CHF, COPD, and other pulmonary conditions are more likely to be discharged to SNFs. Longer acute LOS, older age, being a woman, and HMO membership are also positively associated with SNF use.
Buntin et al202004Medicare: stroke, hip fracture, joint replacement (N=185,608)IRF/SNFAllRand report comparing influence of clinical and nonclinical factors on PAC (1999). Patients, family, doctors, and discharge planners must consider a variety of clinical and nonclinical rationale when determining PAC options. Individual factors include age, sex, race, marital status, functional status, disability level, and medical condition and comorbidities. Hospital-level factors include Medicare volume, size, the percentage of low-income patients, ownership, teaching status, and PAC affiliation. Regional factors include relative community income and PAC supply. Results: PAC availability (distance, supply, and hospital affiliation) is a better predictor than clinical characteristics in many models.
Buntin et al212005Medicare: stroke, hip fracture, joint replacement (N>1,000,000)AllFinancialRand report on BBA effects on PAC access using Medicare claims (1996−2003) as successive PPSs were implemented. Results: In general, each PPS reduced utilization at target sites and increased utilization (substitution) across other settings. Access for severely ill patients was not disproportionately affected.
Buntin et al222006Medicare (N range, 366,145–471,984)IRFFinancialRAND report on impact of IRF PPS using Medicare data (1998−2002). The shift from cost-based to PPS was intended to provide incentives to provide efficient care; however, it also encourages facilities to change their care and coding practices to increase revenues. Results: The number of IRFs and patient discharges increased from 1996−2002. Ratings of functional status decreased (increased severity) across all rehabilitation impairment groups from 1999−2002 (due to both changes in coding rules and upcoming). LOS consistently decreased from 1998−2002.
Carter et al232003Medicare (N=306,255)IRFStructuralRAND report describing case mix at IRFs for CMS review of the 75% rule based on IRF-PAI (2002) data and MEDPAR files (1996−1999). At least 75% of an IRF’s patient mix must be from 1 of 10 specific conditions in order to ensure that IRFs are primarily providing intensive rehabilitation services. Results: Overall IRF case mix did not and does not meet the 75% rule. Only 50% of 2002 IRF patients had a medical condition from the list and only 13% of IRFs were compliant. Compliance percentages vary regionally across the country. If and when enforcement of the 75% rule is increased, >80% of IRFs may be noncompliant (ineligible).
Carter et al242005Medicare (N range, 344,126−390,048)IRFAllRAND report summarizing 4 previous RAND efforts to assist with development of the IRF PPS. Used MEDPAR and FIM data to gather case mix information and develop weighted CMGs for payment purposes.
Cotterill and Gage42002Medicare (N=NA)AllFinancialOverview of PAC utilization before and after the BBA (1994−2000). Results: Acute hospital stays declined and PAC use increased sharply between 1994−1997. The interim payment system for HHAs (1997) followed by a PPS resulted in a sharp decline in home health services. The SNF PPS (1998) resulted in a flat growth curve after 12 straight years of steady growth. As PPSs for IRFs and LTCHs were delayed, PAC at these sites continued to increase between 1997−2000.
Deutsch et al252003All (N=39,562)AllPersonal AttitudinalDescriptive study of patients discharged from subacute facilities (1999) within the UDSmr database. Results: Overall, 65% were women, 91% were white, 62% were 75 years or older, 55% did not live alone before impairment onset, and 95% were admitted from an acute care facility. Impairment categories included stroke (9.1%), lower-limb fracture (12.5%), hip replacement (15.1%), cardiac (8.7%), pulmonary (4.2%), and debility (8.8%). Sixty-seven percent of the patients had Medicare.
Deutsch et al262006Medicare: stroke (N=58,724)IRF/SNFPersonalUDSmr data from 1996−1997 for stroke patients receiving PAC in IRF and SNF facilities. Experts recommend the selection of a rehabilitation setting should be primarily based on a patient’s motor and cognitive function, physical activity endurance, and social support. Results: IRF patients were younger, more likely to be male, more likely to be nonwhite, more likely to be married, and had higher admission FIM motor ratings.
Finch et al271997TBI (N=46)IRFAttitudinalPredictive study to assess influence of functional characteristics at admission on cognitive status at discharge. Results: Median age, 38.5y; median days since injury, 26. Ability to execute “higher cognitive operations” at admission appears to be the most important factor predicting cognitive abilities at discharge.
Foster and Tilse132003TBI (N=NA) AllUsed social pattern theory to describe variation in patterns of PAC referral for TBI. This study provides a conceptual model for PAC referral after TBI. The model includes person-level characteristics (clinical and nonclinical), interpretive activities of health professionals (influence of role and responsibilities of gate keeper), organizational context (characteristics of acute care facility where referral is determined), and health care context (external resources, availability and policy, relationship with these resources).
Gage281999Medicare (N≈100,000)All
Financial

Personal

Reviewed impact of BBA on PAC use by Medicare beneficiaries. Describes how rapid growth of PAC in the early 1990s was reduced after the BBA and sequential setting-specific PPSs. Depicts patient characteristics and PAC site use before the BBA (1995) and presents models to predict utilization based on patient profiles.
Gregory et al292006Stroke (N=12,208)IRF
Financial

Personal

Attitudinal

Used patient characteristics to predict referral to IRF vs “other” setting for stroke rehabilitation in Maryland, a DRG and PAC-PPS exempt state. Results: Factors associated with discharge to IRF included urban living, Medicare coverage, and acute LOS. Hemorrhagic stroke decreased the odds of discharge to IRF. Although race alone was not a significant predictor of IRF referral, 2 interaction terms were: black patients living in an urban setting and black patients who suffered a hemorrhagic stroke were more likely to be discharged to IRF.
Haas301988All (N=NA)AllAttitudinalAlthough many people are believed to be included in discussions regarding PAC admission, it is the acute care physician who often makes the decision. Thus, allocation decisions are often the result of physicians acting as gatekeepers in matching patients’ needs with scarce resources. Medical factors considered in these decisions include diagnosis, prognosis, comorbidities, physiologic instability and complications, along with communication, mobility, self-care, and cognitive abilities. Nonmedical factors include social (family support), vocational, financial (insurance coverage), institutional (specialty, availability), and personal circumstances.
Kane et al311996Medicare (N=2248)All
Structural

Personal

Predicted discharge status of acute care patients (1988−1989) based on patient- and hospital-level characteristics. Results: PAC use varies by region and diagnostic groups. In general, functional status appears more important than disease severity; living status (availability of in-home care) is also an important predictor of discharge destination. PAC facility ownership did not affect referrals. This study found variations in discharge placements using factors considered essential in the discharge planning process.
Kane et al322002Medicare (N≈90,000)All
Structural

Attitudinal

Description of geographic variation in PAC use across the United States (1996−1998). Results: Stroke, hip replacement and fracture, COPD, pneumonia, and CHF are among the most common diagnostic groups referred to PAC. PAC by census division ranged from 67% to 74% for patients within these 6 diagnostic groups. Acute LOS is positively associated with PAC. PAC among regions within diagnostic groups is particularly variable. This variability most likely results from practice styles, supply of services, and local regulatory practices.
Kramer et al331997Medicare: stroke, hip fracture (N=1003)IRF/SNFPersonalAssessed IRF vs SNF for stroke and hip fracture patients (1991−1994). Results: IRF patients were younger, more likely to have a caregiver, were rarely admitted from a nursing home, were more likely to have had a prior amputation (hip fracture only), be less physically and cognitively impaired, had fewer comorbidities, and exhibited better premorbid function (including social and recreational activity) compared with SNF patients. Patients with comparable characteristics, however, were admitted to both settings. The most important factors appeared to be availability of a caregiver, MMSE score, participation in social and recreational activities, the Barthel Index score, and premorbid walking score.
Lin et al342006Medicare (N range, 41,701–42,746)AllFinancialAssessed PAC use after BBA (1996−2000). Results: With successive implementation of PPSs, use decreased at target site and increased at other PAC sites (site substitution). Shift in use occurred in 2 phases that corresponded to payment reforms affecting HHAs and SNFs.
Liu et al351999Medicare (N=NA)AllStructuralU.S. DHHS review of policies that have influenced PAC spending and utilization. Number of SNF and HHA users nearly doubled from 1990−1996. SNF eligibility is limited to people with hospital stays of 3+ days within previous 30 days. Coverage is limited to 100 days and daily coinsurance payments begin on day 20. SNFs must have transfer agreements with a hospital. HHAs are for homebound people needing intermittent skilled nursing and therapy, which must be prescribed by a physician and recertified every 62 days. Eligibility for IRF is physician-determined and patients must need frequent physician involvement, 24-hour nursing, 3+ hours of therapy a day, and multidisciplinary care. IRFs are bound by the 75% rule and require a deductible and daily coinsurance payment per admission. Describes PAC for the 32 most common DRGs relative to age, sex, race, Medicaid eligibility, functional status, living arrangement, health factors, severity of illness, hospital size and teaching status, ownership and affiliation, hospital volume (poor, Medicare), PAC availability, and urban vs rural status.
Liu et al362000Medicare (N=NA)AllFinancialU.S. DHHS review of PAC issues after interview with discharge planners, quality of care experts, consumer groups, researchers, and health policy analysts (1998−1999). Results: Access issues included PPS-related disincentives for SNFs to admit patients with high nontherapy ancillary costs. HHA cuts were greater than intended and were below critical levels necessary to care for high-cost patients (eg, wound care, cardiac, diabetes).
Liu et al372001Medicare (N=203 facilities)LTCHStructuralDescription of LTCH characteristics (1997). LTCHs are defined as hospitals with an average LOS >25 days ; most specialize in respiratory, rehabilitation, or mental care. Seventy-one percent of patients have Medicare, most have several comorbidities, and are generally less stable than patients admitted to other PAC settings.
McCall et al382003Medicare (N range, ≈33,000–32,000)AllFinancialReview of BBA effect on Medicare payments for, and utilization of, PAC (1997−1999). Results: Utilization shifted from HHAs to other PAC settings immediately after passage of the BBA. Relative changes varied by DRG. Demonstrates how Medicare providers alter practice processes relative to financial incentives.
Murray et al392005Medicare (N range, 7006–61,569)SNF
Financial

Personal

Predictive study on likelihood of admission to services received in Ohio SNFs, pre- (1994−1996) and post- (2000−2001) PPS. Results: Select characteristics of post-PPS SNF patients include female sex (67%), white race (92%), married (32%), depression (22%), psychiatric medication (42%), diabetes (28%), CHF (25%), dementia (25%), stroke (20%), pulmonary disease (20%), hip fracture (9%), and functional motor score of 55±17. Many of these characteristics differed significantly from pre-PPS patient profiles. More apparent, however, is the differences in services provided for SNF patients after PPS implementation.
Paddock et al402007Medicare (N range, ≈364,000–446,000)IRFFinancialCompared MEDPAR and UDSmr FIM data (1999) with IRF-PAI data (2002) to assess the impact of PPS on IRF admission and treatment processes. Results: Overall case-mix severity did not change; however, predicted probability of death and resource use relative to case mix decreased after the PPS was implemented. Thus, while IRF admission practices were not affected by the PPS, it appears that patients within payment groups are receiving less care.
Paddock et al412006Medicare (N range, ≈49,000–56,000)IRFFinancialRAND report on changes in severity of patients in IRF after PPS (1999−2002). Results: No change in the percentage of costly patients admitted to IRFs after PPS was implemented. Severity of cases actually increased (slightly). LOS (intensity of care) continued to decrease. Overall, access to IRFs was not dramatically affected by PPS.
Rovinsky421999All (N=NA)AllFinancialBecause the BBA may limit access to PAC, IDSs may be forced to care for PAC patients in their high-cost, acute settings. This, in turn, may cause some IDSs to further limit patient access unless they develop a plan to better deliver cost-effective care across the continuum.
Schoenman432004Medicare (N=NR)AllFinancialNORC Walsh Center for Rural Health Analysis report on BBA effects on PAC transfers from rural acute care hospitals (1998−2001). Results: Neither rural nor urban hospitals appear to have altered PAC transfers based on the new policy despite declines in Medicare revenues. Rural hospitals, with less PAC available, may actually benefit via delivery of a greater proportion of care within the inpatient setting.
Shatto442002Medicare (N=2388)HHA/SNFPersonalMedicare Current Beneficiary Survey data (1999) of HHA vs SNF vs no PAC. Results: Among surveyed beneficiaries, 17% were admitted to an SNF, 28% to an HHA, and 55% received no PAC after acute hospital stays. Those reporting ”poor” health status included 16% no-PAC, 20% SNF, and 25% HHA. Percentages reporting ”chronic conditions” included 82% SNF, 93% no-PAC, and 96% HHA. Fifty-seven percent of no-PAC patients, 82% of HHA patients, and 85% of SNF patients reported difficulties with at least 1 ADL or IADL. About 30% of SNF patients were married vs about 40% of HHA and >50% of no-PAC patients. Thirty-two percent of no-PAC patients lived alone vs 35% of HHA patients and 47% of SNF patients.
Unsworth et al451995Stroke (N=NA)AllAttitudinalUsed social judgment theory to evaluate decision processes of 13 rehabilitation teams. Discharge status of 50 hypothetical stroke patients was determined using 15 pertinent attributes, 8 of which varied across patients (mobility, cognitive ability, patient preference, ADLs, IADLs, general health status, social situation, prior living arrangement), and 7 of which remained constant (age, severity, rehabilitation LOS, sex, communication skills, income, and family preference). Results: mobility, ADLs, and social support appeared to be the most important factors in post-rehabilitation discharge status. Although research teams demonstrated moderate to high internal consistency, inter-team agreement was considerably lower. All 13 teams only agreed on 6 of the 50 cases. Thus, not only is PAC choice not an exact science, but rehab professionals differ in their opinions about status after PAC.
Wrigley et al461994TBI (N=756)HHA/IRF
Structural

Personal

Compared characteristics of TBI patients discharged home vs formal rehabilitation after acute care (1989−1991). Results: discharge status categorized as home with informal care (44%), home with self-care (30%), IRF (16%), additional acute care (7%), or home with skilled care (3%). Factors related to more formal care included presence of rehabilitation specialist, injury-related complications, abnormal CT scans, longer LOS, single, older age, and an unintentional injury.

Abbreviations: ADL, activities of daily living; CABG, coronary artery bypass graft; CHF, congestive heart failure; CMG, case-mix group; CMS, Centers for Medicare & Medicaid Services; COPD, chronic obstructive pulmonary disease; CT, computed tomography; DHHS, Department of Health and Human Services; DRG, diagnosis-related groups; HMO, health maintenance organization; IADL, instrumental activities of daily living; IDS, integrated delivery system; IRF, inpatient rehabilitation facility; IRF-PAI, Inpatient Rehabilitation Facility–Patient Assessment Instrument; LOS, length of stay; MDS, Minimum Data Set; MEDPAR, Medicare Provider Analysis and Review; MMSE, Mini-Mental State Examination; NA, not available; NR, not reported; UDSmr, Uniform Data System for Medical Rehabilitation; TBI, traumatic brain injury.

Back to Article Outline

Results 

Our database searches yielded 823 articles. A scan of article titles and key words reduced the number of potential sources to 49. A subsequent screening of abstracts resulted in 39 relevant articles. After evaluating their full texts, we retained 35 articles for review and summary. Table 1 presents the key elements we extracted from each of the articles.

Based on data from those articles, we classified potential indicators of PAC access to rehabilitation services into 4 categories: financial, structural, personal, and attitudinal. We realize these categories are limited and somewhat arbitrary, but they represent an initial attempt to identify indicators that can eventually form the basis for a national system with which to monitor access to PAC, similar to the systems developed by the Centers for Medicare & Medicaid Services to monitor quality outcomes in nursing homes9 and home health agencies.10

Financial Barriers 

Financial barriers to accessing PAC services include insurance coverage and provider reimbursement rates associated with the different PPSs. Other financial barriers may include out-of-pocket expenses for patients, cost of treatment, and the substantial indirect costs to patients and society of residual disability.

Personal Barriers 

Personal barriers include factors associated with differential access, regardless of diagnosis, and disability or age (eg, education, ethnicity, income). Historically, these factors have influenced providers in their treatment of patients and in the ability of patients to adhere to a prescribed treatment regimen. A lack of awareness of services available in different PAC settings, and scant knowledge about how to access the health care system, are examples. These personal characteristics tend to be correlated with socioeconomic status, which makes it difficult to disentangle them from the effects of living in a particular neighborhood, the lack of insurance coverage, and reliance on public resources for medical care.

Structural Barriers 

Structural barriers include poorly developed referral systems, or inadequate discharge planning, lack of appropriate service providers or facilities in the local area, procedures or policies that limit access to some facilities (eg, 3-hour rule, 75% rule), or inflexible institutional practices—such as referrals restricted to a specific health care network—that limit access even when providers or facilities are locally available. These barriers detract from a patient’s ability to access appropriate PAC.

Attitudinal Barriers 

The attitudes and practice habits of providers can have an influence on patient access to PAC. This statement is applicable to primary, specialty, nursing home, and home health PAC. Attitudes of patients also affect access. For example, research11 suggests that there are significant differences in the use of long-term care facilities across racial and ethnic groups that appear to be based on personal attitudes and preferences.

We synthesized study counts (frequencies) from the 35 studies for patient groups, settings, and access categories to demonstrate with whom, where, and how access to PAC has been approached to date. Many studies included more than 1 category within each grouping; consequently, the summed totals may be greater than 35. Medicare beneficiaries (25 studies) was the most common population group studied, followed by all patients (4 studies) and older adults (1 study). Regarding diagnostic-specific subgroups, patients with stroke (6 studies) were studied most frequently, followed by orthopedic (3 studies), traumatic brain injury (3 studies), and cardiac (1 study) patients. Sixteen studies evaluated access issues related to all venues of PAC. Venue-specific analyses were primarily reported for inpatient rehabilitation facilities (13 studies), but there were also articles on access within SNFs (8 studies) and HHAs (2 studies). Only 1 study focused on access-related issues pertaining specifically to LTCHs. Last, the order (frequency) of access categories was: financial (16 studies), personal (11 studies), structural (9 studies), and attitudinal (8 studies).

Back to Article Outline

Discussion 

Access to health care services in the United States is generally discussed in the context of health insurance. The most recent information from the National Health Interview Survey (January–June 2006) is that 42.4 million Americans of all ages do not have health insurance.3 Working-age adults (range, 18−64y) were almost twice as likely (23.4%) to lack coverage as children under the age of 18 (12.9%). Among unemployed adults 18 to 64 years old, 57.5% were uninsured for at least the first 6 months of 2006. People 65 years and older are currently covered under the various Medicare programs.

The literature concerning the health-related, social, and political limitations of our health insurance system is both plentiful and diverse. The Institute of Medicine’s report, Insuring America’s Health: Principles and Recommendations,2 provides a comprehensive examination of the role of insurance coverage on health status and outcomes. Not surprisingly, financial factors were the most common focus of the studies included in our review; 16 of the 35 studies discussed the financial barriers that restrict access to PAC rehabilitation services.

The isolated description in our study of the financial, structural, personal, and attitudinal barriers reflects the need for a conceptual framework that can provide general direction and help define the parameters for a future access indicator monitoring system. These categories are obviously related to access in complex ways that are not currently understood. Existing conceptual models related to PAC can be examined to determine their fit with rehabilitation services.12 Figure 1 presents a basic conceptual model adapted from the work of Foster and Tilse13 that was originally developed for people with traumatic brain injury. The figure includes characteristics of the general environment and specific facility (eg, number of beds, presence of interdisciplinary teams) and relates these to individual factors such as medical diagnosis and treatment preferences. These environmental, facility, and individual factors interact to determine the patient’s PAC destination and the type of rehabilitation services received. There are complicated practical and political issues associated with the setting in which a service is provided and the level or type of services available. The development of a conceptual model for access to postacute rehabilitation will provide a basic starting point for examining these relationships in future research.

Access to appropriate and effective rehabilitation services is a particularly important issue that has recently received increased attention for 2 reasons: the introduction of mandated PPS for persons receiving PAC services reimbursed by Medicare,4 and the stricter implementation of Medicare’s 75% rule.14 These 2 factors have raised concerns and questions about access, effectiveness, and equity across PAC settings. In addressing these concerns, it is important to understand that access, effectiveness, and equity are related, but also have distinct characteristics. No matter how effective a particular program or intervention may be, a good health outcome cannot be guaranteed. The most important consideration is whether people have the opportunity to achieve a good outcome. When such opportunities are systematically denied or restricted, then there is a problem of access that must be addressed. More than 25% of the studies in our review contained information on the structural barriers to accessing PAC.

The first step in assessing, and ultimately improving, access is to establish a set of indicators that can be used to monitor access to such services. Such a system would be analogous to the well-established system of economic indicators that measure economic health and that includes variables such as unemployment rate, inflation, consumer confidence, and new housing starts. These indicators provide a picture of the state of the economy and how it might be changing. They also reveal potential problem areas and provide information about what interventions may be needed to bring the system back into balance. A similar system using indicators of PAC access would allow us to track whether access to effective rehabilitation services is changing over time. In addition, routinely reported access indicators could stimulate national debate about needed health policy actions, similar to the economic debates that occur when, for example, there is an increase in inflation or unemployment. We identified 4 broad categories that could be developed to provide an initial set of indicators with which to assess and monitor access to PAC services.

The second step in assessing and ultimately improving access is identifying databases and other information sources about access indicators. There are many databases that contain relevant access information that is collected on a regular basis and that could potentially contribute to the establishment of the access monitoring system.1, 9 The expansion of existing health report cards and quality indicator systems to include access indicators, as well as measures of (quality) outcomes, should be explored. We must identify areas of PAC where such services can be shown to maximize health, and we must determine if poor outcomes for some groups can be explained by variables associated with access.

Back to Article Outline

Conclusions 

Access to PAC services involves the interaction of factors that can be identified broadly as financial, structural, personal, and attitudinal. This review and the other articles in the series originating from the State-of-the-Science Symposium on Post-Acute Rehabilitation, provide a starting point for identifying and monitoring issues associated with access to postacute rehabilitation. The state of any science is dynamic and the picture described by the data is incomplete. The articles from the Symposium represent a palate that future researchers will use to produce a more complete picture that clearly identifies the factors that facilitate or impede access to PAC services.

Back to Article Outline

References 

  1. Millman M. Access to healthcare in America. Washington (DC): Natl Acad Pr; 1993;
  2. Committee on the Consequences of UninsuranceInstitute of Medicine. Insuring America’s health: principles and recommendations. Washington (DC): Natl Acad Pr; 2004;
  3. National Health Interview Survey. NHIS. Available at: http://www.cdc.gov/NCHS/about/major/NHIS/released200612.htm. Accessed June 25, 2007.
  4. Cotterill PG, Gage BJ. Overview: Medicare post-acute care since the Balanced Budget Act of 1997. Health Care Financ Rev. 2002;24:1–6
  5. Chan L. The state-of-the-science: challenges in designing postacute care payment policy. Arch Phys Med Rehabil. 2007;88:1522–1525
  6. Kaplan SJ. Growth and payment adequacy of postacute care rehabilitation. Arch Phys Med Rehabil. 2007;88:1494–1499
  7. Beeuwkes Buntin M. Access to postacute rehabilitation. Arch Phys Med Rehabil. 2007;88:1488–1493
  8. Cooper HM, Hedges LV. The handbook of research synthesis. New York: Russell Sage Foundation; 1994;
  9. Medicare. Nursing home compare. Available at: http://www.medicare.gov/NHCompare/home.asp. Accessed June 25, 2007.
  10. Medicare. Home health compare. Available at: http://www.medicare.gov/HHCompare/home.asp. Accessed June 25, 2007.
  11. Markides KS, Miranda MR. Minorities aging and health. Thousand Oaks: Sage; 1997;
  12. Khan AA. Access to health care. Eval Health Prof. 1994;17:60–76
  13. Foster M, Tilse C. Referral to rehabilitation following traumatic brain injury: a model for understanding inequities in access. Soc Sci Med. 2003;56:2201–2210
  14. Braddom RL. Medicare funding for inpatient rehabilitation: how did we get to this point and what do we do now?. Arch Phys Med Rehabil. 2005;86:1287–1292
  15. Angelelli J, Gifford D, Intrator O, Gozalo P, Laliberte L, Mor V. Access to postacute nursing home care before and after the BBA. Health Aff (Millwood). 2002;21:254–264
  16. Angelelli J, Grabowski DC, Mor V. Effect of educational level and minority status on nursing home choice after hospital discharge. Am J Public Health. 2006;96:1249–1253
  17. Bernstein AB, Hing E, Moss AJ, Allen KF, Siller AB, Tiggle RB. Health care in America: trends in utilization. Hyattsville: National Center for Health Statistics; 2003;DHHS Pub No. 2004-1031.
  18. Bronskill SE, Normand SL, McNeil BJ. Post-acute service use following acute myocardial infarction in the elderly. Health Care Financ Rev. 2002;24:77–93
  19. Buczko W. Effects of institutional services and characteristics on use of postacute care settings. J Health Hum Serv Adm. 2001;24:103–132
  20. Buntin MB, Garten AD, Paddock S, Saliba D, Totten M, Escarce JJ. How much is post-acute care affected by its availability?. Santa Monica: RAND Corp; 2004;Working Paper 10424.
  21. Buntin MB, Escarce JJ, Hoverman C, Paddock SM, Totten M, Wynn BO. Effects of payment changes on trends in access to post-acute care. Santa Monica: RAND Corp; 2005;
  22. Buntin MB, Carter GM, Hayden O, Hoverman C, Paddock SM, Wynn BO. Inpatient rehabilitation facility care use before and after implementation of the IRF prospective payment system. Santa Monica: RAND Corp; 2006;
  23. Carter GM, Hayden O, Paddock SM, Wynn BO. Case mix certification rule for inpatient rehabilitation facilities. [draft] Santa Monica: RAND Corp; 2003;DRA-2981-CMS
  24. Carter GM, Buchanan JL, Buntin MB. Executive summary of analyses for the initial implementation of the inpatient rehabilitation facility prospective payment system. Santa Monica: RAND Corp; 2005;
  25. Deutsch A, Fiedler RC, Iwanenko W, Granger CV, Russell CF. The Uniform Data System for Medical Rehabilitation report: patients discharged from subacute rehabilitation programs in 1999. Am J Phys Med Rehabil. 2003;703–711
  26. Deutsch A, Granger CV, Heinemann AW, et al. Poststroke rehabilitation: outcomes and reimbursement of inpatient rehabilitation facilities and subacute rehabilitation programs. Stroke. 2006;37:1477–1482
  27. Finch M, Sandel ME, Spettell C, Mack A, Spivack G. Admission examination factors predicting cognitive improvement during acute brain injury rehabilitation. Brain Inj. 1997;11:713–721
  28. Gage B. Impact of the BBA on post-acute utilization. Health Care Financ Rev. 1999;20:103–126
  29. Gregory PC, Han E, Morozova O, Kuhlemeier KV. Do racial disparities exist in access to inpatient stroke rehabilitation in the State of Maryland?. Am J Phys Med Rehabil. 2006;85:814–819
  30. Haas JF. Admission to rehabilitation centers: selection of patients. Arch Phys Med Rehabil. 1988;69:329–332
  31. Kane RL, Finch M, Blewett L, Chen Q, Burns R, Moskowitz M. Use of post-hospital care by Medicare patients. J Am Geriatr Soc. 1996;44:242–250
  32. Kane RL, Lin WC, Blewett LA. Geographic variation in the use of post-acute care. Health Serv Res. 2002;37:667–682
  33. Kramer AM, Steiner JF, Schlenker RE, et al. Outcomes and costs after hip fracture and stroke (A comparison of rehabilitation settings). JAMA. 1997;277:396–404
  34. Lin WC, Kane RL, Mehr DR, Madsen RW, Petroski GF. Changes in the use of postacute care during the initial Medicare payment reforms. Health Serv Res. 2006;41(4 Pt 1):1338–1356
  35. Liu K, Gage B, Harvell J, Stevensen D, Brennan N. Medicare’s post-acute care benefit: background, trends, and issues to be faced. Washington (DC): U.S. Department of Health and Human Services; 1999;
  36. Liu K, Harvell J, Gage B. Post-acute care issues for Medicare: interviews with provider and consumer groups, and researchers and policy analysts. Washington (DC): U.S. Department of Health and Human Services; 2000;
  37. Liu K, Baseggio C, Wissoker D, Maxwell S, Haley J, Long S. Long-term care hospitals under Medicare: facility-level characteristics. Health Care Financ Rev. 2001;23:1–18
  38. McCall N, Korb J, Petersons A, Moore S. Reforming Medicare payment: early effects of the 1997 Balanced Budget Act on postacute care. Milbank Q. 2003;81:277–303
  39. Murray PK, Love TE, Dawson NV, Thomas CL, Cebul RD. Rehabilitation services after the implementation of the nursing home prospective payment system: differences related to patient and nursing home characteristics. Med Care. 2005;43:1109–1115
  40. Paddock SM, Escarce JJ, Hayden O, Buntin MB. Did the Medicare inpatient rehabilitation facility prospective payment system result in changes in relative patient severity and relative resource use?. Med Care. 2007;45:123–130
  41. Paddock SM, Escarce JJ, Hayden O, Buntin MB. Changes in patient severity following implementation of the inpatient rehabilitation facility prospective payment system. Santa Monica: RAND Corp; 2006;
  42. Rovinsky M. How IDSS can turn BBA postacute care provisions to their advantage. Healthc Financ Manage. 1999;53:31–33
  43. Schoenman JA. Exploring the impact of Medicare’s post-acute care transfer payment policy on rural hospitals. NORC Walsh Center Rural Health Anal. 2004;5:1–6July
  44. Shatto A. Comparing Medicare beneficiaries, by type of post-acute care received: 1999. Health Care Financ Rev. 2002;24:137–142
  45. Unsworth CA, Thomas SA, Greenwood KM. Rehabilitation teams decisions on discharge housing for stroke patients. Arch Phys Med Rehabil. 1995;76:331–340
  46. Wrigley JM, Yoels WC, Webb CR, Fine PR. Social and physical factors in the referral of people with traumatic brain injuries to rehabilitation. Arch Phys Med Rehabil. 1994;75:149–155

 Supported by the National Institute on Aging (grant no. K02-AG019736) and by Advanced Rehabilitation Research Training grant, the National Institute on Disability and Rehabilitation Research (grant no. H133P040003).

 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 author(s) or upon any organization with which the author(s) is/are associated.

PII: S0003-9993(07)00442-X

doi:10.1016/j.apmr.2007.06.761

Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 11 , Pages 1513-1521, November 2007

Article Tools

* Image Usage & Resolution