This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation

The Impact of the 1997 Balanced Budget Amendment's Prospective Payment System on Patient Case Mix and Rehabilitation Utilization in Skilled Nursing

^a California Center for Long Term Care Integration, Andrus Gerontology Center, University of Southern California, Los Angeles
^b Leonard Davis School of Gerontology, University of Southern California, Los Angeles
^c School of Policy, Planning, and Development, University of Southern California, Los Angeles

Correspondence: Judy Y. Yip, PhD, Research Associate, California Center for Long Term Care Integration, Andrus Gerontology Center, University of Southern California, Los Angeles, CA 90089-0191. E-mail: jyip{at}usc.edu.

Decision Editor: Laurence G. Branch, PhD

	Abstract

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 
Purpose: This study examines the impact of the post-acute prospective payment system (PPS) on Medicare-funded rehabilitation services in skilled nursing facilities (SNFs) and whether such impact varies under different payment mechanisms. Design and Methods: We interviewed 214 Medicare beneficiaries admitted to three SNFs in southern California for rehabilitation. We compared patients' admission characteristics and therapy utilization among those receiving post-acute rehabilitation before and after the implementation of PPS. Results: Patients admitted after PPS implementation were more likely to have orthopedic problems or stroke and poorer self-reported physical health. They had significantly shorter lengths of stay in rehabilitation and received significantly less therapy, although those in managed care had less reduction in treatment after SNF-PPS implementation than those in fee-for-service. Implications: After SNF-PPS implementation, rehabilitation treatment levels in the study sites were reduced. Whereas changes in Medicare managed care were comparatively modest, we observed significant changes in intensity and duration of physical and occupational therapies in Medicare fee-for-service.

Key Words: Skilled nursing facility • Prospective Payment System • Patient case mix • Fee-for-service • Managed care

Medicare reimbursement has been shaped by two major departures from traditional fee-for-service (FFS) health care implemented in the early 1980s: prospective payment and Medicare risk contracts. Both of these departures were efforts to change incentives and reduce costs by shifting financial risk from Medicare as the insurer to providers including hospitals, physicians, skilled nursing facilities, and home health agencies.

	The Evolution of Prospective Payment System in Medicare

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 
Medicare's shift to a prospective payment system (PPS) began in 1983, when diagnostic-related groups (DRGs) were introduced to cover inpatient hospitalizations. Literature examining the impact of PPS on hospitals found that Medicare patients had shorter lengths of stay after DRGs were introduced (Carey 2000; Coulam and Gaumer 1991), raising the now familiar concern that patients were discharged sicker and quicker. Despite the implementation in 1992 of a resource-based relative value system (a PPS for physicians), Medicare payments for hospital outpatient services and ambulatory surgery centers rapidly increased as did growth in overall Medicare expenditures (Coddington, Fischer, and Moore 2000). Although Medicare's move to PPS forced hospitals and physicians to become more accountable for how they managed resources and treated patients, utilization and costs rose dramatically in the post-acute care health sector during the 1990s (Kane et al. 1998). For example, in skilled nursing facilities (SNFs), therapy charges increased from 15% of all Part A charges in 1990 to 29% in 1996 (Gage 1999). A report by the General Accounting Office (GAO; U.S. General Accounting Office 2000) noted that some of this growth was related to increases in payments for ancillary services, an area of expenditure where there had been little program oversight.

To contain the rapid growth of post-acute care utilization and expenditures, the Balanced Budget Act of 1997 mandated that Medicare payments for SNFs and home health care be converted from a cost-plus reimbursement to a PPS-based mechanism. For SNFs, resource utilization groups (RUGs) were used to determine prospective payment amounts. In contrast to the cost-plus reimbursement system, which favors treatment, incentives under PPS discourage SNFs from admitting patients whose costs could exceed fixed reimbursements (Arling, Williams, and Kopp 2000; Gage 1999). Not surprisingly, the introduction of PPS in SNF (SNF-PPS) led to questions about how it would impact patient access and utilization (National Health Policy Forum 1999). For example, there has been concern that the various categories of RUGs may not be refined enough to accurately reflect patients' conditions (National Health Policy Forum 1999). In addition, the Office of Inspector General 1999a, Office of Inspector General 1999b and GAO (2000) reported that SNFs were likely to respond to RUGs with several strategies including admitting patients needing fewer resources and avoiding patients who required high intensity care. To the extent that this is the case, patients admitted to SNFs for rehabilitation (a major component in ancillary service) after the implementation of PPS might differ in such characteristics as level of acuity, functional ability, and primary diagnosis and comorbidities.

Thus far, studies examining the impacts of SNF-PPS have focused on patient access and nursing homes' admission practices rather than on the use of rehabilitation services. After the implementation of SNF-PPS, several evaluation studies conducted by the OIG (1999a, 1999b) found that hospital discharge planners and nursing home administrators reported that they experienced no widespread access problems for patients, although hospital discharge planners noted some changes in admission practices in SNFs. For example, SNFs tended to request more detailed clinical information on patients, sometimes even visiting the hospital to conduct patient assessment before making admission decisions. Similarly, findings of a nationwide survey conducted by GAO (2000) suggested that after SNF-PPS, nursing homes were more inclined to admit Medicare patients needing rehabilitation. At the same time, discharge planners noted that SNFs appeared to be more cautious about admitting those who might need high-cost health care services.

	Medicare Managed Care

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 
The second major policy change, the Tax Equity and Fiscal Responsibility Act legislated in 1982, was designed to encourage Medicare's beneficiaries to enroll in Medicare risk plans. Unlike PPS, which is mandatory for Medicare beneficiaries, enrollment in managed care is voluntary. Approximately 14% of Medicare beneficiaries nationwide are enrolled in Medicare-risk managed care, through which providers receive a capitated reimbursement on a per member per month basis (Barry 2001). In Los Angeles, where the present study was conducted, Medicare managed care penetration is approximately 43% and much of the financial risk has been shifted from the health plans to medical groups (Centers for Medicare and Medicaid Services 2001). Although typically medical groups and often physician groups are capitated under Medicare managed care, most SNFs negotiate contracts with managed care plans to pay the facility on a per diem basis for each day that the patient receives in SNF rehabilitation. The entity that bears the risk, most often medical groups in California, manages each patient's care through utilization review to minimize the time spent in rehabilitation (custodial care in SNFs is not covered by Medicare). Thus, under both managed care and PPS, the incentives are to minimize treatments as cost-effective measures.

	Research Questions

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 
This article analyzes patient characteristics and rehabilitation utilization patterns among Medicare beneficiaries who received therapy in SNFs before and after the implementation of SNF-PPS. In addition, it compares similarities and differences between patients in FFS and managed care pre- and post-PPS. Recent studies have found that rehabilitation utilization is associated with patients' sociodemographic characteristics, clinical risk factors, provider types, and payment mechanisms (Angelelli, Wilber, and Myrtle 2000; Evenson, Rosamond, and Luepker 1998; Harada, Chun, Chiu, and Pakalniskis 2000; Kramer et al. 2000). Building on this research, we use a model suggested by Kane 1997 that includes patients' (1) demographic factors (age and sex), (2) risk factors (baseline functioning, primary diagnosis, and co-morbidities), (3) setting, and (4) Medicare payment mechanism (FFS or managed care) to address three research questions:

1. Are there differences in the characteristics of patients who received rehabilitation services in SNFs before and after the implementation of the SNF-PPS? On the basis of what has been learned about how DRGs affect hospital length of stay and other service utilization measures, we hypothesized that differences would be observed in FFS Medicare patient case mix pre- and post-PPS. As suggested by the OIG reports, admission decisions by SNFs for rehabilitation were expected to be more cautious, focusing on patients whose diagnoses have an established protocol for treatment or patients with higher level of functioning.
   
2. Are there differences in rehabilitation utilization among patients before and after the implementation of the SNF-PPS? The change from a cost-based to a fixed, predetermined, per diem-based system is assumed to trigger changes in treatment practice. More specifically, PPS creates an incentive to tighten the monitoring of patient care, to reduce unnecessary care, and to control resource allocation. Hence, we hypothesized that after controlling for patient mix, post-PPS patients would receive less duration (fewer days) and lower intensity (fewer minutes per day) of rehabilitation services than similar pre-PPS patients.
   
3. Do differences in case mix and utilization pattern, if any, vary among patients under FFS and managed care? The fixed payments of PPS encourage providers to more carefully manage patients' care. The incentives created are similar to those under managed care risk contracts (i.e., paying providers on a per diem rate based on patients' level of care). In a study conducted in California before PPS was implemented, Angelelli and colleagues 2000 found that patients who opted for Medicare managed care received less treatment than those under FFS. Inasmuch as SNFs that are experienced with managed care reimbursement may already have had various cost-conscious practices in place prior to SNF-PPS, we hypothesized that there would be more pre- and post-PPS variation in the FFS patient group than in the managed care patient group. Further, it is hypothesized that these variations will be reflected in patient case mix and rehabilitation treatment differences.
   

	Methods

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 
Sample
Data used for this study were from a longitudinal study comparing rehabilitative outcomes among patients under different payment mechanisms (Wilber 2000). Information was collected through face-to-face interviews and the medical records of patients who entered one of the three SNF study sites in southern California for Medicare-funded rehabilitation between March 1998 and September 1999. The study sites were owned by a midsized (12-facility), for-profit, post acute care provider operating in southern California. Selection criteria for the study sites included: sufficient Medicare funded post-acute care beds to ensure a sample of over 200 study participants, a history of post-acute care contracts with FFS and managed care providers, evidence of high quality care as evaluated by the Department of Health Services Licensing and Certification Division, and stable ownership. This last criterion, stability of ownership, was important because SNFs have been buffeted by mergers, acquisitions, and divestitures that have the potential to dramatically alter operations and care within facilities.

The SNF-PPS was implemented January 1, 1999, half way through the original study. Participants screened into the study were English-speaking Medicare beneficiaries at least 65 years of age who lived in the community before they were hospitalized, and who were newly admitted to post-acute care treatment for physical and/or occupational rehabilitation. To become enrolled, participants also had to pass a brief cognitive screen (Morishita et al. 1995). We interviewed eligible patients within one week after admission into therapy. Of 322 eligible patients, 217 (67%) signed a consent form agreeing to participate in the study, and 214 completed baseline interviews. We asked eligible patients who declined to participate to consent to provide demographic information and to permit a member of the research team to access their medical records for clinical data. Bivariate comparisons of patients who participated in the study and those who refused to participate showed no statistically significant differences in age, sex, primary diagnosis, and payment mechanism. However, one site had a significantly smaller proportion of patients who refused to be in the study (16.3%) than the other two sites (47.2% and 46.7%; ²(N = 105)= 28.335, p < .001).

Measurement
Following Kane 1997 model, we grouped variables into patients' demographic characteristics, risk factors, setting, and payment mechanism. Patient demographic information included age and sex. We measured age continuously ranging from 65 to 96 years. Risk factors included clinical information collected from patients' medical records. These included primary diagnosis (orthopedic, stroke, and other diagnoses such as shingles, pneumonia, cellulitus, chronic renal failure, gastric ulcer), number of comorbidities, and hospital length of stay (measured in days).

We used the Short-Form 36 (SF-36) of the Medical Outcomes Study to measure health-related quality of life and functional status. We selected the SF-36 because of its well-tested psychometric properties, excellent track record in research, and relatively low subject burden. The SF-36 has demonstrated high internal consistency and good construct, convergent, discriminant, and postdictive validity (Andresen, Rothenberg, Panzer, Katz, and McDermott 1998; Kazis, Skinner, Miller, Clark, and Lee 1996; McHorney 1996; Ware, Snow, Kosinski, and Gandek 1993). Developed to measure generic health status for the general population, the SF-36 has increasingly been used in research focusing on older adults (McHorney 1996). Its 36 self-report items capture eight health-related subscales. Each of the eight scales has been converted to a range of 0 to 100, with higher scores indicating better health-related quality of life and functional status. From the eight scales, two summary scales (physical health and mental health) were developed using confirmatory factor analysis (Ware et al. 1993). We used summary scales in the regression models in the present study. Internal consistency values for the present study are similar to those reported elsewhere (Ware et al. 1993) and are as follows: physical functioning (PF, = 0.91), role physical (RP, = 0.65), bodily pain (BP, = 0.87), general health (GH, = 0.66), social functioning (SF, = 0.82), role emotional (RE, = 0.89), mental health (MH, = 0.80), and vitality (VT, = 0.78).

Treatment variables included rehabilitation utilization measured by rehabilitation length of stay, therapy days, therapy hours, and therapy intensity. We extracted rehabilitation utilization information from therapy logs in patients' medical charts. We logged therapy utilization in 15-min increments in each therapy day by each therapist who provided rehabilitation services. These increments were then transformed to number of hours by the authors. Rehabilitation length of stay is the number of days a patient received physical and/or occupational therapy. Therapy days are the number of days, coded separately for physical and occupational therapy. Rehabilitation length of stay is always equal to the range of authorized physical and/or occupational therapy days. Therapy hours are the number of hours (converted from minutes) of physical therapy and the number of hours of occupational therapy received. We measured therapy intensity in minutes per day, calculated separately for physical and occupational therapy. The setting variable describes the site where patients received their rehabilitation treatment. We measured site by three dummy variables (Site 1, Site 2, and Site 3). Payment mechanism is a dummy variable, which indicates whether the payment was FFS or Medicare risk (FFS = 1, managed care = 0).

Analysis
We classified patients in one of two groups, based on the date of admission to SNFs: pre-PPS (admitted before January 1, 1999) and post-PPS (admitted on or after January 1, 1999). We used chi-square tests (for discrete variables) and t tests (for continuous variables) to compare patient characteristics and rehabilitation utilization pre- and post-PPS. Significant differences in rehabilitation utilization at the bivariate level were then analyzed using multiple regression. We performed analyses first for the entire sample and then separately for FFS and managed care patients.

	Results

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 
Differences in Patient Characteristics
Table 1 presents bivariate comparisons of characteristics of subjects receiving rehabilitation pre- and post-PPS. We found significant differences in primary diagnosis, site, and functional status. Post-PPS was significantly associated with diagnosis reflected in a higher proportion of orthopedic patients (40.9% vs. 28.3%). Significant site differences also were found pre- and post-PPS, with a higher proportion of pre-PPS participants in Site 1 and a higher proportion of post-PPS participants in Site 3. Moreover, post-PPS participants also had a significantly lower physical functioning score (10.52 vs. 20.10) and physical summary scores (24.11 vs. 26.52) and higher role emotional scores (68.44 vs. 55.83).

Differences in Rehabilitation Utilization
We observed significant differences between pre- and post-PPS in rehabilitation treatments (Table 2 ). On average, post-PPS patients had shorter stays for rehabilitation, especially for physical therapy. Post-PPS patients received, on average, five less physical therapy days (13.09 days) than pre-PPS participants (18.53 days). Further, post-PPS patients received only 46.6% of physical therapy and 54.4% of occupational therapy minutes provided to pre-PPS participants. Also, the number of minutes of physical therapy dropped by 19.08 min per day.

Comparing pre-PPS participants in FFS and managed care, we observed significant differences in the number of days of physical and occupational therapy, total amount of rehabilitation received (minutes in both types of therapy), and rehabilitation intensity. These payment differences disappeared when we compared patients admitted post-PPS. In general, SNF-PPS is associated with reduction of rehabilitation utilization, but the impact is greater for FFS than managed care. Moreover, managed care patients, on average, did not experience significant reduction in rehabilitation utilization (occupational therapy minutes and physical therapy minutes per day) before and after SNF-PPS implementation, whereas patients in FFS showed significant reduction in utilization after SNF-PPS.

Results of ordinary least squares regression, shown in Table 3 , indicate that after controlling for other characteristics (i.e., age, gender, diagnosis, functioning), PPS status persists in reducing rehabilitation utilization. On average, post-PPS subjects received fewer rehabilitation days and had lower rehabilitation treatment intensity. We observed also a significant relationship with payment mechanism and rehabilitation treatments. After controlling for SNF-PPS, fee-for-service payment was associated with more days of physical therapy (duration), greater total number of minutes of physical and occupational therapy received, and more intense physical therapy received per day.

	Discussion

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

The first research question asked whether significant differences existed among patients admitted before and after SNF-PPS implementation. Changes in diagnoses and physical functioning offer partial support for our hypothesis that the implementation of SNF-PPS results in a change of case mix in SNFs. Differences are found in primary diagnosis: there are more patients admitted for stroke or orthopedic conditions and fewer patients admitted with diagnoses of other post-PPS. The significant decline in patient admission with other diagnoses may be related to the fact that care protocols for these patients tend to be less well-defined, making assignment to appropriate case-mix groups difficult and uncertain. This result may reflect the general findings reported by the OIG (1999a) that hospital discharge planners experienced difficulties in placing clinically complex patients in SNFs. We also found significant differences in patients' physical functioning before and after the SNF-PPS implementation. Compared to pre-PPS, post-PPS patients have lower (worse) scores on physical functioning and the SF-36 physical summary score, and higher (better) scores on role emotional.

Multivariate analyses show that changes associated with SNF-PPS persist even after controlling for other variables in the model (demographic, risk, setting, and payment mechanism). In terms of differences by payment mechanism, although the decrease in physical functioning scores appears similar for both groups (approximately 10 points) the difference is statistically significant for managed care patients (p = .004) but not for FFS patients (p = .057). This appears to be a consequence of inadequate statistical power because of the smaller sample size of the FFS group.

The second research question asked if there were differences in rehabilitation utilization among patients before and after the implementation of the SNF-PPS. Results show that PPS reduced both the duration and intensity of rehabilitation treatments. On average, patients admitted post-PPS receive close to 5 days less physical therapy (30% decrease) and 22 min less physical therapy per day (32.4% decrease) than patients admitted pre-PPS. Similar results have been reported by Warren, Wirtalla, and Leibensberger 2001 whose study indicated a 35–37% decrease in utilization by stroke and orthopedic patients after the implementation of the SNF-PPS.

Our last research question focused on whether the effect of SNF-PPS varied with payment mechanism. We hypothesized that there would be more pre- and post-PPS variation (reflected in changes in patient case mix and rehabilitation treatment) in the FFS patient group than in the managed care patient group. Findings shown in Table 2 and Table 4 support this hypothesis. Significant post-PPS change among managed care patients is found only in rehabilitation length of stay, whereas among FFS patients, significant post-PPS changes are found in the duration and intensity of both physical and occupational therapy. Considering the amount of physical therapy received per day, PPS reduces FFS patients an average of 44 min compared to 9 min for managed care patients.

This study provides important findings on the effects of the SNF-PPS on Medicare-funded rehabilitation, particularly in the areas of admission characteristics and utilization of therapy. Not surprisingly, it supports the notion that form follows funding in that as FFS moves to a fixed reimbursement system, it begins to look more like per diem-funded managed care. Nevertheless several limitations must be noted. We drew the sample from three skilled nursing facilities purposely selected because of high quality of care stability and experience with different types of payment mechanisms. It is possible that other facilities may behave differently as they transition to a PPS. Moreover, the study took place in southern California, which has been characterized as a mature managed care market with high managed care penetration among Medicare beneficiaries. Other state and substate regions have different levels of experience with managed care and with SNF cost-containment strategies. Health care delivery systems tend to vary extensively by region, making it problematic to generalize from one region to another. The sample includes all rehabilitation patients within the three facilities who met the inclusion criteria and agreed to participate during an 18-month time period. Nevertheless, the sample size lacks adequate power to perform comparisons of payment mechanism and SNF-PPS implementation.

Variations in the number of patients admitted to one of the three SNFs before and after the implementation of SNF-PPS represent lower enrollment in that site, post PPS. According to management at the SNF, this reduction is a direct result of PPS, based on difficulties the facility experienced in renegotiating several of its provider contracts after SNF-PPS was implemented. To examine the effects on the study results, we excluded participants from this site and reanalyzed the data. There were no differences in the findings.

We did not explore how post-PPS reductions in treatment affect patient outcomes. This is an important area that should be examined in future research. In addition, because our data collection ended in September 1999, we were not able to examine the long-term and possibly less volatile changes resulting from the move to SNF-PPS as facilities became more familiar with the requirements.

A concern related to SNF-PPS implementation and the timeframe of the present study is that the reduction in rehabilitation treatment at the beginning of the implementation period could be a result of phase-in issues on the part of SNF providers unfamiliar with the SNF-PPS system. To examine this further, we conducted a post hoc analysis by dividing the sample into four quarters, two before PPS (March 1–August 31, 1998, and September 1–December 31, 1998) and two post-PPS implementation (January 1–March 31, 1999, and April 1–September 30, 1999). Rehabilitation utilization was compared among patients in each quarter. We found a drop in all significant variables (i.e., physical therapy minutes, occupational therapy minutes, rehabilitation length of stay, physical therapy minutes per day) from the first to the second quarter, a further decline of physical and occupational therapy minutes between the second and the third quarters, and a plateau in the fourth quarter (results not shown).

Statistically significant differences were found between the first and the third quarters, as well as between the first and the fourth quarters. Future studies that examine patients' rehabilitation utilization at various time points from the SNF-PPS implementation would provide a richer understanding of how this policy affects provider behavior and patient outcomes.

	Acknowledgments

 
The research reported in this article was supported by a grant from the John Randolph Haynes and Dora Haynes Foundation. An earlier version of this article was presented at the 53rd Annual Scientific Meeting of The Gerontological Society of America, Washington DC, November, 2000. We thank George Shannon for his helpful comments. We thank also the two anonymous reviewers for valuable comments.

Received for publication September 24, 2001. Accepted for publication April 15, 2002.

	References

TOP Abstract The Evolution of Prospective... Medicare Managed Care Research Questions Methods Results Discussion References

 

• Andresen E. M., Rothenberg B. M., Panzer R., Katz P., McDermott M. P., 1998. Selecting a generic measure of health-related quality of life for use among older adults: A comparison of candidate instruments. Evaluation & the Health Professions 21:244-264.
• Angelelli J. J., Wilber K. H., Myrtle R., 2000. A comparison of skilled nursing facility rehabilitation treatment and outcomes under Medicare managed care and Medicare fee-for-service reimbursement. The Gerontologist 40:646-653. [Abstract/Free Full Text]
• Arling G., Williams A. R., Kopp D., 2000. Therapy use and discharge outcomes for elderly nursing home residents. The Gerontologist 40:587-595. [Abstract/Free Full Text]
• Balanced Budget Act of (1997). Pub. L. No. 105-33 (August 5, 1997).
• Barry, P. (2001, October). Medicare: New road? AARP Bulletin.
• Carey K., 2000. Hospital cost containment and length of stay: An econometric analysis. Southern Economic Journal 67:363-380.
• Centers for Medicare and Medicaid Services2001Retrieved March 2001 from http://www.hcfa.gov/medicare/mpsct1.htm. Medicare managed care market penetration for all medicare plan contractors—Quarterly state/county data files
• Coddington D. C., Fischer E. A., Moore K. D., 2000. Characteristics of successful health care systems. Health Form Journal 43:40-45.
• Coulam R. F., Gaumer G. L., 1991. Medicare's prospective payment system: A critical appraisal. Health Care Financing Review 13: (Annual Suppl.) 45-77.
• Evenson K. R., Rosamond W. D., Luepker R. V., 1998. Predictors of outpatient cardiac rehabilitation utilization: The Minnesota heart survey registry. Journal of Cardiopulmonary Rehabilitation 18:192-198. [Medline]
• Gage B., 1999. Impact of the BBA on post–acute utilization. Health Care Financing Review 20: (4) 103-125. [Medline]
• Harada N. D., Chun A., Chiu V., Pakalniskis A., 2000. Patterns of rehabilitation utilization after hip fracture in acute hospitals and skilled nursing facilities. Medical Care 38:1119-1130. [Medline]
• Kane R. L., 1997. Improving outcomes in rehabilitation: A call to arms (and legs). Medical Care 35: (Suppl. 6) JS21-JS27. [Medline]
• Kane R. L., Chen Q., Finch M., Blewett L., Burns R., Moskowitz M., 1998. Functional outcomes of posthospital care for stroke and hip fracture patients under Medicare. Journal of the American Geriatrics Society 46:1525-1533. [Medline]
• Kazis L. E., Skinner K., Miller D., Clark J., Lee A., 1996. Quality of life for the chronically ill elderly. Romeis J. C., Coe R. M., Morley J. E., , ed.Applying health services research to long-term care 43-75. Springer, New York.
• Kramer A. M., Kowalsky J. C., Lin M., Girsby J., Hughes R., Steiner J. F., 2000. Outcome and utilization differences for older persons with stroke in HMO and fee-for-service systems. Journal of the American Geriatrics Society 48:726-734. [Medline]
• McHorney C. A., 1996. Measuring and monitoring general health status in elderly persons: Practical and methodological issues in using the SF-36 Health Survey. The Gerontologist 36:571-583. [Abstract]
• Morishita L., Boult C., Ebbitt B., Fambel M., Fallstorm K., Booden T., 1995. Concurrent validity of administering the Geriatric Depression Scale and the physical functioning dimension of the sickness impact profile by telephone. Journal of the American Geriatrics Society 43:680-683. [Medline]
• National Health Policy Forum1999. Implementing the BBA: The challenge of moving Medicare post-acute services to PPS (Issue Brief No. 743) George Washington University, Washington, DC.
• Office of Inspector General1999a. Early effects of the prospective payment system on access to skilled nursing facilities (OEI-02-99-00400) U.S. Department of Health and Human Services, Washington, DC.
• Office of Inspector General1999b. Early effects of the prospective payment system on access to skilled nursing facilities: Nursing home administrators' perspective (OEI-02-99-00401) U.S. Department of Health and Human Services, Washington, DC.
• Tax Equity and Fiscal Responsibility Act of (1982). Pub. L. No. 97-248 (September 3, 1982).
• U.S. General Accounting Office2000. Skilled nursing facilities: Medicare payment changes require provider adjustments but maintain access (GAO/HEHS-00–23) Government Printing Office, Washington, DC.
• Ware J. E., Snow K. K., Kosinski M., Gandek B., 1993. SF-36 health survey manual and interpretation guide The Health Institute, New England Medical Center, Boston, MA.
• Warren R. L., Wirtalla C., Leibensberger A., 2001. Preliminary observations on reduced utilization in skilled nursing facility rehabilitation. American Journal of Physical Medicine and Rehabilitation 80:626-633. [Medline]
• Wilber K. H., 2000. The quality of post-acute health care in Los Angeles: Does managed care make a difference? (Final Report to the John Randolph Haynes and Dora Haynes Foundation) Andrus Gerontology Center, University of Southern California, Los Angeles.

This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation