Archives of Physical Medicine and Rehabilitation
Volume 90, Issue 8 , Pages 1260-1268, August 2009

Looking Upstream: Factors Shaping the Demand for Postacute Joint Replacement Rehabilitation

  • Wenqiang Tian, MD, PhD

      Affiliations

    • National Rehabilitation Hospital, Center for Post-acute Studies, Washington, DC
    • ,
  • Gerben DeJong, PhD

      Affiliations

    • National Rehabilitation Hospital, Center for Post-acute Studies, Washington, DC
    • Corresponding Author InformationReprint requests to Gerben DeJong, PhD, Center for Post-acute Studies, National Rehabilitation Hospital, 102 Irving St, Washington, DC 20010
    • ,
  • Michael Brown, BA

      Affiliations

    • National Rehabilitation Hospital, Center for Post-acute Studies, Washington, DC
    • ,
  • Ching-Hui Hsieh, PhD

      Affiliations

    • National Rehabilitation Hospital, Center for Post-acute Studies, Washington, DC
    • ,
  • Zvedomir P. Zamfirov, MD

      Affiliations

    • National Rehabilitation Hospital, Center for Post-acute Studies, Washington, DC
    • ,
  • Susan D. Horn, PhD

      Affiliations

    • Institute for Clinical Outcomes Research, Salt Lake City, UT

Article Outline

Abstract 

Tian W, DeJong G, Brown M, Hsieh C-H, Zamfirov ZP, Horn SD. Looking upstream: factors shaping the demand for postacute joint replacement rehabilitation.

Since 1993, the numbers of hip and knee replacements in the United States have increased 2-fold to 3-fold while lengths of stay in acute care have decreased by about half, leading to a significant growth in the use of postacute rehabilitative care for patients with a joint replacement. To document these trends, this article uses secondary analysis of acute hospital discharge survey data and evaluates projections to 2030. This article uses a market approach to identify 3 sets of factors that influence the use of joint replacements: (1) increasing patient demand, (2) increasing supply of practitioners, and (3) the role of fiscal intermediaries. The article reviews underlying epidemiologic trends, growing numbers of orthopedic surgeons performing the procedure, technologic innovations, changing indications for the procedure, changing payer mix, and the effects of payer attempts to contain joint replacement costs. An unintended effect of Medicare payment policy has been to shift costs from acute care to downstream postacute care. Medicare and private health plan reimbursement policies need to take into account this broader perspective and not examine joint replacement care and payment in isolated care settings. Future research and health policy needs to consider the interdependent features of the health care system by linking changes in postacute care with upstream changes both in society at large and in the organization, delivery, and financing of acute care associated with joint replacement.

Key Words: Arthroplasty, replacement, hip, Arthroplasty, replacement, knee, Rehabilitation

List of Abbreviations: AAOS, American Academy of Orthopaedic Surgeons, AHRQ, Agency for Healthcare Research and Quality, ICD-9-CM, International Classification of Diseases–9th Revision–Clinical Modifications, IRF, inpatient rehabilitation facility, JOINTS, Joint Replacement Outcomes in Inpatient Rehabilitation Facilities and Nursing Treatment Sites, LOS, length of stay, NIS, National Inpatient Sample, PAC, postacute care, SNF, skilled nursing facility

 

IN RECENT YEARS, there has been considerable controversy about where patients with joint replacements should receive their postsurgical rehabilitative care—at home, in an outpatient center, in an SNF, in an IRF, or none of these. The purpose of this special communication is to characterize the changes in the demand for postacute rehabilitation precipitated by upstream changes in the numbers of joint replacements performed in acute care hospitals. These changes are a function of demographic and epidemiologic changes in the American population and changes in the indications for joint replacement, durability of joint replacements, orthopedic practice patterns, and health plan reimbursement policies. Collectively, these upstream changes create a dynamic environment that is not fully understood when considering the future of postacute rehabilitation services for patients who have had a recent joint replacement.

Back to Article Outline

Rapid Increase in Joint Replacements 

Using HCUPnet,1 an online data tool, we reviewed 13 years of data (1993–2005) furnished by the AHRQ to evaluate major trends in the numbers of joint replacements performed in the United States. The AHRQ database, known as NIS, is based on a 20% sample of all hospital discharges in the nation. From 1993 to 2005, there has been nearly a 2.5-fold increase in the number of total knee replacements (identified as ICD-9-CM procedure code 81.54)—from 200,216 to 497,419—and a 1.75-fold increase in the number of total hip replacements (identified as ICD-9-CM procedure code 81.51)—from 135,992 to 237,645 during the same period (fig 1). These data do not include joint replacements that may have been performed in ambulatory surgery centers in recent years. HCUPnet estimates are in keeping with earlier reports, some of which are not based on national probability samples.2, 3 Some studies using different national data sources also showed similar trends.4, 5, 6

As the stock of primary joint replacements increases, we can expect, over time, a corresponding increase in the number of revisions to an original replacement, especially when people are obtaining joint replacements earlier in life and living longer. Improvements in surgical technology, however, may limit the need for revisions and thus moderate future increases in the numbers of revisions. According to a HCUPnet estimate, the number of revised hip replacements increased from approximately 27,200 in 1993 to 38,900 in 2001, and then declined to 37,200 in 2005. Over the same period, the number of revised knee replacements increased from 18,400 to 39,800 between 1993 and 2005 (fig 2). This estimate varies somewhat with previous reports,5, 6 which estimated a higher number of revisions.

Despite these increases in revisions, the revision rate—the number of revisions relative to primary replacements—generally decreased during the 13 years, when the revision rate declined from 20% to 15.7% for total hip replacement, and from 9.2% to 8% for total knee replacement. One explanation for this finding is that the number of primary replacements is increasing more rapidly, making the denominator in the rate calculation larger than the numerator.

Back to Article Outline

Variation in Use of Joint Replacement 

To focus only on the rapid growth of arthroplastic surgeries overlooks important differences across populations and geographic areas. There remain significant differences in sex, race, and geographic location. Some of these disparities, especially those that relate to race and geography, suggest there may be additional increases in demand should these disparities diminish in the future.

Sex 

Women are much more likely to undergo joint replacement surgery than men. Women are more than twice as likely as men to obtain a hip or knee replacement.4, 7, 8 Sex differences in joint replacement rates have been attributed to higher prevalence of osteoarthritis among women.4, 9 Hawker et al9 also suggest that compared with men, women reported more arthritis pain, were more likely to be disabled, and were more likely to seek treatments largely because they were more likely to live alone. Sex differences are also thought to stem from differences between men and women in their preferences for the procedure.10

Race and Ethnicity 

Previous studies report significant disparities along racial and ethnic lines in the use of total knee or hip replacements. Joint arthroplasty was performed predominantly in those who were white and had a higher income.7, 8, 11, 12 Controlling for interactions of sex and race, previous research reported that white men were 2.6 to 5.1 times more likely to undergo total knee replacement than black men,13, 14, 15 and were also 1.9 times more likely to have total hip replacement surgery than black men.6, 15, 16 Compared with black women, white women also underwent both procedures at a higher rate, but the difference was much smaller.3, 13, 15 In addition to differences between black and white subjects, some studies also suggest that the Asian population has the lowest rate of joint replacements, followed by black subjects and Hispanic subjects, while white subjects have the highest rates.15, 17, 18, 19, 20

Geographic Variation 

Previous studies have documented significant geographic variation in the use of total joint replacements. Peterson et al4 found that the highest rates of both types of joint replacement were in the midwest and northwest, and that the lowest rates were in the south and east. Using census regions, Mahomed et al7, 8 found that the west north central area had the highest rates of both types of joint replacements, while the west south central area and the mid-Atlantic area had the lowest rates of total hip replacements and total knee replacements. Katz et al3 observed that the highest rates were in region VIII (CO, MT, ND, SD, UT, WY) and region X (AK, ID, OR, WA), while the lowest rates were in region II (NY, NJ). Using the hospital referral region as the geographic variable in the study, Skinner et al15 observed geographic disparities in use of knee arthroplasty among both blacks and non-Hispanic whites.

Back to Article Outline

Projections of Future Use 

Although the procedure has become increasingly prevalent in the last few decades, researchers expect the trend to continue well into the future. Several studies have been conducted to project the number of primary hip and knee replacements as well as revisions. Based on a 2002 AAOS analysis, Frankowski and Watkins-Castillo21 projected that there will be over 274,000 total hip replacements and 474,000 total knee replacements in 2030. Apparently, this projection greatly underestimated the volume of future joint replacements because the number of procedures in 2005 has already approached or exceeded the expected numbers for 2030.

An argument5 against the AAOS model is that the prediction solely took account of the anticipated growth of the United States population and assumed a constant rate of total hip/knee replacements based on 4 years (1996–1999) of data reported from the National Hospital Discharge Survey. Using both census data and 1990 to 2002 Healthcare Cost and Utilization Project-NIS data, Kurtz et al22 instead projected that total hip replacements will grow 226 percent from 201,000 in 2005 to 453,000 in 2030, and the number of knee replacements will grow approximately 5-fold, from 428,000 in 2005 to 2.16 million procedures by 2030. Their study also projects that hip revisions will grow from 41,000 in 2005 to 98,000 in 2030, and knee revisions will grow from 37,000 to 195,000 between 2005 and 2030. Because the predicted number of total joint replacement procedures performed in 2005 is already lower than the number of actual procedures in 2004, the Kurtz22 projection may still be conservative given current rates of growth.

Back to Article Outline

Dynamics Behind Actual and Projected Growth Rates 

Rapid growth in the actual and projected number of joint replacements poses important questions: what is fueling the growth? What are the implications of this growth for payers (eg, Medicare), downstream PAC providers, and health policy makers? To answer these questions, we use a market approach that examines (1) the demand side, (2) the supply side, and (3) the role of financial intermediaries whose policies modulate the interaction between demand and supply sides of the market.

Back to Article Outline

Demand Side 

Increasing demand for joint replacements from patients is the major engine driving the growth of joint replacements. Growing numbers of candidates for the procedure and the procedure's presumed underuse among selected groups will continue to contribute to increasing demand.

Aging remains the strongest risk factor associated with development of osteoarthritis, the most common condition leading an individual to seek a joint replacement.23, 24, 25 In 1999, about 12% of all American adults over 65 reported having osteoarthritis. Among all patients with osteoarthritis, almost half are over 65 years old.26, 27 The growing elderly population will produce more individuals with osteoarthritis, which is expected to increase demand for total joint replacements.27 During the last several decades, the postwar baby boom and longer life expectancies have contributed significantly to the growth of age groups that produce the most candidates for joint replacements, namely those in their 50s, 60s, and 70s. The actual and projected numbers of those 65 years and older continues to grow. In 1960, there were 16.5 million individuals who were 65 years and older. By 2005, this number had grown to 37 million or 12.4% of the U.S. population, and it is projected to grow to 70 million or 19.7% of the population in 2030.28, 29

The increased prevalence of obesity among Americans is also associated with an increased risk of knee osteoarthritis and, to a lesser degree, with hip osteoarthritis, which leads to increased demand for joint replacements.30, 31, 32, 33, 34 Flegal et al35 reported that, from 1999 to 2000, approximately two thirds of adult Americans were overweight or obese (body mass index ≥30), and one third of those 60 to 79 years old were obese, compared with about 25% in 1988 to 1994. Moreover, obese individuals place greater mechanical forces on their joint replacements. There is substantial evidence indicating that obesity is associated with higher revision rates.36, 37 Hence, increased prevalence of obesity is prone to increase the demand for both primary and revised joint replacements.

Census data indicate that the elderly population is becoming better educated. In 2000, 32% of individuals older than 65 had completed high school as their highest level of education, and 33.5% had some type of college or higher degree—about 2.5 to 3 times higher than in 1960.38, 39 Patients' educational level is a marker for patient familiarity with, and knowledge of, the risks and benefits of medical procedures. This familiarity is known to influence the willingness and decisions of patients to use joint replacement surgery.40 Higher educational achievements among older populations suggest that they will be more likely to agree with their physicians' recommendation for a joint replacement.25

Despite the increasing number of joint replacements, they are thought to be substantially underused. After reviewing 20 years of data, a panel convened by the National Institutes of Health41, 42 estimated that, among those with a potential need for knee replacement, only 13% of women and 9% of men expressed definite willingness to undergo the procedure. Underuse of both knee and hip replacements among America's minorities, noted previously, suggests that unmet need is a possible reason of underuse of total joint replacements in the overall population.4, 9 A population-based survey of 48,218 residents in 2 areas of Ontario, Canada, reported that potential needs for arthroplasty among men and women were unmet, although the rates of both types of total joint replacements among women were higher than among men.9

Unmet need is thought to result from both patients and physicians. Economic barriers, lack of needed information, poor interaction with physicians, and preferences for other treatments may lead patients to choose options other than joint replacement surgery.4, 9, 42, 43 Unmet need for total joint replacements is also attributed to the perspective of physicians, especially primary care physicians. Previous studies suggest that many primary care physicians are not as familiar with joint replacement outcomes as they could be.42 This leads to fewer referrals to orthopedic surgeons. Previous studies also suggest that disagreements persist among both primary care physicians and orthopedists about indications for the surgery.44 This lack of agreement may partially explain the unmet need.4

Some have argued that use of nonoperative treatments would offset the use of joint replacement.45 In the early stage of osteoarthritis, the use of nonoperative care such as oral nonsteroidal anti-inflammatory medications, oral dietary supplements, and intra-articular injections can relieve symptoms and improve function, and in some instances slows progression. The nonoperative care thus may reduce or delay the demand for joint replacement. However, we are not familiar with any empirical study supporting this suggestion.

In short, changes in population demographics have increased the number of candidates for total joint replacements and will produce a larger volume of potential patients undergoing this procedure in the future. Although little research has addressed this issue, unmet need leaves considerable room for increases above those suggested by the population's changing demography alone.

Back to Article Outline

Supply Side 

As demand for total joint replacements increases, changes in supply also are taking place. These changes include increases in the supply of practitioners, advances in surgical technology of joint replacement, and indications for joint replacement. Over the past few decades, these changes have strengthened the provider's capacity to offer more procedures, improved the surgical safety and efficacy, and expanded the target population that could benefit from the procedure.

Supply of Practitioners 

A series of AAOS reports dating back to 1990 documents how orthopedic practice has changed over time. First, the number of certified orthopedists has increased—albeit moderately, by 7.5%, from 16,266 in 2000 to 2001, to 17,486 in 2004 to 2005.46 As a result, the density of orthopedic surgeons increased modestly from 5.8 per 100,000 people to 6.2 per 100,000 people, while the rate was 5.3 in 1990.6, 46 Second, more orthopedic surgeons have focused their practice on joint replacements. It is estimated that the portions of orthopedic surgeons whose focus areas of practice were adult hip and adult knee increased from 31% and 41%, respectively, in 1992 to 1993, to 41% and 52%, respectively, in 2004 to 2005.6 Combined increases in both the number and the density of orthopedic surgeons in the area of joint replacements suggest a growing capacity for performing more joint replacements.

Advances in Technology 

Joint replacement technology has been evolving since the 1960s, when the first widely successful total hip replacement was performed. Improving surgical techniques and mechanical materials decreased operative time, blood loss, surgical risk, and postoperative complications, and increased the durability of implant devices. Improvements in surgery have led to better outcomes, enlarged indications, and decreased average LOS for patients who undergo the procedure.

Better outcomes encourage more primary care physicians to consider the procedure and refer potential patients. In addition, improvements in total joint replacement encourage acceptance by patients and larger communities of interest.

Advances in joint replacements help to reduce average LOS and, in turn, enable hospitals to offer the procedure to more patients. HCUPnet data in figure 3 indicate that the average LOS declined nearly 50% from 1993 to 2005. For 3 years, from 1998 to 2000, the downward trend had flattened to 4.6 days and then resumed its downward trend. The 3-year pause can, perhaps, be attributed in part to Medicare's short-stay transfer policy, which was implemented in September 1998 and exacted penalties for early transfer to PAC for certain conditions. HCUPnet LOS trend data are in keeping with earlier studies. Forrest et al,47 for example, report that the average LOS for patients with total joint replacement declined from 6.4 days in 1995 to 5.1 days in 1997. FitzGerald et al48 report that the mean LOS fell from 5.5 days in January 1996 to 4.7 days in September 1998, and then increased slightly to 4.9 days in 2000.

Indications for Joint Replacement 

In the process of diffusing medical technology, previous research suggests that indications for use of the technology will change and, over time, expand as physicians gain more experience by performing more procedures.49 After several decades of use, indications for use of total joint replacements have broadened to include individuals who were not previously eligible for the procedure because they were more fragile, had more comorbidities, or had more severe diseases.4, 50

In previous decades, nearly all knee replacements were performed on patients between 65 and 75 years old.42 However, a growing number of both younger and older patients are undergoing this procedure,4, 12, 42, 51 which suggests a broadening of indications for this procedure.4, 5, 12 Jain et al12 found that between 1990 and 2000, the proportion of total knee replacements performed increased by 95.2% in the age group 40 to 49 years and by 53.7% in the age group 50 to 59 years. HCUPnet data indicate that, from 1997 to 2005, the portion of patients with total hip replacements younger than 65 years old increased from 26.5% to 43.7%; the portion of total knee replacements in this age group increased from 25.6% to 39.5%. A larger percentage of the older patients is also undergoing joint replacement surgery.4, 5, 42

Likewise, patients with comorbidities, who previously may have been excluded, are now considered eligible for the procedure. For example, Katz et al4 found that the average number of diagnoses, a proxy for comorbidities, increased from 2.6 to 3.1 between 1985 and 1989. The growing list of indications over time has enlarged the eligible population for total joint replacements, and hence has led to more usage.

Back to Article Outline

Role of Financial Intermediaries 

Changing Payer Mix 

Financial intermediaries have played an important and, perhaps, an inadvertent role in the growth of total joint replacement. Because of the age distribution of patients obtaining a joint replacement, Medicare remains the single most important payer and accounted for about 57% of all payment for total joint replacements in 2005. The payer mix for the procedure, however, has changed significantly over the last decade. Medicare's proportion of all charges continues to decline as younger individuals obtain joint replacements. Meanwhile, private health plans and others, which are usually more generous than Medicare, have increased their portion of overall payment. HCUPnet data indicate that between 1997 and 2005, the portion of total hip replacement charges covered by Medicare decreased from 62.6% to 55.8%, while the portion of private health plans increased from 30.4% to 38.2%. For a total knee replacement, the portion of charges covered by Medicare decreased from 68% to 59.4%, and the portion of private health plans increased from 25.5% to 34.7%. Mehrotra et al51 report that the proportion of total charges covered by private insurance increased from 25% to 35%, and the proportion covered by Medicare decreased from 73% to 63% from 1990 through 2000.

Medicare Reimbursement 

Despite the growth of private health plan participation in joint replacements, Medicare's reimbursement policies and cost containment strategies continue to shape practice and provider behavior. Rapid increases in use of joint replacements have prompted Medicare to implement several cost reduction strategies over the years. Researchers noted that Medicare reimbursement to hospitals per total joint replacement declined in the past 2 decades.52, 53 The reimbursement rate to physicians for joint replacement also decreased over time.54, 55 Even though the absolute number of total joint replacements covered by Medicare increased, total aggregate reimbursement for total hip replacement decreased from $1.66 billion in 1997 to $1.40 billion in 2003 (for both primary and revised replacements). For total knee replacement, total reimbursement declined from $2.94 billion to $2.59 billion between 1997 and 2003.53

In addition, the Centers for Medicare and Medicaid Services (formerly the Health Care Financing Administration) launched a series of major changes to force providers to reduce the cost of total joint replacements during the 1990s. These changes include use of clinical pathways, an implant standardization program, and more recently, participating centers of excellence.

Effect on Hospital Costs, Charges, and Margins 

Hospitals have responded by cutting their costs. Healy et al56 reported that from 1992 to 1995, the average hospital cost for total knee replacement declined by 12.9% an admission, from $10,043 to $8747. Although we do not have national estimates of cost trends since then, it is reasonable to believe that because of reduced LOS and pressure of cost containment from health plans, especially Medicare, cost for the procedure in hospitals has decreased, or at least remained at a steady level after adjusting for inflation.

Despite reimbursement and cost reductions, charges for total joint replacement have continued to increase over the past decade. HCUPnet data indicate that the average charge per patient for total hip replacement has been increasing steadily from $22,099 in 1993 to $39,348 in 2005, while the average charge for total knee replacement increased from $21,024 to $35,946 (in 2005 dollars) (fig 4). Ong et al55 report a similar trend in charges for both types of joint replacement between 1997 and 2003.

Total joint replacements are seen as one of hospitals' more lucrative service lines.57 Because costs of the procedure relative to charges and payment have decreased over the last decade, hospital margins have presumably increased. The growing portion of private health plans, moreover, has offered hospitals a more favorable overall reimbursement rate over time. Two studies based on 1993 to 1995 data and 1995 data56, 58 reported that all payers were profitable for hospitals except for Medicaid and some selected managed care contracts. There is no up-to-date research focusing on hospital margins. The financial advantages of this procedure have stimulated hospitals and physicians to perform more joint replacements, and hence have fueled the increase in the number of procedures during the past decade.

Effects on Quality and Volume 

Continuously decreasing reimbursement raises a concern that hospitals may try to reduce their costs by stinting or reducing services as noted in the rapid decline in average LOS for the procedure. Studies report that despite reduction in cost and LOSs, quality of joint replacement and patient satisfaction have not declined. However, these studies generally focused on short-term quality indicators. Studies with longer follow-up periods are needed to examine long-term consequences of Medicare's expenditure reduction efforts.

Other Potential Effects 

To cope with reduced reimbursement, hospitals can, in theory, reduce LOS, reduce unnecessary cost, decrease the portion of patients whose fees are from Medicare, shift from inpatient care to outpatient care or rehabilitation services, and even lose interest in the service.59 Likewise, physicians can reduce the volume of procedures, reduce the number of related or complementary procedures, and/or increase the volume of services to other patients.54, 55 Hospital or physician preference for non-Medicare patients can lead to varying uses of the procedure. Dunlop et al,60 for example, report that use of joint replacements was lower among people who depended solely on Medicare compared with those who also had a supplemental health plan (controlling for difference in demographics, health needs, and economic access).

Back to Article Outline

Implications for Postacute Rehabilitation Care 

Upstream trends in society at large and acute care in particular have material downstream effects on demand for, and supply of, postacute rehabilitation services for individuals who acquire a joint replacement.

Oversimplified, increasing numbers of patients with joint replacement who participate in a postacute bed service rehabilitation program—that is, in SNFs or IRFs—are a function of 2 main variables: (1) volume of joint replacement patients and (2) rates at which they are discharged to a postacute setting. As discussed in this article's opening section, the volume of joint replacements increased rapidly over the last decade. At the same time, the rate of patients with joint replacement who were discharged to postacute facilities also increased.

As shown in figure 5, HCUPnet data indicate that 74.4% of patients who underwent total knee replacement and 77.5% of patients who underwent total hip replacement were discharged to PAC settings (eg, SNFs, IRFs, or home health agencies) in 2005. In 1993, the portions were only 49.1% and 51.5%, respectively. During the same period, the percent discharged from acute care to home declined from 49.7% to 24.8% for total knee replacement, and from 46.7% to 21.6% for total hip replacement. Another study reports that between January 2002 and June 2003, approximately 30% of Medicare patients with joint replacement used SNF care, and 35% used IRF care.61 Increased referral rates to PAC may be a function of the reduced LOSs in acute care and the need for some degree of ongoing medical supervision and rehabilitation in another setting. As the result of both factors—increased number of replacements and increased referral rates to PAC—the number discharged to PAC (skilled nursing, inpatient rehabilitation, home health) increased 4.4 times, from 62,980 to 369,296, for patients with knee replacement, and 5.9 times, from 41,409 to 184,236, for patients with hip replacement between 1993 and 2005 (HCUPnet estimate).

Looking forward, if one were simply to apply the rates of discharge to PAC in 2005 to the numbers of joint replacements in 2030 projected by Kurtz,22 we estimate that 1.22 million patients with joint replacement will be discharged to a postacute facility. We simply do not know whether this will be the case given several mitigating factors such as the increased use of minimally invasive replacements, increased numbers of younger and healthier patients, increased use of more durable implants, and the intended and unintended effects of various government and health plan payment policies. Nonetheless, we can assume that future demand for postacute rehabilitation placement will continue to increase substantially. The challenge for health policy makers, payers, and providers alike is 3-fold: (1) developing clear need-based criteria for postacute placement, (2) determining the optimal capacity and facility-mix needed to meet the growing demand for joint replacement rehabilitation, and (3) developing reimbursement systems that will optimize capacity and patient outcomes.

Volume and capacity in PAC are also changing independent of upstream events mainly because of events related to the Medicare program: (1) reinstatement of the 75% rule (now the 60% rule), (2) local coverage decisions made by Medicare fiscal intermediaries, and (3) retroactive audits. All of these limit the ability of IRFs to accept patients with joint replacement. The number of patients with hip and knee replacement obtaining IRF-level care has declined by 12,869 a quarter, from 31,991 in the fourth quarter of 2003 to 19,122 in the second quarter of 2006.62 On an annualized basis, this represents a decline of 51,476 patients. A recent report notes that there were 8 fewer acute hospital-based rehabilitation units in 2005 than in the previous year.63 This is the first decline in the number of rehabilitation units after decades of growth. More of these patients are presumably going to SNFs, obtaining home health care, or limiting their postacute rehabilitation to outpatient settings. Based on the JOINTS study64 reported elsewhere in this issue of the journal, approximately 3 out of 4 of all patients who received bed service rehabilitation in an SNF or an IRF go on to obtain outpatient therapy as well.

Whether the shifting balance between SNF-level and IRF-level care is appropriate is beyond the scope of this article. Nonetheless, policy makers and payers cannot remain indifferent to what is happening in acute care and what it portends for postacute rehabilitation. As policy makers and health plans seek to reduce expenditures for acute care, acute hospital managers and orthopedic surgeons will do what they can to curtail costs and preserve margins—and if they lose margins, to make it up in volume where they can in order to maintain income.

One way to reduce acute care costs is simply to reduce LOS. Supported by advances in surgical technique, as we discussed earlier, the average hospital stay for joint replacements has continued to decrease since 1983 when the Medicare prospective payment system was implemented. Reducing LOS may also presume that there is a postacute setting to which patients can be discharged in order to provide the level of medical monitoring and therapy that patients may still require. The old adage about health care cost containment very much applies to joint replacement care—that is, as one tries to press down on costs and expenditures in one segment of health care, they are likely to show up in other segments of health care. The desire to rein in acute care expenditures may well have helped to create the demand for postacute rehabilitation care and shift costs from acute to PAC.47, 65, 66

A study48 showed the relationship between shortened hospital stays and increase in PAC use between January 1996 and December 2000 by examining the effects of the short-stay transfer policy. The study found there had been a steady increase in discharge of patients with joint replacement to PAC prior to the short-stay transfer policy and then a 4.6% decrease immediately after the short-stay transfer policy. Similar trends are also suggested in figure 3 and figure 5, which were estimated based on a national patient sample. However, Medicare's short-stay transfer policy was only a blip in the long-term downward trend in acute LOS for joint replacement. The transfer rate to PAC was reversed only for a short time, grew at a slower rate, and then in 2003 accelerated again (see fig 5). The study by FitzGerald et al48 also found that LOS was stable shortly after implementation of the short-stay transfer policy, and the rate of discharge to PAC grew again, albeit at a rate lower than the rate before the short-stay transfer policy.

Significant changes in upstream care for patients with joint replacement may also affect the quality and outcome of rehabilitation care for these patients. The impact of changes in acute care hospitals may have 2 very different consequences for postacute rehabilitation quality and outcome. On the one hand, increased volume of joint replacement may lead to higher quality of rehabilitation care. The positive relationship between patient volume and quality of care has been widely established for acute care, but there is less evidence that the relationship is valid in PAC settings. Future studies are needed to examine the effect of patient volume on the outcome of rehabilitation care. On the other hand, shorter acute LOS may negatively influence the outcome of rehabilitation care among patients with joint replacement who still need careful medical monitoring but are not getting it. It was found that reduced hospital LOS, for example, resulted in a decreased range of motion for knee.67 The influence of acute care changes can be either positive or negative on the outcome of rehabilitation care for patients with joint replacement, but there is little or no empirical evidence that connects the effects of acute and postacute practice on outcomes.

Back to Article Outline

Conclusions 

Total joint replacement provides material benefits for patients who seek substantial pain relief, improved function, and quality of life.68 This article reviews the use of total joint replacement in the United States. The number of joint replacements performed has increased rapidly in the last decade, and the trend is predicted to continue in the future. Three groups of factors—increasing demand, increasing supply, and financial incentives—are associated with the use of the procedure. Despite the increasing volume of hip and knee replacements, great variation in the use of total joint replacement still exists among different demographic (eg, sex, race) and geographic groups. Although extensively studied, reasons for these disparities remain unclear. Further research is needed to understand the variation in use as well as the need for total joint replacement.

Policy makers remain concerned about the growth in Medicare expenditures stemming from the increased number of total joint replacements. Despite Medicare's success in reducing acute care expenditures for joint replacements, there remain many, perhaps unintended, downstream consequences. Increased use of the total joint replacement procedure and declining lengths of acute care stays are strongly associated with the rapid growth in use of PAC rehabilitation for patients with joint replacement. Reimbursement policies of Medicare and health plans need to take into account this broader perspective and not examine joint replacement care in isolated settings without considering other downstream effects.

Very few, if any, studies have linked changes in postacute rehabilitation with upstream changes in the epidemiology of joint disease and changes in acute care hospitals—for example, increased patient volumes, changes in indications, and reduced LOSs in acute care. Most studies, and health policies for that matter, fail to view joint replacement care as part of a larger continuum of care that may include an acute hospital stay, a rehabilitation facility stay, and several weeks of home and/or outpatient care.

Back to Article Outline

References 

  1. HCUPnet, Healthcare Cost and Utilization Project. Agency for Healthcare Research and Quality, Rockville, MD http://hcupnet.ahrq.govAccessed June 14, 2007
  2. Chassin MR, Brook RH, Park RE, et al. Variations in the use of medical and surgical services by the Medicare population. N Engl J Med. 1986;314:285–290
  3. Peterson MG, Hollenberg JP, Szatrowski TP, Johanson NA, Mancuso CA, Charlson ME. Geographic variations in the rates of elective total hip and knee arthroplasties among Medicare beneficiaries in the United States. J Bone Joint Surg Am. 1992;74:1530–1539
  4. Katz BP, Freund DA, Heck DA, et al. Demographic variation in the rate of knee replacement: a multi-year analysis. Health Serv Res. 1996;31:125–140
  5. Kurtz S, Mowat F, Ong K, Chan N, Lau E, Halpern M. Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002. J Bone Joint Surg Am. 2005;87:1487–1497
  6. Kurtz S, Lau E, Ong KL, Mowat FS, Halpern M. The future burden of hip and knee revisions: U.S. projections from 2005 to 2030. American Academy of Orthopaedic Surgeons 73rd Annual Meeting, Chicago, IL, March 22-26, 2006.
  7. Mahomed NN, Barrett JA, Katz JN, et al. Rates and outcomes of primary and revision total hip replacement in the United States Medicare population. J Bone Joint Surg Am. 2003;85-A:27–32
  8. Mahomed NN, Barrett J, Katz JN, Baron JA, Wright J, Losina E. Epidemiology of total knee replacement in the United States Medicare population. J Bone Joint Surg Am. 2005;87:1222–1228
  9. Hawker GA, Wright JG, Coyte PC, et al. Differences between men and women in the rate of use of hip and knee arthroplasty. N Engl J Med. 2000;342:1016–1022
  10. Karlson EW, Daltroy LH, Liang MH, Eaton HE, Katz JN. Gender differences in patient preferences may underlie differential utilization of elective surgery. Am J Med. 1997;102:524–530
  11. Gittelsohn AM, Halpern J, Sanchez RL. Income, race, and surgery in Maryland. Am J Public Health. 1991;81:1435–1441
  12. Jain NB, Higgins LD, Ozumba D, et al. Trends in epidemiology of knee arthroplasty in the United States, 1990-2000. Arthritis Rheum. 2005;52:3928–3933
  13. Wilson MG, May DS, Kelly JJ. Racial differences in the use of total knee arthroplasty for osteoarthritis among older Americans. Ethn Dis. 1994;4:57–67
  14. Petersen LA, Wright SM, Peterson ED, Daley J. Impact of race on cardiac care and outcomes in veterans with acute myocardial infarction. Med Care. 2002;40(1 Suppl):I86–I96
  15. Skinner J, Weinstein JN, Sporer SM, Wennberg JE. Racial, ethnic, and geographic disparities in rates of knee arthroplasty among Medicare patients. N Engl J Med. 2003;349:1350–1359
  16. Baron JA, Barrett J, Katz JN, Liang MH. Total hip arthroplasty: use and select complications in the US Medicare population. Am J Public Health. 1996;86:70–72
  17. Oishi CS, Hoaglund FT, Gordon L, Ross PD. Total hip replacement rates are higher among Caucasians than Asians in Hawaii. Clin Orthop Relat Res. 1998;166–174
  18. Escalante A, Barrett J, del Rincón I, Cornell JE, Phillips CB, Katz JN. Disparity in total hip replacement affecting Hispanic Medicare beneficiaries. Med Care. 2002;40:451–460
  19. Skinner J, Zhou W, Weinstein J. The influence of income and race on total knee arthroplasty in the United States. J Bone Joint Surg Am. 2006;88:2159–2166
  20. Jain N, Pietrobon R, Guller U, Shankar A, Ahluwalia AS, Higgins LD. Effect of provider volume on resource utilization for surgical procedures of the knee. Knee Surg Sports Traumatol Arthrosc. 2005;13:302–312
  21. Frankowski JJ, Watkins-Castillo S. Primary total knee and hip arthroplasty projections for the U.S. population to the year 2030. http://www.aaos.org/wordhtml/research/stats/TJR_projections.pdf2002;Accessed January 20, 2007
  22. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89:780–785
  23. Hart DJ, Doyle DV, Spector TD. Incidence and risk factors for radiographic knee osteoarthritis in middle-aged women: the Chingford Study. Arthritis Rheum. 1999;42:17–24
  24. Sharma L, Kapoor D, Issa S. Epidemiology of osteoarthritis: an update. Curr Opin Rheumatol. 2006;18:147–156
  25. Crowninshield RD, Rosenberg AG, Sporer SM. Changing demographics of patients with total joint replacement. Clin Orthop Relat Res. 2006;(Feb):266–272
  26. Jordan JM, Linder GF, Renner JB, Fryer JG. The impact of arthritis in rural populations. Arthritis Care Res. 1995;8:242–250
  27. Creamer P. Osteoarthritis pain and its treatment. Curr Opin Rheumatol. 2000;12:450–455
  28. U.S. Census Bureau. Dramatic changes in U.S. aging highlighted in new census, NIH report: impact of baby boomers anticipated. http://www.census.gov/Press-Release/www/releases/archives/aging_population/006544.html2006;Accessed June 13, 2007
  29. He W, Sengupta M, Velkoff VA, DeBarros KA. 65+ in the United States, 2005 (U.S. Census Bureau). 2005 http://www.census.gov/prod/2006pubs/p23-209.pdf2005;Accessed June 20, 2007
  30. Felson DT, Anderson JJ, Naimark A, Walker AM, Meenan RF. Obesity and knee osteoarthritis: the Framingham Study. Ann Intern Med. 1988;109:18–24
  31. Felson DT, Zhang Y, Anthony JM, Naimark A, Anderson JJ. Weight loss reduces the risk for symptomatic knee osteoarthritis in women: the Framingham Study. Ann Intern Med. 1992;116:535–539
  32. Spector TD, Hart DJ, Doyle DV. Incidence and progression of osteoarthritis in women with unilateral knee disease in the general population: the effect of obesity. Ann Rheum Dis. 1994;53:565–568
  33. Van Saase JL, Vandenbroucke JP, van Romunde LK, Valkenburg HA. Osteoarthritis and obesity in the general population (A relationship calling for an explanation). J Rheumatol. 1988;15:1152–1158
  34. Tepper S, Hochberg MC. Factors associated with hip osteoarthritis: data from the First National Health and Nutrition Examination Survey (NHANES-I). Am J Epidemiol. 1993;137:1081–1088
  35. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999-2000. JAMA. 2002;288:1723–1727
  36. Foran JR, Mont MA, Etienne G, Jones LC, Hungerford DS. The outcome of total knee arthroplasty in obese patients. J Bone Joint Surg Am. 2004;86-A:1609–1615
  37. Lingard EA, Katz JN, Wright EA, Sledge CB. Predicting the outcome of total knee arthroplasty. J Bone Joint Surg Am. 2004;86-A:2179–2186
  38. Gist YJ, Hetzel LI. We the People: aging in the United States. US Census Bureau Economic and Statistics. http://www.census.gov/prod/2004pubs/censr-19.pdf2004;Accessed May 4, 2007
  39. Bauman KJ, Graf NL. Educational attainment: 2000 (Census 2000 Brief). Department of Education http://www.census.gov/prod/2003pubs/c2kbr-24.pdf2003;Accessed July 12, 2007
  40. Ibrahim SA, Siminoff LA, Burant CJ, Kwoh CK. Differences in expectations of outcome mediate African American/white patient differences in “willingness” to consider joint replacement. Arthritis Rheum. 2002;46:2429–2435
  41. National Institutes of Health. NIH Consensus Development Conference on Total Knee Replacement. http://consensus.nih.gov/2003/2003TotalKneeReplacement117html.htm2003;Accessed July 12, 2007
  42. Vastag B. Knee replacement underused, says panel: useful option when nonsurgical therapies fail. JAMA. 2004;291:413–414
  43. Ibrahim SA, Burant CJ, Siminoff LA, Stoller EP, Kwoh CK. Self-assessed global quality of life: a comparison between African-American and white older patients with arthritis. J Clin Epidemiol. 2002;55:512–517
  44. Tierney WM, Fitzgerald JF, Heck DA, et al. Tricompartmental knee replacement: a comparison of orthopaedic surgeons' self reported performance rates with surgical indications, contraindications, and expected outcomes (Knee Replacement Patient Outcomes Research Team). Clin Orthop Relat Res. 1994;209–217
  45. Buckwalter JA, Stanish WD, Rosier RN, Schenck RC, Dennis DA, Coutts RD. The increasing need for nonoperative treatment of patients with osteoarthritis. Clin Orthop Relat Res. 2001;36–45
  46. Watkins-Castillo S. Orthopedists practice in the US 2005-2006. In: Rosemont: American Academy of Orthopaedic Surgeons; 2006;p. 1–12June 2006
  47. Forrest GP, Roque JM, Dawodu ST. Decreasing length of stay after total joint arthroplasty: effect on referrals to rehabilitation units. Arch Phys Med Rehabil. 1999;80:192–194
  48. FitzGerald JD, Boscardin WJ, Hahn BH, Ettner SL. Impact of the Medicare short stay transfer policy on patients undergoing major orthopedic surgery. Health Serv Res. 2007;42:25–44
  49. Hillman AL, Schwartz JS. The adoption and diffusion of CT and MRI in the United States: a comparative analysis. Med Care. 1985;23:1283–1294
  50. Rand JA, Trousdale RT, Ilstrup DM, Harmsen WS. Factors affecting the durability of primary total knee prostheses. J Bone Joint Surg Am. 2003;85-A:259–265
  51. Mehrotra C, Remington PL, Naimi TS, Washington W, Miller R. Trends in total knee replacement surgeries and implications for public health, 1990-2000. Public Health Rep. 2005;120:278–282
  52. Boardman DL, Lieberman JR, Thomas BJ. Impact of declining reimbursement and rising hospital costs on the feasibility of total hip arthroplasty. J Arthroplasty. 1997;12:526–534
  53. Bernstein J. Policy implications of physician income homeostasis. J Health Care Finance. 1998;24:80–86
  54. Mitchell JM, Hadley J, Gaskin DJ. Physicians' responses to Medicare fee schedule reductions. Med Care. 2000;38:1029–1039
  55. Ong KL, Mowat FS, Chan N, Lau E, Halpern MT, Kurtz SM. Economic burden of revision hip and knee arthroplasty in Medicare enrollees. Clin Orthop Relat Res. 2006;6(May):22–28
  56. Healy WL, Iorio R, Ko J, Appleby D, Lemos DW. Impact of cost reduction programs on short-term patient outcome and hospital cost of total knee arthroplasty. J Bone Joint Surg Am. 2002;84-A:348–353
  57. Haugh R. A joint strategy for orthopedics: hospitals team up with docs to keep a lucrative service line. Hosp Health Netw. 2002;76:54–58
  58. Iorio R, Healy WL, Richards JA. Comparison of the hospital cost of primary and revision total knee arthroplasty after cost containment. Orthopedics. 1999;22:185–189195-9
  59. McGinnity ES, Pluth TE. Managing orthopedics and neurosciences costs through standard treatment protocols. Hosp Technol Ser. 1994;13:1–25
  60. Dunlop DG, Brewster NT, Madabhushi SP, Usmani AS, Pankaj P, Howie CR. Techniques to improve the shear strength of impacted bone graft: the effect of particle size and washing of the graft. J Bone Joint Surg Am. 2003;85-A:639–646
  61. Buntin M, Deb P, Escarce J, Hoverman C, Paddock S, Sood J. Comparison of Medicare spending and outcomes for beneficiaries with lower extremity joint replacements. http://works.bepress.com/cgi/viewcontent.cgi?article=1006&context=melinda_buntin2005;Accessed January 7, 2008
  62. The Moran Co.. Utilization trends in inpatient rehabilitation: update through Q II. http://www.aha.org/aha/content/2006/pdf/2006septmoranreport.pdf2006;Accessed July 20, 2007
  63. Center for Post-acute Studies, National Rehabilitation Hospital. CPS 2008 annual report: trends in post-acute care. http://www.post-acute.org/news/Biannual%20Report%20FINAL.pdf2008;Accessed April 3, 2008
  64. DeJong G, Tian W, Smout R, et al. Use of rehabilitation and other health care services by patients with joint replacement after discharge from skilled nursing and inpatient rehabilitation facilities. Arch Phys Med Rehabil. 2009;90:1297–1305
  65. Munin MC, Rudy TE, Glynn NW, Crossett LS, Rubash HE. Early inpatient rehabilitation after elective hip and knee arthroplasty. JAMA. 1998;279:847–852
  66. Oldmeadow LB, McBurney H, Robertson VJ. Hospital stay and discharge outcomes after knee arthroplasty. J Qual Clin Pract. 2001;21:56–60
  67. Teeny SM, York SC, Benson C, Perdue ST. Does shortened length of hospital stay affect total knee arthroplasty rehabilitation outcomes?. J Arthroplasty. 2005;20(7 Suppl 3):39–45
  68. Buckwalter JA, Lohmander S. Operative treatment of osteoarthritis. J Bone Joint Surg Am. 1994;76-A:1405–1418

 Supported by the HealthSouth Corp, ARA Research Institute of the American Rehabilitation Providers Association, Brooks Health, National Rehabilitation Hospital, American Hospital Association, the Federation of American Hospitals, and others.

 A commercial party having a direct financial interest in the results of the research supporting this article has conferred or will confer a financial benefit on one or more of the authors.

PII: S0003-9993(09)00306-2

doi:10.1016/j.apmr.2008.10.035

Archives of Physical Medicine and Rehabilitation
Volume 90, Issue 8 , Pages 1260-1268, August 2009

Article Tools

* Image Usage & Resolution