SEARCH

SEARCH BY CITATION

Keywords:

  • Total hip replacement;
  • Discharge destination;
  • Rehabilitation facilities

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Objective

To identify the factors associated with discharge to an inpatient rehabilitation facility versus discharge directly to home after elective total hip replacement (THR).

Methods

Data were obtained from a medical record review and a cross-sectional survey of a population-based cohort of patients who received elective primary or revision THR in 1995. Postoperative functional status was measured as the ability to walk independently prior to discharge. A multivariate regression model was developed with discharge to an inpatient rehabilitation facility versus directly to home as the dependent variable. The model adjusted for demographic characteristics, socioeconomic features, and several preoperative and postoperative clinical factors.

Results

We included 1,276 patients age 65–94 years in the analysis. Of these, 58% were discharged from the acute care hospital to a rehabilitation facility. The cohort had mean age of 73 years; 96% were white; 62% were female; 32% were living alone; 38% had an annual income less than $20,000; and 78% were unable to walk independently in the hospital before discharge. For primary THR patients, multivariate analysis showed a significant association between being discharged to a rehabilitation facility and being unable to walk at discharge, older age, obesity, and living alone. For revision THR patients, multivariate analysis also showed a significant association between being discharged to a rehabilitation facility and being unable to walk at discharge.

Conclusion

In a large population-based sample, postoperative functional status, age, obesity, and social support all influenced the discharge destination following elective THR.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Total hip replacement (THR) has revolutionized the management of advanced hip arthritis. In general, patients who undergo THR experience substantial relief of pain and improvement in functional status (1–3). Annually, >140,000 persons in the United States undergo primary THR (4). Of the THRs performed electively in the Medicare population for indications other than fracture, >85% are performed for advanced primary or secondary osteoarthritis, 7% for avascular necrosis, 3% for rheumatoid arthritis, and the remainder for a range of other conditions (5).

Currently, discharge from acute care hospitals following THR typically occurs on the fourth or fifth postoperative day (6, 7). As length of stay has diminished in the last decade, many patients have been discharged to inpatient rehabilitation or nursing facilities to continue their functional recovery (8). There are large variations in practice patterns. On many occasions, the decision to discharge to an inpatient facility is based on an overall impression by the treating team that the patient is not progressing as well as could be expected. Sometimes, patients are discharged to inpatient facilities because they do not have sufficient social support or assistance from friends or family to enable the patient to be discharged directly to home. However, there are no clear factors to guide the decision to discharge patients to home or to an inpatient rehabilitation facility (8).

Several studies have investigated factors associated with discharge to an inpatient rehabilitation facility. Older age, female sex, race, insurance, the number or severity of comorbid conditions, types of arthritis, complications, and living alone have all been associated with discharge to an inpatient rehabilitation facility (8–12). Determinants of home-based rehabilitation have included preference for home-based rehabilitation, male sex, and knowledge regarding total joint replacement (TJR) (13). However, previous studies have had limited sample sizes and have recruited patients from only 1 center, leaving the generalizability of the findings in question. To the best of our knowledge, no population-based studies have been undertaken to examine factors associated with discharge to an inpatient facility following THR, or discharge determinants following revision THR as compared with primary THR.

The specific aim of this study was to use data from a large population-based cohort of Medicare beneficiaries who received THR to identify factors that distinguish individuals who were discharged to an inpatient rehabilitation program from those discharged directly to home after elective THR. We hypothesized that both the patients' physical functional status immediately following surgery and social support at home would influence whether patients were discharged to inpatient rehabilitation facilities.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Patients.

The analyses presented in this article are built upon a population-based sample of patients who received elective primary or revision THR. To select the study cohort, we used Medicare claims data to identify a random sample of patients aged ≥65 years who were resident in 3 states (Ohio, Pennsylvania, and Colorado) and who had elective primary or revision THR in calendar year 1995. Details of the cohort selection are published elsewhere (14).

The primary outcome measure was the discharge destination from the acute care hospital to an inpatient rehabilitation facility versus to home. Inpatient rehabilitation facilities included both inpatient rehabilitation and skilled nursing facilities. For the current analyses, we included all patients who had data on the primary dependent variable (discharge destination).

Data sources.

Data used in this report came from multiple sources, including medical records, Medicare claims, and a patient survey.

The medical record reviews were performed by trained personnel using a standardized data abstraction form. The medical record included data on patients' preoperative clinical characteristics, such as age, sex, weight, height, primary underlying joint disease (osteoarthritis, rheumatoid arthritis, or avascular necrosis), history of previous orthopedic surgery, and comorbid medical conditions. The comorbid medical conditions were extracted and aggregated using the Charlson Index (15). This variable was dichotomized at >2 comorbidities. The data recorded also included the American Society of Anesthesiologists' Physical Status Classification (ASA) before surgery (16). This variable assesses the severity of coexistent conditions and was dichotomized at class I or II versus class III or IV (where class II indicates mild systemic disease with no functional limitation). The type of anesthesia used (general, spinal, combination) and surgical procedure performed (primary versus revision) were recorded. For revision surgery, the medical record was also scrutinized to ascertain whether there had been prior revisions of the index hip. Data on the process of care, such as postoperative control of pain using patient-controlled analgesia (PCA; which allows patients to titrate analgesics in amounts proportional to perceived pain stimulus), was also recorded. Functional status after surgery was assessed by the physical therapist during the last physical therapy session as the ability to walk independently—without assistance of another person—before discharge from the acute-care setting.

Body mass index (BMI) was computed as weight divided by height squared (kg/m2) and was dichotomized at BMI >30 kg/m2 (overweight grade II) (17).

The patient survey was administered postoperatively 3 years after THR. The questionnaire included items about patients' preoperative functional status (including use of supportive devices, limp, stair climbing, and walking distance). These items were adapted from the Harris Hip Score (2, 3, 18). Weights were assigned as in the Harris Hip Score, summed, and standardized to a 0–100 scale, with 100 representing the best preoperative functional status. Preoperative functional status was dichotomized at the lowest quartile. Patients also indicated their education level, socioeconomic status, their living arrangement, and the number of years of formal education completed. Level of education was dichotomized by college education versus less than college level of education; annual income was categorized as greater than or less than $20,000 and as a third category of missing values.

Medicare claims provided data on the volume of primary and revision THRs performed in 1995 categorized by surgeon and hospital; these claims also provided length of stay in the acute-care setting (19). We used as our indicator of hospital and surgeon volume the combined volume of primary plus revision THR. The hospital volume variable was dichotomized at the median as ≤40 versus >40 procedures per year in the Medicare population. The surgeon volume variable was dichotomized at the median as ≤14 or >14 procedures per year in the Medicare population. Length of stay was dichotomized at the lowest quartile as ≤4 days or >4 days of stay in the acute-care setting following THR.

Statistical analysis.

We examined the association between discharge destination and patient sociodemographic and clinical factors, and hospital and surgeon procedure volume. We assessed the bivariate relationship of each explanatory variable to the outcome. Significant covariates were selected for the multivariate model in addition to clinically relevant variables. We used relative risk (RR) to measure the association between the risk factors and the outcome and employed generalized linear models with log link to estimate relative risks and 95% confidence intervals (95% CIs). P values < 0.05 were considered statistically significant. Analysis was performed using SAS 8.0 (SAS Institute, Cary, NC). Due to the large important number of missing values for the independent variable “income,” a separate category with these missing values was included in the analysis.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

In total, 3,507 patients were selected for participation in the parent study: 1,939 who had primary THR in 1995 and 1,568 patients who had revision THR (14). Among the 1,939 primary THR recipients, 32 (1.7%) had died in the 6–12 months between the time we chose the sample and the time we contacted patients. Another 20 (1.0%) could not be located because of an incorrect address. Of the 1,887 surviving patients with accurate addresses, 519 (28%) never responded to the 3 letters of invitation, 338 (18%) refused to participate, and 1,030 (55%) agreed to participate. Of these, 958 (93%) had data available from all 3 sources (medical records, claims, patient survey). Among 1,568 revision THR patients chosen for the cohort, 21 (1.3%) died and 16 (1.0%) could not be located. Among 1,531 survivors with valid addresses, 605 (40%) never responded, 258 (17%) refused to participate, and 669 (44%) agreed to participate. Of these, 595 (89%) returned questionnaires. After patients with missing values on the primary dependent variable were excluded, the final study cohort included 758 patients who had primary THR and 425 who had revision THR.

Subjects who completed the survey were 2 years younger than subjects who either refused or never responded to our invitations (74.1 versus 76.2 years; P = 0.0001). Forty percent of African American patients completed the survey versus 51.3% of white patients (P = 0.03), and 34% of patients who were eligible for Medicaid (indicating poverty level income) completed the survey versus 51% of those with higher incomes (P = 0.004). Only 41% of patients with ≥2 comorbid conditions on the Charlson Index completed the survey versus 52% of those with <2 comorbidities.

Demographic characteristics of the entire group of THR patients are presented in Table 1. In the primary THR group as well as in the revision THR group, 58% were discharged from the acute-care hospital to inpatient rehabilitation facilities, from which 24% were discharged to a skilled nursing facility, intermediate care facility, or transitional care unit and 34% were discharged to a rehabilitation facility. The remaining 42% were discharged directly to their homes.

Table 1. Characteristics of the study sample (n = 1,276)*
 Primary THR, % (n = 810)Revision THR, % (n = 466)
  • *

    THR = total hip replacement; BMI = body mass index; ASA = American Society of Anesthesiologists' Physical Status Classification.

Patient characteristics  
 Sex  
  Female6260
 Age, years  
  ≤725148
  >724952
 Ethnicity  
  White9795
 Living alone3131
 Education  
  At least college2020
  Less than college8080
 Income  
  ≥$20,0004440
  <$20,0003642
  Missing1918
 Comorbidities  
  <27878
  ≥22222
 Prior orthopedic surgery (nonindex joint)  
  No6040
  Yes4060
 BMI  
  <307679
  ≥302421
 Underlying disease  
  Osteoarthritis9562
  Rheumatoid arthritis55
  Avascular necrosis84
 Functional status pre-THR  
  Top 3 quartiles7063
  Lowest quartile3037
 ASA class  
  ≤II5349
  >II4751
 Anesthesia  
  Epidural/spinal3928
  Combination/general6172
 Patient-controlled anesthesia  
  Yes5147
  No4953
 Walk independently before discharge  
  Yes2321
  No7779
 Discharge destination  
  Home4242
  Rehabilitation facility5858
 Length of stay, days  
  >46365
  ≤43735
Hospital characteristics  
 Hospital volume  
  >404164
  ≤405936
 Surgeon volume  
  >143959
  ≤146141

Patients admitted to a rehabilitation facility were older (mean ± SD 74 ± 5.3 years) than patients discharged to home (72 ± 5.06 years; P < 0.0001). Older age was associated with the need for rehabilitation in a linear fashion (P value for trend < 0.0001), with the majority of elderly patients discharged to rehabilitation facilities. Age was dichotomized according to the median for further analyses. Patients discharged to a rehabilitation facility had a shorter length of stay (4.9 ± 2.2 days) than those discharged directly to home (5.8 ± 2.09 days; P < 0.001). Patients discharged to a rehabilitation facility had more comorbidities than patients discharged to home (23% versus 21%; P = 0.29).

In the primary THR sample, 62% were female, 31% were living alone, 20% had a college degree, and 36% had an annual income less than $20,000. In the revision THR sample, 60% were female, 31% were living alone, 20% had a college degree, and 42% had an annual income less than $20,000.

Crude analyses of the primary THR sample showed a statistically significant association between being discharged to a rehabilitation facility and female sex (RR 1.35, 95% CI 1.11–1.64), older age (RR 1.60, 95% CI 1.33–1.93), living alone (RR 1.53, 95% CI 1.27–1.84), low income (RR 1.16, 95% CI 1.08–1.26), obesity (RR 1.35, 95% CI 1.11–1.65), and inability to walk before discharge (RR 6.57, 95% CI 4.14–10.4). Table 2 shows factors associated with discharge to an inpatient rehabilitation facility after primary THR. We did not find statistically or clinically meaningful associations between being discharged to a rehabilitation facility and educational level, preoperative functional status, Charlson Index, previous orthopedic surgeries, primary underlying joint disease, ASA score, type of anesthesia, use of PCA after the procedure, or hospital and surgeon volumes (Table 2).

Table 2. Factors associated with discharge to inpatient rehabilitation facility after primary THR*
 Crude RR (95% CI)Adjusted RR (95% CI)
  • *

    THR = total hip replacement; RR = relative risk; 95% CI = 95% confidence interval; BMI = body mass index; ASA = American Society of Anesthesiologists' Physical Status Classification.

  • Adjusted for age, sex, living alone, level of education, income, and obesity.

Patient characteristics  
 Sex  
  Male1.00 
  Female1.35 (1.11–1.64)1.14 (0.92–1.43)
 Age, years  
  ≤721.00 
  >721.60 (1.33–1.93)1.36 (1.11–1.66)
 Living alone  
  No1.00 
  Yes1.53 (1.27–1.84)1.23 (1.004–1.5)
 Education  
  College1.00 
  Less than college1.12 (0.90–1.38)0.97 (0.74–1.28)
 Income  
  ≥$20,0001.001.00
  <$20,0001.16 (1.08–1.26)1.02 (0.80–1.32)
  Missing1.07 (1.98–1.18)0.90 (0.58–1.41)
 Comorbidities  
  <21.00 
  ≥21.06 (0.86–1.32) 
 Prior orthopedic surgery (nonindex joint)  
  No1.00 
  Yes0.97 (0.81–1.17) 
 BMI  
  <301.00 
  ≥301.35 (1.11–1.65)1.29 (1.05–1.60)
 Underlying disease  
  Osteoarthritis1.00 
  Rheumatoid arthritis0.90 (0.76–1.08) 
  Avascular necrosis1.07 (0.94–1.22) 
 Functional status pre-THR  
  Top 3 quartiles1.00 
  Lowest quartile1.09 (0.91–1.32) 
 ASA class  
  ≤II1.00 
  >II1.13 (0.94–1.35) 
 Anesthesia  
  Epidural/spinal1.00 
  Combination/general0.92 (0.76–1.11) 
 Patient-controlled anesthesia  
  Yes1.00 
  No0.97 (0.81–1.17) 
 Walk independently before discharge  
  Yes1.00 
  No6.57 (4.14–10.4)5.60 (3.52–8.92)
Hospital characteristics  
 Hospital volume  
  >401.00 
  ≤400.97 (0.78–1.19) 
 Surgeon volume  
  >141.00 
  ≤140.98 (0.82–1.17) 

The final multivariate model included age, sex, living arrangement, level of education, obesity, and postoperative functional status. Adjusted analyses confirmed a striking association between the inability to ambulate independently before discharge and discharge to a rehabilitation facility (RR 5.60, 95% CI 3.52–8.92). The adjusted analysis also revealed a significant association between being discharged to a rehabilitation facility and older age (RR 1.36, 95% CI 1.11–1.66), living alone (RR 1.23, 95% CI 1.004–1.5), and obesity (RR 1.29, 95% CI 1.05–1.60). No interaction was found between sex and living arrangement.

Crude analyses of the revision THR sample showed a statistically significant association between being discharged to a rehabilitation facility and female sex (RR 1.53, 95% CI 1.18–1.99), living alone (RR 1.41, 95% CI 1.10–1.80), a low level of education (RR 1.51, 95% CI 1.06–2.15), a low income (RR 1.09, 95% CI 0.99–1.21), and inability to walk after THR (RR 9.09, 95% CI 4.28–19.28). No clinically or statistically meaningful associations were observed between being discharged to a rehabilitation facility and older age, obesity, preoperative functional status, Charlson Index, primary underlying joint disease, previous orthopedic surgeries, ASA score, type of anesthesia, use of PCA after the procedure, and hospital and surgeon volumes (Table 3).

Table 3. Factors associated with discharge to inpatient rehabilitation facility after revision THR*
 Crude RR (95% CI)Adjusted RR (95% CI)
  • *

    For acronym definitions, see Table 2.

  • Adjusted for age, sex, living alone, level of education, income, and obesity.

Patient characteristics  
 Sex  
  Male1.00 
  Female1.53 (1.18–1.99)1.32 (0.98–1.78)
 Age, years  
  ≤721.00 
  >721.23 (0.95–1.58)1.09 (0.84–1.43)
 Living alone  
  No1.00 
  Yes1.41 (1.10–1.80)1.22 (0.93–1.61)
 Education  
  College1.00 
  Less than college1.51 (1.06–2.15)1.24 (0.84–1.84)
 Income  
  ≥$20,0001.001.00
  <$20,0001.09 (0.99–1.21)0.91 (0.65–1.27)
  Missing1.12 (0.99–1.28)0.81 (0.45–1.45)
 Comorbidities  
  <21.00 
  ≥21.04 (0.78–1.38) 
 Prior orthopedic surgery (nonindex joint)  
  No1.00 
  Yes0.97 (0.76–1.24) 
 BMI  
  <301.00 
  ≥300.97 (0.72–1.31)0.96 (0.70–1.31)
 Underlying disease  
  Osteoarthritis1.00 
  Rheumatoid arthritis1.07 (0.86–1.33) 
  Avascular necrosis1.03 (0.82–1.30) 
 Functional status pre-THR  
  Top 3 quartiles1.00 
  Lowest quartile0.98 (0.76–1.25) 
 ASA class  
  ≤II1.00 
  >II1.26 (0.99–1.61) 
 Anesthesia  
  Epidural/spinal1.00 
  Combination/general0.99 (0.75–1.30) 
 Patient-controlled anesthesia  
  Yes1.00 
  No0.91 (0.71–1.16) 
 Walk independently before discharge  
  Yes1.00 
  No9.09 (4.28–19.28)9.13 (4.04–20.61)
Hospital characteristics  
 Hospital volume  
  >401.00 
  ≤400.99 (0.76–1.30) 
 Surgeon volume  
  >141.00 
  ≤140.95 (0.73–1.22) 

The final multivariate model included age, sex, living arrangement, level of education, obesity, and postoperative functional status. As was the case for primary THR, the adjusted analyses showed a striking association between patients' inability to ambulate independently before discharge and going to a rehabilitation facility following revision THR (RR 9.13, 95% CI 4.04–20.61). Patients who lived alone or who had education less than college were also more likely to be discharged to a rehabilitation facility, although these trends did not reach statistical significance.

To address the hypothesis that discharge to rehabilitation facilities was a result of hospitals' own clinical pathway or usual practices after THR, we considered whether there was a hospital effect in our analyses, using the intraclass correlation coefficient (ICC) to assess between and within hospital variability. The ICC for primary THR was 0.1887 and the ICC for revision THR was 0.035.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

We analyzed data from a large population-based cohort to identify factors that differentiated individuals who were discharged to an inpatient rehabilitation program from those who were discharged directly to home after elective THR. The bivariate analyses showed a statistically significant association between discharge to an inpatient rehabilitation facility and worse postoperative functional status, older age, female sex, living alone, lower income, and obesity in the primary THR sample. The multivariate analysis confirmed that worse postoperative functional status before discharge, older age, female sex, living alone, and obesity are independent predictors of discharge to an inpatient rehabilitation facility after a primary THR.

In the revision THR sample, a statistically significant association was also observed between discharge to an inpatient facility and female sex, living alone, low education level, and low income in the bivariate analyses. The multivariate analysis confirmed that worse postoperative functional status before discharge is a strong predictor of discharge to an inpatient rehabilitation facility after a primary THR.

For both primary and revision THR, the most striking predictor of discharge to an inpatient rehabilitation facility was a worse functional status before discharge. Several other studies have addressed the question of which patients go to rehabilitation after being discharged from the acute-care setting. Munin et al found that the factors associated with being discharged to a rehabilitation facility were older age, presence of 2 comorbid conditions, whether the patient lived alone, and the level of pain after surgery (10). Forrest et al found that the factors associated with being discharged to a rehabilitation facility were older age and diabetes mellitus (20). Jones et al found that age was associated with being discharged to inpatient rehabilitation (21). Forrest et al found that age, living alone, and ASA classification were associated with being discharged to a rehabilitation facility (22). In a retrospective cohort study of 146 primary total hip and knee replacements, Mahomed et al found that determinants of home-based rehabilitation included preference for home-based rehabilitation, male sex, and knowledge regarding TJR (13). In general, these studies consistently identify age, measures of comorbidity, and social support as key factors associated with discharge to a rehabilitation facility. Our study confirms the important role of age, sex, and measures of social support (living alone), although our study does not identify comorbidity as a predictor of discharge to an inpatient rehabilitation facility. Our results showed that obesity is an independent predictor of discharge to an inpatient rehabilitation facility after a primary THR. Because obesity is a condition that may impair functional ability, obese patients undergoing THR may need more professional assistance for recovery after surgery.

The most striking finding in our study is that the patients with worse functional status, who were unable to walk without assistance before discharge, were ∼5–9 times more likely to be discharged to a rehabilitation facility. A previous study, in which functional measures before discharge were evaluated by physical therapists (including ambulation for 100 feet), also showed that very few patients with an independent level of activity in the hospital were sent to rehabilitation (10). These studies are consistent in showing that measures of postoperative functional status are strong predictors of discharge destination.

Patients discharged to rehabilitation facilities had shorter lengths of stay than those discharged directly home. Patients considered to be good candidates for inpatient rehabilitation may be sent early to inpatient rehabilitation facilities before they could possibly walk without assistance. Thus, a worse functional status before discharge may, in some cases, be a marker, rather than a reason, for early rehabilitation.

Another factor that may influence whether a patient is discharged to an inpatient facility is the availability of an onsite rehabilitation facility. This could explain a reduced length of stay in the acute-care hospital followed by immediate admission to the onsite inpatient rehabilitation facility. Moreover, there is an increasing focus on reducing total costs in the acute-care hospital and shifting medical services to inpatient rehabilitation facilities (23), particularly since diagnosis related groups were developed in the mid-1980s (24). However, some efforts to reduce costs in the acute-care hospital may be offset by increasing costs outside the hospital (25). Further research should address these economic issues.

Also, we have no information on whether each individual hospital follows a clinical pathway after THR, which would probably affect discharge destination either to an onsite rehabilitation facility or to home. We hypothesized that high-volume hospitals and high-volume surgeons (which usually practice in high-volume hospitals) would have developed and implemented their own clinical pathway, from the immediate postoperative period to the discharge destination; therefore, we looked at that variable. In addition, high-volume hospitals tend to have onsite inpatient rehabilitation facilities, in contrast to small-volume hospitals. To address this hypothesis, we considered whether there was a hospital effect in our analyses, using the ICC to assess between and within hospital variability. The low ICC for primary THR and for revision THR does not support the hypothesis that discharge to rehabilitation facilities was a result of hospitals' own clinical pathway or usual practices. In addition, of 29 hospitals with a volume >30, only 14% had a uniform discharge pattern of all patients in these hospitals from our sample. For the remaining 86%, the probability of going to rehab ranged from 5 to 81% for primary THR.

An important question is whether the use of inpatient rehabilitation facilities improves patient outcome beyond acute hospital care, as the decision regarding discharge destination following elective THR should be directed at improving functional status in the immediate postoperative period. Studies evaluating patients with hip fracture showed early benefits from inpatient rehabilitation programs; however, outcome measures at 6 months and later were not significantly different from those of controls (26, 27). A study evaluating the determinants of rehabilitation setting discharge after TJR and its influence on early functional outcomes found that at a mean followup of 8 months after the surgical procedure there were no significant differences between the inpatient and home-based rehabilitation groups with respect to the Western Ontario and McMasters Universities Osteoarthritis Index, Short Form 36, and satisfaction scores, supporting continued use of home-based rehabilitation (13). Two other studies, however, showed that patients discharged to inpatient rehabilitation following hip replacement had more favorable functional outcomes than patients discharged directly home (28, 29). Another important aim of discharge to rehabilitation facilities may be to ensure appropriate postoperative care to individuals who do not have sufficient social support.

A major difference between previous studies and ours is that fewer patients were discharged to rehabilitation facilities in prior investigations, in contrast with 58% in our population. Our patient population was all covered by Medicare, which reimburses the use of rehabilitation facilities for THR patients relatively easily, whereas the other studies included patients with different types of insurance. In addition, Medicare patients are older, on average, than a typical group of hip replacement patients, and thus more likely to use inpatient rehabilitation.

We acknowledge several limitations in the study. First, the low response rate is a major limitation and the results may therefore not be generalizable to the whole Medicare population. Also, potentially important predictors were not measured, including patient preference and a wider range of functional milestones, such as muscle strength, range of motion, sit-to-stand transfers, supine-to-sit transfers, and lower limb dressing. In addition, we do not have information on the number of physical therapy sessions attended by patients or on the ambulation distance attained before discharge, and we were unable to discriminate what level of rehabilitation the patients received. Preoperative function was measured retrospectively, introducing potential inaccuracy (30). Also, income and living arrangement were assessed 3 years postoperatively. In addition, the prominent effect of age in this sample may be unique to the fact that these are all Medicare patients older than 65 years. Finally, the generalizability of the study is limited to the Medicare population, which comprises about two-thirds of all elective THRs. Overall, about one-third of THRs are done in patients younger than 65 years.

On the other hand, major advantages include the prospective design of the study, its population-based nature, and size of the sample, which is the largest addressing this issue. To the best of our knowledge, no population-based studies have been undertaken to examine factors associated with discharge to an inpatient facility following THR, or discharge determinants of revision THR compared with primary THR. In addition, patients were evaluated by physical therapists for ambulation without assistance after surgery, providing an independent assessment of function prior to discharge.

We conclude that factors associated with being discharged to a rehabilitation facility after elective primary THR are older age, obesity, and social support. However, the most important factor defining discharge destination following elective primary and revision THR is functional status prior to discharge from the acute-care setting. The results of this study have important implications for discharge planning in patients with THR. Indicators of worse postoperative functional status, socioeconomic factors, and social support all influence the discharge destination and must be considered in efforts to optimize and streamline discharge planning. Future research should be directed toward improving functional status in the immediate postoperative period and clarifying the type of rehabilitation and social support needed throughout the continuum of care after elective THR.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES
  • 1
    NIH Consensus Development Panel on Total Hip Replacement. NIH consensus conference: total hip replacement. JAMA 1995; 273: 19506.
  • 2
    Harris WH, Sledge CB. Total hip and total knee replacement. I. N Engl J Med 1990; 323: 72531.
  • 3
    Harris WH, Sledge CB. Total hip and total knee replacement. II. N Engl J Med 1990; 323: 8017.
  • 4
    Gillum BS, Graves EJ, Wood E. National hospital discharge survey. Vital Health Stat 13 1998; 133: 151.
  • 5
    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: 702.
  • 6
    Williamson V. Clinical pathways for a patient with a total joint replacement. Orthop Nurs 1997; 16: 415.
  • 7
    Weingarten S, Riedinger M, Conner L, Siebens H, Varis G, Alter A, et al. Hip replacement and hip hemiarthroplasty surgery: potential opportunities to shorten lengths of hospital stay. Am J Med 1994; 97: 20813.
  • 8
    Morrow-Howell N, Proctor E. Discharge destinations of Medicare patients receiving discharge planning: who goes where? Med Care 1994; 32: 48697.
  • 9
    Kwoh CK, Petrick MA, Munin MC. Inter-rater reliability for function and strength measurements in the acute care hospital after elective hip and knee arthroplasty. Arthritis Care Res 1997; 10: 12834.
  • 10
    Munin MC, Kwoh CK, Glynn N, Crossett L, Rubash HE. Predicting discharge outcome after elective hip and knee arthroplasty. Am J Phys Med Rehabil 1995; 74: 294301.
  • 11
    Sharma L, Sinacore J, Daugherty C, Kuesis DT, Stulberg SD, Lewis M, et al. Prognostic factors for functional outcome of total knee replacement: a prospective study. J Gerontol A Biol Sci Med Sci 1996; 51: M1527.
  • 12
    Wolfe F, Nietfields X, Hedrick R, McElrath T, Ross K. Length of hospital stay and complications of total joint replacement are unrelated to functional status, pain, or psychological variables [abstract]. Arthritis Rheum 1993; 36: S59.
  • 13
    Mahomed NN, Koo M, Levesque J, Lan S, Bogoch ER. Determinants and outcomes of inpatient vs. home based rehabilitation following elective hip and knee replacement. J Rheumatol 2000; 27: 17538.
  • 14
    Katz JN, Phillips CB, Baron JA, Fossel AH, Mahomed NN, Barrett J, et al. Association of hospital and surgeon volume of total hip replacement with functional status and satisfaction three years following surgery. Arthritis Rheum 2003; 48: 5608.
  • 15
    Charlson ME, Sax FL, MacKenzie CR, Fields SD, Braham RL, Douglas RG Jr. Assessing illness severity: does clinical judgment work? J Chronic Dis 1986; 39: 43952.
  • 16
    Keats AS. The ASA classification of physical status: a recapitulation. Anesthesiology 1978; 49: 2336.
  • 17
    National Institutes of Health Consensus Development Conference Statement. Health implications of obesity. Ann Intern Med 1985; 103: 14751.
  • 18
    Mahomed NN, Arndt DC, McGrory BJ, Harris WH. The Harris hip score: comparison of patient self-report with surgeon assessment. J Arthroplasty 2001; 16: 57580.
  • 19
    Katz JN, Losina E, Barrett J, Phillips CB, Mahomed NN, Lew RA, et al. Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population. J Bone Joint Surg Am 2001; 83: 16229.
  • 20
    Forrest G, Fuchs M, Gutierrez A, Girardy J. Factors affecting length of stay and need for rehabilitation after hip and knee arthroplasty. J Arthroplasty 1998; 13: 18690.
  • 21
    Jones CA, Voaklander DC, Johnston DW, Suarez-Almazor ME. The effect of age on pain, function, and quality of life after total hip and knee arthroplasty. Arch Intern Med 2001; 161: 45460.
  • 22
    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: 1924.
  • 23
    Munin MC, Rudy TE, Glynn NW, Crossett LS, Rubash HE. Early inpatient rehabilitation after elective hip and knee arthroplasty. JAMA 1998; 279: 84752.
  • 24
    Zuckerman JD. Inpatient rehabilitation after total joint replacement. JAMA 1998; 279: 880.
  • 25
    Weingarten S, Riedinger MS, Sandhu M, Bowers C, Ellrodt AG, Nunn C, et al. Can practice guidelines safely reduce hospital length of stay? Results from a multicenter interventional study. Am J Med 1998; 105: 3340.
  • 26
    Jette AM, Harris BA, Cleary PD, Campion EW. Fucntional recovery after hip fracture. Arch Phys Med Rehabil 1987; 68: 73540.
  • 27
    Koval K, Su E, Aharonoff G, Zuckerman J. The effectiveness of inpatient rehabilitation after hip fracture. J Bone Joint Surg Am 1998; 80: 35764.
  • 28
    Bitzer EM, Dorning H, Schwartz FW. Effects of rehabilitation after hip replacement surgery on postoperative complaints regarding the disease and limitation of function. Rehabilitation (Stuttg) 2001; 40: 439.
  • 29
    Walker WC, Keyser-Marcus LA, Cifu DX, Chaudhri M. Inpatient interdisciplinary rehabilitation after total hip arthroplasty surgery: a comparison of revision and primary total hip arthroplasty. Arch Phys Med Rehabil 2001; 82: 12933.
  • 30
    Lingard EA, Wright EA, Sledge CB, and the Kinemax Outcomes Group. Pitfalls of using patient recall to derive preoperative status in outcome studies of total knee arthroplasty. J Bone Joint Surg Am 2001; 83: 114956.