Abstract
Background: Most joint-replacement surgeries are currently performed in community hospitals. We sought to determine whether the functional outcomes of joint-replacement surgery differ between academic and community hospitals.
Methods: We surveyed 471 patients for demographic data, Western Ontario McMaster University Osteoarthritis Index (WOMAC) scores and Medical Outcomes Study Short Form 36 (SF-36) scores at baseline and at 3-month and 1-year follow-up. We assessed patient satisfaction at 1 year with a single survey question.
Results: Community hospital patients (n = 269) were significantly older and had greater comorbidity than academic hospital patients (n = 202; p < 0.05). We found no difference in WOMAC scores, SF-36 scores or in patient satisfaction between hospitals at 1-year follow-up (p > 0.05). Adjusted analysis showed that patients undergoing surgery in an academic or community hospitals have the same functional outcomes.
Conclusion: There is no significant difference in the functional outcomes of joint-replacement surgery between academic and community hospitals. Further work will involve evaluating cost of care differences between these types of hospitals.
Total joint arthroplasty (TJA) has been shown to be an effective procedure for relieving pain and improving quality of life.1–3 Literature from both the United States and Canada projects that the number of joint replacements being performed will continue to rise owing to an aging population and a greater proportion of young patients who are undergoing these operations.4–6
Most joint-replacement surgeries are currently being performed in community hospitals, and therefore it is important to evaluate patient outcomes from these hospitals.7,8 Presently, there are few published reports on joint-replacement surgery outcomes from nonteaching hospitals.9–13 Moreover, there are few studies evaluating the costs and resource utilization for TJAs performed at academic and community hospitals.14
The primary goal of our study was to determine whether patient-reported functional outcomes and satisfaction after hip- and knee-replacement surgery differ between academic and community hospitals.
Methods
We performed a case–control study from previously and prospectively collected data to compare patient outcomes for joint-replacement surgery after patient participation in a home-based or inpatient rehabilitation program.15 The original study collected data from 10 randomly selected hospitals across the Greater Toronto Area (GTA; 5 academic and 5 community hospitals). All hospitals were considered to be high-volume centres, performing at least 500 hip- or knee-replacement surgeries each year. All surgeons from the academic hospitals had arthroplasty fellowship training, and all surgeons from all 10 hospitals had been practising in the field for 5 years or more. The study involved patients aged 19 years or older who were undergoing primary hip- or knee-replacement surgery.
We recorded baseline demographic data, including age, sex, diagnosis and comorbidity; comorbidity was defined by the 14 categories of chronic illness adapted from the Cumulative Illness Rating Scale.16,17 The scale covers the domains of cardiac, vascular, hematological, respiratory, otorhinolaryngological and ophthalmological, upper gastrointestinal, lower gastrointestinal, hepatic and pancreatic, renal, genitourinary, musculoskeletal and tegmental, neurologic, endocrine, metabolic and breast, and psychiatric systems. We assessed patients’ functional status and quality of life preoperatively and at 3 months and 1 year postsurgery based on the Western Ontario McMaster University Osteoarthritis Index (WOMAC)18 and the Medical Outcomes Study Short Form 36 (SF-36),19–21 respectively.
We assessed patient satisfaction at 1 year postsurgery with a single question scored on a 4-point Likert scale: “How satisfied were you with the results of your surgery?” We collapsed responses of “very satisfied” and “somewhat satisfied” into a “satisfied” group, and we collapsed responses of “somewhat dissatisfied” and “very dissatisfied” into a “not satisfied” group.
We compared continuous data such as age, body mass index (BMI), comorbidities, SF-36 and WOMAC scores between groups using Student t tests. Means and standard deviations are reported for all continuous variables. Categorical data such as sex and satisfaction are reported with frequencies, and we compared groups using the Fisher exact test.
We performed multivariate linear regression modelling to determine the impact of hospital status on 1-year WOMAC and SF-36 scores. We created separate models for each dependent variable; the relevant covariates entered in the models were age, sex, BMI, comorbidities and the relevant preoperative score.
A minimally clinically significant change on the WOMAC score has been defined as a change of 7.4 points with a standard deviation (SD) of 14 by Bellamy and colleagues.22 A sample size of 471 patients in our study, assuming a type I error rate of 5%, yielded 99% power to detect this difference.
We performed all statistical analyses with SPSS version 13.0 (SPSS Inc.). Beta coefficients for regression modelling and their 95% confidence intervals (CI) are reported. All reported p values are 2-tailed with an α of 0.05.
Results
In our study, there were 202 patients from the academic hospitals and 269 patients from the community hospitals. Three-month follow-up data were available for 384 of 471 (81.5%) patients. In total, 228 of the 384 (75.1%) participants who had complete 3-month follow-up data also had complete 1-year follow-up data.
Patients with complete data did not differ significantly from those with incomplete data for age, sex, comorbidity, baseline total WOMAC or total SF-36 scores.
The patients from the community hospitals were older by 3.2 years and had significantly greater comorbidities than the patients from the academic hospitals (p < 0.05) There were no differences in sex distribution between groups (Table 1).
We found no differences in mean WOMAC and SF-36 scores preoperatively, at 3 months or at 1 year postsurgery between the 2 groups of patients (Table 2). Similarly, we found no significant difference in patient satisfaction between patients at academic and community hospitals at 1 year, with 95.3% and 95.8% reporting satisfaction, respectively (Table 2).
Linear regression modelling showed that hospital status was not a significant predictor of a better 1-year total WOMAC scores or SF-36 scores adjusting for age, sex, comorbidity and the relevant preoperative score (Table 3).
Discussion
The results of our study show that there is no difference in functional outcomes, quality of life or satisfaction between patients undergoing hip- or knee-replacement surgery in an academic institution versus a community hospital in the GTA. This finding is consistent with that of others who have shown that clinical outcomes of joint-replacement surgery are no different between academic and community hospitals.11–13
The costs of care have been shown to be higher in teaching hospitals than in nonteaching hospitals.23–26 One group reported a 22% increase in resource consumption for joint-replacement surgery in an academic hospital versus a nonacademic hospital.14 One hypothesis proposed to explain these increased costs is that the patients treated in an academic centre have greater medical comorbidities.25 Our study demonstrated the opposite finding, in that the patients from the community hospitals had greater medical comorbidities than the patients from the academic hospitals.
One of the strengths of our paper is that we collected data from 10 different hospitals (5 academic and 5 community), which increases the external validity of our study. One potential limitation of our study is that the 1-year response rate was only 228 of 384 (75.1%), despite 2 follow-up phone calls from research staff. However, we found no difference between responders and nonresponders in terms of age, sex, comorbidity or baseline WOMAC or SF-36 scores, and we believe our conclusions remain valid and generalizable.
In conclusion, we have shown that there are no differences in functional outcomes for joint-replacement surgery when performed in academic or community hospitals. Future work will be directed toward understanding the differences in resource utilization between these types of hospitals with the goal of improving the efficiency of care while maximizing resident education, clinical outcomes and patient satisfaction.
Footnotes
Competing interests: None declared.
Contributors: All authors designed the study. Drs. Davis and Mahomed acquired the data, which all authors analyzed. Dr. Gandhi and Ms. Tso wrote the article, which Drs. Davis and Mahomed reviewed. All authors approved publication.
- Accepted September 3, 2008.