Translate this page into:
Patient-reported outcome measures for hip and knee arthroplasty in Ontario, Canada
, Stephen Petersena, Calum Thompsona, James Waddellc, Brent Lantingb, Sarah Wardc, Erik Hellstena
⁎Corresponding author: Steven Habbous. shabbous@uwo.ca
-
Received: ,
Accepted: ,
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Abstract
Increasing uptake of hip and knee arthroplasty to outpatient settings is an attractive option to shorten wait-times, reduce health system costs, and relieve hospital pressures, but the impact on patient-reported outcome measures (PROMs) is unclear.
We examine the population-level uptake of orthopedic PROMs (EuroQoL-5D-5L; Oxford Hip/Knee Score) in Ontario, Canada since April 1, 2019. We explored factors associated with completion of pre-operative survey; PROMs scores; and reaching a minimal clinically significant improvement (MCSD).
From 01Apr2024-31Mar2025, 41% (n = 19130) completed a pre-operative survey within 2 months of surgery, 14% completed a 3-5 month postoperative survey, and 15% completed a 9-15 month postoperative survey. Substantial hospital-level variation in reporting was observed. Pre-operative Oxford scores were higher (better) among outpatients (mean difference 1.8 points), and lower for females (−2.4 points), patients undergoing hip replacement (−2.1 points), and patients having higher comorbidity. A MCSD was observed for 53% of patients 3-5 months post-operatively and was more likely among outpatients [OR 1.28 (1.12-1.45)], hip replacement patients, rural residents, and among those having a higher baseline EQ5D score. Patients were less likely to reach the MCSD if they had a comorbidity score ≥4, if they resided in a long-term care setting, or resided in areas with higher concentrations of racialized or newcomer populations.
Efforts to improve orthopedic PROMs reporting have had some success, but completion rates fall short of the 60% target. There was no evidence that patients undergoing outpatient hip/knee arthroplasty experienced worse QoL than inpatients, supporting ongoing efforts to promote outpatient surgery.
Keywords
Hip arthroplasty
Knee arthroplasty
Inpatient
Ambulatory care
Patient-reported outcome measures
PROMs
1 Introduction
Hip and knee arthroplasties are among the most common surgical procedures performed in hospitals, demonstrating year-over-year increases between 2010 and 2022.1 While waiting for surgery, patients experience worse health-related quality-of-life (HR-QoL) compared with population controls, with some experiencing deterioration while they wait.2,3
The use of standardized and validated tools to assess HR-QoL has been identified by a Delphi panel of experts as important for quality improvement, supported by high-quality evidence.4 Since 2019, the Ontario Ministry of Health has been working with Ontario Health to support hospitals implementing and collecting patient-reported outcome measures (PROMs) data. However, little is known about the uptake of orthopedic PROMs in the province. Moreover, with increased utilization of outpatient hip/knee arthroplasty, little is known about whether PROMs differ between these two settings. There is some evidence that the outpatient experience is generally positive, but previous studies comparing QoL between outpatients and inpatients have been limited by small sample sizes and were not population-based.5–9.
In this study, we 1) examine the uptake of hip and knee QoL instruments in Ontario over time; 2) report the percentage of patients providing a QoL assessment before and after elective hip and knee arthroplasty; and 3) compare QoL reported by patients undergoing surgery in outpatient and inpatient settings.
2 Methods
2.1 Setting
All elective hip and knee arthroplasties were performed in a publicly-funded Ontario hospital with data collection mandated by the Canadian Institute for Health Information (CIHI). Day surgeries were captured using the National Ambulatory Care Reporting System (CIHI-NACRS) or the Discharge Abstract Database (CIHI-DAD) with a same-day discharge. All other procedures captured using the CIHI-DAD were inpatient procedures.
2.2 PROMs data collection
The Ministry of Health has proactively expanded its musculoskeletal strategy by implementing systematic collection of PROMs for patients undergoing elective hip/knee replacement surgery in Ontario. This initiative was part of the orthopedic Quality Based Procedure (QBP) program, a volume-based funding model that remunerates hospitals for each primary hip/knee replacement surgery adjusted to reflect patient complexity. Hospitals were required to collect PROMs data to be eligible for QBP funding (all 57 hospitals with an orthopedic clinic in Ontario as of April 1, 2019). During the study period, there were three required time-points for PROMs collection: 1) within 2 months before surgery; 2) within 3-5 months after surgery; and 3) 9-15 months after surgery.
Patients complete the PROMs surveys using Ontario's Interactive Symptom Assessment and Collection (ISAAC) Patient Portal, which was launched in 2007 to facilitate PROMs data collection for patients diagnosed with cancer.10 PROMs surveys can be completed using mobile devices, paper, tablets, web forms, email, or in-clinic kiosks.
2.2.1 Instruments
The Canadian version of the EQ-5D-5L and visual analog scale (VAS) were used as generic measures of health. The EQ-5D-5L is a five-question QoL survey that asks about mobility, self-care, usual activities, pain/discomfort, and anxiety/depression, ranging from 1 (lowest severity) to 5 (highest severity). A higher score indicates worse QoL. The EQ-5D-5L-VAS is a single question asking raters to assess their current health using a thermometer ranging from 0 (worst health imaginable) to 100 (best health imaginable). A higher score indicates better QoL.
The Oxford Hip Score (OHS) and Oxford Knee Score (OKS) are joint-specific instruments for hip/knee conditions, are specific to a single laterality, and have high validity.11,12 These are 12-question surveys that include questions about pain, trouble with daily activities, and self-care specific to the joint in question. Each question is ranked from 0 (most severe symptoms) to 4 (few or no symptoms). A higher score indicates better QoL. The OHS and OKS can be mapped to the EQ-5D-5L as an estimate of utility to enable cost-effectiveness analyses.13,14 The Oxford scores were further separated into their pain and function subscales based on the factor analysis performed by Harris et al.15,16 We used the factor loading scores to weigh each question and normalized the score by dividing by the sum of the maximum possible weight so the maximum value is 100.
All orthopedic PROMs surveys were extracted from ISAAC on July 8, 2025 with a survey date April 1, 2019 or later. Surveys were deduplicated (the most recent survey per patient per day per joint per laterality was retained). To allow for sufficient follow-up, the most recent surgery permitted was September 30, 2024 and December 31, 2023 for estimating 3-5 month and 9-15 month PROM survey completion, respectively.
2.3 Covariates
Sociodemographic characteristics included age at surgery and sex from DAD/NACRS. Charlson comorbidity index was estimated using a 3-year look-back period from the date of surgery.17 Patients’ postal code at the time of surgery was linked to the Postal Code Conversion File (PCCF+) to ascertain rurality and to the Ontario Marginalization Index (2021) to ascertain area-level quintiles of material resources and the proportion of the neighbourhood comprised of racialized persons or newcomers.18 Obesity, residence in long-term care, and orthopedic surgeon visits were captured using the Ontario Health Insurance Program (OHIP) database (eTable S1).
2.4 Statistical methods
This study was intended to be primarily descriptive. For exploratory analysis, we used logistic regression to examine which factors were associated with PROMs completion and linear regression to measure factors associated with PROMs score. All regression models were adjusted for patient age, patient sex, comorbidity score, outpatient/inpatient setting, joint type (hip/knee), bilateral/unilateral, obesity, long-term care residence, rurality, and area-level material resource and racialized/newcomer quintiles, unless otherwise stated. Since small changes in PROMs may be statistically significant without clinical relevance, we also explored factors associated with achieving a minimal clinically significant difference (MCSD), defined as an improvement of at least 5 points on the same instrument.19 To estimate hospital-to-hospital variability in PROMs completion, we estimated the intercluster correlation coefficient using a random intercepts model. P-values <0.05 were considered statistically significant. All analyses were conducted using Statistical Analysis Software v9.4 (SAS Institute Inc., Cary NC).
2.5 Privacy and ethics
Research ethics was not required as per the Ontario Health privacy assessment as this work was performed for the purpose of quality improvement and no identifying information was obtained. This study was compliant with section 45(1) of PHIPA (Ontario Health is a prescribed entity); thus, patient consent was not required.
3 Results
3.1 PROMs survey completion
The number of PROMs surveys increased over time despite a pause during the COVID-19 pandemic (Fig. 1).
Linking with hospital records, a total 413,906 hip and knee arthroplasties were observed for 324,227 unique patients (eFigure S1). Year-over-year, the proportion of patients proceeding to surgery after completing a PROMs survey increased from 79% in FY2019/20 to 98% in FY2022/23 and remained stable since then (Table 1).
| 2019/20 | 2020/21 | 2021/22 | 2022/23 | 2023/24 | 2024/25 | |
| EQ-5D-5L Survey | ||||||
| Surgery after EQ-5D-5La | 11902 (79%) | 8150 (87%) | 9086 (89%) | 11816 (98%) | 13005 (96%) | 7782 (96%) |
| EQ-5D-5L relative to surgery | ||||||
| Pre-op (2 months)a | 10170 (25%) | 10985 (28%) | 13176 (34%) | 18925 (35%) | 21059 (35%) | 19130 (41%) |
| Post-op 1 (3-5 months)b | 1950 (4.0%) | 2732 (6.9%) | 3478 (8.9%) | 5464 (10%) | 5827 (9.8%) | 4430 (14%) |
| Post-op 2 (9-15 months)c | 3442 (7.0%) | 4023 (10%) | 4812 (12%) | 6756 (13%) | 6570 (15%) | - |
| Post-op (0-15 months)c | 7155 (15%) | 6794 (17%) | 7950 (20%) | 12202 (23%) | 9743 (22%) | - |
| Oxford Hip/Knee Survey | ||||||
| Pre-op (2 months)a | 10318 (26%) | 11046 (28%) | 13475 (34%) | 19489 (36%) | 21535 (36%) | 19520 (42%) |
| Post-op 1 (3-5 months)b | 1967 (4.0%) | 2747 (7.0%) | 3494 (8.9%) | 5504 (10%) | 5830 (9.8%) | 4439 (14%) |
| Post-op 2 (9-15 months)c | 3447 (7.0%) | 4028 (10%) | 4843 (12%) | 6812 (13%) | 6607 (15%) | - |
Preoperative EQ-5D-5L survey completion increased from 25% in FY2019/20 to 41% by FY2024/25 (Table 1). Post-operative survey participation also increased over time but remained low, increasing from 4.0% in FY2019/20 to 14% in FY2024/25 for the 3-5 months post-operative timepoint. The 1-year post-operative survey response rate was higher and increased from 7.0% in FY2019/20 to 15% in FY2023/24. Similar trends were observed for Oxford Hip/Knee Surveys (Table 1). Only 8503/238,280 (3.6%) completed the EQ-5D-5L for all three time-points (surgeries performed between January 1, 2019 and December 31, 2023).
The median time until surgery from the most recent EQ-5D-5L survey was 13 days. The time from surgery until the first post-operative EQ-5D-5L was not a smooth curve, with surges of patients receiving their first post-operative survey 2-months or 3-months after surgery with no clear pattern until the 1-year time point (Fig. 2).
The distribution of the Oxford scores on the pain and function subscales by survey time point are shown in Fig. 3, demonstrating improvement over time and a potential ceiling effect: by 9-15 months post-operatively, 3568 (13%) reported a perfect score on the Oxford scale, 6825 (25%) on the pain subscale, and 1428 (5.2%) on the function subscale.
3.1.1 Factors associated with pre-operative PROMs completion
During FY2024/25, completion of a pre-operative PROMs survey was significantly less likely among older adults, particularly for octogenarians [OR 0.79 (0.70-0.88)] and nonagenarians [OR 0.63 (0.48-0.82)] compared with patients aged 18-49 years. Completion was more likely among females [OR 1.09 (1.05-1.13)], and patients having obesity [OR 1.20 (1.10-1.31)] (Table 2). Residents living in rural areas [OR 1.09 (1.03-1.15)] and areas with the higher levels of material resources were significantly more likely to complete a pre-operative PROMs.
| Surgeries performed June 1, 2019 – March 31, 2025 (N = 295,243) | Surgeries performed in FY2024/25 (n = 63,002) | |||||
| Age group, years | N (%) | OR (95% CI)a | p-value | N (%) | OR (95% CI)a | p-value |
| 18-49 | 8963 (3%) | 1.0 (ref) | <0.0001 | 1766 (3%) | 1.0 (ref) | <0.0001 |
| 50-59 | 42,838 (14%) | 1.04 (0.99-1.10) | 7984 (13%) | 1.09 (0.98-1.22) | ||
| 60-69 | 106,247 (36%) | 1.08 (1.03-1.14) | 22,347 (35%) | 1.05 (0.95-1.17) | ||
| 70-79 | 103,289 (35%) | 1.05 (1.00-1.10) | 23,127 (36%) | 1.02 (0.91-1.13) | ||
| 80-89 | 32,134 (11%) | 0.92 (0.87-0.97) | 7409 (12%) | 0.79 (0.70-0.88) | ||
| 90+ | 1772 (1%) | 0.78 (0.69-0.88) | 368 (1%) | 0.63 (0.48-0.82) | ||
| Female vs Male | 169,831 (58%) | 1.03 (1.01-1.05) | 0.0006 | 36,334 (58%) | 1.09 (1.05-1.13) | <0.0001 |
| Outpatient vs Inpatient | 96,281 (33%) | 1.20 (1.18-1.22) | <0.0001 | 29,654 (47%) | 1.01 (0.97-1.04) | 0.74 |
| Hip vs Knee | 114,751 (39%) | 1.10 (1.08-1.12) | <0.0001 | 23,783 (38%) | 1.11 (1.07-1.15) | <0.0001 |
| Bilateral vs Unilateral | 4352 (1.5%) | 0.72 (0.67-0.78) | <0.0001 | 646 (1.0%) | 0.80 (0.68-0.96) | 0.01 |
| Rurality | ||||||
| Urban | 245,579 (83%) | 1.0 (ref) | 0.57 | 52123 (83%) | 1.0 (ref) | 0.002 |
| Rural | 47,005 (16%) | 1.01 (0.98-1.03) | 9928 (16%) | 1.09 (1.03-1.15) | ||
| Missing | 2659 (1%) | - | 951 (2%) | - | ||
| Comorbidity | ||||||
| 0 | 245,149 (83%) | 1.0 (ref) | 0.07 | 53,161 (84%) | 1.0 (ref) | 0.72 |
| 1 | 26,747 (9%) | 1.04 (1.01-1.07) | 5064 (8%) | 1.04 (0.98-1.11) | ||
| 2 | 14,915 (5%) | 1.01 (0.97-1.04) | 3076 (5%) | 1.00 (0.93-1.08) | ||
| 3 | 4860 (2%) | 0.97 (0.91-1.03) | 935 (1%) | 1.04 (0.91-1.20) | ||
| 4+ | 3572 (1%) | 0.96 (0.89-1.04) | 766 (1%) | 1.01 (0.86-1.18) | ||
| Obesity | 10,134 (3.4%) | 1.21 (1.16-1.27) | <0.0001 | 2406 (3.8%) | 1.20 (1.10-1.31) | <0.0001 |
| Long-term care | 1719 (0.6%) | 0.79 (0.70-0.89) | 0.0002 | 383 (1%) | 0.92 (0.73-1.17) | 0.51 |
| Material resources | ||||||
| 1 (highest) | 62,499 (21%) | 1.0 (ref) | <0.0001 | 13546 (22%) | 1.0 (ref) | <0.0001 |
| 2 | 60,757 (21%) | 0.89 (0.87-0.91) | 12965 (21%) | 0.92 (0.88-0.97) | ||
| 3 | 57,224 (19%) | 0.90 (0.88-0.92) | 12209 (19%) | 0.90 (0.86-0.95) | ||
| 4 | 51,319 (17%) | 0.82 (0.80-0.84) | 10781 (17%) | 0.88 (0.84-0.93) | ||
| 5 (lowest) | 43,825 (15%) | 0.76 (0.74-0.78) | 8930 (14%) | 0.83 (0.78-0.88) | ||
| Missing | 19,619 (7%) | - | 4571 (7%) | - | ||
| Ethnic concentration | ||||||
| 1 (lowest) | 69,977 (24%) | 1.0 (ref) | <0.0001 | 14870 (24%) | 1.0 (ref) | <0.0001 |
| 2 | 63,956 (22%) | 1.11 (1.09-1.14) | 13493 (21%) | 1.17 (1.11-1.23) | ||
| 3 | 55,852 (19%) | 1.16 (1.13-1.19) | 11880 (19%) | 1.21 (1.14-1.28) | ||
| 4 | 48,933 (16%) | 1.11 (1.08-1.14) | 10387 (16%) | 1.16 (1.10-1.23) | ||
| 5 | 36,906 (12%) | 0.93 (0.90-0.96) | 7801 (12%) | 1.01 (0.95-1.07) | ||
| Missing | 19,619 (7%) | - | 4571 (7%) | - | ||
| Physician sex | ||||||
| Male | 273,281 (93%) | 1.0 (ref) | <0.0001 | 58,195 (92%) | 1.0 (ref) | 0.82 |
| Female | 13,126 (4%) | 1.38 (1.33-1.43) | 3017 (5%) | 0.99 (0.92-1.07) | ||
| Missing | 8836 (3%) | - | 1790 (3%) | - | ||
| Physician age | ||||||
| N | 286,407 | - | 61,212 | - | ||
| Mean (SD) | 50.3 (9.7) | - | 50.2 (9.5) | - | ||
| - | ||||||
| 30 to 39 years | 48,683 (16%) | 1.0 (ref) | <0.0001 | 10,851 (17%) | 1.0 (ref) | <0.0001 |
| 40 to 49 years | 97,596 (33%) | 1.07 (1.04-1.10) | 20,599 (33%) | 1.08 (1.02-1.13) | ||
| 50 to 59 years | 89,678 (30%) | 0.94 (0.92-0.96) | 18,870 (30%) | 1.00 (0.95-1.05) | ||
| 60+ years | 50,450 (17%) | 0.74 (0.72-0.76) | 10,892 (17%) | 0.67 (0.63-0.71) | ||
| Missing | 8836 (3%) | - | 1790 (3%) | - | ||
Patients undergoing hip replacements [OR 1.11 (1.07-1.15)] were more likely to have completed a PROMs survey, and those that received a bilateral replacement [OR 0.80 (0.68-0.96)] or had a surgeon aged 60+ years [OR 0.67 (0.63-0.71) versus 30-39 years] were less likely to have completed a survey. Outpatients were more likely to complete a pre-operative PROMs survey over the entire study cohort [OR 1.20 (1.18-1.22)], an effect that was lost when restricting to surgeries performed in FY2024/25 [OR 1.01 (0.97-1.04)].
3.1.2 Factors associated with post-operative PROMs completion
Octogenarians, nonagenarians, patients having the highest comorbidity score (4+), and rural residents were less likely to complete either post-operative survey (eTable S2). Completion of post-operative surveys was more likely for patients who completed the preoperative surveys, with 10-fold higher odds of completing the 3-5 month follow-up [OR 10.2 (9.39-11.2)] and over 7-fold higher odds of completing the 9-15 month follow-up [OR 7.60 (7.13-8.09)]. Outpatient surgery exhibited qualitatively different results: outpatients were less likely to complete the 3-5 month follow-up survey [OR 0.80 (0.74-0.86), p < 0.0001] but more likely to complete the 9-15 month follow-up survey [OR 1.13 (1.06-1.20), p < 0.0001]. Patients were more likely to complete the follow-up surveys if they also had a visit with an orthopedic surgeon during that time [OR3-5mos 1.27 (1.18-1.37); OR9-15mos 2.02 (1.91-2.15)]. Despite this, only 10728 (34%) of patients had a visit with an orthopedic surgeon during the 3-5 months follow-up period, 1740 (16%) of whom completed a PROMs survey during that time period. Similarly, 20725 (47%) had a visit with an orthopedic surgeon during the 9-15 months follow-up window, with a PROMs completion rate of 19% (n = 4018). Expanding the window to 2-15 months after surgery, 81% had a visit with an orthopedic surgeon, but only 22% completed a postoperative PROMs survey. There was substantial regional variability in the completion of pre-operative PROM surveys (Fig. 4). After adjustment, hospitals accounted for 57% of the total variation in the relative odds of preoperative PROMs completion.
3.2 PROMs by hospital setting
3.2.1 Pre-operative
Pre-operative PROMs on all instruments demonstrated better QoL among patients receiving outpatient arthroplasty, although differences were small (Table 3; Fig. 5). Using the Oxford score as the outcome, we examined factors associated with a higher (better) score using linear regression. Excluding bilateral surgeries, the QoL was higher by 1.8 points for outpatients and higher by 2.8 points for patients aged 60-89 compared with <60 years of age (p < 0.0001). QoL was worse among females (−2.4 compared with males), hip replacements (−2.1 points), residents of long-term-care (−3.1 points), patients having obesity (−1.7 points), among patients having greater comorbidity (−2.8 points for a comorbidity score 4+), and patients residing in areas having fewer material resources (−2.1 points for the highest quintile of marginalization) (Table 4).
| EQ5D score (lower better) | YHT score (higher better) | Oxford score (higher better) | ||||
| Left hip | Right hip | Left knee | Right knee | |||
| Setting (whole cohort) | ||||||
| Inpatient | 13.8 (3.7) | 64.9 (20.4) | 19.6 (9.9) | 19.5 (9.9) | 21.5 (9.1) | 21.3 (9.2) |
| Outpatient | 12.9 (3.5) | 69.6 (19.2) | 22.5 (10.1) | 22.3 (10.2) | 23.4 (9.1) | 23.4 (9.2) |
| Setting (FY2024/25) | ||||||
| Inpatient | 13.7 (3.8) | 64.7 (20.3) | 20.4 (10.5) | 20.3 (10.4) | 22.2 (9.6) | 22.1 (9.6) |
| Outpatient | 12.8 (3.5) | 69.6 (19.2) | 23.1 (10.3) | 22.7 (10.4) | 23.9 (9.1) | 23.8 (9.1) |
| Preoperative EQ-5D score (surgeries performed in FY2024/25) | Reaching MCSD within 15 months follow-up | |||
| Age group | Effect (95% CI)a | p-value | OR (95% CI)b | p-value |
| 18-49 | 0 (ref) | <0.0001 | 1.0 (ref) | 0.71 |
| 50-59 | 0.6 (−0.1, 1.3) | 0.95 (0.66-1.38) | ||
| 60-69 | 1.7 (1.0, 2.4) | 1.08 (0.76-1.52) | ||
| 70-79 | 2.8 (2.1, 3.5) | 1.10 (0.77-1.55) | ||
| 80-89 | 2.8 (2.1, 3.6) | 1.00 (0.68-1.49) | ||
| 90+ | −1.1 (−2.9, 0.7) | 1.44 (0.48-4.31) | ||
| Female vs male | −2.4 (−2.6, −2.2) | <0.0001 | 0.94 (0.84-1.07) | 0.35 |
| Outpatient vs inpatient | 1.8 (1.6, 2.0) | <0.0001 | 1.28 (1.12-1.45) | 0.0002 |
| Hip vs knee | −2.1 (−2.3, −1.8) | <0.0001 | 1.47 (1.30-1.67) | <0.0001 |
| Bilateral vs unilateral | - | - | 1.99 (0.99-4.00) | 0.05 |
| Comorbidity | ||||
| 0 | 0 (ref) | <0.0001 | 1.0 (ref) | 0.004 |
| 1 | −1.7 (−2.1, −1.3) | 0.82 (0.66-1.03) | ||
| 2 | −0.8 (−1.3, −0.2) | 1.00 (0.75-1.35) | ||
| 3 | −2.4 (−3.4, −1.5) | 0.95 (0.55-1.63) | ||
| 4+ | −2.8 (−3.8, −1.7) | 0.33 (0.18-0.61) | ||
| Obesity | −1.7 (−2.3, −1.2) | <0.0001 | 0.82 (0.57-1.17) | 0.26 |
| Long-term care | −3.1 (−4.6, −1.6) | <0.0001 | 0.27 (0.08-0.90) | 0.03 |
| Rural vs urban | −0.2 (−0.5, 0.1) | 0.22 | 1.31 (1.09-1.59) | 0.005 |
| Material resources | ||||
| 1 | 0 (ref) | <0.0001 | 1.0 (ref) | 0.82 |
| 2 | −0.6 (−0.9, −0.3) | 0.95 (0.81-1.13) | ||
| 3 | −0.9 (−1.2, −0.6) | 0.93 (0.78-1.12) | ||
| 4 | −1.7 (−2.0, −1.3) | 0.89 (0.74-1.08) | ||
| 5 | −2.1 (−2.4, −1.7) | 0.93 (0.76-1.14) | ||
| Racialized/Newcomer | ||||
| 1 | 0 (ref) | 0.0007 | 1.0 (ref) | 0.0004 |
| 2 | −0.2 (−0.5, 0.1) | 1.03 (0.86-1.23) | ||
| 3 | 0.0 (−0.4, 0.4) | 1.14 (0.93-1.39) | ||
| 4 | 0.1 (−0.2, 0.5) | 0.96 (0.78-1.19) | ||
| 5 | −0.7 (−1.1, −0.3) | 0.67 (0.53-0.86) | ||
| Baseline EQ5D | - | - | 1.75 (1.70-1.80) | <0.0001 |
3.2.2 Minimal clinically significant difference
A MCSD (a reduction by more than 5 points in the EQ-5D-5L) was observed for 7373/14,014 of patients (53%) by 3-5 months postoperatively (Fig. 6) and 10879/19039 (57%) by 1-year. Among 8503 patients with all three surveys, a total 1219 (14%) reached the MCSD by 1-year and not 3-5 months; 645 (8%) reached the MCSD by 3-5 months, which was lost by 1-year; 2697 (32%) did not reach the MCSD at either time-point, and 3942 (46%) reached the MCSD at 3-5 months and stayed there at 1-year.
Considering the 5806/8503 (68%) patients who reached the MCSD at either the 3-5 month or 1-year time-points as the outcome, patients were more likely to reach the MCSD if they were outpatients [OR 1.28 (1.12-1.45)], received hip replacement [OR 1.47 (1.30-1.67)], lived in a rural area [OR 1.31 (1.09-1.59)], and had a higher (worse) baseline EQ-5D-5L score [1.75 (1.70-1.80) per point] (Table 4). Patients were less likely to reach the MCSD if they had a comorbidity score of 4 or higher [OR 0.33 (0.18-0.61)], if they resided in a LTC setting [OR 0.27 (0.08-0.90)], or resided in an area having the highest density of racialized persons or newcomers [OR 0.67 (0.53-0.86)].
4 Discussion
In this study we observed increasing completion of PROMs by hip and knee arthroplasty patients over time, but completion rates were below the 60% target recommended by the International Society of Arthroplasty Registries.20 We also found that outpatients were more likely to improve than inpatients, and the majority who achieved a clinically significant improvement did so within 5 months of surgery.
One single-centred study in the United States reported a response rate of 61% for matched pairs (pre and post), exceeding the 50% quality-based funding target for that study population.21 Another single-centred study in Michigan (United States) reported that despite increased reporting between 2021 and 2022, the proportion of completed matched pairs was 22%, less than the 33% target reported by the authors.22,23 In our study, access to follow-up appointments with orthopedic surgeons may only partly explain low follow-up response rates, but even among patients who did have a follow-up orthopedic surgical visit completion rates were only 16-19%. Substantial regional variation was observed, suggesting that differences in hospital culture toward PROMs, availability of resources to facilitate PROMs completion, or perceived importance of PROMs collection by patients or providers may be contributing factors.24,25
Because of the financial burden of data collection and the observation that most of the clinically relevant improvements occurred within the first 6 months after surgery, one study recommended collecting data once and only up to 6 months post-operatively.26 An Australian study estimated a cost of US $5-12 for two PROMs surveys per surgery following the implementation of an electronic PROMs collection system.27 One systematic review found evidence that there were clinically important changes after six months in England, more notable for hip replacements, although their definition of a minimal important difference differed from ours.28 Others recommend using a 1-year time point following total knee replacement.29 The International Society of Arthroplasty Registries recommends 1-year follow-up after hip and 18 months follow-up after knee replacement, but we acknowledge that earlier follow-up may be associated with less loss to follow-up and can provide earlier opportunity for intervention if patients are not improving as quickly as expected.20
4.1 Further support for outpatient hip and knee arthroplasty
In addition to being cost saving with no demonstrable adverse clinical outcomes, outpatient hip/knee arthroplasty remains a favourable choice among appropriately selected patients.1,30,31 By FY2024/25, nearly half of all hip and knee arthroplasty patients in the province received outpatient surgery. Given the dearth of information about QoL across these settings in the literature, our findings continue to support outpatient arthroplasty, but we interpret this cautiously owing to low response rates and the presence of unmeasured confounders. Previous studies were limited to small sample sizes or were not population-based. One randomized controlled study in London, Ontario randomized patients to the outpatient setting (n = 56) or the inpatient setting (n = 49) and reported that outpatients experienced similar pain and stiffness scores but a better functional score at 12 weeks using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) but a non-significant difference in MCSD.6 Another small (n = 84) randomized controlled trial from China reported no difference in the OHS after total hip arthroplasty.7 One single-centred study from New York (n = 963) reported no difference in the change of PROMs by setting.8,9
4.2 Future directions
Measures of healthcare quality tend to focus on process indicators because they are timely, objective, and can often be measured using administrative data.32 However, the patient's perspective should be considered more directly.4,33 Previous population-based data from Ontario and the United States report that <2% of patients experience an adverse outcome within 1 year (composite of revision, infection, or all-cause mortality).1,34 Thus, PROMs can be used to estimate safety, effectiveness, or cost-effectiveness of interventions, such as evaluating specific modifications to the surgical approach.35–37 One study from the United Kingdom reported improved PROMs over time, suggesting that instruments be used that better discriminate among patients reporting high scores.38 This may be particularly important for the OHS subscales, which may demonstrate some ceiling effects post-operatively.39 Hospitals underreporting PROMs should be the focus of quality improvement, which could potentially influence QBP funding.25 With 48% of patients visiting an orthopedic surgeon 9-15 months after surgery, there is potential opportunity to support surgeons and primary care providers to provide PROMs at preschedule postoperative visits.40
Long wait-times for hip/knee replacement, the demonstrable success with outpatient surgery, and cost-savings associated with avoiding postoperative hospitalization has enabled hip/knee replacement surgery to be considered a candidate procedure for independent health facilities.31,41,42 The call for applications for licenses also mandates PROMs collection and reporting as well as data entry into the Canadian Joint Replacement Registry, which is crucial to ensure quality of care is provided in hospitals and surgical centres alike.20,43 Reporting through the Wait Time Information System should also be mandated to allow policymakers to understand the impact of independent health facilities on surgical wait-times in a universal health system.
4.3 Limitations
Owing to non-response bias, we may have a non-representative sample of survey respondents and the true difference in QoL between outpatient and inpatient hip/knee replacement remains uncertain.44–46 Moreover, given the subjective nature of PROMs, it is also possible that responses may be influenced by perceptions associated with the treatment setting.47 For example, a patient may indicate better health if they believe they are healthy enough to receive outpatient surgery.
4.4 Conclusion
Efforts should prioritize improving PROMs completeness and aligning data collection windows with relevant periods of recovery. To balance costs, administrative burden, low completion rates, and clinical utility, our findings support a six-month target for PROMs collection following hip and knee arthroplasty. This is supported because the majority of patients reaching a MCSD did so by the 3-5 month time-point and a ceiling effect was more prevalent with longer follow-up. Earlier follow-up (e.g., six months) can allow modification of rehabilitation protocols for patients identified early who could benefit from it, while those who are likely to achieve perfect pain/function scores could be provided with more sensitive instruments at subsequent visits.
5 Data source disclosures
Parts of this material are based on data and information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed herein are those of the author, and not necessarily those of the CIHI. Parts of this publication are based on data provided by the ICES. However, the views expressed in this publication are those of the researcher and do not necessarily represent those of the ICES. This report was produced with the support of the Ontario Ministry of Health. However, the views expressed herein are those of the author and not necessarily those of the Ontario Ministry of Health or the Government of Ontario.
Author statement - CRediT
Conceptualization (SH, EH); Writing – original draft (SH); Writing – review and editing (all authors); Methodology (all authors); Formal Analysis (SH); Data Curation (SH); Investigation and Validation (all authors).
Funding
No specific sources of funding were used.
References
- The successful and safe conversion of joint arthroplasty to same-day surgery: a necessity after the COVID-19 pandemic. PLoS One. 2023;18(11)
- [Google Scholar]
- The Manitoba arthroplasty waiting list: impact on health-related quality of life and initiatives to remedy the problem. J Eval Clin Pract. 2009;15(1):208-211.
- [Google Scholar]
- Health-related quality of life in patients waiting for major joint replacement. A comparison between patients and population controls. Health Qual Life Outcome. 2006;4
- [Google Scholar]
- Development of quality indicators for hip and knee arthroplasty rehabilitation. Osteoarthr Cartil. 2018;26(3):370-382.
- [Google Scholar]
- Outpatient total hip and knee arthroplasty – patient expectations versus experience. J Orthop. 2024;51:109-115.
- [Google Scholar]
- The cost of outpatient versus inpatient total hip arthroplasty: a randomized trial. Can J Surg. 2022;65(5):E553-E561.
- [Google Scholar]
- Cost-effectiveness of same-day discharge surgery for primary total hip arthroplasty: a pragmatic randomized controlled study. Front Public Health. 2022;10
- [Google Scholar]
- Similar outcomes after hospital-based same-day discharge vs inpatient total hip arthroplasty. Arthroplast Today. 2020;6(3):451-456.
- [Google Scholar]
- A formal same-day discharge total joint arthroplasty program may not be necessary: a propensity-matched cohort study. J Arthroplast. 2022;37(8):S823-S829.
- [Google Scholar]
- Selecting, implementing and evaluating patient-reported outcome measures for routine clinical use in cancer: the cancer care Ontario approach. J Patient-Rep Outcomes. 2020;4(1)
- [Google Scholar]
- Assessing responsiveness of the EQ-5D-3L, the Oxford hip score, and the Oxford knee score in the NHS patient-reported outcome measures. J Orthop Surg Res. 2021;16(1)
- [Google Scholar]
- Systematic review of measurement properties of patient-reported outcome measures used in patients undergoing hip and knee arthroplasty. Patient Relat Outcome Meas. 2016;7:101-108.
- [Google Scholar]
- Mapping the Oxford hip score onto the EQ-5D utility index. Qual Life Res. 2013;22(3):665-675.
- [Google Scholar]
- Mapping analysis to predict EQ-5D-5 L utility values based on the oxford hip score (OHS) and oxford knee score (OKS) questionnaires in the Spanish population suffering from lower limb osteoarthritis. Health Qual Life Outcome. 2020;18(1)
- [Google Scholar]
- Can pain and function be distinguished in the Oxford knee score in a meaningful way? An exploratory and confirmatory factor analysis. Qual Life Res. 2013;22(9):2561-2568.
- [Google Scholar]
- Can pain and function be distinguished in the Oxford hip score in a meaningful way?: an exploratory and confirmatory factor analysis. Bone Joint Res. 2014;3(11):305.
- [Google Scholar]
- Charlson comorbidity index: ICD-9 update and ICD-10 translation. Am Health Drug Benefits. 2019;12(4):188.
- [Google Scholar]
- The 2011 and 2016 iterations of the Ontario marginalization index: updates, consistency and a cross-sectional study of health outcome associations. Can J Public Health. 2022;113(2):260-271.
- [Google Scholar]
- Meaningful changes for the Oxford hip and knee scores after joint replacement surgery. J Clin Epidemiol. 2015;68(1):73-79.
- [Google Scholar]
- Patient-reported outcome measures in arthroplasty registries: report of the patient-reported outcome measures working group of the international society of arthroplasty registries: part II. Recommendations for selection, administration, and analysis. Acta Orthop. 2016;87(Suppl 1):9.
- [Google Scholar]
- Evaluating patient-reported outcome measure collection and attainment of substantial clinical benefit in total joint arthroplasty patients. J Arthroplast. 2025;40(6):1452-1459.
- [Google Scholar]
- Challenges in meeting centers for medicare and medicaid services patient-reported outcome measures collection requirements and patient predictors of substantial clinical benefit achievement in total joint arthroplasty. J Arthroplast. 2025;40(9S1):S131-S136.
- [Google Scholar]
- Achieving the centers for medicare and medicaid services defined substantial clinical benefit following total knee arthroplasty and total hip arthroplasty in the Michigan arthroplasty registry collaborative quality initiative. J Arthroplast. 2025;40(9S1):S92-S99.
- [Google Scholar]
- Patient experiences with patient-reported outcome measures: an interview study of patients undergoing total Hip– and knee arthroplasty. J Patient-Rep Outcomes. 2023;7(1)
- [Google Scholar]
- Low participation in voluntary period for patient-reported outcome performance measures for hip and knee replacement. Health Affairs Scholar. 2025;3(6):qxaf111.
- [Google Scholar]
- Patient-reported outcome measures in total joint arthroplasty: defining the optimal collection window. Arthroplast Today. 2020;6(1):62-67.
- [Google Scholar]
- National implementation of an electronic patient-reported outcome measures program for joint replacement surgery: pilot study. JMIR Form Res. 2022;6(4)
- [Google Scholar]
- What is the optimal time point to assess patient-reported recovery after hip and knee replacement? A systematic review and analysis of routinely reported outcome data from the English patient-reported outcome measures programme. Health Qual Life Outcome. 2013;11(1):128.
- [Google Scholar]
- One-year Oxford knee scores should be used in preference to 6-month scores when assessing the outcome of total knee arthroplasty. Knee Surg Relat Res. 2020;32(1)
- [Google Scholar]
- Safety and efficacy of outpatient hip and knee arthroplasty: a systematic review with meta-analysis. Arch Orthop Trauma Surg. 2022;142(8):1775-1791.
- [Google Scholar]
- Cost savings of outpatient versus inpatient hip and knee arthroplasty in Ontario, Canada. PLoS One. 2025;20(5)
- [Google Scholar]
- Quality indicators in primary elective total hip and knee arthroplasty. Int Orthop. 2023;47(3):647-658.
- [Google Scholar]
- How would we know whether joint replacements are successful if we do not ask patients? Healthc Q. 2024;26(4):10-13.
- [Google Scholar]
- Patient reported outcome measures (PROMs) as primary and secondary outcomes in total hip and knee arthroplasty randomized controlled trials: a systematic review. Arch Orthop Trauma Surg. 2024;144(5):2257-2266.
- [Google Scholar]
- Effects of surgical approach and tourniquet use on patient-reported outcomes following total knee arthroplasty: a pilot randomized clinical trial. Orthop Res Rev. 2022;14:407-417.
- [Google Scholar]
- Curved versus straight stem uncemented total hip arthroplasty osteoarthritis multicenter trial (CUSTOM): design of a prospective blinded randomised controlled multicentre trial. BMJ Open. 2016;6(3)
- [Google Scholar]
- Robotic-arm assisted total knee arthroplasty is associated with comparable functional outcomes but improved forgotten joint scores compared with conventional manual total knee arthroplasty at five-year follow-up. Knee Surg Sports Traumatol Arthrosc. 2023;31(12):5453-5462.
- [Google Scholar]
- Early patient-reported outcomes from primary hip and knee arthroplasty have improved over the past seven years an analysis of the NHS PROMs dataset. Bone Jt J. 2022;104 B(6):687-695.
- [Google Scholar]
- Floor and ceiling effects in the OHS: an analysis of the NHS PROMs data set. BMJ Open. 2015;5(7)
- [Google Scholar]
- The use of patient-reported outcome measures in primary care: applications, benefits and challenges. J Patient-Rep Outcomes. 2021;5(Suppl 2):84.
- [Google Scholar]
- Hip fracture outcomes, risk prediction, and hospital comparisons: a population-based study in Ontario Canada. Injury. 2025;56(8)
- [Google Scholar]
- The Canadian joint replacement Registry—What have we learned? Acta Orthop. 2010;81(1):119.
- [Google Scholar]
- Linking patient-reported outcome measure scores to adverse event data to gain insight into overestimation of postoperative patient-reported outcome measure improvement after total hip arthroplasty and total knee arthroplasty due to selective nonresponse. J Arthroplast. 2025;40(1):45-52.
- [Google Scholar]
- What can legacy patient-reported outcome measures tell us about participation bias in patient-reported outcomes measurement information system scores among lumbar spine patients? Neurospine. 2022;19(2):307-314.
- [Google Scholar]
- A narrative literature review of bias in collecting patient reported outcomes measures (PROMs) Int J Environ Res Publ Health. 2021;18(23)
- [Google Scholar]
- Patient-reported outcome measures (PROMs) as proof of treatment efficacy. BMJ Evid Based Med. 2022;27(3):153-155.
- [Google Scholar]