ABSTRACT
Background
Osteoarthritis (OA) affects over 700,000 individuals in NSW, with projections estimating an increase to 1 million by 2030. This has led to a significant rise in joint replacement surgeries and associated healthcare costs. Despite guidelines recommending non‐surgical care as first line treatment, many patients remain on surgical waiting lists without accessing these treatments.
Aim
The aims of this study were to describe the characteristics and clinical outcomes of people with hip and knee OA attending two Sydney Local Health District (SLHD) Osteoarthritis Chronic Care Program (OACCP) clinics, and to identify whether baseline measures are associated with achieving Minimal Clinically Important (MCI) difference in outcomes.
Methods
This retrospective study (n = 1402) evaluated the OACCP's implementation within two SLHD hospitals between Oct 2018 and Sept 2023, describing patient characteristics, patient reported outcomes, physical performance measures, and allied health access. Changes in clinical outcomes in the short‐term (3 months) and long‐term (12 months) assessment as well as associations of patient's baseline measures with MCI improvements in clinical outcomes were analysed.
Results
Most self‐reported outcomes and physical performance measures improved at both 3 and 12 months. 38.6% of patients reached the MCI threshold for KOOS total score at 3 months, which rose to 52.1% at 12 months. None of the baseline characteristics predicted outcomes.
Conclusion
Following multidisciplinary non‐surgical care, people with hip and/or knee OA experience improved quality of life and function that is sustained for up to 12 months. However, analysis was limited by low long‐term follow‐up rates and no control group.
Keywords: hip, HOOS, knee, KOOS, OACCP, osteoarthritis
1. Introduction
It is estimated that over 700,000 of the 8.2 million residents of New South Wales (NSW), Australia have osteoarthritis (OA) and this is expected to rise to 1 million by 2030 due to the ageing population (NSW Agency for Clinical Innovation 2019). There has been a 50% increase in joint replacement surgery in the 10‐year period between 2005 and 2015 in Australia, contributing to over $3.75 billion annually in healthcare expenditure (NSW Agency for Clinical Innovation 2019). Concerningly, 7 in 10 people on waiting lists for joint replacement surgery have not accessed guideline‐recommended non‐surgical care (NSW Agency for Clinical Innovation 2019).
A recent systematic review of 20 clinical practice guidelines (Conley et al. 2023) for OA concluded that key consistent treatment recommendations for hip and knee OA were exercise, education, and advice on diet and exercise to achieve weight loss (where appropriate). In NSW, the Osteoarthritis Chronic Care Program (OACCP) was developed by the NSW Agency for Clinical Innovation (ACI) and implemented across the NSW public health system as part of the Ministry of Health Leading Better Value Care initiative. The OACCP Model of Care is a comprehensive multidisciplinary model incorporating clinical care standards designed to improve the care of people with hip and knee OA. It promotes a whole person assessment, with coordination of non‐operative care, regular review, and shared decision making regarding joint replacement surgery (NSW Agency for Clinical Innovation 2019).
Participants are encouraged to access OACCP dietetics, physiotherapy, medical, and where appropriate clinical psychology services. All participants who are overweight or obese are encouraged to lose a minimum of 5% of their body weight through a combination of diet and exercise, which have been found to be clinically effective and cost‐effective treatment options (Losina et al. 2019; Messier et al. 2004; Shahid et al. 2024). All OACCP participants are screened for psychological distress and those found to have moderate or high levels of depression, anxiety or stress are encouraged to access clinical psychology services (NSW Agency for Clinical Innovation 2012). Participants may also choose to access clinical psychology services to assist with lifestyle behaviour change. The OACCP clinics receive referrals from orthopaedic clinics (i.e. patients deemed not appropriate for joint replacement surgery at this time, or those on the surgical wait list) as well as from rheumatology clinics.
1.1. Rationale for Performing the Study and Study Objectives
The delivery of the OACCP Model of Care within the Sydney Local Health District (SLHD) has not been formally evaluated. Therefore, there are no published data regarding the characteristics of patients attending the service, and their clinical outcomes.
The aim of this study were:
Describe patient characteristics, patient reported outcomes, physical performance measures and allied health access of patients attending the OACCP.
Describe the changes in clinical outcomes in the short‐term (3 months) and long‐term (12 months) assessment for self‐reported OA status: Knee injury and osteoarthritis outcome score (KOOS), Hip injury and osteoarthritis outcome score (HOOS), quality of life (PROMIS‐29), physical performance measures 40‐m fast paced walk test and Timed Up and Go (TUG) and BMI.
Identify patient characteristics (baseline HOOS/KOOS score, baseline pain scores, BMI category, age, sex) that are associated with minimal clinically important (MCI) improvements in clinical outcomes at 3 and 12 months.
2. Methods
2.1. Design
We conducted a retrospective review and analysis of routinely collected data and patient medical records. This study follows the recommendations of the Reporting of studies Conducted using Observational Routinely collected health Data (RECORD) statement (Benchimol et al. 2015).
2.2. Ethics Approval
Ethics approval was granted by the SLHD for protocol X23‐0112 and 2023/ETH00672: Evaluating outcomes of the OACCP.
2.3. Setting and Participants
This study was conducted at two public hospitals within SLHD in Sydney, NSW, Australia: The Canterbury Hospital (TCH) and Royal Prince Alfred Hospital (RPAH). Eligible participants included those who attended an initial assessment at an OACCP clinic, from 15th October 2018 to 25th September 2023.
2.4. Data Collection
Routinely collected data collected at OACCP clinical visits were extracted from the SLHD OACCP REDCap (Harris et al. 2009) databases and the NSW Health digital platform HOPE (Health Outcomes and Patient Experience) databases.
2.5. Outcome Measures
Patient‐reported outcome measures include the KOOS (rated 0–100, where 0 represents extreme knee problems and 100 represents no knee problems) for knee OA and the HOOS (rated 0–100) for hip OA. Both KOOS and HOOS contain five subcategories (pain, other symptoms, function in activities of daily living (ADL), function in sport and recreation, and knee or hip related Quality of Life (QOL)). PROMIS‐29 (Patient‐Reported Outcomes Measurement Information System) scores are also collected across multiple domains: depression, anxiety, physical function, pain interference, fatigue, sleep disturbance, and ability to participate in social roles and activities are rated on a scale from 4 to 20, and global pain is rated from 0 to 10.
Physical performance measures collected by OACCP physiotherapists included the 40‐m fast paced walking test, and the Timed Up and Go (TUG) test. BMI (height and weight) was also measured.
Allied health access was extracted by manual review of a subset of electronic medical records. This included a number of consultations with a physiotherapist, clinical psychologist, and dietician, as well as attendance at individual or group exercise sessions or hydrotherapy, over a 12‐month period from initial attendance.
2.6. Data Analysis
Quantitative data were analysed using STATA version 16.1 statistical software (StataCorp, College Station, TX). Descriptive statistics were used for patient demographics and clinical characteristics. Categorical variables were described with frequencies (%), and continuous variables with means and standard deviations (SD) or median and interquartile range (IQR). Differences between initial assessment and the follow‐up time‐points were calculated using paired t‐tests, with outcomes considered statistically significant if alpha was < 0.05.
The MCI in clinical outcomes was used to determine the proportion of patients who achieved an MCI improvement in each outcome in the short‐term (3 months) and long‐term (12 months) assessments. The specific cut‐offs that are considered clinically relevant are population‐ and context‐based (Cronström et al. 2023); therefore, the best available metrics for those undergoing non‐surgical treatment were used.
MCI difference in clinical outcomes were set as 10‐point increase for the overall KOOS (Torstensen et al. 2023), 18‐point increase for HOOS (Hung et al. 2018; Jayakumar et al. 2022), a one unit improvement in global pain rating (Cronström et al. 2023), 1.10 s improvement in TUG (Alghadir et al. 2015), 0.1 m/s improvement for the 40m walk test (Chui et al. 2012), and a minimum weight loss of 5% (Bliddal et al. 2014). Participants in the underweight or healthy weight ranges were excluded from the weight loss analysis.
Multinomial mixed‐effects logistic regression was used to investigate whether the pre‐selected individual patient characteristics were associated with achieving the MCI difference at both short‐term and long‐term time‐point assessments. Potential variables were force entered into the model based on the literature (Holden et al. 2023; Teoh et al. 2017).
3. Results
3.1. Patient Characteristics
A total of 1402 patient medical records were analysed. Of these, 71.6% (n = 1004) were from the RPAH, and 28.4% (n = 398) from TCH. Median patient age was 66 years (IQR 59–74, range 30–88). 66.5% (n = 932) of patients identified as female, 33.2% (n = 466) male, 0.1% as neither, and 0.2% had missing data regarding sex. Knee OA was the presenting reason for 86.5% of patients (n = 1213), hip OA for 12.6% (n = 176), and both knee and hip for 0.9% (n = 13). BMI data was available for 63.2% (n = 886) of patients, with a mean BMI of 31.5 (SD 7.1). Detailed patient characteristics for each hospital are presented in Table 1.
TABLE 1.
Baseline demographic data (age, gender, BMI, joint, referral source).
| Variable | All participants (n = 1402) | RPAH participants (n = 1004) | TCH participants (n = 398) |
|---|---|---|---|
| Age (y), median (IQR; range) | 66 (59–74; 39–88) | 65 (58–74; 39–88) | 69 (63–76; 44–90) |
| Gender, n (%) | |||
| Female | 932 (66.5) | 682 (67.9) | 250 (62.8) |
| Male | 466 (33.2) | 320 (31.9) | 146 (36.7) |
| Neither | 1 (0.1) | 0 (0) | 1 (0.3) |
| Missing | 3 (0.2) | 2 (0.2) | 1 (0.3) |
| BMI (category), n (%) | |||
| Underweight | 2 (0.1) | 2 (0.2) | 0 (0) |
| Healthy | 141 (10.1) | 106 (10.6) | 35 (8.8) |
| Overweight | 284 (20.3) | 213 (21.2) | 71 (17.9) |
| Obese class I | 232 (16.6) | 157 (15.6) | 75 (18.9) |
| Obese class II | 120 (8.6) | 89 (8.9) | 31 (7.8) |
| Obese class III | 107 (7.6) | 82 (8.2) | 25 (6.3) |
| Missing | 515 (36.8) | 355 (35.4) | 161 (40.5) |
| Joint, n (%) | |||
| Knee | 1213 (86.5) | 872 (86.9) | 341 (85.7) |
| Hip | 176 (12.6) | 119 (11.9) | 57 (14.3) |
| Both hip and knee | 13 (0.9) | 13 (1.3) | 0 (0.0) |
| Referral source, n (%) | |||
| Physiotherapist | 24 (1.7) | 24 (2.4) | 0 (0) |
| Orthopaedic clinic | 514 (36.7) | 514 (51.2) | 0 (0) |
| Rheumatology clinic | 197 (14.1) | 161 (16.0) | 36 (9.1) |
| GP | 107 (7.6) | 42 (4.2) | 65 (16.3) |
| Orthopaedic waitlist | 300 (21.4) | 3 (0.3) | 297 (74.6) |
| Orthopaedic non‐waitlist | 2 (0.1) | 2 (0.2) | 0 (0) |
| Other | 30 (2.1) | 30 (3.0) | 0 (0) |
| Missing | 228 (16.3) | 228 (22.7) | 0 (0) |
The main source of referrals was from the orthopaedic clinics but the two hospitals differed in their referral criteria. At Royal Prince Alfred Hospital, the referrals were predominantly patients deemed not requiring a joint replacement or patients requesting non‐surgical treatments (n = 514, 26.2%). In contrast, Canterbury Hospital received most referrals from the joint replacement waitlist (n = 297, 74.6%).
3.2. Baseline Patient‐Reported Outcomes
Baseline KOOS scores were available for 946/1213 (78.0%) of patients with knee OA, with a mean total score of 43.3 (SD 18.2). Baseline HOOS scores were available for 146/176 (83.0%) patients with hip OA, with a mean total score of 41.7 (SD 19.6). For both KOOS and HOOS, the lowest scoring subcategories (greater disability) were for sports and recreation, and QOL.
PROMIS‐29 raw scores included mean physical function 11.2 (SD 3.8), anxiety 10 (SD 4.5), depression 9.0 (SD 4.6), fatigue 11.8 (SD 4.3), sleep 12.2 (SD 4.0), and social 11.0 (SD 4.3) dimensions. Pain interference was rated at mean of 13.8 (SD 4.3), and global pain was rated at a mean of 6.2 (SD 2.3). Detailed baseline outcomes are presented in Table 2.
TABLE 2.
KOOS, HOOS, PROMIS‐29, objective measures raw scores for baseline, 3 months and 12 months, and the mean difference (95% CI) from baseline at 3 months and 12 months (combined Royal Prince Alfred Hospital and Canterbury Hospital).
| Outcome measures | Baseline | 3‐month (short‐term) | 12‐month (long‐term) | ||
|---|---|---|---|---|---|
| KOOS, n, mean (SD) | MD (95% CI) | MD (95% CI) | |||
| Symptoms | 946, 49.0 (20.1) | 485, 54.5 (19.6) | 5.4 (3.9–6.9) | 196, 58.2 (20.5) | 6.4 (3.8–8.9) |
| Pain | 942, 46.4 (19.4) | 484, 54.4 (19.2) | 7.0 (5.5–8.5) | 195, 60.2 (20.3) | 9.6 (7.1–12.0) |
| ADLs | 940, 49.2 (21.2) | 485, 57.0 (21.1) | 7.0 (5.5–8.5) | 195, 63.5 (22.5) | 8.9 (6.5–11.4) |
| Sport and recreation | 846, 24.7 (25.3) | 428, 30.3 (25.1) | 6.2 (3.6–8.9) | 171, 40.3 (28.2) | 13.6 (9.2–18.1) |
| QOL | 936, 27.5 (19.1) | 484, 36.7 (20.9) | 8.2 (6.5–9.9) | 196, 43.9 (21.6) | 13.3 (10.4–16.1) |
| Total | 946, 43.3 (18.2) | 486, 50.8 (18.5) | 6.8 (5.6–8.1) | 196, 57.1 (20.2) | 9.6 (7.6–11.9) |
| HOOS, n, mean (SD); | |||||
| Symptoms | 146, 46.0 (20.4) | 76, 51.4 (20.9) | 7.0 (3.0–10.9) | 20, 54.8 (20.0) | 6.8 (−0.1–13.7) |
| Pain | 144, 43.7 (20.1) | 76, 51.8 (20.0) | 8.4 (4.8–12.0) | 20, 59.1 (19.9) | 13.1 (5.5–20.6) |
| ADLs | 144, 46.5 (21.7) | 76, 54.9 (21.4) | 9.5 (4.9–14.0) | 20, 62.6 (21.5) | 11.8 (3.1–20.6) |
| Sport and recreation | 138, 30.1 (23.0) | 74, 37.4 (24.2) | 9.2 (3.3–15.1) | 18, 44.7 (20.2) | 5.8 (−4.6–16.2) |
| QOL | 146, 27.3 (20.8) | 76, 34.5 (22.5) | 9.2 (4.4–14.0) | 20, 42.2 (21.4) | 12.3 (4.2–20.4) |
| Total | 146, 41.7 (19.6) | 76, 49.5 (19.7) | 8.8 (5.0–12.6) | 20, 56.7 (19.7) | 11.3 (4.0–18.5) |
| PROMIS, n, mean (SD) | |||||
| Physical | 1,041, 11.2 (3.8) | 526, 12.7 (3.8) | 1.1 (0.9–1.4) | 192, 13.8 (3.9) | 1.4 (0.9–1.9) |
| Anxiety | 1,038, 10.0 (4.5) | 524, 9.1 (4.3) | −0.6 (−0.9 to −0.2) | 191, 8.7 (4.3) | −0.6 (−1.0 to −0.1) |
| Depression | 1,033, 9.0 (4.6) | 524, 8.1 (4.5) | −0.4 (−0.7 to −0.1) | 192, 8.0 (4.5) | 0.2 (−0.3–0.7) |
| Fatigue | 1,028, 11.8 (4.3) | 522, 10.8 (4.2) | −0.7 (−1.0 to −0.4) | 187, 10.6 (4.4) | −0.8 (−1.4 to −0.3) |
| Sleep | 1,023, 12.2 (4.0) | 521, 11.4 (3.8) | −0.7 (−0.9 to −0.4) | 187, 11.0 (4.2) | −0.5 (−1.0 to 0.0) |
| Social | 1,020, 11.0 (4.3) | 519, 12.5 (4.3) | 1.4 (1.0–1.7) | 189, 13.3 (4.4) | 1.3 (0.8–1.9) |
| Pain interference | 1,030, 13.8 (4.3) | 526, 11.8 (4.5) | −1.7 (−2.0 to −1.4) | 187, 10.5 (4.7) | −2.1 (−2.7 to −1.6) |
| Pain (global) | 1,033, 6.2 (2.3) | 529, 5.3 (2.4) | −0.7 (−0.9 to −0.6) | 190, 4.7 (2.6) | −0.9 (−1.3 to −0.6) |
| Objective measures, n, mean (SD) | |||||
| TUG test, secs | 851, 12.2 (7.6) | 495, 10.0 (5.0) | −2.1 (−2.7 to −1.5) | 241, 9.0 (4.8) | −2.4 (−3.0 to −1.8) |
| 40m walk Test, m/s | 841, 1.1 (0.4) | 491, 1.2 (0.4) | 0.1 (0.7–0.1) | 240, 1.3 (0.4) | 0.2 (0.1–0.2) |
| BMI | 886, 31.5 (7.1) | 522, 31.3 (7.7) | — | 243, 30.3 (6.9) | — |
| % Body weight change (n, mean SD) | — | 429, −0.7 (2.3) | — | 190, −1.0 (5.0) | — |
Note: PROMIS scores include subscales for physical function, depression, fatigue, sleep disturbance, ability to perform social roles and pain interference rated on a scale from 4 to 20, global pain is rated from 0 to 10. HOOS and KOOS Scores includes subscales for pain, symptoms, function in activities of daily living, joint‐related quality of life, and function in sports and recreation, with scores on each subscale ranging from 0 (worst) to 100 (best).
Abbreviations: BMI, Body Mass Index; CI, Confidence Interval; GP, General Practitioner; HOOS, Hip dysfunction and Osteoarthritis Outcome Score; KOOS, Knee injury and Osteoarthritis Outcome Score; MD, Mean Difference; n, number; SD, Standard deviation; TUG, Timed Up and Go test.
3.3. Baseline Physical Performance Measures
Physical performance measures were available for 851/1402 (60.1%) of patients. TUG took a mean of 12.2 (SD 7.6) seconds, and the mean 40‐m walk test was 1.1 m/s (SD 0.4).
3.4. Allied Health
Allied health access data were available for a subset of patients, n = 345/1402 (24.6%). At both hospitals, 82.0% (n = 283/345) consulted a dietician, 42.3% (n = 146/345) consulted a clinical psychologist, and 100% (n = 345/345) consulted a physiotherapist.
At the RPAH, there were similar proportions of patients choosing to complete their exercise programme independently (n = 96/183, 52.5%) versus in supervised gym sessions (n = 87/183, 47.5%). Patients attended a mean of 9 (SD 8) individual sessions with a physiotherapist or exercise physiologist versus a mean of 12 (SD 9) group gym sessions. Hydrotherapy was accessed by 14.7% (n = 27/183) of patients over a mean of 18 (SD 12) sessions. Those who consulted a dietician had a mean of 5 (SD 3) sessions. Those that consulted a clinical psychologist had a mean of 3 (SD 4) sessions.
Detailed patient demographics, referral sources, and clinical outcomes are presented in Tables 1 and 2.
3.5. Short‐ and Long‐Term Clinical Outcomes
Follow‐up rates for KOOS and HOOS at 3 and 12 months are presented in Figure 1. Detailed clinical outcomes are presented in Figure 2a,b, and Table 2 for both hospital sites. Supporting Information S1: Tables S1 and S2 present site‐specific outcomes for RPAH and TCH, respectively.
FIGURE 1.
Flow of participants through the study. Data availability for KOOS and HOOS scores.
FIGURE 2.
(a) KOOS dimensions at baseline, 3 months and 12 months follow‐up. (b). HOOS dimensions at baseline, 3 months and 12 months follow‐up.
3.6. KOOS
At 3 and 12 months from baseline, all KOOS domains and the total score had mean improvements that were statistically significant (Table 2). For KOOS total score at 3 months, 38.6% (n = 186/482) of patients across all sites reached the MCI threshold, which rose to 52.1% (n = 101/194) at 12 months.
3.7. HOOS
At 3 and 12 months from baseline, all HOOS domains and the total score were considered to have a mean improvement that was statistically significant (Table 2). For the HOOS total score 25.7% (n = 18/70) of patients reached the MCI threshold at 3 months and 30% (n = 6/20) at 12 months.
3.8. PROMIS‐29
At 3 months, physical function improved a mean of 1.1 (95% CI 0.9 to 1.4) points, anxiety decreased a mean of −0.6 (95% CI −0.9 to −0.2), depression decreased a mean of −0.4 (95% CI −0.7 to −0.1), fatigue decreased a mean of −0.7 (95% CI −1.0 to −0.4), sleep disturbance decreased a mean of −0.7 (95% CI −0.9 to −0.4), social function improved a mean of 1.4 (95% CI 1.0 to 1.7), pain interference decreased by a mean of −1.7 (95% CI −2.0 to −1.4), and global pain rating decreased by a mean of −0.7 (95% CI −0.9 to −0.6). All mean differences were considered statistically, but not clinically, significant.
At 12 months, physical function improved a mean of 1.4 (95% CI 0.9–1.9) points, anxiety decreased a mean of 0.6 (95% CI 1.0 to 0.1), depression decreased a mean of 0.2 (95% CI −0.3 to 0.7), fatigue decreased a mean of −0.8 (95% CI −1.4 to −0.3), sleep disturbance decreased a mean of −0.5 (95% CI −1.0 to −0.0), social function improved a mean of 1.3 (95% CI 0.8–1.9), pain interference decreased a mean of −2.1 (95% CI 2.7 to 1.6), and global pain rating decreased a mean of −0.9 (95% CI −1.3 to −0.6). All mean differences were considered statistically, but not clinically, significant.
3.9. Physical Performance Measures
At 3 months from baseline, the mean difference in TUG was 2.1 s faster (95% CI −2.7 to −1.5), and at 12 months, 2.4 s faster (95% CI −3.0 to −1.8), which was statistically and clinically significant at both time‐points. At 3 months 49% (n = 240/495) and at 12 months 37% (n = 71/193) of patients reached the MCI threshold for improvement in TUG.
At 3 months from baseline, the mean difference in the 40‐m walk speed was 0.1 m/s faster (95% CI 0.7 to 0.1) and at 12 months 0.2 m/s faster (95% CI 0.1–0.2), which was statistically and clinically significant at both time points. At 3 months, 46% (n = 225/491) and at 12 months 58% (n = 111/190) of patients reached the MIC threshold for improvement in walking speed.
3.10. Weight Loss
Participants in the overweight or obese BMI categories (n = 743/886, 83.9% of participants with BMI data at baseline) lost a mean of 0.7% (SD 2.3) of their bodyweight at 3 months and 1.0% (SD 5.0) at 12 months, which was not considered clinically significant. Only 13% of all patients in the overweight or obese BMI categories at baseline were able to achieve weight loss of at least 5%. However, there was a wide range in overall weight change in the cohort, from a loss of 17% to a gain of 14% at 12 months.
3.11. Associations With Clinical Improvement
Multinomial mixed‐effects logistic regression was used to analyse associations with clinical improvement, set as reaching the MCI thresholds for improvement in KOOS scores at either 3 or 12 months. Patient age, sex, BMI category and baseline scores (pain intensity and interference, KOOS) were not associated with reaching MCIs due to wide confidence intervals (Supporting Information S1: Tables S3 and S4). Logistic regression was not performed for HOOS scores due to low available data.
4. Discussion
4.1. Principal Findings
This study evaluated the outcomes of the OACCP in two metropolitan hospital clinics, focusing on patient characteristics, clinical outcomes, and factors associated with clinically important improvement. The results demonstrate that after completing a non‐surgical, multidisciplinary model of care people with hip and knee OA can significantly improve self‐reported outcomes and objective measures of physical performance. Very few (∼13%) of all patients in the overweight or obese BMI categories at baseline were able to achieve weight loss of at least 5%. None of the baseline measures were significantly predictive of outcomes.
4.2. Strengths and Limitations
The study has a large sample size from two metropolitan teaching hospitals. Data collection was concurrent with clinical reviews during the 12 months of the programme incorporating routinely collected data, medical records and patient‐reported outcomes. Objective measures were taken using standardised protocols.
Due to the nature of a retrospective analysis of real‐world routinely collected data, there are missing follow up data from patients that did not attend their 3‐ and 12‐month reviews. COVID‐19 related restrictions on attendance of face‐to‐face clinical appointments also impacted follow up. Long term KOOS scores for knee OA were only available for 21% of patients, and HOOS scores for 14% of hip OA patients, which is on par with similar studies (Cronström et al. 2023). This could reflect a wide range of clinical scenarios and include patients managing well with conservative care independently or those who chose to undergo surgery. There was also a small subsample of people with hip OA, compared to knee OA; therefore, an analysis on associations with clinical improvement was not conducted for the hip OA group. Interpretation of change in PROMIS‐29 scores is also limited as it may have poor responsiveness to change in a chronic OA population, with large ceiling effects (Katz et al. 2017). PROMIS‐29 raw scores were used because of a lack of a suitable reference population.
4.3. Comparison to Similar Studies
Summary data from the HOPE database are available regarding results of the OACCP at other hospital sites in New South Wales (n = 5600 for 19 sites in 2021) (NSW Agency for Clinical Innovation 2023). Baseline global pain ratings at SLHD sites were comparable to the statewide cohort (6.2 vs. 6.1). However, mean global pain rating improved more in the SLHD cohort (0.9 points) than in the NSW cohort (0.2 points). Baseline KOOS sub‐scores were also comparable to the statewide cohort. However, long‐term outcomes were considerably better in the SLHD cohort. Mean improvement at discharge across all KOOS domains in the NSW cohort was 17%, compared with 35% in the SLHD cohort. It is unclear if this is due to differences in patient case‐mix, implementation of the programme or other factors that differ between sites that is delivery of care, patient demographics and variability in referral sources across hospital sites. Mean improvements in objective physical function measures (e.g. 40m walk test) were the same as those in a longitudinal study of GLA:D program participants (n = 32,708) in three countries (Australia, Canada, Denmark) (Roos et al. 2021).
No associations were found between pre‐treatment characteristics and clinical outcomes, which is consistent with prior studies (Eyles et al. 2016).
4.4. Clinical Implications and Future Research
The results of this study support the positive impact of the OACCP model of care on QOL and function, emphasising the importance of a tailored multidisciplinary approach being available to all patients for the management of hip and knee OA. The study will contribute to continued improved service delivery and inform future randomised controlled trials of the service.
Future studies should assess outcomes beyond 12 months to provide insights into the sustainability of improvements and potential long‐term benefits of the programme. Additionally, investigation to identify if transition from hospital to community‐based exercise programs could provide for prolonged and greater improvements in individuals undertaking non‐surgical management for knee or hip OA is desirable. Comparative studies with other conservative care approaches can clarify the unique contributions of the OACCP model of care. Important considerations for future research include the optimal timepoint to offer this model of care in the trajectory of OA as well as programme duration and intensity. Future OACCP outcome measures could include body composition (muscle and body fat mass) to add to the clinical picture beyond BMI alone. Additionally, recent studies (Bliddal et al. 2024; Wei et al. 2024) have shown that anti‐obesity medication is safe and effective for moderate weight loss in a population with knee OA. Therefore, integration of pharmacologically assisted weight loss in the OACCP population is warranted.
5. Conclusion
The results demonstrate that following a multidisciplinary non‐surgical care programme, people with hip and knee OA can improve QOL and function after three months, and improvement was sustained up to 12 months in this patient cohort. However, analysis was limited by low long‐term follow‐up rates and no control group.
Author Contributions
F.M. and C.N. conceived the research evaluation. M.M., K.B., R.D., S.H., F.M. and C.N. collected data, provided discipline‐specific expertise, and reviewed manuscript drafts. A.M., T.F.D.C., J.Z., G.C.M., and C.G.M. completed data analysis. All authors contributed to and reviewed the final manuscript.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting Information S1
Acknowledgements
Open access publishing facilitated by The University of Sydney, as part of the Wiley ‐ The University of Sydney agreement via the Council of Australian University Librarians.
Melman, Alla , Menz Fred, Needs Chris, et al. 2025. “The Clinical Course of Hip and Knee Osteoarthritis in Adults Attending Two Hospital Outpatient Physiotherapy Clinics in NSW: An Evaluation of the Osteoarthritis Chronic Care Program (OACCP).” Musculoskeletal Care: e70174. 10.1002/msc.70174.
Funding: The authors received no specific funding for this work.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- Alghadir, A. , Anwer S., and Brismée J. M.. 2015. “The Reliability and Minimal Detectable Change of Timed up and Go Test in Individuals With Grade 1‐3 Knee Osteoarthritis.” BMC Musculoskeletal Disorder 16, no. 1: 174. 10.1186/s12891-015-0637-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Benchimol, E. I. , Smeeth L., Guttmann A., et al. 2015. “The Reporting of Studies Conducted Using Observational Routinely‐Collected Health Data (RECORD) Statement.” PLoS Medicine 12, no. 10: e1001885. 10.1371/journal.pmed.1001885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bliddal, H. , Bays H., Czernichow S., et al. 2024. “Once‐Weekly Semaglutide in Persons With Obesity and Knee Osteoarthritis.” New England Journal of Medicine 391, no. 17: 1573–1583. 10.1056/NEJMoa2403664. [DOI] [PubMed] [Google Scholar]
- Bliddal, H. , Leeds A. R., and Christensen R.. 2014. “Osteoarthritis, Obesity and Weight Loss: Evidence, Hypotheses and Horizons ‐ A Scoping Review.” Obesity Reviews 15, no. 7: 578–586. 10.1111/obr.12173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chui, K. , Hood E., and Klima D.. 2012. “Meaningful Change in Walking Speed.” Topics in Geriatric Rehabilitation 28, no. 2: 97–103. 10.1097/TGR.0b013e3182510195. [DOI] [Google Scholar]
- Conley, B. , Bunzli S., Bullen J., et al. 2023. “Core Recommendations for Osteoarthritis Care: A Systematic Review of Clinical Practice Guidelines.” Arthritis Care & Research 75, no. 9: 1897–1907. 10.1002/acr.25101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cronström, A. , Ingelsrud L. H., Nero H., et al. 2023. “Interpretation Threshold Values for Patient‐Reported Outcomes in Patients Participating in a Digitally Delivered First‐Line Treatment Program for Hip or Knee Osteoarthritis.” Osteoarthritis and Cartilage Open 5, no. 3: 100375. 10.1016/j.ocarto.2023.100375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eyles, J. P. , Mills K., Lucas B. R., et al. 2016. “Can We Predict Those With Osteoarthritis Who Will Worsen Following a Chronic Disease Management Program?” Arthritis Care & Research 68, no. 9: 1268–1277. 10.1002/acr.22836. [DOI] [PubMed] [Google Scholar]
- Harris, P. A. , Taylor R., Thielke R., Payne J., Gonzalez N., and Conde J. G.. 2009. “Research Electronic Data Capture (REDCap)—A Metadata‐Driven Methodology and Workflow Process for Providing Translational Research Informatics Support.” Journal of Biomedical Informatics 42, no. 2: 377–381. 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Holden, M. A. , Hattle M., Runhaar J., et al. 2023. “Moderators of the Effect of Therapeutic Exercise for Knee and Hip Osteoarthritis: A Systematic Review and Individual Participant Data Meta‐Analysis.” Lancet Rheumatology 5, no. 7: e386–e400. 10.1016/S2665-9913(23)00122-4. [DOI] [PubMed] [Google Scholar]
- Hung, M. , Bounsanga J., Voss M. W., and Saltzman C. L.. 2018. “Establishing Minimum Clinically Important Difference Values for the Patient‐Reported Outcomes Measurement Information System Physical Function, Hip Disability and Osteoarthritis Outcome Score for Joint Reconstruction, and Knee Injury and Osteoarthritis Outcome Score for Joint Reconstruction in Orthopaedics.” World Journal of Orthopedics 9, no. 3: 41–49. 10.5312/wjo.v9.i3.41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Jayakumar, P. , Galea V. P., Geng M., Moore M. G., Bozic K. J., and Koenig K. M.. 2022. “Functional Outcomes of a Musculoskeletal Integrated Practice Providing Comprehensive Whole Person Care for Hip Osteoarthritis.” Supplement, Journal of Arthroplasty 37, no. S7: S471–S478.e471. 10.1016/j.arth.2022.03.019. [DOI] [PubMed] [Google Scholar]
- Katz, P. , Pedro S., and Michaud K.. 2017. “Performance of the Patient‐Reported Outcomes Measurement Information System 29‐Item Profile in Rheumatoid Arthritis, Osteoarthritis, Fibromyalgia, and Systemic Lupus Erythematosus.” Arthritis Care & Research 69, no. 9: 1312–1321. 10.1002/acr.23183. [DOI] [PubMed] [Google Scholar]
- Losina, E. , Smith K. C., Paltiel A. D., et al. 2019. “Cost‐Effectiveness of Diet and Exercise for Overweight and Obese Patients With Knee Osteoarthritis.” Arthritis Care & Research 71, no. 7: 855–864. 10.1002/acr.23716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Messier, S. P. , Loeser R. F., Miller G. D., et al. 2004. “Exercise and Dietary Weight Loss in Overweight and Obese Older Adults With Knee Osteoarthritis: The Arthritis, Diet, and Activity Promotion Trial.” Arthritis & Rheumatism 50, no. 5: 1501–1510. 10.1002/art.20256. [DOI] [PubMed] [Google Scholar]
- NSW Agency for Clinical Innovation . 2012. Musculoskeletal Network Osteoarthritis Chronic Care Program Model of Care. https://aci.health.nsw.gov.au/__data/assets/pdf_file/0003/165306/Osteoarthritis‐Chronic‐Care‐Program‐Mode‐of‐Care‐High‐Resolution.pdf.
- NSW Agency for Clinical Innovation . 2019. Osteoarthritis Chronic Care Program. https://aci.health.nsw.gov.au/statewide‐programs/lbvc/osteoarthritis‐chronic‐care‐program.
- NSW Agency for Clinical Innovation . 2023. Patient‐Reported Outcome Measures: Methods for Analysis and Reporting. https://aci.health.nsw.gov.au/__data/assets/pdf_file/0009/925659/ACI‐Patient‐reported‐outcome‐measures‐methods‐for‐analysis‐and‐reporting.pdf.
- Roos, E. M. , Grønne D. T., Skou S. T., et al. 2021. “Immediate Outcomes Following the GLAD® Program in Denmark, Canada and Australia. A Longitudinal Analysis Including 28,370 Patients With Symptomatic Knee or Hip Osteoarthritis.” Osteoarthritis and Cartilage 29, no. 4: 502–506. 10.1016/j.joca.2020.12.024. [DOI] [PubMed] [Google Scholar]
- Shahid, A. , Thirumaran A. J., Christensen R., et al. 2024. “Comparison of Weight Loss Interventions in Overweight and Obese Adults With Knee Osteoarthritis: A Systematic Review and Network Meta‐Analysis of Randomized Trials.” Osteoarthritis and Cartilage 33, no. 4: 518–529. 10.1016/j.joca.2024.08.012. [DOI] [PubMed] [Google Scholar]
- Teoh, L. S. G. , Eyles J. P., Makovey J., Williams M., Kwoh C. K., and Hunter D. J.. 2017. “Observational Study of the Impact of an Individualized Multidisciplinary Chronic Care Program for Hip and Knee Osteoarthritis Treatment on Willingness for Surgery.” International Journal of Rheumatic Diseases 20, no. 10: 1383–1392. 10.1111/1756-185x.12950. [DOI] [PubMed] [Google Scholar]
- Torstensen, T. A. , Østerås H., LoMartire R., Rugelbak G. M., Grooten W. J. A., and Äng B. O.. 2023. “High‐ Versus Low‐Dose Exercise Therapy for Knee Osteoarthritis : A Randomized Controlled Multicenter Trial.” Annals of Internal Medicine 176, no. 2: 154–165. 10.7326/m22-2348. [DOI] [PubMed] [Google Scholar]
- Wei, J. , Hunter D., Lane N. E., et al. 2024. “Weight Loss Induced by Antiobesity Medications and All‐Cause Mortality Among Patients With Knee or Hip Osteoarthritis.” Arthritis & Rheumatology 76, no. 4: 577–586. 10.1002/art.42754. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information S1
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.