Advertisement

Knee and hip osteoarthritis are more alike than different in baseline characteristics and outcomes: a longitudinal study of 32,599 patients participating in supervised education and exercise therapy

  • E.M. Roos
    Correspondence
    Address correspondence and reprint requests to: E.M. Roos Department of Sports Science and Clinical Biomechanics University of Southern Denmark Campusvej 55 DK-5230 Odense, Denmark. Tel.: 45-60-11-43-41.
    Affiliations
    Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark
    Search for articles by this author
  • D.T. Grønne
    Affiliations
    Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark
    Search for articles by this author
  • J.B. Thorlund
    Affiliations
    Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark

    Research Unit for General Practice, Department of Public Health, University of Southern Denmark
    Search for articles by this author
  • S.T. Skou
    Affiliations
    Research Unit for Musculoskeletal Function and Physiotherapy, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark

    The Research Unit PROgrez, Department of Physiotherapy and Occupational Therapy, Næstved-Slagelse-Ringsted Hospitals, Denmark
    Search for articles by this author
Open AccessPublished:February 14, 2022DOI:https://doi.org/10.1016/j.joca.2022.02.001

      Summary

      Objective

      Compare baseline characteristics and change in outcomes in patients with symptomatic knee or hip OA participating in patient education and exercise therapy.

      Design

      Longitudinal cohort study. Good Life with osteoArthritis in Denmark (GLA:D®) is an 8-week patient education and supervised exercise program delivered by certified clinicians. Changes in pain intensity, Knee injury/Hip disability Osteoarthritis Outcome Scores’ subscale Quality of Life (K/HOOS QOL), EuroQoL 5-Dimensions 5-Level (EQ-5D) and 40 m walk test at ∼3 and 12 months were compared between knee and hip patients.

      Results

      24,241 knee and 8,358 hip patients were included, with response rates of 75% and 60% at ∼3 and 12 months. Age, gender, symptom duration, pain medication use, pain intensity, physical function and quality of life were alike. More knee than hip patients were obese and had bilateral symptoms. At 3 months, clinically relevant improvements were seen in both knee and hip OA patients with clinically irrelevant between groups differences; 2.1 (1.5; 2.8) mm in pain intensity, −1.1 (−1.5; −0.7) point in K/HOOS QOL score, −0.010 (−0.013; −0.007) in EQ-5D index score and −0.02 (−0.02; −0.01) m/sec in walking speed. At 12 months the slight immediate differences were equalized.

      Conclusion

      Patients presenting with knee and hip OA in primary care were on average more alike than different. Following treatment, clinically relevant improvements were seen in both knee and hip OA patients at 3 and 12 months. Patients with knee and hip OA should be prioritized alike for treatment with patient education and supervised exercise therapy.

      Keywords

      Introduction

      Osteoarthritis (OA) affects 528 million people worldwide with the greatest burden due to OA of the knee and hip
      • Diseases G.B.D.
      • Injuries C.
      Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019.
      . OA can be arrived at via different pathways including one or more risk factors, shown to differ between knee and hip OA. For example, overweight and obesity are associated with a much greater risk to develop radiographic knee OA than hip OA
      • Richmond S.A.
      • Fukuchi R.K.
      • Ezzat A.
      • Schneider K.
      • Schneider G.
      • Emery C.A.
      Are joint injury, sport activity, physical activity, obesity, or occupational activities predictors for osteoarthritis? A systematic review.
      . As a result, the clinical presentation of patients with knee and hip OA may differ, but little comparative data is available
      • Hall M.
      • van der Esch M.
      • Hinman R.S.
      • Peat G.
      • de Zwart A.
      • Quicke J.G.
      • et al.
      How does hip osteoarthritis differ from knee osteoarthritis?.
      .
      Exercise therapy is a cornerstone in OA treatment, and systematic reviews of the available data from randomized trials suggest the pain-relieving effect immediately after the treatment being larger in patients with knee OA, effect size −0.49 [−0.59, −0.39]
      • Fransen M.
      • McConnell S.
      • Harmer A.R.
      • Van der Esch M.
      • Simic M.
      • Bennell K.L.
      Exercise for osteoarthritis of the knee.
      , compared to hip OA, effect size −0.38 [−0.55, −0.20]
      • Fransen M.
      • McConnell S.
      • Hernandez-Molina G.
      • Reichenbach S.
      Exercise for Osteoarthritis of the Hip.
      . It is however only few of the trials included in these systematic reviews that have included both patients with knee or hip OA, which would allow for a direct comparison of treatment result
      • van Baar M.E.
      • Dekker J.
      • Lemmens J.A.
      • Oostendorp R.A.
      • Bijlsma J.W.
      Pain and disability in patients with osteoarthritis of hip or knee: the relationship with articular, kinesiological, and psychological characteristics.
      • Hopman-Rock M.
      • Westhoff M.H.
      The effects of a health educational and exercise program for older adults with osteoarthritis for the hip or knee.
      • Fransen M.
      • Nairn L.
      • Winstanley J.
      • Lam P.
      • Edmonds J.
      Physical activity for osteoarthritis management: a randomized controlled clinical trial evaluating hydrotherapy or Tai Chi classes.
      • Abbott J.H.
      • Robertson M.C.
      • Chapple C.
      • Pinto D.
      • Wright A.A.
      • Leon de la Barra S.
      • et al.
      Manual therapy, exercise therapy, or both, in addition to usual care, for osteoarthritis of the hip or knee: a randomized controlled trial. 1: clinical effectiveness.
      • Foley A.
      • Halbert J.
      • Hewitt T.
      • Crotty M.
      Does hydrotherapy improve strength and physical function in patients with osteoarthritis--a randomised controlled trial comparing a gym based and a hydrotherapy based strengthening programme.
      . Thus, it is difficult to determine if the difference in pain relieving effect reported in systematic reviews is due to different response to exercise therapy in patients with knee and hip OA, or if it is due to program variables. Also, there is a paucity of studies reporting results from patient education and exercise therapy in terms of quality of life (QOL) and objectively assessed function such as walking speed. Knowledge on the comparable treatment effects when knee and hip patients have participated in the same program and exercised together may help guide prioritization in the health care system.
      Good Life with osteoArthritis in Denmark (GLA:D®) is a patient education and supervised exercise therapy program for people with knee or hip OA launched in Denmark in 2013
      • Skou S.T.
      • Roos E.M.
      Good Life with osteoArthritis in Denmark (GLA:D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
      to facilitate implementation of the clinical guidelines for knee OA released in 2012
      . Both knee and hip OA patients participate in the same program and exercise together in groups. Patient- and clinician-reported outcomes are collected in a clinical registry. Participating in GLA:D® is associated with an on average 25% pain relief, a 10% faster walking speed, a decrease in days on sick leave compared to the year prior to participating in GLA:D® and 1/3 stopping with pharmacological pain relievers
      • Skou S.T.
      • Roos E.M.
      Good Life with osteoArthritis in Denmark (GLA:D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
      ,
      • Thorlund J.B.
      • Roos E.M.
      • Goro P.
      • Ljungcrantz E.G.
      • Gronne D.T.
      • Skou S.T.
      Patients use fewer analgesics following supervised exercise therapy and patient education: an observational study of 16 499 patients with knee or hip osteoarthritis.
      .
      The aim of the study was to compare clinical characteristics and change in outcomes after participating in GLA:D® in patients with symptomatic knee or hip OA as their primary complaint.

      Patients and methods

      Design

      This was a longitudinal, register-based study using data from the GLA:D®-registry in Denmark holding information on patients with pain and/or functional limitations associated with knee or hip OA participating in the GLA:D®-program. The study compared clinical characteristics and change in outcomes immediately following the intervention (∼3 months) and at 12 months. The GLA:D® intervention consists of 2–3 patient education sessions and 12 clinician-supervised exercise therapy sessions. The inclusion criteria for GLA:D® are ‘Joint problems from knee and/or hip that have resulted in contact with the health care system’. The exclusion criteria are: another reason than OA for the problems, for example, tumour; inflammatory joint disease, or sequelae after hip fracture; other symptoms that are more pronounced than the OA problems, for example, chronic, generalized pain, or fibromyalgia and unable to read and understand Danish'. The program was previously described in detail
      • Skou S.T.
      • Roos E.M.
      Good Life with osteoArthritis in Denmark (GLA:D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
      . This report conforms to the STROBE statement for reporting observational studies. According to the ethics committee of the North Denmark Region, ethics approval of GLA:D® was not needed. GLA:D® was approved by the Danish Data Protection Agency (SDU; 10.084). According to the Danish Data Protection Act patient consent was not required as personal data was processed exclusively for research and statistical purposes.

      Participants

      The current study included patients enrolled in the GLA:D®-program between February 2014, when the EuroQoL 5-Dimensions 5-Level (EQ-5D) questionnaire (EQ-5D) was added to the registry, and December 2018. Patients with symptoms from the knee or hip joint as primary complaint and available self-reported baseline data on the self-reported outcomes were included. Analysis of change in the 40 m Fast-paced Walk Test is based on clinician entered data and was included for those who underwent the test at the clinical baseline visit. 12-month follow-up data collected before the end of February 2020 i.e., 13 months after the last patient for this study was enrolled, were included.

      Outcomes

      Outcomes evaluated in the study at all three timepoints included pain intensity during last month evaluated on a 100 mm Visual Analog Scale (VAS) scale (0–100, worst to best), Knee injury and Osteoarthritis Outcome Score (KOOS)/Hip disability and Osteoarthritis Outcomes Score (HOOS) joint-related QOL subscale score (0–100, worst to best)
      • Roos E.M.
      • Roos H.P.
      • Lohmander L.S.
      • Ekdahl C.
      • Beynnon B.D.
      Knee injury and osteoarthritis outcome score (KOOS)--development of a self-administered outcome measure.
      ,
      • Collins N.J.
      • Prinsen C.A.
      • Christensen R.
      • Bartels E.M.
      • Terwee C.B.
      • Roos E.M.
      Knee Injury and Osteoarthritis Outcome Score (KOOS): systematic review and meta-analysis of measurement properties.
      and generic health-related QOL measured with the EQ-5D converted into an index score using time-trade-off based weights from the Danish crosswalk value set (−0.624 to 1, worst to best)
      • Wittrup-Jensen K.U.
      • Lauridsen J.
      • Gudex C.
      • Pedersen K.M.
      Generation of a Danish TTO value set for EQ-5D health states.
      ,
      • Bilbao A.
      • García-Pérez L.
      • Arenaza J.C.
      • García I.
      • Ariza-Cardiel G.
      • Trujillo-Martín E.
      • et al.
      Psychometric properties of the EQ-5D-5L in patients with hip or knee osteoarthritis: reliability, validity and responsiveness.
      . Walking speed evaluated by the 40 m Fast-paced Walk Test (m/sec., worst to best) was measured by the therapist at baseline and immediately following the intervention
      • Dobson F.
      • Hinman R.S.
      • Roos E.M.
      • Abbott J.H.
      • Stratford P.
      • Davis A.M.
      • et al.
      OARSI recommended performance-based tests to assess physical function in people diagnosed with hip or knee osteoarthritis.
      .

      Statistical analysis

      Baseline characteristics are presented for patients with their knee or hip as primary complaint respectively and to explore statistical differences in baseline characteristics between knee and hip patients two-sided unpaired t-test, Wilcoxon Rank Sum Test and chi
      • Richmond S.A.
      • Fukuchi R.K.
      • Ezzat A.
      • Schneider K.
      • Schneider G.
      • Emery C.A.
      Are joint injury, sport activity, physical activity, obesity, or occupational activities predictors for osteoarthritis? A systematic review.
      -test were applied as appropriate.
      The difference in change between knee and hip patients in self-reported outcomes (pain intensity, KOOS/HOOS QOL and EQ-5D) from baseline to immediately after treatment and at 12 months was estimated using linear mixed effect regression (maximum likelihood estimation (ML)) with random intercept, random slope and an unstructured covariate structure for each outcome of interest. Time of measurement (categorical, baseline, immediately after treatment at ∼3 months, 12 months) and primary joint (categorical, knee/hip) and the interaction between time and primary joint were fixed effects and patients nested in clinics were random effects. The impact of knee/hip joint on change in walking speed from baseline to after treatment was estimated using a similar approach, but with only two available data points (i.e., baseline and immediately after treatment at ∼3 months). Crude and adjusted estimates with 95% confidence intervals (95% CI) were reported. Age, gender and BMI were included as covariates in the adjusted analysis as they, based on theoretically driven directed pathways, were considered as potential confounders. Assumptions underlying the models were met.
      The potential influence of joint replacement during follow-up was explored in a sensitivity analysis repeating all linear mixed regression analyses of the self-reported outcomes excluding patients who at 12 months follow-up self-reported to have had knee or hip replacement. To explore if concomitant involvement from both knee and hip joints would affect the result, we repeated all linear mixed regression analyses for the self-reported outcomes categorizing the patients into three groups: 1. Knee involvement only; 2. Hip involvement only; 3. Concomitant involvement of knee and hip.
      To support interpretation of the group mean results we calculated the proportions of patients with knee or hip as their primary complaint who improved at least 15 mm in pain intensity at ∼3 and at 12 months. Several thresholds for a clinically relevant improvement in pain, ranging from 11 to 20 mm, are suggested in the literature and an at least 15 mm improvement is suggested to represent a ‘good, satisfactory effect with occasional episodes of pain or stiffness’ in patients with knee or hip OA
      • Tubach F.
      • Ravaud P.
      • Baron G.
      • Falissard B.
      • Logeart I.
      • Bellamy N.
      • et al.
      Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement.
      . In addition, we calculated the proportions of knee and hip patients who could be categorized as responders as follows; for KOOS/HOOS QOL those reaching an improvement of 15 points
      • Monticone M.
      • Ferrante S.
      • Salvaderi S.
      • Motta L.
      • Cerri C.
      Responsiveness and minimal important changes for the Knee Injury and Osteoarthritis Outcome Score in subjects undergoing rehabilitation after total knee arthroplasty.
      , for EQ-5D reaching an improvement of 0.05
      • Bilbao A.
      • García-Pérez L.
      • Arenaza J.C.
      • García I.
      • Ariza-Cardiel G.
      • Trujillo-Martín E.
      • et al.
      Psychometric properties of the EQ-5D-5L in patients with hip or knee osteoarthritis: reliability, validity and responsiveness.
      and for walking speed reaching an improvement of 0.095 m/s
      • Gilbert A.L.
      • Song J.
      • Cella D.
      • Chang R.W.
      • Dunlop D.D.
      What is an important difference in gait speed in adults with knee osteoarthritis?.
      as these levels are suggested to represent clinically relevant improvements. For this study we were interested in the difference in the proportions of patients with knee and hip OA, respectively, that were responders. This difference in proportions should be large enough to suggest clinicians prioritizing treating one group over the other in case of limited resources. Since we were unable to find any such consensus-based or otherwise established estimates we decided, a priori, that an absolute 10% between-group difference in proportions of responders would not be sufficient, that a 20% difference would probably be sufficient and that a 30% difference would definitely be sufficient to consider prioritizing treatment resources to one type of OA patients over the other.

      Results

      A flow chart of included participants is presented in Fig. 1. 75% of the 32,599 patients provided data on self-reported outcomes after treatment at ∼3 months and 60% at 12 months, and 55 % had complete self-reported data at all three time points. At baseline, information on walking speed was available for 95% of those with self-reported data. Follow-up data for walking speed from immediately after treatment was available for 67% of those with available baseline data. Slightly more participants with incomplete self-reported data had markers of worse health status (e.g., higher BMI, lower educational level, living alone, more co-morbidities, more pain medication, higher pain intensity, lower walking speed, lower physical activity level, lower QOL), these differences were largely similar for those with knee and hip OA as their primary complaint, Supplementary Table S1.
      Baseline characteristics of 24,241 knee patients and 8,358 hip patients included in the study are presented in Table I. On average, patients with knee or hip as their primary complaint had similar distributions of e.g., age, gender, duration of symptoms, use of pain medication, pain intensity, physical function and QOL related to the symptomatic joint and general health. A larger proportion of knee patients than hip patients were obese, and a larger proportion reported bilateral symptoms. Symptoms from the hips in knee patients were less prevalent than symptoms from the knees in hip patients. Previous injury and surgery were far more prevalent in knee patients compared to hip patients, Table I.
      Table IBaseline clinical characteristics in knee and hip patients
      Knee patients

      (n: 24,241)
      Hip patients

      (n: 8,358)
      Difference (95% CI)/P-value
      Evaluated with two-sided unpaired t-test, Wilcoxon Rank Sum Test and chi2-test as appropriate.
      Age (years), mean (SD)64.6 (9.7)66.0 (9.5)1.3 (1.1; 1.6)
      Gender (Female), % (n)71.5 (17,327)72.3 (6,042)0.8 (−0.5; 2.1)
      BMI, % (n)
       <18.5 kg/m20.4 (85)0.8 (63)0.4 (0.2; 0.6)
       ≥18.5 and < 25 kg/m225.1 (6,073)36.0 (3,004)10.9 (9.7; 12.1)
       ≥25 and < 30 kg/m239.1 (9,481)39.9 (3,336)0.8 (0.4; 2.0)
       ≥30 and < 35 kg/m223.3 (5,636)16.9 (1,411)−6.4 (−7.3; −5.4)
       ≥35 kg/m212.2 (2,966)6.5 (544)−5.7 (−6.4; −5.0)
      Symptom duration (mths.), median (IQR)18 (6–48)18 (6–40)0.867
      Previous x-ray
      Self-reported previous X-ray.
      , % (n)
       No13.4 (3,241)11.5 (958)−1.9 (−2.7; −1.1)
       Yes, more than 6 months ago29.9 (7,241)27.0 (2,256)−2.9 (−4.0; −1.8)
       Yes, within last 6 months56.0 (13,560)61.0 (5,094)5.0 (3.8; 6.2)
       Do not know0.7 (165)0.5 (41)−0.2 (−0.4; 0.0)
      Radiographic signs of OA
      Self-reported results from X-ray among those having had X-ray taken.
      , % (n)
       Yes91.3 (18,998)92.5 (6,800)1.1 (0.4; 1.9)
       No4.6 (962)4.5 (333)−0.1 (−0.7; 0.5)
       Do not know4.0 (840)3.0 (217)−1.1 (−1.6; −0.6)
      Previous surgery in worst joint, % (n)28.2 (6,833)4.3 (363)−23.9 (−24.6; −23.1)
      Pain medication
      Self-reported use of pain medication during last 3 months. NSAIDs include systemic or topical drugs and opioids include Tramadol, Morphine, Codeine or other opioids.
      , % (n)
       Overall63.6 (15,404)67.2 (5,614)3.6 (2.4; 4.8)
       Paracetamol53.0 (12,848)57.6 (4,815)4.6 (3.4; 5.8)
       NSAIDs37.6 (9,104)36.2 (3,024)−1.4 (−2.6; −1.8)
       Opioids7.1 (1,713)8.0 (669)0.9 (0.3; 1.6)
      30 s Chair Stand Test (no. rises), mean (SD)11.9 (3.8)12.2 (3.9)0.4 (0.3; 0.5)
      Walking speed (m/sec)
      Walking speed measured with the 40 m Fast-paced Walk test.
      , mean (SD)
      1.48 (0.33)1.48 (0.34)0.0 (0.0; 0.0)
      Marital status, % (n)
       Married or living with others74.7 (18,116)72.7 (6,078)−2.0 (−3.1; −0.9)
       Single25.3 (6,123)27.3 (2,279)2.0 (0.9; 3.1)
       Born in Denmark, % (n)95.9 (23,249)96.5 (8,063)0.6 (0.0; 1.0)
      Educational level, % (n)
       Primary school17.9 (4,337)17.9 (1,492)0.0 (−0.1; 0.1)
       Secondary school11.2 (2,724)10.5 (876)−0.8 (−1.5; 0.0)
       Short-term education19.9 (4,819)19.3 (1,617)−0.5 (−1.5; 0.5)
       Middle-term education39.7 (9,621)40.4 (3,373)0.7 (−0.5; 1.9)
       Long-term education11.3 (2,740)12.0 (1,000)0.7 (−0.1; 1.5)
      Number of comorbidities
      Number of comorbidities calculated from self-report of the following conditions: hypertension, cardiovascular diseases, lung diseases, diabetes, stomach diseases, liver- or kidney diseases, blood diseases, cancer, depression, rheumatoid arthritis, neurological disorders, other medical diseases.
      , % (n)
       037.1 (8,736)38.0 (3,093)0.9 (−0.3; 2.1)
       135.8 (8,433)35.5 (2,885)−0.4 (−1.6; 0.8)
       217.9 (4,207)17.2 (1,400)−0.7 (−2.9; 1.6)
       3 or more9.2 (2,154)9.3 (753)0.1 (−0.6; 0.8)
      Physical activity level
      Physical activity level recorded on the UCLA Activity Scale. The 10 levels were combined to 5 levels for reporting purposes.
      , % (n)
       Physically inactive2.6 (621)2.3 (192)−0.3 (−0.6; 0.1)
       Low physical activity level29.8 (7,211)30.0 (2,503)0.2 (−0.9; 1.3)
       Moderate physical activity level34.2 (8.292)34.8 (2,905)0.5 (−0.6; 1.7)
       High physical activity level27.2 (6,592)26.8 (2,237)−0.4; (−1.5; 0.7)
       Very high physical activity level6.2 (1,507)6.2 (517)0.1 (−0.6; 0.6)
      Previous injury in worst joint
      Self-reported previous injury in worst joint that caused the patient to consult a doctor.
      , % (n)
      52.1 (11,986)33.2 (2,624)−18.9 (−20.2; −17.7)
      Bilateral symptoms, % (n)45.1 (10,930)25.4 (2,123)−19.7 (−20.8; −18.6)
      Hip/knee symptoms
      Hip symptoms in knee patients and knee symptoms in hip patients.
      , % (n)
      17.8 (4,305)35.1 (2,933)17.3 (16.2; 18.5)
      Pain intensity (VAS 0–100), mean (SD)47.6 (22.1)47.6 (21.6)0.0 (−0.5; 0.6)
      KOOS/HOOS QOL, mean (SD)45.1 (15.2)47.1 (15.9)2.1 (1.7; 2.5)
      EQ-5D Index score, median (IQR)0.732 (0.699–0.780)0.719 (0.650–0.755)<0.001
      Missing values: Symptom duration missing 3,175 knee patients and 1,065 hip patients mainly due to technical problems. X-Ray missing 34 knee patients and 9 hip patients. Chair stand test missing 900 knee patients and 312 hip patients. Walk test missing 1,351 knee patients and 438 hip patients. Marital status missing 2 knee patients and 1 hip patient. Born in Denmark missing 3 knee patients. Number of comorbidities missing 711 knee patients and 227 hip patients mainly due to late introduction to the registry. Physical activity level missing 18 knee patients and 4 hip patients. Previous injury missing 1,251 knee patients and 454 hip patients mainly due to late introduction to the registry.
      Evaluated with two-sided unpaired t-test, Wilcoxon Rank Sum Test and chi2-test as appropriate.
      Self-reported previous X-ray.
      Self-reported results from X-ray among those having had X-ray taken.
      § Self-reported use of pain medication during last 3 months. NSAIDs include systemic or topical drugs and opioids include Tramadol, Morphine, Codeine or other opioids.
      Walking speed measured with the 40 m Fast-paced Walk test.
      Number of comorbidities calculated from self-report of the following conditions: hypertension, cardiovascular diseases, lung diseases, diabetes, stomach diseases, liver- or kidney diseases, blood diseases, cancer, depression, rheumatoid arthritis, neurological disorders, other medical diseases.
      # Physical activity level recorded on the UCLA Activity Scale. The 10 levels were combined to 5 levels for reporting purposes.
      †† Self-reported previous injury in worst joint that caused the patient to consult a doctor.
      ‡‡ Hip symptoms in knee patients and knee symptoms in hip patients.
      At baseline, knee patients reported on average slightly worse joint-related QOL (45.1 (44.9; 45.4)) compared to hip patients (47.1 (46.7; 47.5)) but slightly better health-related QOL 0.711 (0.709; 0.714) vs 0.697 (0.694; 0.700), while pain intensity and walking speed were similar, Fig. 2 and Supplementary Table S2.
      Fig. 2
      Fig. 2Results from the adjusted mixed effects models for pain intensity (a), Joint related quality of life (b), Generic quality of life (c) and walking speed (d). Results are shown for knee and hip patients as means with 95% Confidence Intervals.
      Immediately after treatment at ∼3 months, slightly on average larger improvements were consistently seen for knee patients compared to hip patients; 2.1 (1.5; 2.8) mm greater improvement in pain intensity, 1.1 (0.7; 1.5) point greater improvement in KOOS QOL score, 0.010 (0.007; 0.013) greater improvement in EQ-5D index score and 0.02 (0.01; 0.02) m/sec greater improvement in walking speed. At 12 months, where only self-reported outcomes were collected, the slight differences between patients with their knee or hip as their primary complaint were equalized, Fig. 2 and Supplementary Table S2.
      We conducted a sensitivity analysis for the 12-month self-reported data where participants reporting to have had a total joint replacement in any knee or hip joint during follow up were excluded. Excluding TJR patients revealed small but consistent differences in improvement favoring the knee OA patients, 3.5 (2.7; 4.3) mm greater improvement in pain intensity, 1.9 (1.3; 2.5) point greater improvement in KOOS QOL score, and 0.019 (0.015; 0.023) greater improvement in EQ-5D index score, Supplementary Table S3. Repeating the analyses for the three groups with knee involvement only, hip involvement only and those with involvement from knee and hip did not reveal any substantially different results (Supplementary appendix, Table S4).
      To facilitate interpretation of the group mean analysis we conducted responder analysis for all outcomes (Supplementary appendix, Table S5). Immediately after the program at ∼3 months, 44% (7898 of 18,054) patients with knee OA as their primary complaint and 40% (2485 of 6248) of patients with hip OA as their primary complaint experienced a decrease in pain intensity of 15 mm or more. At 12 months the corresponding numbers were 44% for both knee and hip OA patients (6380 of 14,416 knee patients and 2171 of 4983 hip patients). When participants reporting to have had total joint replacement in any knee or hip joint during the 12 months follow up were excluded, 43% (5666 of 13,293) knee OA patients and 37% (1546 of 4127) hip OA patients reported an improvement of at least 15 mm at 12 months. In line with this, responder analyses for KOOS/HOOS QOL, EQ5D index score and walking speed showed only minor differences between knee and hip patients (Supplementary appendix, Table S5). The greatest absolute difference in any analysis in proportion of responders between knee and hip patients was 6%, which did not exceed the a priori defined absolute difference of 20% suggested to possibly affect clinical prioritization.

      Discussion

      This is the first large scale study to compare clinical characteristics and one of the first to separately report change in outcomes in patients with knee or hip OA treated in primary care
      • Dell'Isola A.
      • Jonsson T.
      • Ranstam J.
      • Dahlberg L.E.
      • Ekvall Hansson E.
      Education, home exercise, and supervised exercise for people with hip and knee osteoarthritis as part of a nationwide implementation program: data from the better management of patients with osteoarthritis registry.
      ,
      • Holm I.
      • Pripp A.H.
      • Risberg M.A.
      The active with OsteoArthritis (AktivA) Physiotherapy implementation model: a patient education, supervised exercise and self-management program for patients with mild to moderate osteoarthritis of the knee or hip joint. A national register study with a two-year follow-up.
      . In addition to the commonly reported treatment results for pain and function we also present data on joint- and generic health-related QOL and walking speed. Clinicians should expect patients presenting with symptomatic knee and hip OA as their primary complaints in primary care to be on average more alike than different. While they on average differed in patterns of obesity, previous injury and surgery and other affected knee and hip joints, they were on average alike in other studied clinical characteristics including self-reported outcomes. Immediately after treatment, similar pain reductions of 13.0 and 10.9 mm were seen for patients with knee and hip OA. At 12 months, we found sustained pain reduction in both groups. Excluding those having had their knee or hip replaced during the 12-months, the mean difference in pain relief between patients with knee and hip OA was 3.5 mm, and 6% more of the knee patients reached the threshold for a clinically relevant change in absolute comparison to the hip patients. These differences are still very small and should not impact on clinical prioritizations. Differences in proportions having surgery, timing of surgery and length of post-operative rehab may contribute to explain the minor difference noted between knee and hip OA patients in our study.
      There is a paucity of high-quality studies comparing the treatment results from exercise therapy in patients with knee and hip OA. While 54 and 10 randomized trials were included in the most recent Cochrane-reports of exercise for knee and hip OA
      • Fransen M.
      • McConnell S.
      • Harmer A.R.
      • Van der Esch M.
      • Simic M.
      • Bennell K.L.
      Exercise for osteoarthritis of the knee.
      ,
      • Fransen M.
      • McConnell S.
      • Hernandez-Molina G.
      • Reichenbach S.
      Exercise for Osteoarthritis of the Hip.
      , respectively, only five trials included both knee and hip patients
      • van Baar M.E.
      • Dekker J.
      • Lemmens J.A.
      • Oostendorp R.A.
      • Bijlsma J.W.
      Pain and disability in patients with osteoarthritis of hip or knee: the relationship with articular, kinesiological, and psychological characteristics.
      • Hopman-Rock M.
      • Westhoff M.H.
      The effects of a health educational and exercise program for older adults with osteoarthritis for the hip or knee.
      • Fransen M.
      • Nairn L.
      • Winstanley J.
      • Lam P.
      • Edmonds J.
      Physical activity for osteoarthritis management: a randomized controlled clinical trial evaluating hydrotherapy or Tai Chi classes.
      • Abbott J.H.
      • Robertson M.C.
      • Chapple C.
      • Pinto D.
      • Wright A.A.
      • Leon de la Barra S.
      • et al.
      Manual therapy, exercise therapy, or both, in addition to usual care, for osteoarthritis of the hip or knee: a randomized controlled trial. 1: clinical effectiveness.
      • Foley A.
      • Halbert J.
      • Hewitt T.
      • Crotty M.
      Does hydrotherapy improve strength and physical function in patients with osteoarthritis--a randomised controlled trial comparing a gym based and a hydrotherapy based strengthening programme.
      . Only one trial was adequately powered to potentially detect differences (n = 181)
      • van Baar M.E.
      • Dekker J.
      • Lemmens J.A.
      • Oostendorp R.A.
      • Bijlsma J.W.
      Pain and disability in patients with osteoarthritis of hip or knee: the relationship with articular, kinesiological, and psychological characteristics.
      , and no trial reported results separately for knee and hip OA in the original reports. In the Cochrane reports, after obtaining data from the respective authors, the calculated pain-relieving effect in these five studies was always larger for patients with hip OA compared to patients with knee OA. This is contrary to the overall synthesis of the results, were exercise for knee OA was associated with a larger pain relieving effect than hip OA with standardized mean differences of 0.49 and 0.38, respectively
      • Fransen M.
      • McConnell S.
      • Harmer A.R.
      • Van der Esch M.
      • Simic M.
      • Bennell K.L.
      Exercise for osteoarthritis of the knee.
      ,
      • Fransen M.
      • McConnell S.
      • Hernandez-Molina G.
      • Reichenbach S.
      Exercise for Osteoarthritis of the Hip.
      . In support of the individual trial results from the Cochrane reports, the pain relieving effect was moderate in an exercise study powered to allow for stratified analysis (n = 165) in patients with hip OA awaiting total hip replacement (SMD 0.57 (0.20, 0.94)), and only small in patients with knee OA awaiting total knee replacement (0.15 (−0.21, 0.52))
      • Villadsen A.
      • Overgaard S.
      • Holsgaard-Larsen A.
      • Christensen R.
      • Roos E.M.
      Immediate efficacy of neuromuscular exercise in patients with severe osteoarthritis of the hip or knee: a secondary analysis from a randomized controlled trial.
      .
      The current observational study is by far the largest where knee and hip OA patients have had patient education and exercised together in blended groups, allowing for direct comparison of change in outcomes when having had the same intervention. In GLA:D®, all treating clinicians have had 2 days of theoretical and practical training in delivering the same 1-h neuromuscular exercise program to knee and hip patients, with starting level and progression tailored to the individual patient supported by the use of a pain monitoring scale and an exercise diary
      • Skou S.T.
      • Roos E.M.
      Good Life with osteoArthritis in Denmark (GLA:D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
      ,
      • Ageberg E.
      • Link A.
      • Roos E.M.
      Feasibility of neuromuscular training in patients with severe hip or knee OA: the individualized goal-based NEMEX-TJR training program.
      .
      Two other nationwide osteoarthritis care programs have previously reported outcomes for knee and hip patients separately
      • Dell'Isola A.
      • Jonsson T.
      • Ranstam J.
      • Dahlberg L.E.
      • Ekvall Hansson E.
      Education, home exercise, and supervised exercise for people with hip and knee osteoarthritis as part of a nationwide implementation program: data from the better management of patients with osteoarthritis registry.
      ,
      • Holm I.
      • Pripp A.H.
      • Risberg M.A.
      The active with OsteoArthritis (AktivA) Physiotherapy implementation model: a patient education, supervised exercise and self-management program for patients with mild to moderate osteoarthritis of the knee or hip joint. A national register study with a two-year follow-up.
      . These programs are less well-suited for the purpose of comparing outcomes for knee and hip patients since the exercise therapy programs delivered to the individual knee or hip OA patient may vary with regard to content, dose and delivery mode to a larger extent than in GLA:D®. While the educational component is very similar across the three programs, the exercise programs in BOA and AktivA are not as standardized as in GLA:D®. In the Swedish BOA program, physiotherapists were asked to follow neuromuscular principles but were not trained in delivering one specific exercise program. There is no report on what type of exercises (aerobic, strength, neuromuscular) were used. The mode of delivery (home, individual or group) and number of completed sessions varied widely, with less than 25% participating in at least 10 supervised sessions
      • Limbäck Svensson G.
      • Wetterling K.
      • Anderson L.
      BOA Årsrapport 2019.
      .
      The Norwegian AktivA exercise program is tailored by the treating clinician from 15 suggested neuromuscular and strength exercises to meet each individual patient's needs and wishes
      • Holm I.
      • Pripp A.H.
      • Risberg M.A.
      The active with OsteoArthritis (AktivA) Physiotherapy implementation model: a patient education, supervised exercise and self-management program for patients with mild to moderate osteoarthritis of the knee or hip joint. A national register study with a two-year follow-up.
      . It is therefore not known to which extent the number and type of exercises chosen for knee and hip patients are the same. Also, the duration of each session was not reported. Similar to the current study, adherence in AktivA was high and 78% completed at least 10 supervised sessions
      • Holm I.
      • Pripp A.H.
      • Risberg M.A.
      The active with OsteoArthritis (AktivA) Physiotherapy implementation model: a patient education, supervised exercise and self-management program for patients with mild to moderate osteoarthritis of the knee or hip joint. A national register study with a two-year follow-up.
      .
      Differences in treatment response between patients with knee and hip OA has been noted for the pharmacological agent Naproxen. Patients with knee OA improved their self-reported pain and function with about 5 mm more than patients with hip OA at 6 weeks following Naproxen treatment, corresponding to large effect sizes of about 0.8 for knee OA patients and moderate effect sizes between 0.5 and 0.6 for hip OA patients
      • Svensson O.
      • Malmenas M.
      • Fajutrao L.
      • Roos E.M.
      • Lohmander S.
      Greater reduction of knee than hip pain in osteoarthritis treated with naproxen, as evaluated by WOMAC and SF-36.
      . This finding supports that treatment results may differ somewhat between patients with knee and hip OA when enrolled in the same study and subjected to same treatment. Differing treatment response in knee and hip OA patients could be related to differing pain-related factors in patients with knee or hip OA. A recent meta-analysis found moderate quality evidence that knee pain was associated with radiographic disease, effusion, meniscal damage and older age while hip pain was associated with large bone marrow lesions and chronic widespread pain
      • Sandhar S.
      • Smith T.O.
      • Toor K.
      • Howe F.
      • Sofat N.
      Risk factors for pain and functional impairment in people with knee and hip osteoarthritis: a systematic review and meta-analysis.
      . Differing treatment response warrant further study, and reporting treatment response separately for knee or hip OA patients facilitates such work.
      Major strengths of this study are the large sample size, and that all patients were treated together in groups following the same program supervised by certified clinicians who deliver the program in clinical practices situated in rural and urban areas nationwide. The major limitation is the lack of a control group, and thus the improvements seen cannot be attributed to the program per se. Another limitation is that not all patient provided complete data or attended all follow-ups. However these limitations affect the results for patients with knee and hip OA alike and are therefore not expected to impact on the comparison.

      Conclusion

      Clinicians should expect patients presenting with symptomatic knee and hip OA in primary care to be on average more alike than different. Clinically relevant improvements were seen in both knee and hip OA patients. The differences in change in outcomes between types of OA patients were small and should not impact on prioritizing one type of OA patients over the other for treatment with patient education and supervised exercise therapy.

      Author contributions

      Study conception and design: Roos, Skou, Grønne.
      Acquisition of data: Roos, Skou.
      Analysis and interpretation of data: Roos, Skou, Grønne, Thorlund.
      Drafting the article or revising it critically for important intellectual content: Roos, Skou, Grønne, Thorlund.
      Final approval of the article: Roos, Skou, Grønne, Thorlund.
      Roos and Skou takes responsibility for the integrity of the work as a whole, from inception to finished article.

      Conflict of interest

      Dr. Roos is deputy editor of Osteoarthritis and Cartilage, the developer of the KOOS and several other freely available patient-reported outcome measures and co-founder of Good Life with Osteoarthritis in Denmark (GLA:D®), a not-for profit initiative hosted at University of Southern Denmark aimed at implementing clinical guidelines for osteoarthritis in clinical practice.
      Dr. Skou is associate editor of the Journal of Orthopaedic & Sports Physical Therapy, has received grants from The Lundbeck Foundation, personal fees from Munksgaard, all of which are outside the submitted work. He is co-founder of GLA:D®.
      Dr. Thorlund report a research grant from Pfizer outside the submitted work.
      MSc Grønne declares no conflict of interest.

      Funding sources

      The Danish Physiotherapist Association's fund for research, education and practice development; the Danish Rheumatism Association; and the Physiotherapy Practice Foundation supported the start-up phase of GLA:D®.
      Dr. Skou is currently funded by a grant from Region Zealand (Exercise First) and two grants from the European Union's Horizon 2020 research and innovation program, one from the European Research Council (MOBILIZE, grant agreement No 801790) and the other under grant agreement No 945377 (ESCAPE).
      None of the funding sources had any role in the study other than to providing funding.

      Acknowledgments

      The authors would like to thank the clinicians and patients involved in collecting data for GLA:D®.

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • Diseases G.B.D.
        • Injuries C.
        Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019.
        Lancet. 2020; 396: 1204-1222
        • Richmond S.A.
        • Fukuchi R.K.
        • Ezzat A.
        • Schneider K.
        • Schneider G.
        • Emery C.A.
        Are joint injury, sport activity, physical activity, obesity, or occupational activities predictors for osteoarthritis? A systematic review.
        J Orthop Sports Phys Ther. 2013; 43: 515-B519
        • Hall M.
        • van der Esch M.
        • Hinman R.S.
        • Peat G.
        • de Zwart A.
        • Quicke J.G.
        • et al.
        How does hip osteoarthritis differ from knee osteoarthritis?.
        Osteoarthritis Cartilage. 2022; 30: 32-41
        • Fransen M.
        • McConnell S.
        • Harmer A.R.
        • Van der Esch M.
        • Simic M.
        • Bennell K.L.
        Exercise for osteoarthritis of the knee.
        Cochrane Database Syst Rev. 2015; 1: CD004376
        • Fransen M.
        • McConnell S.
        • Hernandez-Molina G.
        • Reichenbach S.
        Exercise for Osteoarthritis of the Hip.
        Cochrane Database Syst Rev, 2014: CD007912
        • van Baar M.E.
        • Dekker J.
        • Lemmens J.A.
        • Oostendorp R.A.
        • Bijlsma J.W.
        Pain and disability in patients with osteoarthritis of hip or knee: the relationship with articular, kinesiological, and psychological characteristics.
        J Rheumatol. 1998; 25: 125-133
        • Hopman-Rock M.
        • Westhoff M.H.
        The effects of a health educational and exercise program for older adults with osteoarthritis for the hip or knee.
        J Rheumatol. 2000; 27: 1947-1954
        • Fransen M.
        • Nairn L.
        • Winstanley J.
        • Lam P.
        • Edmonds J.
        Physical activity for osteoarthritis management: a randomized controlled clinical trial evaluating hydrotherapy or Tai Chi classes.
        Arthritis Rheum. 2007; 57: 407-414
        • Abbott J.H.
        • Robertson M.C.
        • Chapple C.
        • Pinto D.
        • Wright A.A.
        • Leon de la Barra S.
        • et al.
        Manual therapy, exercise therapy, or both, in addition to usual care, for osteoarthritis of the hip or knee: a randomized controlled trial. 1: clinical effectiveness.
        Osteoarthritis Cartilage. 2013; 21: 525-534
        • Foley A.
        • Halbert J.
        • Hewitt T.
        • Crotty M.
        Does hydrotherapy improve strength and physical function in patients with osteoarthritis--a randomised controlled trial comparing a gym based and a hydrotherapy based strengthening programme.
        Ann Rheum Dis. 2003; 62: 1162-1167
        • Skou S.T.
        • Roos E.M.
        Good Life with osteoArthritis in Denmark (GLA:D): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide.
        BMC Muscoskel Disord. 2017; 18: 72
      1. Knæartrose – Nationale Kliniske Retningslinjer Og Faglige Visitationsretningslinjer. København: Sundhedsstyrelsen, 2012
        • Thorlund J.B.
        • Roos E.M.
        • Goro P.
        • Ljungcrantz E.G.
        • Gronne D.T.
        • Skou S.T.
        Patients use fewer analgesics following supervised exercise therapy and patient education: an observational study of 16 499 patients with knee or hip osteoarthritis.
        Br J Sports Med. 2021 Jun; 55 (Epub 2020 Sep 21. PMID: 32958468): 670-675https://doi.org/10.1136/bjsports-2019-101265
        • Roos E.M.
        • Roos H.P.
        • Lohmander L.S.
        • Ekdahl C.
        • Beynnon B.D.
        Knee injury and osteoarthritis outcome score (KOOS)--development of a self-administered outcome measure.
        J Orthop Sports Phys Ther. 1998; 28: 88-96
        • Collins N.J.
        • Prinsen C.A.
        • Christensen R.
        • Bartels E.M.
        • Terwee C.B.
        • Roos E.M.
        Knee Injury and Osteoarthritis Outcome Score (KOOS): systematic review and meta-analysis of measurement properties.
        Osteoarthritis Cartilage. 2016; 24: 1317-1329
        • Wittrup-Jensen K.U.
        • Lauridsen J.
        • Gudex C.
        • Pedersen K.M.
        Generation of a Danish TTO value set for EQ-5D health states.
        Scand J Publ Health. 2009; 37: 459-466
        • Bilbao A.
        • García-Pérez L.
        • Arenaza J.C.
        • García I.
        • Ariza-Cardiel G.
        • Trujillo-Martín E.
        • et al.
        Psychometric properties of the EQ-5D-5L in patients with hip or knee osteoarthritis: reliability, validity and responsiveness.
        Qual Life Res : Int J Qual Life Aspects Treat Care Rehabilit. 2018; 27: 2897-2908
        • Dobson F.
        • Hinman R.S.
        • Roos E.M.
        • Abbott J.H.
        • Stratford P.
        • Davis A.M.
        • et al.
        OARSI recommended performance-based tests to assess physical function in people diagnosed with hip or knee osteoarthritis.
        Osteoarthritis Cartilage. 2013; 21: 1042-1052
        • Tubach F.
        • Ravaud P.
        • Baron G.
        • Falissard B.
        • Logeart I.
        • Bellamy N.
        • et al.
        Evaluation of clinically relevant changes in patient reported outcomes in knee and hip osteoarthritis: the minimal clinically important improvement.
        Ann Rheum Dis. 2005; 64: 29-33
        • Monticone M.
        • Ferrante S.
        • Salvaderi S.
        • Motta L.
        • Cerri C.
        Responsiveness and minimal important changes for the Knee Injury and Osteoarthritis Outcome Score in subjects undergoing rehabilitation after total knee arthroplasty.
        Am J Phys Med Rehabil. 2013; 92: 864-870
        • Gilbert A.L.
        • Song J.
        • Cella D.
        • Chang R.W.
        • Dunlop D.D.
        What is an important difference in gait speed in adults with knee osteoarthritis?.
        Arthritis Care Res. 2021 Apr; 73: 559-565https://doi.org/10.1002/acr.24159
        • Dell'Isola A.
        • Jonsson T.
        • Ranstam J.
        • Dahlberg L.E.
        • Ekvall Hansson E.
        Education, home exercise, and supervised exercise for people with hip and knee osteoarthritis as part of a nationwide implementation program: data from the better management of patients with osteoarthritis registry.
        Arthritis Care Res. 2020; 72: 201-207
        • Holm I.
        • Pripp A.H.
        • Risberg M.A.
        The active with OsteoArthritis (AktivA) Physiotherapy implementation model: a patient education, supervised exercise and self-management program for patients with mild to moderate osteoarthritis of the knee or hip joint. A national register study with a two-year follow-up.
        J Clin Med. 2020; 9
        • Villadsen A.
        • Overgaard S.
        • Holsgaard-Larsen A.
        • Christensen R.
        • Roos E.M.
        Immediate efficacy of neuromuscular exercise in patients with severe osteoarthritis of the hip or knee: a secondary analysis from a randomized controlled trial.
        J Rheumatol. 2014; 41: 1385-1394
        • Ageberg E.
        • Link A.
        • Roos E.M.
        Feasibility of neuromuscular training in patients with severe hip or knee OA: the individualized goal-based NEMEX-TJR training program.
        BMC Muscoskel Disord. 2010; 11: 126
        • Limbäck Svensson G.
        • Wetterling K.
        • Anderson L.
        BOA Årsrapport 2019.
        2020
        • Svensson O.
        • Malmenas M.
        • Fajutrao L.
        • Roos E.M.
        • Lohmander S.
        Greater reduction of knee than hip pain in osteoarthritis treated with naproxen, as evaluated by WOMAC and SF-36.
        Ann Rheum Dis. 2006 Jun; 65: 781-784https://doi.org/10.1136/ard.2005.040519
        • Sandhar S.
        • Smith T.O.
        • Toor K.
        • Howe F.
        • Sofat N.
        Risk factors for pain and functional impairment in people with knee and hip osteoarthritis: a systematic review and meta-analysis.
        BMJ Open. 2020; 10e038720