The clinical profile of individuals with symptoms of knee osteoarthritis referred to secondary care in Denmark: Across-sectional study of 282 people.

Abstract

Osteoarthritis (OA) is a leading global contributor to years lived with disability and accounts for the largest global increase in years lived with disability during the past 30 years among musculoskeletal conditions (Cieza et al., 2020). International guidelines recommend exercise therapy alongside patient education and weight management (if needed) as first-line treatment for all individuals with knee OA (KOA) (Bannuru et al., 2019). However, clinical care pathways for KOA do not seem to reflect these recommendations and community-based OA care is generally of suboptimal quality (Hagen et al., 2016). This is mirrored in a Danish health care setting, where only about 1/3 of patients who were referred to an orthopaedic assessment in secondary care (the hospital) had received exercise therapy and patient education during the preceding year in the period March 2018 to February 2019 (Ingelsrud et al., 2020). These numbers give reasons to believe that a substantial part of the current KOA waiting lists for orthopaedic assessment in secondary health care is comprised of inappropriate referrals (Mikkelsen et al., 2019), that is, artificially inflating waiting lists and wasting both the patient's and the orthopaedic surgeon's time in an inefficient use of the scarce resources in secondary health care. These scarce resources in secondary health care were evident during the extraordinary events of the COVID-19 pandemic, which created a big backlog of KOA patients waitlisted for orthopaedic assessment in secondary health care around Europe (Hampton et al., 2021; Uimonen et al., 2021). In Denmark, this led to a COVID-19 enforced initiative at one public hospital, introducing orthopaedic assessments by specially trained physiotherapists to reduce waiting lists and provide quicker referrals to appropriate KOA care. Using data from this clinical initiative, this report aims to describe patient and clinical characteristics along with patients' treatment preferences and expectations of treatment outcomes among wait-listed KOA patients to better understand referral patterns and health care seeking behaviour in current clinical KOA practice in Denmark. This was an observational cohort study conforming to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement for reporting of observational studies. This study complied with the principles of the declaration of Helsinki and was waived from applying for approval with the Scientific Ethics committee of Region Zealand, Denmark (J.nr. 20-000013). The study was approved by the Danish Data Protection Agency (REG-149-2021). All patients received written and oral information about the study and use of patient data for analytic purposes and signed an informed consent. The study is based on cohort data collected as part of a COVID-19 enforced clinical initiative to reduce waiting lists for people with symptoms of KOA referred to the Orthopaedic Department at Næstved Hospital, Region Zealand, Denmark. Individuals with symptoms of KOA referred from primary care practice to secondary care for an orthopaedic assessment were distributed between orthopaedic surgeons and physiotherapists (PT), specially trained in knee assessments. For most initial referrals from primary care, it was not possible to derive information on prior treatment received. Thus, most referrals were distributed at random between orthopaedic surgeons and PTs. The data reported in this study is based on those KOA patients that were distributed to an orthopaedic assessment performed by the PT. The PT first performed an extensive objective assessment of the knee as well as collecting self-reported outcomes. Based on the assessment, the patient and the PT then decided whether he or she should start with an exercise therapy programme (either tele-based or in-person in the municipalities) or should be referred to the orthopaedic surgeon. Outcomes were collected and stored in two ways: (a) Through data entries in the national registry of Good Life with Osteoarthritis in Denmark (GLA:D®), which is a non-profit initiative, providing exercise and education for patients with knee and hip OA symptoms hosted at the University of Southern Denmark (Skou and Roos, 2017). (b) Through data entries in the electronic journal system of Region Zealand (“Sundhedsplatformen”). Additionally, as a standard part of their visit at the hospital, antero-posterior radiographs of the patient's index knee were obtained and subsequently evaluated. We included variables collected by the PT during the initial knee assessment, covering pain, function, prior treatment, the patient's treatment preferences and expectations of treatment outcomes along with structural OA (radiographs). We also included an overview of any re-referrals to the Orthopaedic Department at Næstved Hospital within a 6-month window from the initial knee assessment. Age, sex, and body mass index (BMI [kg/m2]). Pain (numeric rating scale (NRS) 0–10) during the past week, at rest, during the initiation of movement, during stair climbing and level walking. Subscale function from the short-form version of the Knee injury and Osteoarthritis Outcome Score (KOOS-12) (Roos et al., 2007). Structural OA severity (Kellgren & Lawrence grade, ranging from 0 (none) to 4 (severe)) (Kellgren and Lawrence, 1957), clinical OA diagnosis, using the European League Against Rheumatism (EULAR) scoring criteria (Zhang et al., 2010). Quality and appropriateness of prior treatment using the osteoarthritis quality indicator (OA-QI) (Østerås et al., 2013), and patient preferences for treatment (single-item answers) and expectations (5-point Likert scale; very small to very strong). Data were analysed using descriptive statistics presented as means (with standard deviations (SD)), Confidence Intervals (95% CI) or frequencies (with proportions, n (%)). During the period from May 2020 (start of clinical initiative) to May 2021 (end of clinical initiative), 2224 individuals with KOA symptoms were referred to an orthopaedic assessment at Næstved Hospital. Of these, 282 individuals (13%) were assessed by one of two specially trained PTs. The skewed distribution of referrals is due to the fact that Næstved Hospital has a high-volume orthopaedic department with ≥10 orthopaedic consultants. The 282 individuals were split into two groups depending on whether they required a referral to the orthopaedic surgeon (n = 25) or not (n = 257) (Figure 1, flowchart). Reasons for referring to the orthopaedic surgeon were multifactorial, based on the clinical assessment along with patients' treatment preferences and other conditions/symptoms requiring a second opinion by the orthopaedic surgeon. The group of patients, not referred to the orthopaedic surgeon were referred to exercise therapy online or onsite (Holm et al., 2023). For these patients, we searched for re-referrals to the orthopaedic surgeon in the electronic journal system up to November 2021 (final possible 6-month re-referral). A total of 74 patients (29%) were referred to the orthopaedic surgeon during this period, constituting planned (n = 17, 7%) and not planned (n = 57, 22%) re-referrals at initial PT-led orthopaedic assessment (Figure 1, flowchart). Flowchart of patients assessed by physiotherapists after referral to orthopaedic assessment. Overall, the mean age (SD) was 67.5 years (9.7), mean BMI (SD) was 29.8 kg/m2 (4.8) and 150 (53%) were female. Most patients had a radiographic KL grade of 3 (n = 125, 44%), followed by 2 (n = 77, 27%) and 4 (n = 69, 25%). 158 (56%) patients were identified as having a EULAR clinical diagnosis of KOA. With few exceptions, the two groups (referral to orthopaedic surgeon or not) had similar individual and clinical characteristics (Table 1). Overall, 214 patients (75%) chose exercise therapy with a PT as their preferred treatment. 181 patients (85%) who preferred exercise therapy with a PT as treatment had a moderate or strong expectation that this would reduce their pain and 190 (89%) had a moderate or strong expectation that it would improve their function. 214 patients (76%) had not received information about KOA in the last year and 199 (70%) had not consulted a PT within the last year. There were no noteworthy differences in prior treatment, treatment preferences and expectations among the two groups (referral to orthopaedic surgeon or not) (Table 2). The mean pass rate for OA-QI was 30% (proportion of ‘yes’ answers out of all ‘yes’/‘no’ answers throughout the 15 OA-QI questions) (Table 3). Considering that 9 out of 10 patients (91%) assessed by a PT in our study did not require a referral to the orthopaedic surgeon, our findings indicate that most of the assessed referrals from primary care to secondary hospital care were inappropriate. This is supported by the fact that 3 out of 4 patients had not received the recommended first line OA care (exercise and education) within the preceding year. These numbers clearly demonstrate an alarming underutilisation of recommended first-line OA care for patients with symptoms of KOA. Based on the OA-QI questionnaire, our included sample of patients had a mean pass rate of 30%, indicating that roughly two-thirds of OA care quality indicators are not achieved prior to orthopaedic referral. This number is similar to previous numbers reported for Denmark (median 23%–29%) (Ingelsrud et al., 2020; Østerås et al., 2015) but somewhat lower than for Norway (median 50%), Portugal (median 48%) and the UK (median 55%) (Østerås et al., 2015). Although numbers differ by country (at least partly due to different health systems), there is still overall room for improvement in care quality across these countries (Østerås et al., 2015). One driving factor of inappropriate referrals and inferior quality of care for KOA patients may be the attitudes and beliefs of the referrer, as highlighted in the scoping review by Nissen, Holm and colleagues (Nissen et al., 2021). Especially, general practitioners seem to have a prevailing outdated narrative of KOA as a wear and tear disease with knee replacement surgery as an inevitable end-stage treatment (Nissen et al., 2021). Coupled with a lingering less-than-positive attitude towards exercise as a treatment option (Nissen et al., 2021), this severely restricts referrals to exercise and education as first-line KOA treatment, despite existing clinical guidelines. Notably, our findings of roughly 3 out of 4 patients preferring PT-led exercise therapy as treatmente is somewhat in contrast to the findings by Ingelsrud and colleagues, who found that two-thirds of patients expected to be waitlisted for surgery (Ingelsrud et al., 2020). Although somewhat speculatively, our findings seem to question the weight of patient preferences in current treatment pathways, seemingly pointing to other reasons for inconsistent referral patterns. Patients also seemed to generally have a strong belief that exercise therapy could improve their symptoms—reflected by approximately 9 in 10 patients (85%–89%) having moderate to very strong expectations that exercise therapy would reduce their knee pain and improve function. 183 patients or 7 out of 10 patients who were referred to exercise therapy did not have a re-referral in the public system within 6 months (see Figure 1). Although we have no data on exercise therapy adherence rates or other referrals (e.g., private care referrals), this may imply that exercise therapy has been an appropriate first-line treatment for most of the assessed patients. Interestingly, individual characteristics, clinical characteristics, or treatment preferences did not seem to differ between those patients who did or did not require further referral to the orthopaedic surgeon upon the initial PT-led orthopaedic assessment. Since initial referrals were distributed predominantly at random between assessment with orthopaedic surgeons or PTs, we believe that the current sample of KOA patients referred to secondary public hospital care provides a representative outlook of current referral patterns. The included sample of patients with KOA symptoms were all initially assessed by a PT, who also performed the screening of all self-reported outcomes, including treatment preferences and expectations. Even though PTs were instructed not to initially advise in favour of any treatment, this setting may have introduced some desirability bias, where patients were inclined to favour exercise therapy with a PT over other treatments due to the screening context. This may be one underlying factor for the high number of patients preferring this treatment and needs to be taken into consideration. These numbers are based on a selected sample of patients (n = 282), representing 13% of a 1-year caseload to one Danish public hospital in one of five Danish Regions (Region Zealand); thus, we cannot necessarily generalise these findings to the entire Danish public health care sector or internationally. Based on our data of 282 patients, 9 out of 10 with symptoms of KOA are inappropriately referred to an orthopaedic assessment in secondary health care before receiving recommended first-line treatment. Coupled with the fact that 3/4 patients prefer a PT-led exercise therapy programme over surgery and other treatments, our report reinforces the urgent call for optimising current KOA care pathways to ensure that treatment is based on both evidence-informed recommendations and patient values and preferences. Study conception and design: Hinz, Bruhn, Tang, Skou, Holm. Recruitment of patients: Nyberg, Holm. Acquisition of data: Hinz, Nyberg, Holm. Analysis and interpretation of data: Hinz, Bruhn, Holm. Draughting the article or revising it critically for important intellectual content: Hinz, Bruhn, Tang, Nyberg, Skou, Holm. Final approval of the article: Hinz, Bruhn, Tang, Nyberg, Skou, Holm. All authors had full access to all the data (including statistical reports and tables) in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. We would like to thank physiotherapists, doctors, nurses, secretaries and other staff at the Department of Physiotherapy and Occupational Therapy and the Orthopedic Department at Næstved, Slagelse and Ringsted (NSR) Hospitals involved in collecting data for this study. We would also like to thank Dorte T. Grønne, database manager of GLA:D®, for preparing and providing data from the GLA:D® registry. Finally, we would like to thank all the individuals with knee osteoarthritis who provided data for this study. This study was funded by a program grant from Region Zealand (Exercise First). Tang is funded by The Danish Health Confederation through the Development and Research Fund for financial support (project no. 2703) and Næstved-Slagelse-Ringsted Hospitals research fund, Denmark (project no. A1277). Holm is funded by a program grant from Region Zealand (Exercise First). Skou is currently funded by 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). These institutions had no role in the study design, collection, analysis and interpretation of data, in the writing of the manuscript, or in the decision to submit the manuscript for publication. Skou is co-founder of GLA:D®, associate editor of the JOSPT and has received grants from the Lundbeck Foundation and personal fees from Munksgaard, TrustMe-Ed and Nestlé Health Science, outside the submitted work. Data is available for sharing upon reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.