Who seeks physiotherapy or exercise treatment for hip and knee osteoarthritis? A cross‐sectional analysis of the English Longitudinal Study of Ageing

To determine the characteristics of individuals with hip and/or knee osteoarthritis who are recommended to seek physiotherapy or exercise treatment, and to explore which people are more or less likely to follow such recommendations.


| INTRODUC TI ON
Osteoarthritis is a disabling musculoskeletal disease which poses a significant impact on those who have the disease and society. 1 It is associated with pain, reduced function, independence and decreased quality of life. 2 In developed countries, the resultant socioeconomic burden has been estimated to be between 1% and 2.5% of gross domestic product. 3 Current recommendations for treatment are centered on education and exercise, pharmacological support and weight loss if overweight or obese, with end-stage management being joint replacement. [4][5][6] Exercise should be a core component of non-pharmacological management in combination with information and education on osteoarthritis. 4,6 Little research has been presented exploring the factors associated with who is referred or recommended to seek physiotherapy for chronic musculoskeletal conditions. Previous data have indicated that a number of variables are significantly associated with patient adherence to physiotherapy once treatment has commenced. Opseth et al 7 reported that patient perception of their own poor general health was significantly associated with regular attendance to physiotherapy, but reported that factors such as age, gender, education and employment status were not associated with adherence in physiotherapy. Similarly, Al-Eisa's 8 clinical audit of physiotherapy attendance for individuals with low back pain reported that older age, higher initial pain intensity and subjective reports of importance to their condition were significant factors to repeat adherence to physiotherapy. Lyngcolm et al 9 reported that subjective and objective indicators for improvement in hand function were also significant predictors to attend hand therapy in people following distal radius fracture.
Given the current and projected burden which osteoarthritis has on primary and secondary care services, 3  The purpose of this study was therefore primarily to determine the characteristics of individuals with hip and/or knee osteoarthritis who are recommended to seek physiotherapy or exercise treatment compared to those who are not, and secondly to explore which people are more or less likely to follow such recommendations and seek physiotherapy or exercise treatment.

| ME THODS
This cross-sectional study has been reported in accordance with the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) reporting guidelines. 10

| Cohort
Data were taken from the English Longitudinal Study of Ageing (ELSA). This is a nationally representative, prospective, populationbased cohort study consisting of 11 391 adults aged 50 years and older living in England. 11 The cohort commenced in 2002 and has been subsequently followed every 2 years.

Ethics approval was obtained from the London Multi-Centre
Research Ethics Service (MREC/01/2/91) and written informed consent obtained from all participants. The UK Data Service provided anonymized unlinked data for this study.

| Participants
Participants were eligible if they reported a diagnosis of hip and/or knee osteoarthritis with a visual analog scale (VAS) pain score of one or more and responded to the question as to whether physiotherapy or exercise had been recommended for their osteoarthritic symptoms.

| Data collection
All data were obtained from Wave 4 of the ELSA cohort (2008-2009). Data were collected from a self-completed questionnaire and nurse assessment visit where objective measures of anthropometric characteristics and physical function were made.

| Dependent variables
To assess whether participants were recommended physiotherapy, the question "has physiotherapy or exercise treatment been recommended to you for your hip or knee osteoarthritis" was asked.
Participants were also asked whether they had taken-up these treatment recommendations and sought advice from a physiotherapist or exercise professional.

| Covariates
Covariates included were: participant age, gender, ethnic classification (defined in ELSA as: White/non-White), whether participants were in paid work or not, whether they had access to a car and National Statistics-Socio-Economic Classification scheme status (NS-SEC) which is a validated measure of an individual's social position determined using the nature of their employment. 12 Self-reported health status was also recorded.
Pain measurements included: hip and knee VAS pain score, duration of hip and/or knee osteoarthritis. Location of osteoarthritis was categorised as isolated hip, isolated knee or hip and knee osteoarthritis.
Physical activity participation was determined using the self-reported ELSA physical activity questionnaire (ELSA-PAQ) where participants were asked how often they engaged in vigorous, moderate or mild physical activity. 13,14 This valid method has been previously used to determine the level of physical activity participation undertaken by older people. [13][14][15] Cognitive status was evaluated using the ELSA index of executive function. 11 This is based on 2 brief tests of executive function: verbal fluency and letter cancelation. These have demonstrated reliability and validity in assessing executive function. [16][17][18] Objectively measured physical function was assessed during the nurse assessment visit. These included data on gait speed with an 8-feet (2.4 m) walking test performed at normal walking pace and timed chair raises to complete 5 and 10 chair raises. Impairment of activities of daily living was assessed when participants were asked to report the level of impairment for 18 personal and extended activities of daily living 11 as itemised in Table 1.

| Data analysis
Demographic characteristics were reported with mean and standard deviation values and frequencies for whether physiotherapy or exercise treatment were recommended to participants and up-take was determined. Initially, the frequency and characteristics of those who TA B L E 1 Demographic characteristics of individuals who reported being recommended to those not being recommend to physiotherapy for hip and/or knee osteoarthritis

| RE SULTS
As presented in Figure 1 cise. Subsequently, 83% of those who were recommended to seek physiotherapy or exercise therapy reported that they acted on this recommendation ( Table 1).
The characteristics of the two groups are presented in Table 1 (Table 1).
When these variables were included in the logistic regression model (

| D ISCUSS I ON
These findings indicate that of people who have a clinical indication for physiotherapy or exercise treatment with hip and/or knee osteoarthritis, only 41% of individuals are recommended these interventions. Those who were more likely to be recommended to physiotherapy or exercise treatment were younger individuals, those with isolated knee pain and those with poorer self-reported general health. There is therefore a health inequality where those who are older, have multiple joint pain but better self-reported health, are  (Table 2). There was no evidence that level of impairment was an important predictor to whether people were recommended to physiotherapy. This may be regarded as surprising given that it may be inherent that individuals should be advised to seek physiotherapy for impairments in activities which are meaningful to them. 20 Based on these data, factors such as overall general health may be more meaningful to individuals which may indirectly be influenced by their musculoskeletal disabilities.
These results provide an indication as to which individuals are recommended to seek physiotherapy or exercise treatment. Based on this, older people, those with multiple joint pain and those with better self-reported health are less likely to be recommended physiotherapy. Given that physiotherapy has been shown to improve symptom management for those with hip and knee osteoarthritis, 4  Osteoarthritis has a complex biopsychosocial presentation for patients and for healthcare professionals to assess and develop management plans. 19 Future research aimed at understanding the complex clinical reasoning strategies undertaken by healthcare professionals when making management decisions will further support high-quality treatment. Finally, participants were asked to report whether they had been recommended to seek treatment on physiotherapy or exercise. While exercise is a core component to physiotherapy management for this population, it is not the only intervention. 4 Similarly participants may seek advice on exercise treatment either formally through physiotherapy, sport and exercise clinicians or more informally through gym instructors, walking group leaders or online forum. Due to the nature of the questions posed, it is not possible to differentiate this, but would prove a useful context for future recommendation on treatment provision and decision-making options in managing osteoarthritis in the real-world.

| CON CLUS ION
The minority (41%) of people with hip and/or knee osteoarthritis are recommended to seek physiotherapy or exercise treatment.
Older people, with multi-joint pain and better self-reported health status are less likely to be recommended to seek physiotherapy or exercise treatment. Strategies are now required to identify such subgroups within society and provide them with opportunities to improve symptom management. This will provide a major benefit in reducing healthcare inequalities in people with hip and knee osteoarthritis.

DATA AVA I L A B I LIT Y
All data are available through the UK Datalink. Access to the dataset used in this analysis can be sought through the corresponding author via the UK Datalink service.

ACK N OWLED G EM ENTS
Ethics approvals: Ethics approval was obtained from the London Multi-Centre Research Ethics Service (MREC/01/2/91).

CO N FLI C T O F I NTE R E S T
None.

AUTH O R CO NTR I B UTI O N
The