Effectiveness and cost-effectiveness of a blended exercise intervention for patients with hip and/or knee osteoarthritis: study protocol of a randomized controlled trial

A stratified random sample of 800 physical therapy practices in three provinces of the Netherlands (e.g. Noord-Holland, Utrecht and Gelderland) will be invited by letter to participate in the study. Contact information of physical therapy practices will be obtained from the national database for physical therapists of the Netherlands Institute for Health Services Research (NIVEL). Additionally, a recruitment advertisement will be placed in the online newsletter of The Royal Dutch Society for Physical Therapy (KNGF). Each participating physical therapy practice will be asked to enroll one or two physical therapists. The researchers will recruit 100 physical therapists. Inclusion criteria for physical therapists will concern: (i) practicing in primary care, (ii) treating at least six patients with OA of the hip and/or knee each year. Physical therapists practicing in another physical therapy practice participating in the study will be excluded.

In order to include 200 participants, each physical therapist is requested to recruit about two patients. Since the study of Veenhof et al. [27] showed that recruitment of OA patients in the physical therapy practice is difficult and research has shown that different recruitment strategies do not affect treatment outcomes [28], this study uses various recruitment strategies. First, patients with hip and/or knee OA who visit a physical therapy practice will be invited to participate in the study. Second, recruitment advertisements will be placed in local newspapers. Third, information letters and flyers will be sent to general practitioners. Responders to these articles and flyers will be allocated to the nearest participating physical therapist. Eligibility criteria of patients interested in the study concern: (i) age 40–80 years, (ii) OA of the hip and/or knee according to the clinical criteria of the American College of Rheumatology (ACR) [29]. Exclusion criteria will include: (i) being on a waiting list for a hip or knee replacement surgery, (ii) being contra-indicated for PA without supervision, (iii) being sufficiently physically active according to the physical therapist, (iv) participation in a physical therapy and/or PA program in the last six months, (v) no access to internet, (vi) inability to understand the Dutch language. The diagnostic ACR clinical criteria for knee OA are: knee pain and at least three of the following six criteria: age > 50 years, morning stiffness <30 minutes, crepitation, bony tenderness, bony enlargement and no palpable warmth. Diagnostic ACR clinical criteria for hip OA are: hip pain and hip internal rotation < 15 degree and hip flexion ≤ 115 degree; or hip internal rotation ≥ 15 degree and pain on hip internal rotation and morning stiffness of the hip ≤ 60 minutes and age > 50 years [29]. Contra-indications for PA will be determined using the adapted Physical Activity Readiness Questionnaire (PAR-Q) [30]. This questionnaire is used to identify patients for whom PA is inappropriate.

The 3-month program e-Exercise is based on the Dutch guideline for physical therapists (10) and is a combination of (i) maximum five face-to-face sessions with a physical therapist, and (ii) a web-based PA intervention. Table 1 provides an overview of the program content of e-Exercise.

During the first face-to-face session (week 1), physical therapists will provide information about OA, the importance of PA and the relation of PA with pain. Together with their physical therapist, patients choose one physical activity, for example, walking, cycling or swimming. Physical therapists select and instruct four strength & stability exercises. Patients are instructed to perform the first module of the web-based part of the intervention. In this module, the patients will be asked to determine their physical load ability based on a 3-day self-test. The second assignment is the execution of strength & stability exercises. During the second face-to-face session (week 2), patients’ physical load ability will be discussed and personal short and long-term goals will be formulated according to the principles of Goal Setting, which is based on the idea that goals can affect action [31]. The strength & stability exercises will be trained again. After the second appointment, patients are instructed to perform four online modules for the duration of four weeks. In week 6, a third face-to-face treatment takes place. Patients’ progress will be discussed, based on an online report which is automatically sent to the physical therapists. This report contains a summary of website-visits and patients’ experiences with the strength and stability exercises. After the third face-to-face treatment, patients perform another six online modules. The final face-to-face appointment will take place in week 12. In this final treatment physical therapists will support and encourage patients to maintain a physically active lifestyle. If necessary, physical therapists can plan an additional fifth session. This optional session is especially for patients who are less capable to perform unsupervised physical exercises. Physical therapists are recommended to treat patients according to the e-Exercise protocol, however, with respect to their clinical competences, physical therapist are free to deviate from the protocol.

The web-based part of e-Exercise is based on the web-based intervention Join2move [32] and consists of three topics: (i) Graded Activity; the duration of patients’ chosen physical activity (e.g. walking, cycling, swimming) will gradually be increased until patients reach their personal short-term goal. (ii) Strength & Stability; each module contains two exercises. The number of repeats will gradually increase per 4 weeks. (iii) Information; topics about OA, PA, aetiology of OA, pain-management, weight-management, motivation, medication and social influences on pain will be discussed. Automatic emails are generated if participants do not visit the website once a week.

Patients in the control group will receive usual care. For the current study, usual care is defined as any treatment provided by physical therapists. Physical therapists will be encouraged to practice according to the “KNGF guideline OA Hip-Knee” [10]. According to the guideline, the physical therapy treatment comprises the same three elements as e-Exercise: (i) information, (ii) physical exercise and (iii) strength and stability exercises. Practical content considerations can be made by therapists themselves. The number of sessions will differ per patient. From the NIVEL Primary Care Database we know that the average number of physical therapy sessions in patients with OA is 17.1 [33].

Physical functioning will be assessed subjectively with the subscale ‘function in daily living’ of the Hip OA Outcome Score (HOOS) [34] and/or the Knee Injury and OA Outcome Score (KOOS) [35], depending their affected joint. The HOOS and the KOOS assess 5 indicators: pain, symptoms, physical function, sport and recreation function and quality of life, in relation to patients’ hip or knee complaints. Each indicator is scored on a 5-point Likert scale (0 = extreme symptoms/problems; 4 = no symptoms/problems). A lower score indicates respectively more pain, symptoms, problems in physical functions, problems in sports and recreation activities and a lower quality of life. In addition, objectively physical functioning will be measured by the physical therapist with timed “Up & Go” test (TUG) [36]. In this easily administered test, the patient is requested to rise from an arm chair, walk three meters, turn, walk back again and sit down. Meanwhile, the physical therapist observes the patient and measures the time.

Physical activity will be measured subjectively with the SQUASH [37]. The questionnaire measures habitual PA during a normal week over the last few months. The total score is expressed as minutes per week. In addition, data can also be analysed according to whether the activity is light, moderate or vigorous. Objective PA will be measured through ActiGraph GT3X tri-axial accelerometers. Participants will be instructed to wear the monitor on a belt around their waist for five executive days [38], except during sleeping, showering or swimming. In addition, participants will be requested to fill out a short activity diary. This diary contains questions about wearing time, unusual activities and reasons for device removal. When accelerometers and diaries are returned by post, data can be downloaded, processed and subsequently analyzed. The widely accepted PA thresholds of Freedson et al. [39] will be used: 0–99 counts for sedentary activities, 100–1951 for light PA, 1952–5724 moderate PA, 5725–9498 for vigorous PA and 9499- max for very vigorous activities. The total time spent in light, moderate and vigorous PA will be summed and subsequently divided by the number of days worn to compute the daily average time spent in total activity. For analysis, data will be recorded at 1-minute intervals.

Information on the patients’ healthcare utilization, (unpaid) productivity losses, and sports costs due to OA will be gathered with four retrospective 3-month cost questionnaires that cover the full 12-months of the program. Healthcare utilization due to OA comprises of visits to a physical therapist, general practitioner, massage therapist, alternative therapist, medical specialist, as well as informal care, hospital care, the use of both prescribed and over the counter drugs and medical devices. Healthcare utilization will be valued using Dutch standards costs [40]. If these are unavailable, prices reported by professional organizations will be used. Medication use will be valued using unit prices derived from the “Royal Dutch Society of Pharmacy” [41]. Unpaid productivity losses will be valued in accordance with the “Dutch Manual of Costing” [40]. Paid productivity losses comprise of both sickness absence and presenteeism (i.e. reduced productivity while at work). Sickness absence will be valued in accordance with the “Friction Cost Approach” (FCA), with a friction period of 23 weeks and an elasticity of 0.8, using age- and gender-specific price weights [40]. The FCA assumes that production losses are confined to the “friction period” (i.e. time needed to replace a sick worker) and that a 100 percent loss of labour input corresponds with an 80 percent reduction in productivity (i.e. an elasticity of 0.8) [42]. The participants’ level of presenteeism will be measured using the “World Health Organization – Work Performance Questionnaire” as well as the “Productivity and Disease Questionnaire”, and valued using age- and gender-specific price weights [40, 43‐45]. The cost of the e-Exercise intervention will be estimated using a bottom-up micro-costing approach [46].

Health Related Quality of Life will be measured with the EuroQol-5D (EQ-5D) [47]. This questionnaire comprises of 5 dimensions i.e., mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Per dimension, patients are asked to indicate their health state on a 3-point Likert scale (1 = no problems; 3 = extreme problems). The questionnaire differentiates between 245 health states. These health states will be converted into utility units by using the Dutch tariff [48]. Utilities represent quality of life into a single number that ranges from 0 (death) to 1(full health). Quality adjusted life years (QALY’s) will subsequently be calculated by multiplying the participants’ health state utilities by the duration of time they spent in that particular health state.

Pain and tiredness will be measured with a numeric rating scale (NRS; 0 is no pain/not tired and 10 is worst possible pain/very tired). Furthermore, pain will be assessed with the pain subscale of the HOOS and/or the KOOS [34, 35].

Self-efficacy will be measured by the Arthritis Self-efficacy Scale (ASES) [49]. Subscales for the ASES are pain, symptoms and physical functioning, the 19 statements can be scored on a 5 point-Likert scale (1 = fully disagree; 5 = fully agree). A higher score indicates more self-efficacy.

To determine the short (baseline-3 months) and long term (baseline −12 months) effectiveness of e-Exercise on primary and secondary outcomes, multi-level modelling of repeated measures will be performed controlling for baseline values and relevant confounders such as age, OA location and gender. With multilevel modelling of repeated measures it is possible to correct on one side for dependency of observations within subjects and, on the other side, to take into account the variation between physical therapists [50, 51]. The three-level hierarchy will exist of repeated measurements (level 1), nested within patients (level 2), nested within physical therapists (level 3).

A cost-utility analysis (CUA) and a cost-effectiveness analysis (CEA) will be performed from the societal and the healthcare perspective. From the societal perspective all costs will be taken into account irrespective of who pays or benefits, whereas solely those borne by the healthcare sector will be included when the healthcare perspective is applied [40]. For the CUA and CEA, missing cost and effect data will be imputed using multiple imputation [52]. The results of the imputed datasets will be pooled using Rubin’s rules [52]. In order to account for the highly skewed nature of cost data, bias-corrected and accelerated bootstrapping with 5000 replications will be used to estimate 95% confidence intervals around the mean differences in costs between the study groups. Incremental Cost-Effectiveness Ratios (ICERs) will subsequently be calculated by dividing the differences in costs between study groups by their respective differences in QALYs for the CUA. For the CEA, ICERs will be calculated by dividing the difference in costs by the difference in PA and physical functioning. The uncertainty surrounding the ICERs will be graphically illustrated by plotting bootstrapped incremental cost-effect pairs on cost-effectiveness planes [53]. Moreover, cost-effectiveness acceptability curves (CEACs) will be constructed to provide a summary measure of the joint uncertainty of costs and effects. CEACs indicate the probability of the e-Exercise intervention being cost-effective in comparison to usual care at different willingness-to-pay values [54]. To test the robustness of the study results, several sensitivity analyses will be performed.