P072 Evaluation of two electronic-rehabilitation programmes for persistent knee pain: a randomised feasibility trial

Abstract

Background/Aims Persistent knee pain is a common cause of disability. Internet-delivered education and exercise may provide an appropriate and scalable treatment option. This trial evaluated the feasibility and acceptability of two electronic-rehabilitation interventions: ‘MyKneeUK’, based on the similar Australian intervention MyKneeExercise, and ‘Group E-Rehab’, developed from a previously tested one-to-one internet-delivered programme. Methods We conducted a non-blinded, randomised feasibility trial with three parallel groups. Eligible participants were aged 45 years or older, with a history of persistent knee pain consistent with clinical diagnosis of knee osteoarthritis. Participants were randomly assigned in a 1:1:1 ratio to either the intervention groups (MyKneeUK, Group E-Rehab) or the control group. The MyKneeUK group received a self-directed unsupervised internet-based home exercise programme plus short message service support (targeting exercise behaviour change) for 12-weeks; the Group E-Rehab group received 7 group-based (5-7 participants/group) physiotherapist-prescribed home exercises delivered via videoconferencing accompanied by internet-interactive educational sessions over 12-weeks; the control arm received usual physiotherapy or continued their usual self-management. Feasibility variables, patient-reported outcomes and clinical findings were measured at baseline, 3 and 9-months. Results Of 149 people screened, 97 were eligible (65.1%) and 90 were randomised (92.8%). Common reasons for ineligibility included knee pain <3-months (n = 13;25%), knee pain during walking ≤4 on 11-point NRS (n = 25;48.1%) and no activity-related joint pain (n = 18;34.6%). Participants had a mean (S.D.) age of 58.2 (7.4) years, 53 were female (58.9%), 70 (77.8%) were using medication for their knee pain and most had pain in multiple joints (mean [S.D] painful joints: 2.9 [2.5]). Of those enrolled, 74 (82.2%) completed 3-month follow-up (Usual Care, 27/30 (90.0%); MyKneeUK, 27/30 (90.0%); Group E-Rehab 20/30 (66.7%)) and 62 (68.9%) completed 9-months (Usual Care, 25/30 [83.3%]; MyKneeUK, 23/30 [76.7%]; Group E-Rehab 14/30 [46.7%]). Compared to usual care, WOMAC pain and function scores were reduced between baseline and 3-months for both MyKneeUK (treatment difference [85%CI], WOMAC pain: -0.57 [-1.71,0.57], WOMAC function, -3.40 [-6.75,-0.06]) and Group E-rehab (WOMAC pain: 0.68 [-2.01,0.65], WOMAC function, -4.38 [-8.07,-0.69]), which was sustained to 9-months. Improvements in favour of both interventions were also seen for other outcomes. There were no serious adverse events and 15 adverse events in 11 participants, reasonably balanced across arms (MyKneeUK 7, Group E-Rehab 4, Usual Care 4). Sample size calculations suggest a definitive trial would be feasible (sample sizes for WOMAC pain primary outcome, 80% power with 20% loss-to-follow-up: MyKneeUK, 144, Group E-rehab, 136). Conclusion Both electronic-rehabilitation interventions met pre-specified feasibility criteria for progression to a full RCT, with the exception of moderately high attrition in the Group E-Rehab arm. This may relate to delays for some participants in commencing the intervention due to group sessions requiring a minimum number randomised to this arm before they could start, which could be addressed with minor protocol changes in a full RCT. Disclosure S.R. Kingsbury: None. D. Groves-Williams: None. G.A. McHugh: None. E.M. Hensor: None. C. Comer: None. M. Conner: None. R.K. Nelligan: None. R.S. Hinman: None. K.L. Bennell: None. P.G. Conaghan: None.