Aim To investigate associations between within-person change in patient reported outcomes (PROMs) and wish for joint surgery during up to 12-month participation in a digital first-line intervention for knee and hip osteoarthritis (OA) comprising exercise and education.

Methods Retrospective observational registry data. All participants enrolling between June 1st 2018 and September 30th 2021 with follow-up data and with a change in their wish for surgery (dichotomous, becoming wishing or unwishing) were included. Analysed PROMS comprised pain (numeric rating scale (NRS), 0-10), 5 level EuroQol - 5-dimension (EQ5D-5L, 0.243-0.976), overall health (NRS, 0-10), activity impairment (NRS, 0-10), walking difficulties (yes/no), fear of movement (yes/no), function and quality of life subscales of Knee/Hip injury and Osteoarthritis Outcome Score 12 Items (KOOS-12/HOOS-12, both 0-100). To capture effects of changes in PROMs over time on probability of wish for surgery, the investigators used fixed effect (conditional) logistic regressions.

Inclusion Criteria:

• Clinical diagnosis of knee or hip osteoarthritis
• Date of enrolment between June 1st 2018 and September 30th 2021 (study period)
• Consent to the use of the data at the moment of enrolment (not withdrawn during the study period)

Exclusion Criteria:

• No follow-up data
• No change in wish for surgery (participants remained either wishing or not wishing to undergo surgery for the whole treatment period). This exclusion was due to the focus of the present study on within-person relationships between PROMs and wish for surgery. In this scenario, those with no change in their wish for surgery do not provide any information to the analysis.

People with osteoarthritis of the knee or hip ascertained by a clinicians who enrolled in a first-line digital intervention for OA including exercise and education.