Background The digitalisation of health care has accelerated in recent years, driven by the World Health Organisation's Global Strategy on Digital Health. Musculoskeletal conditions include diseases and disorders affecting the joints, bones, muscles and related connective tissues. These conditions can affect multiple body areas or systems. Furthermore, the conditions are often characterised by pain and limitations in mobility and activities of daily living. Globally, musculoskeletal conditions are estimated to affect 1.7 billion people. Musculoskeletal conditions represent approximately two-thirds of all adults requiring rehabilitation. Physiotherapy plays a central role in the management and rehabilitation of musculoskeletal conditions. When provided as first-line primary care, physiotherapy has proven both efficient and cost-effective . Furthermore, video-based physiotherapy has been associated with improved attendance and adherence, factors that are critical for treatment effectiveness.
Video-based assessment in physiotherapy has been shown to be partially valid and reliable, but the methods remain poorly standardised, and the use of approaches and technical tools varies considerably. Overall, the evidence is limited, and existing descriptions are insufficient to support implementation in clinical practice. The transition to video-based care is described as a paradigm shift, and the profession expresses uncertainty regarding its effectiveness.
Despite the promising potential of digital health encounters, several challenges remain. A substantial proportion of patients initially assessed via video subsequently require in-person visits. Moreover, the therapeutic relationship may be affected by the video-based format. Taken together, these factors highlight an urgent need for robust, long-term evaluations of video-based first-contact physiotherapy.
Methods Study design A pragmatic randomised controlled trial (RCT) with a non-inferiority design will be conducted. The study will evaluate whether a video-based first-contact consultation with a physiotherapy yields outcomes that are not clinically inferior to those of a standard in-person first-contact physiotherapy consultation in primary care. The study will be reported in accordance with the CONSORT 2010 Statement for non-inferiority trials.
A non-inferiority analysis will be conducted, as the study aims to determine whether video-based first-contact physiotherapy consultation is not clinically inferior to an in-person first-contact physiotherapy. The analysis will use a predefined margin (∆) based on both statistical considerations and clinical judgement. The minimal clinically important difference and change for the PSFS was considered to define the non-inferiority margin. Previous literature reports that the minimal clinically important change for the PSFS ranges from approximately 1.3 and 2.3 points, depending on the diagnostic area. To ensure that any potential difference between groups is not clinically relevant, the non-inferiority margin has been set at ∆ = 1.5 points, which is also consistent with previous research in digital physiotherapy. The standard deviation (SD) will be set to 2, based on previous studies .
Sample size calculations based on the primary outcome variable PSFS, with ∆ = 1.5, indicate that 78 participants per group are required to achieve 80% statistical power with a one-sided alpha of 2.5%. Assuming a 25 % dropout rate, 104 participants will be included in each group, for a total of 208 participants.
Participant Selection and Recruitment Participants aged 18 years or older with musculoskeletal symptoms will be recruited. The study will be conducted at approximately six regional primary care rehabilitation units in Region Västra Götaland, Sweden, representing both geographic and demographic variation.
Potential participants will be identified through advertisement via social media, through posters and telephone. Potential participants will be screened to ensure that they meet the inclusion and exclusion criteria. They will receive written participant information digitally, will have the opportunity to ask questions, and will be offered verbal information. Informed consent will be obtained digitally.
Participants will be stratified by age, randomised, and scheduled for the type of consultation (video or in-person) corresponding to their allocated group with participating physiotherapist. All collected data will be handled in accordance with the principal investigator's data management procedures and applicable legislation.
Procedures The video-based first-contact consultation and the in-person first-contact consultation will be conducted within routine care. Participating physiotherapists will undergo approximately five hours of training in a structured model for conducting video-based physiotherapeutic assessments.
Participants will complete digital questionnaires before the first consultation, immediately after the first consultation, and at 6 weeks, 3 months, 6 months, and 12 months. The estimated time required to complete the digital questionnaires is approximately 10-20 minutes. Medical records and registry data will be collected in 6, 12 and 24 months.
Statistical Analysis The analyses will be conducted using IBM SPSS Statistics for Windows, (IBM Corp., Armonk, NY, USA). Descriptive statistics will be presented as means, standard deviations, and medians for continuous variables, and as frequencies for categorical variables.
Comparisons of improvement levels will be conducted using regression models, controlling for potential confounders (e.g., age, gender, trust). Comparisons of proportions showing improvement will be performed using a two-proportion Z-test. Exploratory analyses will be conducted to identify predictors of outcomes within the video consultation group.
For the health economic analysis, QALYs will be calculated from EQ-5D-5L data, and cost data will be obtained from healthcare registers and patient records. Incremental cost-effectiveness ratios (ICERs) will be calculated, with uncertainty analyses performed using bootstrapping and cost-effectiveness acceptability curves.
Estimates of carbon dioxide emissions will be performed from a life-cycle perspective, based on patient travel and estimates from the Swedish Transport Administration, healthcare utilisation based on estimates from the National Health Service in England, and the use of digital technology.
Comparisons between groups and over time will be conducted using appropriate parametric or non-parametric tests and regression models, depending on the characteristics of the data. Matched analyses will be applied where relevant.
