Methods

Reporting of the study is in compliance with the relevant guidelines.21,22 The trial protocol and analysis plans were published prior to completion of data collection and locking of the database.23,24

The study design was a pragmatic, parallel arm, RCT. The trial was conducted at 11 sites in Scotland and England:

• Ninewells Hospital, NHS Tayside, Dundee
• Western General Hospital, NHS Lothian, Edinburgh
• Queen Victoria Infirmary, Newcastle upon Tyne Hospital NHS Foundation Trust, Newcastle
• The Walton Centre NHS Foundation Trust, Liverpool
• Royal Hallamshire Hospital, Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield
• Queen’s Medical Centre, Nottingham University Hospitals NHS Trust, Nottingham
• Salford Royal, Salford Royal NHS Foundation Trust, Salford
• Southmead Hospital, North Bristol NHS Trust, Bristol
• Dorset County Hospital, Dorset County Hospital NHS Foundation Trust, Dorset
• King’s College Hospital, King’s College NHS Foundation Trust, London
• St George’s University Hospital, St George’s NHS Foundation Trust, London.

The participants, adults diagnosed with FMD by a neurologist, were randomised (1 : 1) to receive the trial intervention (specialist physiotherapy), or TAU (referral to community physiotherapy). The eligibility criteria and trial treatments are described in more detail in the protocol.23

Specialist physiotherapy was a protocolised intervention comprising nine sessions and a 3-month follow-up. The nine sessions were scheduled within a 3-week block to provide a minimum level of intensity. The intervention was guided by an interactive workbook. The physiotherapists delivering this treatment completed a 5-day comprehensive training programme. The workbook and physiotherapy training manual are available to download.25,26

Participants randomised to receive TAU were referred to the local community physiotherapy service suitable for people with neurological conditions. Given the pragmatic nature of the study, there was no attempt to control the treatment received in this group. We found that participants randomised to TAU received a median of four sessions [interquartile range (IQR) 2–7].

Follow-up of both groups was conducted remotely at 6 and 12 months. Participants could choose their preferred method of completing the study outcome measures from an online form, pen and paper via return mail, or by telephone with a blinded research assistant. We also received digital data from NHS England and NHS Scotland on the number of admissions and hospital appointments attended by each participant in the 12 months prior to and 12 months post randomisation. In addition, between 6 and 12 months post randomisation, participants completed a telephone interview which collected details of the content, frequency and duration of physiotherapy sessions, as well as satisfaction with treatment.

The original target sample size was 264, which was powered to detect a 9-point difference in the SF-36 Physical Functioning domain, at the 5% level of significance with 90% power. In February 2020, the target sample size was increased to an upper threshold of 300, allowing for a greater potential dropout rate. The sample size was increased again, as described below, to mitigate the impact of the COVID-19 pandemic.

In February 2023, towards the end of the data collection period, the physiotherapists delivering specialist physiotherapy attended an in-person debrief meeting, which included a Nominal Group Process that explored potential barriers and facilitators to implementing the intervention across the NHS.

The COVID-19 pandemic

The full impact of the COVID-19 pandemic on the trial and the mitigating strategies are described in a separate publication.24 To summarise, the UK national lockdown was instigated on 23 March 2020. Trial recruitment and the physiotherapy treatment were suspended. However, follow-up of participants was remote and therefore continued as described in the protocol. At the point of suspension, the recruitment total was 267. Among the recruits, there were 89 participants who had completed baseline assessment but were waiting to receive their randomised treatment. These participants either did not receive treatment during the follow-up period, or their treatment was substantially delayed so that it was received close to the time of the primary outcome. The TAU group were more highly represented within these 89 participants due to the longer wait experienced by this group to start treatment (n = 27 specialist physiotherapy and n = 62 TAU).

To mitigate the impact of the pandemic on the study results, the trial was extended, and an additional 88 participants were recruited. We continued to collect follow-up data from the 89 participants who had not received their trial-allocated treatment prior to lockdown. However, as per our prespecified analysis plans,24 these data were treated as missing in the primary analysis. A series of sensitivity analyses explored the impact of excluding these participants, and the wider impact of the pandemic on the trial.

Summary of results

Recruitment occurred between 19 October 2018 and 31 January 2022, with a 17-month break during the COVID-19 pandemic. In total, 355 participants were recruited into the trial from 11 sites; 179 were assigned to specialist physiotherapy and 176 to TAU. After excluding the 89 participants who did not receive treatment prior to COVID-19 lockdown, there were n = 152 participants in specialist physiotherapy and n = 114 in TAU. The baseline characteristics of the participants are presented in Tables 1 and 2. Retention at 12 months was n = 138 (91%) and n = 103 (90%), respectively. The trial profile is presented in Figure 1.

FIGURE 1 Trial profile. Adapted from Nielsen et al.27 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) licence, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original text. (Figure not reproduced here; see the original article.)

The primary outcome, SF-36 Physical Functioning, was not significantly different between the groups at 12 months. The adjusted difference was 3.5 [95% confidence interval (CI) −2.3 to 9.3].

In secondary outcomes, between group differences were significant in favour of specialist physiotherapy at 12 months for the Clinical Global Impression Improvement (CGI-I) – participant perception of improvement in their motor symptoms [odds ratio (OR) 2.3, 95% CI 1.4 to 3.9); the SF-36 Mental Health domain (adjusted mean difference 5.4, 95% CI 0.9 to 9.8); confidence in the diagnosis; two out of nine scales of the Revised Illness Perception Questionnaire (Personal Control and Illness Coherence). There were no significant differences for the remaining outcomes [the remaining domains of the SF-36, Functional Mobility Scale, Hospital Anxiety and Depression Scale (HADS), a fatigue 5-point scale and digital data on health service use]. The primary and secondary outcomes at 12 months are presented in Tables 3–6.

In post hoc analysis, we calculated the number of participants who made a 10-point improvement in the primary outcome, which we determined to be a clinically significant change. We found 67 of 138 (49%) in specialist physiotherapy made a clinically significant improvement at 12 months, compared to 39 of 103 (38%) in TAU.

At the 6-month assessment, the outcome measures that were significantly better for specialist physiotherapy compared to TAU were:

• CGI-I (OR of improvement 4.7, 95% CI 2.6 to 8.6)
• SF-36 Physical Role Limitations (adjusted difference 10.1, 95% CI 3.7 to 16.5)
• SF-36 Social Functioning (adjusted difference 8.9, 95% CI 2.6 to 15.2)
• SF-36 Mental Health (adjusted difference 5.0, 95% CI 0.5 to 9.6)
• Confidence in the diagnosis (adjusted difference 1.1, 95% CI 0.5 to 1.6)
• HADS, Anxiety (adjusted difference −1.0, 95% CI −2.0 to −0.1)
• Three domains of the Revised Illness Perception Scale (Timeline Cyclical, Personal Control and Illness Coherence).

Outcomes that showed no difference at 6 months were the remaining domains of the SF-36, the Functional Mobility Scale, HADS Depression, a fatigue 5-point scale, six out of nine scales of the Revised Illness Perception Questionnaire. No outcomes were significantly different in favour of TAU. The complete 6-month data are published in the supplementary appendix of the main trial report.27

More participants randomised to specialist physiotherapy were satisfied with their treatment (96.7% vs. 65.3%). Specialist physiotherapy participants rated the knowledge of their physiotherapist more highly (median 10/10 vs. 8/10), and more were likely to recommend their treatment to others who needed similar treatment (99.2% vs. 68.0%).

In the economic evaluation, at 12 months, the mean total adjusted health and social care costs in the specialist physiotherapy group was £3814 (95% CI £3194 to £4433), while the total for TAU was £3670 (95% CI £2931 to £4410); adjusting for baseline, specialist physiotherapy cost £143 (95% CI −£825 to £1112) more than TAU.

Participants in the specialist physiotherapy group had significantly higher EQ-5D-5L utility scores at 6 months (difference 0.054 95% CI 0.002 to 0.107), but not 12 months. There was no significant difference in QALYs over the 12-month follow-up period (0.030 95% CI −0.007 to 0.067). From a health and social care perspective, the mean incremental cost per QALY gained for specialist physiotherapy compared to TAU was £4133. The probability of cost-effectiveness at the £20,000 threshold was 86%. The complete health economic data are published in a separate report.28