The use of large-scale Randomised Controlled Trials (RCTs) is fast becoming "the gold standard" of testing the causal effects of policy, social, and educational interventions. RCTs are typically evaluated — and ultimately judged — by the economic, educational, and statistical significance of the Average Treatment Effect (ATE) in the study sample. However, many interventions have heterogeneous treatment effects across different individuals, not captured by the ATE. One way to identify heterogeneous treatment effects is to conduct subgroup analyses, such as focusing on low-income Free School Meal pupils as required for projects funded by the Education Endowment Foundation (EEF) in England. These subgroup analyses, as we demonstrate in 48 EEF-funded RCTs involving over 200,000 students, are usually not standardised across studies and offer flexible degrees of freedom to researchers, potentially leading to mixed results. Here, we develop and deploy a machine-learning and regression-based framework for systematic estimation of Individualised Treatment Effect (ITE), which can show where a seemingly ineffective and uninformative intervention worked, for whom, and by how much. Our findings have implications for decision-makers in education, public health, and medical trials.

Abstract
. Background
. In a cluster randomised controlled trial (CRCT), randomisation units are “clusters” such as schools or GP practices. This has methodological implications for study design and statistical analysis, since clustering often leads to correlation between observations which, if not accounted for, can lead to spurious conclusions of efficacy/effectiveness. Bayesian methodology offers a flexible, intuitive framework to deal with such issues, but its use within CRCT design and analysis appears limited. This review aims to explore and quantify the use of Bayesian methodology in the design and analysis of CRCTs, and appraise the quality of reporting against CONSORT guidelines.
.
. Methods
. We sought to identify all reported/published CRCTs that incorporated Bayesian methodology and papers reporting development of new Bayesian methodology in this context, without restriction on publication date or location. We searched Medline and Embase and the Cochrane Central Register of Controlled Trials (CENTRAL). Reporting quality metrics according to the CONSORT extension for CRCTs were collected, as well as demographic data, type and nature of Bayesian methodology used, journal endorsement of CONSORT guidelines, and statistician involvement.
.
. Results
. Twenty-seven publications were included, six from an additional hand search. Eleven (40.7%) were reports of CRCT results: seven (25.9%) were primary results papers and four (14.8%) reported secondary results. Thirteen papers (48.1%) reported Bayesian methodological developments, the remaining three (11.1%) compared different methods. Four (57.1%) of the primary results papers described the method of sample size calculation; none clearly accounted for clustering. Six (85.7%) clearly accounted for clustering in the analysis. All results papers reported use of Bayesian methods in the analysis but none in the design or sample size calculation.
.
. Conclusions
. The popularity of the CRCT design has increased rapidly in the last twenty years but this has not been mirrored by an uptake of Bayesian methodology in this context. of studies using Bayesian methodology, there were some differences in reporting quality compared to CRCTs in general, but this study provided insufficient data to draw firm conclusions. There is an opportunity to further develop Bayesian methodology for the design and analysis of CRCTs in order to expand the accessibility, availability, and, ultimately, use of this approach.
.

BACKGROUND: Current NHS policy encourages an integrated approach to provision of mental and physical care for individuals with long term mental health problems. The 'PARTNERS2' complex intervention is designed to support individuals with psychosis in a primary care setting. AIM: the trial will evaluate the clinical and cost-effectiveness of the PARTNERS2 intervention. DESIGN & SETTING: This is a cluster randomised controlled superiority trial comparing collaborative care (PARTNERS2) with usual care, with an internal pilot to assess feasibility. The setting will be primary care within four trial recruitment areas: Birmingham & Solihull, Cornwall, Plymouth, and Somerset. GP practices are randomised 1:1 to either (a) the PARTNERS2 intervention plus modified standard care ('intervention'); or (b) standard care only ('control'). METHOD: PARTNERS2 is a flexible, general practice-based, person-centred, coaching-based intervention aimed at addressing mental health, physical health, and social care needs. Two hundred eligible individuals from 39 GP practices are taking part. They were recruited through identification from secondary and primary care databases. The primary hypothesis is quality of life (QOL). Secondary outcomes include: mental wellbeing, time use, recovery, and process of physical care. A process evaluation will assess fidelity of intervention delivery, test hypothesised mechanisms of action, and look for unintended consequences. An economic evaluation will estimate its cost-effectiveness. Intervention delivery and follow-up have been modified during the COVID-19 pandemic. CONCLUSION: the overarching aim is to establish the clinical and cost-effectiveness of the model for adults with a diagnosis of schizophrenia, bipolar, or other types of psychoses.

. Background
. There is modest evidence that exercise referral schemes increase physical activity in inactive individuals with chronic health conditions. There is a need to identify additional ways to improve the effects of exercise referral schemes on long-term physical activity.
.
.
. Objectives
. To determine if adding the e-coachER intervention to exercise referral schemes is more clinically effective and cost-effective in increasing physical activity after 1 year than usual exercise referral schemes.
.
.
. Design
. A pragmatic, multicentre, two-arm randomised controlled trial, with a mixed-methods process evaluation and health economic analysis. Participants were allocated in a 1 : 1 ratio to either exercise referral schemes plus e-coachER (intervention) or exercise referral schemes alone (control).
.
.
. Setting
. Patients were referred to exercise referral schemes in Plymouth, Birmingham and Glasgow.
.
.
. Participants
. There were 450 participants aged 16–74 years, with a body mass index of 30–40 kg/m2, with hypertension, prediabetes, type 2 diabetes, lower limb osteoarthritis or a current/recent history of treatment for depression, who were also inactive, contactable via e-mail and internet users.
.
.
. Intervention
. e-coachER was designed to augment exercise referral schemes. Participants received a pedometer and fridge magnet with physical activity recording sheets, and a user guide to access the web-based support in the form of seven ‘steps to health’. e-coachER aimed to build the use of behavioural skills (e.g. self-monitoring) while strengthening favourable beliefs in the importance of physical activity, competence, autonomy in physical activity choices and relatedness. All participants were referred to a standard exercise referral scheme.
.
.
. Primary outcome measure
. Minutes of moderate and vigorous physical activity in ≥ 10-minute bouts measured by an accelerometer over 1 week at 12 months, worn ≥ 16 hours per day for ≥ 4 days including ≥ 1 weekend day.
.
.
. Secondary outcomes
. Other accelerometer-derived physical activity measures, self-reported physical activity, exercise referral scheme attendance and EuroQol-5 Dimensions, five-level version, and Hospital Anxiety and Depression Scale scores were collected at 4 and 12 months post randomisation.
.
.
. Results
. Participants had a mean body mass index of 32.6 (standard deviation) 4.4 kg/m2, were referred primarily for weight loss and were mostly confident self-rated information technology users. Primary outcome analysis involving those with usable data showed a weak indicative effect in favour of the intervention group (n = 108) compared with the control group (n = 124); 11.8 weekly minutes of moderate and vigorous physical activity (95% confidence interval –2.1 to 26.0 minutes; p = 0.10). Sixty-four per cent of intervention participants logged on at least once; they gave generally positive feedback on the web-based support. The intervention had no effect on other physical activity outcomes, exercise referral scheme attendance (78% in the control group vs. 75% in the intervention group) or EuroQol-5 Dimensions, five-level version, or Hospital Anxiety and Depression Scale scores, but did enhance a number of process outcomes (i.e. confidence, importance and competence) compared with the control group at 4 months, but not at 12 months. At 12 months, the intervention group incurred an additional mean cost of £439 (95% confidence interval –£182 to £1060) compared with the control group, but generated more quality-adjusted life-years (mean 0.026, 95% confidence interval 0.013 to 0.040), with an incremental cost-effectiveness ratio of an additional £16,885 per quality-adjusted life-year.
.
.
. Limitations
. A significant proportion (46%) of participants were not included in the primary analysis because of study withdrawal and insufficient device wear-time, so the results must be interpreted with caution. The regression model fit for the primary outcome was poor because of the considerable proportion of participants [142/243 (58%)] who recorded no instances of ≥ 10-minute bouts of moderate and vigorous physical activity at 12 months post randomisation.
.
.
. Future work
. The design and rigorous evaluation of cost-effective and scalable ways to increase exercise referral scheme uptake and maintenance of moderate and vigorous physical activity are needed among patients with chronic conditions.
.
.
. Conclusions
. Adding e-coachER to usual exercise referral schemes had only a weak indicative effect on long-term rigorously defined, objectively assessed moderate and vigorous physical activity. The provision of the e-coachER support package led to an additional cost and has a 63% probability of being cost-effective based on the UK threshold of £30,000 per quality-adjusted life-year. The intervention did improve some process outcomes as specified in our logic model.
.
.
. Trial registration
. Current Controlled Trials ISRCTN15644451.
.
.
. Funding
. This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 24, No. 63. See the NIHR Journals Library website for further project information.
.

Abstract
. Background
. Symptoms arising from vestibular system dysfunction are observed in 49–59% of people with Multiple Sclerosis (MS). Symptoms may include vertigo, dizziness and/or imbalance. These impact on functional ability, contribute to falls and significant health and social care costs. In people with MS, vestibular dysfunction can be due to peripheral pathology that may include Benign Paroxysmal Positional Vertigo (BPPV), as well as central or combined pathology. Vestibular symptoms may be treated with vestibular rehabilitation (VR), and with repositioning manoeuvres in the case of BPPV. However, there is a paucity of evidence about the rate and degree of symptom recovery with VR for people with MS and vestibulopathy. In addition, given the multiplicity of symptoms and underpinning vestibular pathologies often seen in people with MS, a customised VR approach may be more clinically appropriate and cost effective than generic booklet-based approaches. Likewise, BPPV should be identified and treated appropriately.
.
. Methods/ design
. People with MS and symptoms of vertigo, dizziness and/or imbalance will be screened for central and/or peripheral vestibulopathy and/or BPPV. Following consent, people with BPPV will be treated with re-positioning manoeuvres over 1–3 sessions and followed up at 6 and 12 months to assess for any re-occurrence of BPPV. People with central and/or peripheral vestibulopathy will be entered into a randomised controlled trial (RCT). Trial participants will be randomly allocated (1:1) to either a 12-week generic booklet-based home programme with telephone support or a 12-week VR programme consisting of customised treatment including 12 face-to-face sessions and a home exercise programme. Customised or booklet-based interventions will start 2 weeks after randomisation and all trial participants will be followed up 14 and 26 weeks from randomisation. The primary clinical outcome is the Dizziness Handicap Inventory at 26 weeks and the primary economic endpoint is quality-adjusted life-years. A range of secondary outcomes associated with vestibular function will be used.
.
. Discussion
. If customised VR is demonstrated to be clinically and cost-effective compared to generic booklet-based VR this will inform practice guidelines and the development of training packages for therapists in the diagnosis and treatment of vestibulopathy in people with MS.
.
. Trial registration
. ISRCTN Number: 27374299
. Date of Registration 24/09/2018
. Protocol Version 15 25/09/2019
.

Background: People severely impaired with progressive multiple sclerosis spend much of their day sitting, with very few options to improve motor function. As a result, secondary physical and psychosocial complications can occur. Effective and feasible self-management strategies are needed to reduce sedentary behaviour and enhance motor function. In this study, we aimed to assess the clinical and cost effectiveness of a home-based, self-managed, standing frame programme. Methods: SUMS was a pragmatic, multicentre, randomised controlled superiority trial of people with progressive multiple sclerosis and severe mobility impairment, undertaken in eight centres from two regions in the UK. The study had assessor-blinded outcome assessments with use of clinician-rated and patient-rated measures at baseline, 20 weeks, and 36 weeks. After baseline assessment, participants were randomised (1:1) by computer-generated assignment to either a standing frame programme plus usual care or usual care alone. The intervention consisted of two home-based physiotherapy sessions (60 min each) to set up the standing frame programme, supported by six follow-up telephone calls (15 min per call). Participants were asked to stand for 30 min, three times per week over 20 weeks, and encouraged to continue in the longer term, although no further physiotherapy support was provided. The primary clinical outcome was motor function measured by the Amended Motor Club Assessment (AMCA) score at week 36, analysed in the modified intention-to-treat population (excluding only patients who were deemed ineligible after randomisation, those who withdrew from the trial and were unwilling for their previously collected data to be used, or those who did not provide baseline and week 36 measurements). A 9-point AMCA score change was considered clinically meaningful a priori. Adverse events were collected through a daily preformatted patient diary throughout the 36 weeks and analysed in the modified intention-to-treat population. An economic assessment established the resources required to provide the standing frame programme, estimated intervention costs, and estimate cost effectiveness. This trial is registered with the International Standard Randomised Controlled Trials, number ISRCTN69614598. Findings: Between Sept 16, 2015, and April 28, 2017, 285 people with progressive multiple sclerosis were screened for eligibility, and 140 were randomly assigned to either the standing frame group (n=71) or the usual care group (n=69). of these, 122 completed the primary outcome assessment (61 participants in both groups) for the modified intention-to-treat analysis. The use of the standing frame resulted in a significant increase in AMCA score compared with that for usual care alone, with a fully adjusted between-group difference in AMCA score at 36 weeks of 4·7 points (95% CI 1·9–7·5; p=0·0014). For adverse events collected through patient diaries, we observed a disparity between the two groups in the frequency of short-term musculoskeletal pain (486 [41%] of 1188 adverse events in the standing frame group vs 160 [22%] of 736 adverse events in the usual care group), which was potentially related to the intervention. The musculoskeletal pain lasted longer than 7 days in five participants (two in the standing frame group and three in the usual care group). No serious adverse events related to the study occurred. The standing frame group had a mean 0·018 (95% CI −0·014 to 0·051) additional quality-adjusted life-years (QALYs) compared with those of the usual care group, and the estimated incremental cost-per-QALY was approximately £14 700. Interpretation: the standing frame programme significantly increased motor function in people with severe progressive multiple sclerosis, although not to the degree that was considered a priori as clinically meaningful. The standing frame is one of the first physiotherapy interventions to be effective in this population. We suggest that the programme is feasible as a home-based, self-managed intervention that could be routinely implemented in clinical practice in the UK. Funding: UK National Institute of Health Research.

AbstractBackground:Epistaxis is a common condition that can be associated with significant morbidity, and it places a considerable burden on our healthcare system. This national audit of management sought to assess current practice against newly created consensus recommendations and to expand our current evidence base.Methods:The management of epistaxis patients who met the inclusion criteria, at 113 registered sites across the UK, was compared with audit standards during a 30-day window. Data were further utilised for explorative analysis.Results:Data for 1826 cases were uploaded to the database, representing 94 per cent of all cases that met the inclusion criteria at participating sites. Sixty-two per cent of patients were successfully treated by ENT clinicians within 24 hours. The 30-day recurrent presentation rate across the dataset was 13.9 per cent. Significant event analysis revealed an all-cause 30-day mortality rate of 3.4 per cent.Conclusion:Audit findings demonstrate a varying alignment with consensus guidance, with explorative analysis countering some previously well-established tenets of management.