Generalized method for adaptive randomization in clinical trials

142

BackgroundGroup cognitive stimulation therapy programmes can benefit cognition and quality of life for people with dementia. Evidence for home-based, carer-led cognitive stimulation interventions is limited.ObjectivesTo evaluate the clinical effectiveness and cost-effectiveness of carer-delivered individual cognitive stimulation therapy (iCST) for people with dementia and their family carers, compared with treatment as usual (TAU).DesignA multicentre, single-blind, randomised controlled trial assessing clinical effectiveness and cost-effectiveness. Assessments were at baseline, 13 weeks and 26 weeks (primary end point).SettingParticipants were recruited through Memory Clinics and Community Mental Health Teams for older people.ParticipantsA total of 356 caregiving dyads were recruited and 273 completed the trial.InterventioniCST consisted of structured cognitive stimulation sessions for people with dementia, completed up to three times weekly over 25 weeks. Family carers were supported to deliver the sessions at home.Main outcome measuresPrimary outcomes for the person with dementia were cognition and quality of life. Secondary outcomes included behavioural and psychological symptoms, activities of daily living, depressive symptoms and relationship quality. The primary outcome for the family carers was mental/physical health (Short Form questionnaire-12 items). Health-related quality of life (European Quality of Life-5 Dimensions), mood symptoms, resilience and relationship quality comprised the secondary outcomes. Costs were estimated from health and social care and societal perspectives.ResultsThere were no differences in any of the primary outcomes for people with dementia between intervention and TAU [cognition: mean difference –0.55, 95% confidence interval (CI) –2.00 to 0.90;p-value = 0.45; self-reported quality of life: mean difference –0.02, 95% CI –1.22 to 0.82;p-value = 0.97 at the 6-month follow-up]. iCST did not improve mental/physical health for carers. People with dementia in the iCST group experienced better relationship quality with their carer, but there was no evidence that iCST improved their activities of daily living, depression or behavioural and psychological symptoms. iCST seemed to improve health-related quality of life for carers but did not benefit carers’ resilience or their relationship quality with their relative. Carers conducting more sessions had fewer depressive symptoms. Qualitative data suggested that people with dementia and their carers experienced better communication owing to iCST. Adjusted mean costs were not significantly different between the groups. From the societal perspective, both health gains and cost savings were observed.ConclusionsiCST did not improve cognition or quality of life for people with dementia, or carers’ physical and mental health. Costs of the intervention were offset by some reductions in social care and other services. Although there was some evidence of improvement in terms of the caregiving relationship and carers’ health-related quality of life, iCST does not appear to deliver clinical benefits for cognition and quality of life for people with dementia. Most people received fewer than the recommended number of iCST sessions. Further research is needed to ascertain the clinical effectiveness of carer-led cognitive stimulation interventions for people with dementia.Trial registrationCurrent Controlled Trials ISRCTN65945963.FundingThis project was funded by the National Institute of Health Research (NIHR) Health Technology Assessment (HTA) programme and will be published in full inHealth Technology Assessment; Vol. 19, No. 64. See the NIHR Journals Library website for further information.

Section: Randomisationmentioning

confidence: 99%

Individual cognitive stimulation therapy for dementia: a clinical effectiveness and cost-effectiveness pragmatic, multicentre, randomised controlled trial

Orgeta

Leung

Yates

et al. 2015

142

“…We used a password-protected website that accessed an adaptive algorithm to protect against subversion while ensuring that each trial arm was balanced by centre. 60 To validate each request for randomisation, the website asked:…”

Section: Randomisationmentioning

confidence: 99%

Comparison Of iNfliximab and ciclosporin in STeroid Resistant Ulcerative Colitis: pragmatic randomised Trial and economic evaluation (CONSTRUCT)

Williams

Alam

Alrubaiy

et al. 2016

BackgroundThe efficacy of infliximab and ciclosporin in treating severe ulcerative colitis (UC) is proven, but there has been no comparative evaluation of effectiveness.ObjectiveTo compare the clinical effectiveness and cost-effectiveness of infliximab and ciclosporin in treating steroid-resistant acute severe UC.MethodBetween May 2010 and February 2013 we recruited 270 participants from 52 hospitals in England, Scotland and Wales to an open-label parallel-group, pragmatic randomised trial. Consented patients admitted with severe colitis completed baseline quality-of-life questionnaires before receiving intravenous hydrocortisone. If they failed to respond within about 5 days, and met other inclusion criteria, we invited them to participate and used a web-based adaptive randomisation algorithm to allocate them in equal proportions between 5 mg/kg of intravenous infliximab at 0, 2 and 6 weeks or 2 mg/kg/day of intravenous ciclosporin for 7 days followed by 5.5 mg/kg/day of oral ciclosporin until 12 weeks from randomisation. Further treatment was at the discretion of physicians responsible for clinical management. The primary outcome was quality-adjusted survival (QAS): the area under the curve (AUC) of scores derived from Crohn’s and Ulcerative Colitis Questionnaires completed by participants at 3 and 6 months, and then 6-monthly over 1–3 years, more frequently after surgery. Secondary outcomes collected simultaneously included European Quality of Life-5 Dimensions (EQ-5D) scores and NHS resource use to estimate cost-effectiveness. Blinding was possible only for data analysts. We interviewed 20 trial participants and 23 participating professionals. Funded data collection finished in March 2014. Most participants consented to complete annual questionnaires and for us to analyse their routinely collected health data over 10 years.ResultsThe 135 participants in each group were well matched at baseline. In 121 participants analysed in each group, we found no significant difference between infliximab and ciclosporin in QAS [mean difference in AUC/day 0.0297 favouring ciclosporin, 95% confidence interval (CI) –0.0088 to 0.0682;p = 0.129]; EQ-5D scores (quality-adjusted life-year mean difference 0.021 favouring ciclosporin, 95% CI –0.032 to 0.096;p = 0.350); Short Form questionnaire-6 Dimensions scores (mean difference 0.0051 favouring ciclosporin, 95% CI –0.0250 to 0.0353;p = 0.737). There was no statistically significant difference in colectomy rates [odds ratio (OR) 1.350 favouring infliximab, 95% CI 0.832 to 2.188;p = 0.223]; numbers of serious adverse reactions (event ratio = 0.938 favouring ciclosporin, 95% CI 0.590 to 1.493;p = 0.788); participants with serious adverse reactions (OR 0.660 favouring ciclosporin, 95% CI 0.282 to 1.546;p = 0.338); numbers of serious adverse events (event ratio 1.075 favouring infliximab, 95% CI 0.603 to 1.917;p = 0.807); participants with serious adverse events (OR 0.999 favouring infliximab, 95% CI 0.473 to 2.114;p = 0.998); deaths (all three who died received infliximab;p = 0.247) or concomitant use of immunosuppressants. The lower cost of ciclosporin led to lower total NHS costs (mean difference –£5632, 95% CI –£8305 to –£2773;p < 0.001). Interviews highlighted the debilitating effect of UC; participants were more positive about infliximab than ciclosporin. Professionals reported advantages and disadvantages with both drugs, but nurses disliked the intravenous ciclosporin.ConclusionsTotal cost to the NHS was considerably higher for infliximab than ciclosporin. Nevertheless, there was no significant difference between the two drugs in clinical effectiveness, colectomy rates, incidence of SAEs or reactions, or mortality, when measured 1–3 years post treatment. To assess long-term outcome participants will be followed up for 10 years post randomisation, using questionnaires and routinely collected data. Further studies will be needed to evaluate the efficacy and effectiveness of new anti-tumour necrosis factor drugs and formulations of ciclosporin.Trial registrationCurrent Controlled Trials ISRCTN22663589.FundingThis project was funded by the NIHR Health Technology Assessment programme and will be published in full inHealth Technology Assessment; Vol. 20, No. 44. See the NIHR Journals Library website for further project information.

“…The randomisation was performed by dynamic allocation 302 to ensure that good balance in the allocation ratio of 1 : 1 was maintained, both within each stratification variable and across the trial. Participants were stratified by (1) hospital and (2) gender.…”

Section: Randomisationmentioning

confidence: 99%

Developing a multidisciplinary rehabilitation package following hip fracture and testing in a randomised feasibility study: Fracture in the Elderly Multidisciplinary Rehabilitation (FEMuR)

Williams

Roberts

Din

et al. 2017

BackgroundProximal femoral fracture is a major health problem in old age, with annual UK health and social care costs of £2.3B. Rehabilitation has the potential to maximise functional recovery and maintain independent living, but evidence of clinical effectiveness and cost-effectiveness is lacking.ObjectivesTo develop an enhanced community-based rehabilitation package following surgical treatment for proximal femoral fracture and to assess acceptability and feasibility for a future definitive randomised controlled trial (RCT) and economic evaluation.DesignPhase I – realist review, survey and focus groups to develop the rehabilitation package. Phase II – parallel-group, randomised (using a dynamic adaptive algorithm) feasibility study with focus groups and an anonymised cohort study.SettingRecruitment was from orthopaedic wards of three acute hospitals in the Betsi Cadwaladr University Health Board, North Wales. The intervention was delivered in the community following hospital discharge.ParticipantsOlder adults (aged ≥ 65 years) who had received surgical treatment for hip fracture, lived independently prior to fracture, had mental capacity (assessed by the clinical team) and received rehabilitation in the North Wales area.InterventionsParticipants received usual care (control) or usual care plus an enhanced rehabilitation package (intervention). Usual care was variable and consisted of multidisciplinary rehabilitation delivered by the acute hospital, community hospital and community services depending on need and availability. The intervention was designed to enhance rehabilitation by improving patients’ self-efficacy and increasing the amount and quality of patients’ practice of physical exercise and activities of daily living. It consisted of a patient-held information workbook, a goal-setting diary and six additional therapy sessions.Main outcome measuresThe primary outcome measure was the Barthel Activities of Daily Living (BADL) index. The secondary outcome measures included the Nottingham Extended Activities of Daily Living (NEADL) scale, EuroQol-5 Dimensions, ICEpop CAPability measure for Older people, General Self-Efficacy Scale, Falls Efficacy Scale – International (FES-I), Self-Efficacy for Exercise scale, Hospital Anxiety and Depression Scale (HADS) and service use measures. Outcome measures were assessed at baseline and at 3-month follow-up by blinded researchers.ResultsSixty-two participants were recruited (23% of those who were eligible), 61 were randomised (control,n = 32; intervention,n = 29) and 49 (79%) were followed up at 3 months. Compared with the cohort study, a younger, healthier subpopulation was recruited. There were minimal differences in most outcomes between the two groups, including the BADL index, with an adjusted mean difference of 0.5 (Cohen’sd = 0.29). The intervention group showed a medium-sized improvement on the NEADL scale relative to the control group, with an adjusted mean difference between groups of 3.0 (Cohen’sd = 0.63). There was a trend for greater improvement in FES-I and HADS in the intervention group, but with small effect sizes, with an adjusted mean difference of 4.2 (Cohen’sd = 0.31) and 1.3 (Cohen’sd = 0.20), respectively. The cost of delivering the intervention was £231 per patient. There was a possible small relative increase in quality-adjusted life-years in the intervention group. No serious adverse events relating to the intervention were reported.ConclusionsTrial methods were feasible in terms of eligibility, recruitment and retention, although recruitment was challenging. The NEADL scale was more responsive than the BADL index, suggesting that the intervention could enable participants to regain better levels of independence compared with usual care. This should be tested in a definitive Phase III RCT. There were two main limitations of the study: the feasibility study lacked power to test for differences between the groups and a ceiling effect was observed in the primary measure.Trial registrationCurrent Controlled Trials ISRCTN22464643.FundingThis project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full inHealth Technology Assessment; Vol. 21, No. 44. See the NIHR Journals Library for further project information.