A phase 3 trial of Rivastigmine to prevent falls in Parkinson's Disease

DESIGN: A multicentre, phase 3, double-blind, RCT of rivastigmine versus placebo to prevent falls in PD. SETTING: Secondary care movement disorder services. POPULATION: For inclusion, participants must be over 18, have a diagnosis of idiopathic PD, Hoehn and Yahr stage 1-4, not be taking a ChEi, have fallen in the previous year and be able to walk >10m without aids or assistance. Participants will be excluded if there is any absolute contraindication or they have been previously treated with a ChEi, have dementia or are falling >4x per day. HEALTH TECHNOLOGY: Intervention is rivastigmine patches. 4.6mg (4 weeks); 9.5mg (26 weeks); 13.3mg (22 weeks). Control is matched placebo patches. MEASUREMENT OF COSTS AND OUTCOMES: Falls (primary outcome) will be assessed prospectively using gold standard methods with diaries and phone calls. The primary analysis will be based on the rate of falls for each participant as the unit of analysis and will follow the intention to treat principle. A linear regression model of log transformed fall rates, adjusted for participants' age, cognition and fall history at baseline, will estimate the treatment effect as a percentage difference in average fall rate. Secondary outcome measures are validated in PD. NHS, social service, and informal care costs will be collected using monthly diaries with hospital episode statistics recording hospitalisations. Quality of life will be assessed using EQ5D-5L to calculate QALYs. We will use a well-being measure ICECAP-O, to capture the broader impact of PD-falls on patients and the Carer Experience Scale for carers. In primary economic analysis, we will estimate the cost per QALY gained of rivastigmine from as NHS and social services perspective. Uncertainty will be summarised using cost effectiveness acceptability curves and by calculating incremental net benefit statistics. If the intervention is potentially, cost-effective at 12 months, we will develop a simple extrapolation model to assess cost-effectiveness over a lifetime horizon. SAMPLE SIZE: We will recruit 600 people and assuming a 20% drop out (n=120), we will have 480, 240 per arm, to detect a 25% difference in mean fall rate between groups with 90% power and a two-sided 5% significance level. This clinically important difference equates to preventing on average 5 falls/ patient/ year. TIMETABLE and RECRUITMENT: Total trial duration will be 4 yrs, 3 months comprising 9 months set up, 3 years recruitment and follow-up and 6 months analysis and dissemination. Accrual is based on staggered adoption of 26 sites recruiting 1.25 patients site/ month or 9.25 patients/ site/ year. An internal pilot of recruitment for 9 months with prespecified stop/ amend/ go criteria will be undertaken. EXPERTISE: EH is a PD Consultant Clinician who successfully delivered the phase 2 trial and is supported by a Senior clinician (AW), Senior Trials Methodologists (CM, WH, YBS) and Senior Clinical Epidemiologist (YBS), Senior Statistician (CM) and Senior Health Economist (WH). The TSC will comprise Prof C Clarke (Chair) who has expertise in running large PD trials (Birmingham), YBS and EH, Prof J Close, Prof S Lord (Neuroscience Research Australia) and F Lindop (Derby) as gait and falls experts, a person with PD and a statistician. Collaborations with experts in cognition, (Dr Yarnall, Newcastle), and quality of life measures (Prof Coast, Bristol) strengthens the team.