A parallel group, double-blind, randomised, placebo-controlled trial comparing the effectiveness and cost-consequence and cost effectiveness of low dose oral modified release morphine (MRM) versus placebo on the intensity of worst breathlessness in people with chronic breathlessness.

BACKGROUND: Disabling chronic breathlessness often persists despite optimal treatment of cardiorespiratory disease(s). It is associated with poorer quality of life, shorter survival, limits in daily activities, unplanned hospital attendances/admissions. Some data support short-term use of regular, low-dose morphine for chronic breathlessness but longer term data are lacking. AIM: Test the effectiveness, cost-consequence and cost-effectiveness of regular, low dose oral modified release morphine (MRM) for breathlessness intensity compared with placebo. METHODS: A mixed-methods, phase 3, parallel arm, randomised placebo controlled trial in opioid naïve, ambulant people with moderate/severe (mMRC 3 or 4) breathlessness due to advanced disease. An internal pilot will test recruitment, and proceed to phase 3 if a priori criteria are met. Participants will be randomised (stratified for causal disease) to i) 5mg twice daily oral modified-release morphine and laxative or ii) placebo modified-release morphine and placebo laxative. Non-responders (<1 point improvement 0-10 numerical rating scale [NRS] worst breathlessness2) who are tolerating the treatment at day 7 will receive 10mg twice daily (or placebo) modified-release morphine. At 2 months, participants may enter an open label pharmacovigilance phase until study completion. Carers will be invited to participate. A normalisation process theory (NPT) study will address clinician and patient factors affecting trial equipoise and safe prescribing/monitoring. Outcomes: PRIMARY: breathlessness (NRS worst breathlessness/24 hours at 4 weeks) measured at baseline, weeks 1, 2, 3 and 4. SECONDARY: distress due to breathlessness/24 hours (NRS); breathlessness mastery (Chronic Respiratory Questionnaire); quality of life (SF-12; EQ-5D-5L); sleep (Epworth); harms (respiratory, sedation, delirium, nausea, constipation); function (Australian Karnofsky Performance Scale); capability (ICECAP-SCM); healthcare utilisation, costs and cost-effectiveness; survival; carer burden (Zarit 6); safe prescribing. STATISTICAL ANALYSES: The intention-to-treat analysis of the primary outcome will use a repeated measures linear model (weeks 1 to 4), adjusted for baseline covariates and causal disease stratification. Secondary outcomes will be analysed as per the primary outcome with an appropriate generalised linear model. Time to event data will use a suitable survival model. Robustness of findings to missing data will be assessed. The health-economic analysis will report cost consequences and the incremental cost-effectiveness ratio. SAMPLE SIZE: For 90% power and 5% level of significance detection of a 0.4 effect size in the primary outcome will need 264 participants. The expected standard deviation is around 2.5 so an effect size of 0.4 translates to an absolute mean difference of 1 in NRS (moderate change). For a post randomisation serial correlation of 0.5 on 4 measures of outcome, this sample size reduces to 168. Further adjustment for baseline covariates (correlation with outcome of 0.5) reduces this to 126. Allowing for 20% attrition increases the sample size to 158. These assumptions will be checked after 50 randomised achieving mature outcome data. A NORMALISATION PROCESS THEORY STUDY will explore attitudes and practices about morphine use (domains: coherence with existing practices; and for clinicians, patients and carers cognitive participation and collective action and reflexive monitoring). A modified Normalisation Measurement Instrument survey will inform interviews (clinicians, patients and carers) and field observations.