129 ACCEPTANCE protocol 5 Table 2. Scenario power calculations Drop-out rate 25% 18% 16% 10% Power 90% 121 per arm 111 per arm 108 per arm 101 per arm Power 80% 91 per arm 83 per arm 81 per arm 76 per arm Statistical analysis The statistical analysis plan is presented in Supplementary File E3. Data will be analysed using the intention-to-treat principle. In addition, a per-protocol analysis will be performed for the primary outcome. Baseline demographic and clinical characteristics will be summarised using means and SD, or medians and IQRs, where appropriate. To test the effect of intervention condition on medication adherence and on changes in medication adherence over time, a multilevel linear mixed-model analysis will be performed. The model will include weekly adherence rates per patient from baseline to T12 (i.e., recorded as a percentage). A precise definition and of medication adherence and how it is calculated is provided in the statistical analysis plan (Supplementary File E3). Medication adherence data around visits will be disregarded to minimise bias. The mixed-model will include a random intercept per general practice. A correlation structure will be chosen for the repeated measurements on the level of patients by selecting the best fitting variance-covariance matrix. The model will include fixed effects for treatment (intervention or control), time, their interaction, age and baseline adherence. Assumptions for mixed models will be investigated beforehand to check that these are met. The mixed-effect model will provide valid statistical inferences in the presence of missing outcome data, which can be explained by covariates in the model (i.e., treatment, age, time). To analyse the effect of intervention on secondary outcomes over time, a similar approach as for the primary outcome will be used. A linear mixed model will be used to assess whether the effect of the intervention on medication adherence and asthma control at 12 months is modified by patient characteristics (i.e., self-efficacy, attitude, medication beliefs, illness perception and eHealth literacy). Sensitivity analysis will be performed using all medication adherence data (i.e., including medication adherence data measured around follow-up moments) and including patients with doctor diagnosed asthma only (see the Route C (recruitment via pharmacies) section). No interim analysis will be performed. Statistical analyses will be carried out using R version 4.1.1.(57 and the R Studio IDE version 1.3.1073 (or higher versions of the programmes)(58). P-values below 0.05 are considered statistically significant. Cost-effectiveness analysis A cost-effectiveness analysis will be performed alongside the trial to compare the costs and outcomes of the smart inhaler programme with the control group. A costeffectiveness model will be used to explore long-term effects. Cost-effectiveness will
RkJQdWJsaXNoZXIy MTk4NDMw