128 Chapter 5 5 care unit days and length of stay, emergency department visits because of asthma, asthma-related visits and phone calls to the general practice, and medical specialist visits because of asthma. Data on medication use will be retrieved from the patients’ main pharmacist dispense system at T12. Absenteeism and presenteeism The Work Productivity and Activity Impairment instrument (WPAI) will be completed by patients to measure absenteeism and presenteeism. The questionnaire consists of nine questions in three domains (work impairment, school impairment and activity impairment)(51,52). Outcomes on absenteeism, presenteeism, work productivity loss and activity impairment are expressed as percentages, with higher numbers indicating greater impairment and/or less productivity. Cost-effectiveness analysis The cost-effectiveness of the smart inhaler programme will be assessed by comparing the costs and benefits of the programme (i.e., intervention group) with usual care (i.e., control group) in a cost-effectiveness analysis. Sample size The power calculation is based on the primary outcome: medication adherence over 12 months, as measured by electronic monitoring of inhaler actuations. The treatment effect is expressed as the absolute difference in mean medication adherence between the intervention group and the control group. The sample size is based on an absolute difference in mean medication adherence between the groups of 15% (effect size), based on an expected adherence rate of 65% in the control group (22,53,54), and the target of a mean adherence of 80% in the intervention group. A SD of 0.30 is used (22). A design effect of 1.075 is used, which is based on an intra-cluster correlation coefficient of 0.025 (55), the INCA® (INhaler Compliance Assessment), and a cluster size of 4. The cluster size is based on (1) the average number of asthma patients in a Dutch general practice; (2) data on age, asthma control level and medication use (56); (3) the assumption that 40% of the patients is non-adherent; and (4) recruitment rates in previous primary care asthma trials. To detect an absolute difference of 15% in mean medication adherence with 90% power and a 5% significance level, a sample size of 242 patients (121 per arm) across approximately 30 clusters in each arm is needed. Given the COVID-19 circumstances and the substantial impact on recruitment pace and strain on healthcare, it is difficult to predict recruitment and drop-out rates. Therefore, we explored different scenarios based on a power of 80% and varying drop-out rates based on literature (Table 2).
RkJQdWJsaXNoZXIy MTk4NDMw