Efficacy of CareAide® App in Improving Adherence in Adults With Chronic Diseases

According to WHO and Ministry of Health (MOH) Malaysia 2020 report, expenditure for chronic diseasesestimated involved around 0.65% of Malaysia's gross domestic product (GDP) which is about RM8.91 billion (Ministry of Health Malaysia, 2020). This showed that chronic diseases imposed a significant healthburden to our country. Malaysian population having good and easy access to effective medical therapyunder the public healthcare system, approximately 46% to 56% of patients with hypertension were non-adherent in terms of their medications. Therefore, CareAide App will be used to study and identify the effect, clinical outcome and cost effectiveness of mobile health app in managing chronic diseases.

The Comprehensive Adherence-enhancing Reminder and Engagement Aide (CAREAide) Trial aims to assess the effect of CareAide® app on adherence, clinical outcomes of patients with chronic diseases and user's perception and satisfaction with the app.

A 3-and 6- month open-label multicentre randomized controlled trial with two parallel groups will be conducted. This study will take place in the outpatient setting of the three hospitals [i.e., Hospital Putrajaya, Hospital Pulau Pinang and Universiti Malaya Medical Center (UMMC)) in Malaysia]. Both interventional and control group will be allocated to 1:1 ratio. Randomization number will be generated by using online number generated programme. Consultants reviewing patients at medical, cardiology,diabetes mellitus and asthma clinics will provide referral and notify independent researchers for eligibility and group assignments. Eligible patients will then be instructed to see pharmacists/researchers atoutpatient pharmacies to refill their prescriptions and subsequently enrol in the assigned study group either to receive an intervention with CareAide® or to receive normal clinical care. Information and training on the app will be provided by these researchers to the participants.Patients will be recruited at therespective outpatient clinics. At the usual care clinic appointments, patients have their demographics and baseline data of blood pressure, blood sugar, basic blood profiles, asthma control and review of theirdisease management collected by consultants/researchers. Patients who meet the study criteria will then be instructed to see pharmacists/researchers at outpatient pharmacy to refill their medications. The pharmacists/researchers will screen patients' inclusion and exclusion criteria for this study and will explainthat they may or may not be selected for the study and if they agree, proceed to provide written informedconsent. The adherence data from the screening will be collected to gauge the general adherence rate inthese group of patients. An independent researcher, who is not involved in the intervention and controlgroups, will use the random number generator software for randomisation sequences. The random numbers will be placed in opaque envelopes, which subsequently will be opened by pharmacists/researchers at the outpatient pharmacies. Participants and pharmacists/researchers are unblinded to group assignment and are aware of the intervention, but the study hypotheses are not disclosed to the participants. The assessor and statisticians are blinded to participant allocation. The study will be evaluating usual care compared with usual care plus pharmacist-driven patient-specific adherence interventions - CareAide® medication adherence digital application. Usual care provided bythese hospitals comprises clinic appointments every 3 to 6 months. Medication adherence: CareAide® effectiveness is measured with a medication adherence tool at baseline and follow-up time periods at 3months and 6 months using Morisky Medication Adherence Scale (MMAS) and proportion of days covered (PDC). MMAS-8 is a licensed validated questionnaire to assess medication adherence, where a higher score represents higher medication adherence and vice versa. Comprising of 8 items, the score ranges from 0 to 8, where a higher score represents a higher medication adherence. The internal consistency (Cronbach's alpha) of MMAS-8 is 0.83 along with a sensitivity of 93% and a specificity of 57%.

The PDC method utilises pharmacy refill records. Fasting blood sugar, HbA1C, cardiovascular events (heart failure admission, mortality), asthma control and blood pressure, hospitalizations and emergency visits measurements will be based on the clinics attended by the participants. Health care utilization, quality of life and health care expenditures willbe measured for all the participants.

Intention to treat (ITT) analysis will be applied to evaluate the results to avoid any bias. The data analysiswill be using the SPSS version 29. Data will be coded and analysed. Descriptive or categorical data will beexpressed as mean ± standard deviation (SD) unless otherwise stated. Student t-test will be used foranalysis of normally distributed or continuous variables. Mann-Whitney test will be used for non-normallydistributed data. Categorical data will be analysed using Chi-square or Fischer's exact test. A value of P <0.05 is considered statistically significant.

The prevalence of medication compliant is quantified at 3 months and 6 months follow-up. Independent t-test or Mann-Whitney test will be used to compare MMAS score rankings between the intervention andcontrol groups. Chi-Square test of independence will be used to determine the difference in number ofpatients with medication adherence between the intervention and control groups at 3 months and 6months follow-up. Ordinal logistic regression, with the MMAS adherence score ranking, will be used to examine the association between intervention and control groups.

For the perception and satisfaction towards the app, the frequency and percentage of respondents in each category (strongly disagree to strongly agree) will be calculated. The median of each statement will then be calculated to give an overview of perception and satisfaction towards the CareAide® app. Chi-square, Mann-Whitney U, Kriskall Wallis, Chi-square tests and logistic regression will be used where appropriate. Linear regression analysis will be conducted to determine the independent predictors of HRQoL.

An economic evaluation will be conducted from a payer and societal perspective. Costs and outcomes will be discounted at 3% per year as the results will be extrapolated to examine costs and outcome over a year. Cost effectiveness analysis will be estimated based on the results of cost data and the incremental cost per quality adjusted life year (QALY) gained. Total incremental cost effectiveness ratio (ICER) will be calculated by dividing the difference between the cost of CareAide intervention group and control group by the difference between the QALYs of CareAide intervention group and control group. Sensitivity analysis will be performed by bootstrapping technique to ensure or examine the robustness of ICER. Then, cost effectiveness acceptability curve (CEAC) will be created to estimate the proportion of the result being cost-effective in relation to the Willingness-to-Pay (WTP) threshold.