Improving medication adherence with adjuvant aromatase inhibitor in women with breast cancer: study protocol of a randomised controlled trial to evaluate the effect of short message service (SMS) reminder

Medication adherence refers to whether a patient takes medication according to the frequency prescribed, or continues to take a prescribed medication. Inadequate adherence to medication may cause alterations in risk-benefit ratios, resulting in reduced benefits, increased risks or both, and is significantly associated with adverse clinical outcomes and higher healthcare costs. We aim to examine the effect of a computer generated short message service (SMS) reminder in improving medication adherence, and inhibiting the aromatisation process amongst breast cancer women receiving oral aromatase inhibitor therapy. In this randomised controlled trial, eligible patients will be equally allocated to receive either SMS reminder or standard care. The former receives weekly SMS reminder to take medication while the latter does not receive any. The primary endpoint of medication adherence at 1-year is assessed using the Simplified Medication Adherence Questionnaire, and compared using the χ2 test. Adjustment for baseline covariate and potential confounders will be made using the logistic regression. Secondary outcomes involving estrone and androstenedione levels will be compared using the analysis of covariance, whereas the estradiol levels (< 18.4 pmol/L versus ≥18.4 pmol/L) will be compared using the χ2 test, and the logistic regression. Further, the assessment of knowledge, attitude, behaviour, and barriers and facilitating factors of medication adherence will be made via logistic regression. This will be the first study to evaluate short-term clinical outcomes from SMS reminder for breast cancer patients on aromatase inhibitor therapy. Random allocation to SMS reminder or control arm ensures that patients in both arms will be comparable with respect to demographic and clinical characteristics, and any difference in outcomes can be attributed to the intervention. Participants are not blinded to the assignment of intervention, thus there may be potential for bias in outcome assessments. NCT02524548 . Retrospectively registered on 17 August 2015.