OBJECTIVES: Although self-reported questionnaires are commonly used to measure medication adherence, self-reported adherence may be influenced by socially desirable answers resulting in untruthful reporting. Such untruthful reporting can have serious consequences for reporting on treatment (cost)effectiveness. This study investigated how to induce truthful self-reported adherence and how this can explain heterogeneity in treatment preferences.

METHODS: A survey measuring self-reported medication adherence was administered immediately after a discrete choice experiment regarding preferences for Multiple Sclerosis (MS) treatments. Data was collected among MS patients in the Netherlands, France and the United Kingdom. Half of the sample was randomized to receive choice-matching which was used to induce truthful responses towards questions on adherence. Choice-matching (financially) incentivizes respondents depending on individuals’ predictions about others’ responses in the same sample, and the choice predictions of others with the same adherence score. Descriptive statistics and mixed logit models with interaction effects were conducted to determine the influence of choice-matching on self-reported adherence and to assess whether truthful adherence explains preference heterogeneity.

RESULTS: The sample comprised 758 MS-patients (mean age 40.8 years, 67% female). On average, respondents reported a higher adherence if they only received the adherence questions compared to respondents also receiving choice-matching (always adherent: 46.6% vs. 39.3%). The mixed logit model showed statistically significant interaction effects between the self-reported adherence and preferences for treatment modality. Only in the choice-matching group, higher adherence resulted in lower utility scores for implants compared to injections or pills (p=0.01).

CONCLUSIONS: Choice-matching encouraged respondents to report their true medication adherence. Linking truthful adherence to preferences allows for a better understanding of preference heterogeneity, better (cost)effectiveness assessments of treatments and can help patients and physicians with making a shared treatment decision that fits the patients’ true preferences.