The challenges of assessing patients' medication beliefs: a qualitative study

Background: An estimated 50% of patients do not take their medication as prescribed, with medication adherence associated with adverse outcomes and higher costs of care. The Necessity-Concerns Framework identified individual's beliefs about their medication as playing a key role in adherence, and UK Clinical Adherence Guidelines recommend eliciting and incorporating individual's perceptions of their medication within the consultation. The Beliefs about Medicines Questionnaire (BMQ) is widely used to assess medication beliefs, however, given the condition-specific nature of some self-management regimens, it is unknown whether this tool is able to fully capture beliefs about more complex medication regimens. Methods: We examined the challenges of assessing medication beliefs using the BMQ in 20 people with a complex relapsing-remitting condition recruited from community sources. Data were collected from people with psoriasis; a patient group characterised by complex medication regimens, which include therapies that are applied topically, phototherapy/photochemotherapy, and therapies that are administered orally or via subcutaneous or intravenous injections. Semi-structured cognitive interviews were undertaken, with responses coded using established schedules and analysed using Content analysis. Results: Individual's beliefs about their condition specific therapies were not accurately captured by the BMQ. Medication beliefs as expressed during 'real-time' completion of the BMQ were underestimated, or failed to be captured, by the corresponding scores given by participants. There was mismatch between the terminology used in the scale and individuals perceptions of their condition and the complexity of its management and treatment outcomes. Currently the BMQ cannot represent beliefs about medicines underuse, even though some individuals with psoriasis viewed access to therapies as overly restrictive. Some the BMQ items were misinterpreted in part due to ambiguous item wording or due to misreading by participants. Conclusions: This is the first study to identify general and condition-specific difficulties experienced by individuals completing the BMQ in 'real time'. The main implication of this research is the need to develop condition-specific versions of the BMQ in order that this important instrument can capture the full range of medication beliefs in individuals living with a complex relapsing-remitting condition. Access to condition-specific versions could significantly increase our understanding of beliefs which facilitate or reduce medication adherence.