Intentional non-adherence to official medical recommendations: An irrational choice or negative experience with the healthcare system?

Abstract

Abstract Background: The World Health Organization recognizes non-adherence to treatment recommendations as a growing global problem. Questionnaires typically focus on only one non-adhering behavior, e.g., medication-taking, and target people with specific health conditions. In this preregistered study, we aimed to (1) develop a novel instrument suitable to the general population, consisting of non-adherence to prescribed therapy, self-medication, and avoidance to seek medical treatment, capturing intentional non-adherence (iNAR), (2) differentiate it from habitual non-adherence, such as smoking, and (3) relate iNAR to patient-related factors, such as sociodemographics, health status, and endorsement of irrational beliefs (conspiratorial thinking and superstitions), and to a set of healthcare-related beliefs and experiences. Methods: First, medical doctors provided a list of non-adherence behaviors in a focus group. To this list we added behaviors that fitted our definition of intentional non-adherence, identified in a thorough literature search. This initial list of 22 behaviors was further refined by a public health expert. The instrument was then tested on a sample of 583 participants, sufficiently powered to detect effects for all preregistered statistical analysis. Participants were recruited using an online snowballing procedure and via social networks. Results: Factor analysis yielded a one-factor solution, and the final 12-item iNAR questionnaire had satisfactory internal reliability (Cronbach’s alpha = .72). A hierarchical linear regression showed that, as expected, health condition variables and healthcare-related beliefs and experiences accounted for 14% of the variance of iNAR behaviors, whereas sociodemographics and irrational beliefs did not additionally contribute. In comparison, the same regression model with smoking (representative of habitual non-adherence) as a criterion variable, accounted for less than 3% of the variance, with education as the only significant predictor. Conclusions: We have constructed a brief, comprehensive, and reliable measure of iNAR behaviors. Normalization of patients’ passivity and, in particular, negative experiences with the healthcare system contributed to intentional, but not habitual non-adherence. We believe that public health interventions designed to discourage intentional non-adherence should foremost attempt to improve all patients' experiences with the system and build trust with their healthcare practitioners rather than aim at specific demographic groups or at correcting patients' unfounded beliefs.