Louise Ann Rabbitt
Background Medication non-adherence is prevalent and clinically significant in patients attending specialist hypertension services, yet it remains poorly detected in clinical practice. Chemical adherence testing (CAT) using liquid chromatography with high-resolution mass spectrometry (LC-HRMS) is an objective method of adherence detection, but the evidence base is heterogeneous, patient perspectives are largely absent, and the economic implications of non-adherence and its detection are incompletely characterised. Aims This thesis investigates the detection, psychosocial correlates, and economic implications of antihypertensive medication non-adherence in specialist hypertension clinic settings, using a multi-methods approach comprising evidence synthesis, observational study, qualitative enquiry, and health economic evaluation. Methods A scoping review examined the existing literature on CAT in hypertension management. A prospective cross-sectional study (the UNEARTH study) was conducted at two Irish tertiary hypertension centres (n = 171), using unannounced LC-HRMS urine testing alongside a battery of validated psychosocial questionnaires. A qualitative interview study used fictional vignettes to explore patient perspectives on CAT. A micro-costing analysis estimated the cost to the Irish health service of delivering specialist hypertension clinic care (n = 65). A pilot cross-sectional study developed and deployed a data collection instrument to quantify private out-of-pocket costs incurred by patients attending hospital outpatient clinics (n = 42). Results The scoping review (48 studies) found the CAT literature to be predominantly observational, methodologically heterogeneous, and largely confined to specialist settings, with near-complete absence of patient perspectives and limited adoption of standardised reporting frameworks. In the UNEARTH study; CAT-confirmed non-adherence, defined as non-detection of at least two prescribed antihypertensive agents, was identified in 6.9% of participants; significantly lower than anticipated from prior literature. Non-adherent participants had substantially higher blood pressure, and sceptical medication beliefs were the most consistent psychosocial correlate of non-adherence. Qualitatively, patients considered CAT most acceptable when embedded within a relational, patient-centred approach to care. The micro-costing analysis estimated the mean total cost per patient at approximately €3,277 (range €1,446 to €5,644 according to diagnosis). The pilot study found a median private cost per visit of €131, with potential gender disparities in carer burden. Conclusions This thesis provides the first multi-methods investigation of medication non-adherence in specialist hypertension care integrating clinical, psychosocial, qualitative, and health economic perspectives. The findings support a circumscribed but important role for CAT at high-stakes clinical decision points, embedded within a relational approach that addresses the attitudinal and practical determinants of non-adherence. The economic analyses suggest that improved early detection of non-adherence has potential to generate cost savings within current care pathways.