Evidence-based medicine and the patient—the example of cardiovascular disease prevention

Abstract

Background
Patient ideas about cardiovascular diseases (CVDs) differ widely from those of clinicians. The failure to understand these explanatory models has been suggested to be a key barrier to effective CVD prevention. Nonetheless, currently available decision-support interventions for cvd prevention have had limited patient involvement in their development.
Aims
Through four studies, this thesis seeks to use qualitative and quantitative methods to identify barriers and recommend best practices for supporting decisions in cvd prevention: both for education, and communicating the risk of benefit and harm from treatment.
Methods
Study i analyses data from the South London Stroke Register, examining factors associated with the diagnosis and treatment of risk factors prior-to-stroke. Study ii systematically reviews qualitative studies of patient perspectives on hypertension and medication taking. Study iii systematically reviews randomized controlled trials (RCTs) and qualitative studies examining the effectiveness of different strategies for communicating cvd risk. Study iv is a qualitative study of patient perspectives on CVD risk in two south London general practices.
Results
Study i found low but increasing prescribing rates of all preventative medication classes; prescribing did not differ by ethnicity or socio-economic status. Study ii included 52 qualitative studies. Participants experienced hypertension as symptomatic and strongly associated with stress. Many actively avoided medication, or self-adjusted medication use at times of lower stress or symptoms; concerns about serious adverse effects were widespread. Study iii included 23 RCTs, and found communicating CVD risk did not affect clinical outcomes, and only modestly improved decision quality; different formats produced similar results. Four qualitative studies found the risks were widely perceived as too small or distant to merit taking action. Study iv found many did not trust cvd risk estimates, perceiving they were not applicable to them as individuals, and omitted pertinent personal characteristics.

Date of Award	1 Aug 2016
Original language	English
Awarding Institution	King's College London
Supervisor	Christopher McKevitt (Supervisor) & Charles Wolfe (Supervisor)