Evaluating Diuretics in Normal Care (EVIDENCE): a feasibility report of a pilot cluster randomised trial of prescribing policy in primary care to compare the effectiveness of thiazide-type diuretics in hypertension

Abstract

Abstract Background Obtaining evidence on comparative effectiveness and safety of widely prescribed drugs in a timely and cost-effective way is a major challenge for healthcare systems. Here, we describe the feasibility of the Evaluating Diuretics in Normal Care (EVIDENCE) study that compares a thiazide and thiazide-like diuretics for hypertension as an exemplar of a more general framework for efficient generation of such evidence. In 2011, the UK NICE hypertension guideline included a recommendation that thiazide-like diuretics (such as indapamide) be used in preference to thiazide diuretics (such as bendroflumethiazide) for hypertension. There is sparse evidence backing this recommendation, and bendroflumethiazide remains widely used in the UK. Methods Patients prescribed indapamide or bendroflumethiazide regularly for hypertension were identified in participating general practices. Allocation of a prescribing policy favouring one of these drugs was then randomly applied to the practice and, where required to comply with the policy, repeat prescriptions switched by pharmacy staff. Patients were informed of the potential switch by letter and given the opportunity to opt out. Practice adherence to the randomised policy was assessed by measuring the amount of policy drug prescribed as a proportion of total combined indapamide and bendroflumethiazide. Routinely collected hospitalisation and death data in the NHS will be used to compare cardiovascular event rates between the two policies. Results This pilot recruited 30 primary care practices in five Scottish National Health Service (NHS) Boards. Fifteen practices were randomised to indapamide (2682 patients) and 15 to bendroflumethiazide (3437 patients), a study population of 6119 patients. Prior to randomisation, bendroflumethiazide was prescribed to 78% of patients prescribed either of these drugs. Only 1.6% of patients opted out of the proposed medication switch. Conclusion The pilot and subsequent recruitment confirms the methodology is scalable within NHS Scotland for a fully powered larger study; currently, 102 GP practices (> 12,700 patients) are participating in this study. It has the potential to efficiently produce externally valid comparative effectiveness data with minimal disruption to practice staff or patients. Streamlining this pragmatic trial approach has demonstrated the feasibility of a random prescribing policy design framework that can be adapted to other therapeutic areas. Trial registration ISRCTN Registry, ISRCTN46635087. Registered on 11 August 2017