Randomised controlled trial of two brief interventions against long-term benzodiazepine use: Cost-effectiveness

Abstract

Previous findings have indicated that a letter from a patient's General Practitioner (GP) and a short GP consultation leads to reduced intake among long-term benzodiazepine (BZD) users. To compare the cost-effectiveness and potential cost savings of these two brief interventions. Economic evaluation conducted alongside a prospective randomised controlled trial from the perspective of the NHS. A total of 273 long-term BZD users (=>6 mos) at seven general practices and regarded by their GPs as suitable to take part in the study within the Newcastle and North Tyneside District Health Authority were identified from repeat prescription computer records. Patients were randomised to usual GP care + assessment only or the offer of a short consultation (12 mins approx) with the patient's GP (or practice pharmacist/practice nurse) or a letter signed by the GP advising gradual reduction in BZD intake. Economic measures taken were: costs of intervention; savings (costs) of changes in health service use from before to after intervention; savings to the NHS from reductions in drug use and dispensing costs; total costs of brief intervention; simulations of savings (costs) extrapolated to the District Health Authority. The letter was the more cost-effective intervention when taking into account changes in health service use and savings to the drugs bill. If all GPs in Newcastle and North Tyneside screened long-term BZD users on their lists and sent the letter studied here to those considered suitable to receive it, it is estimated that savings to the District Health Authority would be a minimum of £4.9 million per annum. Routine implementation of the letter intervention in general practice throughout the UK would result in large financial gains to the NHS. These savings represent a conservative estimate of savings to the public sector, as wider savings to the social care system may also be expected as a result of the policy.