Journal of General Practice

Feasibility of a Decision Support Tool for the Pharmacological Management of Depression

Abstract

Stephen Kellett1,2*, Melanie Simmonds-Buckley1,2, Mike Seneviratne1, Katherine Morgan2, Mark Boon3 and Ian Seymour4

Objective: The accurate prescription of antidepressants is technically complex and difficult to achieve. This project therefore sought to evaluate the feasibility of integrating an evidence-based decision support tool (Predictix) into General Practice. Methods: This was a real-world mixed methods evaluation in N=2 General Practices of the experiences of patients (n=24), General Practitioners (GPs; n=4) and Predictix administrators (n=2). Suitable patients were assessed using Predictix and this produced a recommendation for the likely most efficacious anti-depressant. Two patient study groups were created (a) antidepressant recommended by Predictix and (b) a GP clinical opinion override group. Both groups of patients were followed up at 12-weeks. The PHQ-9 was embedded in the Predictix assessment, and the measure was recompleted at 12-weeks follow-up. Qualitative interviews were conducted with GPs, patients and administrators and a reflexive thematic analysis (RTA) conducted. Results: It took 20-minutes to administer Predictix and then 3-days for the prescription to be issued. N=19 (88%) received a Predictix selected antidepressant (2 participants dropped out during the Predictix assessment), of whom 8/19 recovered. The N=3 (14%) in the GP override group all recovered. The common qualitative themes across GPs, patients and administrators were managing expectations, individualized care and burden. Conclusion: Whilst decision support tools have the potential for increasing prescribing accuracy for antidepressants, practical and psychological barriers to implementation need to be identified and resolved.