Deprescribing medicines in older people living with multimorbidity and polypharmacy: the TAILOR evidence synthesis

Tackling problematic polypharmacy requires tailoring the use of medicines to individual needs and circumstances. This may involve stopping medicines (deprescribing) but patients and clinicians report uncertainty on how best to do this. The TAILOR medication synthesis sought to help understand how best to support deprescribing in older people living with multimorbidity and polypharmacy.

We identified two research questions: (1) what evidence exists to support the safe, effective and acceptable stopping of medication in this patient group, and (2) how, for whom and in what contexts can safe and effective tailoring of clinical decisions related to medication use work to produce desired outcomes? We thus described three objectives: (1) to undertake a robust scoping review of the literature on stopping medicines in this group to describe what is being done, where and for what effect; (2) to undertake a realist synthesis review to construct a programme theory that describes 'best practice' and helps explain the heterogeneity of deprescribing approaches; and (3) to translate findings into resources to support tailored prescribing in clinical practice.

Experienced information specialists conducted comprehensive searches in MEDLINE, Cumulative Index to Nursing and Allied Health Literature, Web of Science, EMBASE, The Cochrane Library (Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials), Joanna Briggs Institute Database of Systematic Reviews and Implementation Reports, Google (Google Inc., Mountain View, CA, USA) and Google Scholar (targeted searches).

The scoping review followed the five steps described by the Joanna Briggs Institute methodology for conducting a scoping review. The realist review followed the methodological and publication standards for realist reviews described by the Realist And Meta-narrative Evidence Syntheses: Evolving Standards (RAMESES) group. Patient and public involvement partners ensured that our analysis retained a patient-centred focus.

Our scoping review identified 9528 abstracts: 8847 were removed at screening and 662 were removed at full-text review. This left 20 studies (published between 2009 and 2020) that examined the effectiveness, safety and acceptability of deprescribing in adults (aged ≥ 50 years) with polypharmacy (five or more prescribed medications) and multimorbidity (two or more conditions). Our analysis revealed that deprescribing under research conditions mapped well to expert guidance on the steps needed for good clinical practice. Our findings offer evidence-informed support to clinicians regarding the safety, clinician acceptability and potential effectiveness of clinical decision-making that demonstrates a structured approach to deprescribing decisions. Our realist review identified 2602 studies with 119 included in the final analysis. The analysis outlined 34 context-mechanism-outcome configurations describing the knowledge work of tailored prescribing under eight headings related to organisational, health-care professional and patient factors, and interventions to improve deprescribing. We conclude that robust tailored deprescribing requires attention to providing an enabling infrastructure, access to data, tailored explanations and trust.

Strict application of our definition of multimorbidity during the scoping review may have had an impact on the relevance of the review to clinical practice. The realist review was limited by the data (evidence) available.

Our combined reviews recognise deprescribing as a complex intervention and provide support for the safety of structured approaches to deprescribing, but also highlight the need to integrate patient-centred and contextual factors into best practice models.

The TAILOR study has informed new funded research tackling deprescribing in sleep management, and professional education. Further research is being developed to implement tailored prescribing into routine primary care practice.

This study is registered as PROSPERO CRD42018107544 and PROSPERO CRD42018104176.

This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 26, No. 32. See the NIHR Journals Library website for further project information.