443 Selection of Outcome Measurement Instruments for a Core Outcome Set for trials aimed at improving appropriate polypharmacy in older people in primary care

Abstract

Abstract Introduction Different outcome measurement instruments (OMIs) have been used to measure similar outcomes in intervention studies focusing on improving appropriate polypharmacy in older people (1). It is crucial to reach consensus on which OMIs should be used for each outcome in a core outcome set (COS) to promote synthesis of trial findings and help identify effective interventions. A polypharmacy COS was previously developed, consisting of seven outcomes, including serious adverse drug reactions (ADRs), medication appropriateness, falls, medication regimen complexity, quality of life, mortality and medication side effects (1). To date, there has been no consensus as to which OMIs should be used for these outcomes. Aim To select OMIs for the seven outcomes in the COS developed to improve appropriate polypharmacy in older people in primary care. Methods This study followed the COSMIN (Consensus-based Standards for the selection of health Measurement Instruments) initiative’s guidelines for selecting OMIs for a COS (2), which include finding existing OMIs from a systematic review, conducting a quality assessment of the OMIs, selecting one OMI for each outcome in the COS, and running a Delphi consensus survey (two rounds) on OMIs for each outcome. For the Delphi exercise, researchers focusing on care of older people or clinicians who had experience in prescribing for this population in primary care, together with public advocates for older people were invited by email to participate. Following provision of consent, a questionnaire was developed and distributed by the SoGoSurvey® platform on two occasions. The two rounds were distributed on June 21st 2022 and July 11th 2022, respectively. Each round remained open for 14 days to allow participants to take part in the study. The questionnaire comprised a list of seven OMIs along with their definitions and illustrative examples. Stakeholders were asked to choose one option (‘agree’, ‘disagree’ or ‘unsure’) for every OMI. Response rate and distribution of scores for the seven OMIs were calculated following each round. Consensus for the inclusion of an OMI in the COS was reached if 70% or more participants selected ‘agree’ and 15% or less participants selected ‘disagree’. Results One hundred and eighty-eight emails were sent to relevant stakeholders (176 researchers or clinicians, 12 public representatives) inviting them to take part in the study. Fifty-seven (30.32%) stakeholders (56 researchers and one public representative) agreed to participate. Fifty (87.72%; all participants were researchers or clinicians) completed Rounds 1 and 2 of the Delphi survey. Following Rounds 1 and 2, consensus (≥70% participants agreed) was reached for three OMIs – ‘the number of ADRs’ (98%), ‘the number of deaths’ (76%) and ‘the number of patients who fell’ (70%) – to measure the ‘serious ADRs’, ‘mortality’ and ‘falls’ outcomes, respectively. Consensus was not reached for the remaining four OMIs. Conclusion This study led to the selection of OMIs for outcomes in a COS aimed at improving appropriate polypharmacy in older people. Although this study planned to include public participants, none agreed to participate. Future work should determine the most appropriate OMIs for the other four outcomes in the COS.