Systematic reviews were the perceived most important source of information for updating a potentially inappropriate medication list for the elderly: An online survey

Polypharmacy in the elderly is a recognized public health problem. For many diseases several drugs are available. Thus, it is important to rule out drugs with an unfavorable risk-benefit profile in older patients, and to suggest safer alternatives. This is often done by developing and publishing lists of potentially inappropriate medications (PIM). Making evidence-based decisions for the elderly is often challenging as randomized controlled trials frequently do not include older adults with multimorbidity and are not sufficiently powered for analyzing harms. Despitemany PIM lists having been published, there is no methodological guidance on how they should be developed, what we recognized while preparing our update of the German PRISCUS list.1 Most PIM lists use the Delphi method to find consensus.2,3 When using a Delphi study to develop a PIM list it is crucial to consider what information will be provided to the participants of the Delphi study to formulate recommendations. There is no research on what type of information experts need or prefer to support rating of PIM criteria or derive recommendations for PIMs. Our aim was to investigate the process of providing information and the importance of different sources of information in the context of developing a PIM list. Our study was carried out as part of updating the Potentially inappropriate medications list in the elderly in Germany (henceforth termed “the PRISCUS list”).1 This included a three-round Delphi process to derive the list of PIMs. In total, 95 experts from clinical practice and research from general medicine, geriatrics, clinical pharmacy, pharmacology, psychiatry, palliativemedicine, and internal medicine evaluated whether selected substances are PIM for the elderly. Due to limited resources, we were only able to conduct 13 de novo SRs. Altogether, 187 substances were classed as PIM.4 We conducted a closed follow-up survey of all experts who took part in the assessment of PIM for the update of the PRISCUS list. The surveywas administered inGerman via SurveyMonkey (https://de. surveymonkey.com/).Wesentout an invitationemail including ahyperlink to all expertswho took part in the assessment for the update of the PRISUCS list. The follow-up survey took place between May and June 2021. One reminder was sent via email to all experts. Participation was voluntary and anonymous. We did not collect any personal or identifying information. The survey included 10 closed


Systematic reviews were the perceived most important source of information for updating a potentially inappropriate medication list for the elderly: An online survey
Polypharmacy in the elderly is a recognized public health problem. For many diseases several drugs are available. Thus, it is important to rule out drugs with an unfavorable risk-benefit profile in older patients, and to suggest safer alternatives. This is often done by developing and publishing lists of potentially inappropriate medications (PIM).
Making evidence-based decisions for the elderly is often challenging as randomized controlled trials frequently do not include older adults with multimorbidity and are not sufficiently powered for analyzing harms. Despite many PIM lists having been published, there is no methodological guidance on how they should be developed, what we recognized while preparing our update of the German PRISCUS list. 1 Most PIM lists use the Delphi method to find consensus. 2,3 When using a Delphi study to develop a PIM list it is crucial to consider what information will be provided to the participants of the Delphi study to formulate recommendations. There is no research on what type of information experts need or prefer to support rating of PIM criteria or derive recommendations for PIMs. Our aim was to investigate the process of providing information and the importance of different sources of information in the context of developing a PIM list.
Our study was carried out as part of updating the Potentially inappropriate medications list in the elderly in Germany (henceforth termed "the PRISCUS list"). 1 This included a three-round Delphi process to derive the list of PIMs. In total, 95 experts from clinical practice and research from general medicine, geriatrics, clinical pharmacy, pharmacology, psychiatry, palliative medicine, and internal medicine evaluated whether selected substances are PIM for the elderly. Due to limited resources, we were only able to conduct 13 de novo SRs. Altogether, 187 substances were classed as PIM. 4 We conducted a closed follow-up survey of all experts who took part in the assessment of PIM for the update of the PRISCUS list.
The survey was administered in German via SurveyMonkey (https://de. surveymonkey.com/). We sent out an invitation email including a hyperlink to all experts who took part in the assessment for the update of the  list. This is a pity as we found the usual GRADE criteria to be in need of modifications to meet the purpose of developing a PIM list. 6 The second most valued information was about anticholinergic effects, which is a specific indicator for potential side effects affecting cognition and resulting in a diverse range of symptoms all of them of particular relevance for older adults. 9,10 Thus, it is not surprising, that this information was also highly appreciated and is considered to be particularly relevant in addition to SRs.
PIM lists provide recommendations for practice. In this perspective, they are very similar to clinical practice guidelines (CPGs). However, CPGs tend to be based on SRs including GRADE assessments. Transparency is called for when developing CPGs to ensure trustworthiness.
Along with the recommendations all details such as literature reviews and GRADE assessments are usually reported. We do not see any reason why this should not also be the case for future PIM lists because they are very similar in their intended use. This appears particularly important because of their high impact on practice and ability to reduce the number of PIMs prescribed. [11][12][13] Overall, the respondents were very positive how the information was presented to them. Nevertheless, there still seems to be room for improvement. There is no knowledge how experts derive decisions based on the information provided to them. The clinical expertise of the experts is also of importance and plays a major role. There was no one participating in the survey who fully agreed that they would have made the same decisions without the provided information. Understanding GRADE SoF tables can be challenging for readers. Thus, ways of presenting SoF tables are also subject of research. 14,15 Our results might be limited by a small number of respondents.