A Systematic Review of Published Algorithms for Selecting an Inhaled Delivery System in COPD.

RATIONALE Medication for treatment of COPD is available in many different delivery systems; however, national and international guidelines do not provide recommendations on how to select the optimal system for an individual patient. OBJECTIVES To perform a systematic review of published algorithms for inhaler selection in out-patients with COPD. METHODS PubMed, EMBASE, PsycINFO, Cochrane, and Google Scholar were search for articles on inhaler selection published between January 1, 1990 and March 10, 2021. The results were reviewed for articles containing an algorithm for inhaler selection. The quality of publications containing an algorithm was assessed using the JBI SUMARI text and opinion critical appraisal checklist. Individual steps recommended in the algorithms and the order in which they were considered were extracted independently by the two authors using the JBI text and opinion data extraction tool. Textual syntheses and a table of factors included were used to appraise and compare algorithms. RESULTS The search identified 1016 publications. After removing duplicate studies (n = 409), 607 abstracts were examined. Nine different algorithms or hierarchical recommendations for device selection were identified. All nine publications were considered of good quality. Most algorithms contain only a few decision steps. There were significant differences between the algorithms. None of the algorithms have been validated. Three domains for factors included in the algorithms were identified: patient factors, device attributes, and HCP factors. Patient factors were considered most frequently (19 times) compared with device attributes (10 times) and HCP factors (7 times). Five specific attribute/factors with at least three rankings in different algorithms, were identified as key factors for device selection. CONCLUSION Although the algorithms generally provide step-by-step approaches based on a literature review and/or the experiences of the different authors, none were developed using item generation/reduction methodology nor included input from patients with COPD. There were considerable differences between the algorithms; however, the review identified key factors that should be considered by HCPs when selecting therapy. Registration: PROSPERO (CRD42021244475).