Key points are not available for this paper at this time.
Abstract Background Prescribing cascades, where one medication is used to treat/prevent the adverse effects of another, have been the subject of increased research focus. However, few studies have confirmed potential prescribing cascades identified via administrative or dispensed medicines records with general practice patient record data. This study examined the incidence of ThinkCascades, a list of nine prescribing cascades of clinical relevance in older adults developed via international expert consensus, using general practice data in The Netherlands. Methods A retrospective cohort study examined prescribing records for adults aged ≥ 65 years captured within the FaMe-Net general practice database in The Netherlands for the period 2011–2021. The primary exposures were incident use of Drug A within each ThinkCascades dyad, with the primary outcomes defined as incident use of the corresponding Drug B within 365 days. Independent clinical review of identified potential prescribing cascades was conducted by a general practitioner and pharmacist using an approach consistent with methods applied in earlier prescribing cascade research. Results The eligible cohort comprised 710 incident users of any ThinkCascades Drug A. Their mean age was 75.8 (SD = 6.1) years; 53.5% (n = 380) were female. Overall, 21 ThinkCascades dyads were identified in 18 participants, representing a one-year incidence proportion of 2.5% (N = 710). Only six of nine ThinkCascades were identified amongst study participants, most commonly the calcium channel blocker to diuretic prescribing cascade. Just over one-quarter of identified potential prescribing cascades (6/21; 28.6%) were supported as true prescribing cascades based on independent clinical record review. True cascade likelihood was indeterminable for three cases. Conclusions Prescribing cascades, defined by ThinkCascades, were relatively uncommon over the eight-year study of older people attending Dutch general practice. Three of nine ThinkCascades were not identified. Only one in four identified prescribing cascades had evidence supporting a true prescribing cascade following independent clinical record review. Further research to characterise the prevalence of confirmed prescribing cascades is required. Recent research recommends expanding the number of prescribing cascade dyads which may impact identification rates. Tools to support prescribing cascade awareness, identification and deprescribing need to incorporate shared decision-making with patients and demonstrate clinical utility in supporting medication reviews in primary care.
Doherty et al. (Tue,) studied this question.