Assessing incidence of clinically significantly prescribing cascades in older adults: a nationwide retrospective cohort study using the French National Health Data System.

BACKGROUND

Prescribing cascades occur when an adverse drug event is misinterpreted as a new medical condition, leading to the prescription of additional medications. They represent an important and preventable source of potentially inappropriate prescribing in older adults, yet their population-level incidence remains poorly documented in France.

METHODS

We conducted a nationwide retrospective cohort study in 2022 using the French National Health Data System, including all individuals aged ≥65 years in 2022. Clinically important prescribing cascades were identified using a consensus-based list (ThinkCascades). Two complementary approaches were applied: a conservative longitudinal analysis and prescription sequence symmetry analysis. Incidence ranges were estimated, and subgroup analyses were performed to identify factors associated with cascade initiation.

RESULTS

Among 16.6 million older adults, four of seven predefined potential prescribing cascades showed significant positive signals. The strongest associations were observed for benzodiazepine-antipsychotic (adjusted sequence ratio [aSR] 2.48, 99% CI 2.44-2.51) and antipsychotic-antiparkinsonian cascades (aSR 2.01, 99% CI 1.91-2.11). Incidence estimates ranged from 0.69% to 3.29%. Female sex, advanced age (≥85 years), hospital initiation of the index drug and lack of prescriber continuity were associated with higher risks for specific cascades.

CONCLUSIONS

Clinically important potential prescribing cascades are common among older adults in France. These findings suggest potential prescribing cascades as a persistent and preventable contributor to medication-related harm and underscore the importance of systematic medication review, particularly during transitions of care and when new prescriptions shortly follow treatment initiation, as potential signals of prescribing cascades.