Quantifying Problematic Prescribing Cascades Relevant for Clinical Practice with Prescription Sequence Symmetry Analysis

Background: A prescribing cascade occurs when medication causes an adverse drug reaction (ADR) leading to the prescription of additional medication. A prescribing cascade is problematic when the benefits of the prescribing cascade do not outweigh the risks. An overview of problematic prescribing cascades that need to be addressed in clinical practice is currently lacking.

Objectives: To identify and quantify problematic prescribing cascades relevant for clinical practice.

Methods:

Design: A mixed-methods study was conducted in which prescribing cascades were identified from the literature and grouped to create distinct prescribing cascades. An expert panel assessed in a structured consensus procedure which of these were considered problematic. Next, a cohort study was performed to quantify problematic prescribing cascades using prescription sequence symmetry analysis (PSSA).
Setting: Prescribing cascades identified from the literature were included when the ADR could be confirmed in Micromedex. The expert panel consisted of 16 pharmacists and physicians from Dutch primary and secondary care. Dispensing data for adult patients were retrieved from January 2015 to December 2020 from a national database including 30% of community pharmacies covering different regions in The Netherlands. To identify incident users a washout window of 12 months was applied.
Exposure: Incident users being dispensed the index and marker medication of a prescribing cascade were included. The exposure window for identification was set at 12 months.
Main outcome measure: Prescribing cascades with sequence ratio adjusted for temporal prescribing trends (aSR) significantly larger than 1.
Statistical analysis: PSSA was used to calculate the aSRs. A negative control was included.

Results: Seventy-six distinct prescribing cascades were identified in literature and experts provided 3 additional prescribing cascades. Of these, 66 (83.5%) were considered problematic. A significant aSR >1 was found for 41 (62.1%) problematic prescribing cascades. The highest aSR was found for amiodarone potentially causing hypothyroidism treated with thyroid hormones (aSR 4.63 [95% confidence interval 4.40-4.85]), based on 565 incident users. The prescribing cascade with a significant aSR and the largest number of incident users (n=34,645) was angiotensin converting enzyme-inhibitors potentially causing urinary tract infections treated with antibiotics. The negative control showed an insignificant aSR.

Conclusions: A significant positive aSR was found for 41 of 66 problematic prescribing cascades based on community pharmacy dispensing data in The Netherlands. These results can support healthcare providers in identifying problematic prescribing cascades in clinical practice.