(A-140) Prescribing Cascades Potentially Associated with Harms in Long-term Care: A Prescription Sequence Symmetry Analysis

Background: Prescribing cascades occur when a new medicine is prescribed to treat an adverse drug reaction (ADR) and can result in increased risk of harms. The older population, particularly those in long-term care, are at increased risk of prescribing cascades due to polypharmacy, multimorbidity and dementia, which increase the likelihood of ADRs.

Objectives: To quantify prescribing cascades potentially associated with harm following entry to long-term care.

Methods: A retrospective population-based cohort study was conducted using data from the Registry of Senior Australians (ROSA) of all individuals aged 65 years or older who entered long-term care between January 2016 and June 2020. A total of 95 prescribing cascades identified from literature review and expert review of likelihood of harms were included. Prescription sequence symmetry analysis was conducted using national dispensing data of newly initiated index medicines (medicines associated with ADR) and marker medicines (medicines used to treat ADR), with a 12-month exposure period of the index and marker medicine. Incident medicine use was defined as no prior use in the 90 days before entry to long-term care. Adjusted sequence ratios (aSR) adjusting for temporal trends were calculated. Analyses were stratified by the presence of dementia.

Results: 166193 long-term care residents were included (female=60.5%, median age=86 yrs [IQR 10 yrs], median comorbid conditions=5, with dementia=50.4%). Significant prescribing cascades were identified for 28 (29.5%) of the examined cascades. Cascades with the most users were benzodiazepine (index) and antipsychotic (marker) n=6403/4198, aSR 1.52 (95%CI 1.46,1.58); PPI and antibiotic n=5938/4082, aSR 1.43 (95%CI 1.38,1.49); statin and antibiotic n=3315/999, aSR 3.19 (95%CI 2.97,3.43); and ACE-inhibitor and antibacterial n=2429/1720, aSR 1.38 (95%CI 1.30,1.47). The cascades with the largest aSR were PPI and intestinal anti-infectives (aSR 3.68; 95%CI 2.30,5.89), statin and antibiotic (aSR = 3.19; 95%CI 2.97,3.43), and statin and antipsychotic (aSR 3.03; 95%CI 2.66,3.44). Most significant cascades were similar among those with and without dementia, except for PPI and intestinal anti-infectives (with dementia aSR 2.38, 95%CI 1.14,4.96 vs. without dementia aSR 4.41, 95%CI 2.37,8.20) and statin and antipsychotic (with dementia aSR 2.55, 95%CI 2.19,2.97 vs. without dementia aSR 4.43, 95%CI 3.48,5.64).

Conclusions: This national population-based study highlights prescribing cascades likely associated with harms following entry to long-term care, including people with dementia. Healthcare providers should be vigilant in identifying and addressing these cascades as potential deprescribing targets to reduce medication burden and harms.