Association of low-income subsidy, medicaid dual eligibility, and disability status with high-risk medication use among Medicare Part D beneficiaries

Mapping potentially inappropriate medications in older adults using the Anatomical Therapeutic Chemical (ATC) classification system

BACKGROUND

Potentially inappropriate medications (PIMs) in older adults are medications in which risks often outweigh benefits and are suggested to be avoided. Worldwide, many distinct guidelines and tools classify PIMs in older adults. Collating these guidelines and tools, mapping them to a medication classification system, and creating a crosswalk will enhance the utility of PIM guidance for research and clinical practice.

METHODS

We used the Anatomical Therapeutic Chemical (ATC) Classification System, a hierarchical classification system, to map PIMs from eight distinct guidelines and tools (2019 Beers Criteria, Screening Tool for Older Person's Appropriate Prescriptions [STOPP], STOPP-Japan, German PRISCUS, European Union-7 Potentially Inappropriate Medication [PIM] list, Centers for Medicare & Medicaid Services [CMS] High-Risk Medication, Anticholinergic Burden Scale, and Drug Burden Index). Each PIM was mapped to ATC Level 5 (drug) and to ATC Level 4 (drug class). We then used the crosswalk (1) to compare PIMs and PIM drug classes across guidelines and tools to determine the number of PIMs that were index (drug-induced adverse event) or marker (treatment of drug-induced adverse event) drug of prescribing cascades, and (2) estimate the prevalence of PIM use in older adults continuously enrolled with fee-for-service Medicare in 2018 as use cases. Data visualization and descriptive statistics were used to assess guidelines and tools for both use cases.

RESULTS

Out of 480 unique PIMs identified, only three medications-amitriptyline, clomipramine, and imipramine and two drug classes-N06AA (tricyclic antidepressants) and N06AB (selective serotonin reuptake inhibitors), were noted in all eight guidelines and tools. Using the crosswalk, 50% of classes of index drugs and 47% of classes of marker drugs of known prescribing cascades were PIMs. Additionally, 88% of Medicare beneficiaries were dispensed ≥1 PIM across the eight guidelines and tools.

CONCLUSION

We created a crosswalk of eight PIM guidelines and tools to the ATC classification system and created two use cases. Our findings could be used to expand the ease of PIM identification and harmonization for research and clinical practice purposes.