LAUSR

Every 3 years, the American Geriatrics Society updates its Beers Criteria, a list of medications that are potentially inappropriate to prescribe to older adults. The list includes medications that can be harmful to older adults and medications that interact with other medications and/or common comorbidities in this population. The use of potentially inappropriate medications (PIMs) is associated with an increased risk of falls, cognitive decline, frailty, hospitalizations, and mortality.2–4 Examples of PIMs include anticholinergics, antipsychotics, benzodiazepines, and opioids. Despite widespread familiarity with the Beers Criteria in clinical settings, the proportion of older adults in the United States who are prescribed PIMs is increasing, from 57% in 1999 to 94% in 2017. Since 2009, the Centers for Medicare and Medicaid Services (CMS) have taken steps to discourage prescription of PIMs to older patients. Medicare prescription drug plans are held accountable to several quality metrics that assess the proportion of patients who are prescribed certain PIMs, including anticholinergics, benzodiazepines, and opioids. One of these measures, for example, “Concurrent Use of Opioids and Benzodiazepines,” assesses the percentage of insured adults who have had concurrent prescriptions for opioids and benzodiazepines for ≥30 cumulative days, excluding patients who have cancer, sickle cell disease, or are in hospice. This measure seeks to address evidence that when opioids and benzodiazepines are used concurrently, they increase the risk of overdose, hospitalizations, and death.6–9 Overdose mortality rates among concurrent users of opioids and benzodiazepines are estimated to be 10 times compared with opioid-only users. About 3% of older adults have concurrent prescriptions for opioids and benzodiazepines. This population may face even greater risks from the concurrent use of opioids and benzodiazepines, since both medications increase the risk of cognitive impairment, delirium, and falls. There is increasing evidence that PIMs are prescribed differentially based on race, ethnicity, and socioeconomic status, including from the accompanying article by Niznik and colleagues.11–15 While the CMS quality metrics are a good first step in encouraging clinical teams to avoid PIMs, we propose that any PIM deprescribing policies and interventions must be designed to promote equity. Pharmacoequity is a term that refers to equity in access to high-quality pharmacotherapy, regardless of race, sex, age, and other socioeconomic factors. The pharmacoequity literature so far has highlighted the importance of ensuring equitable access to guidelineconcordant therapies, lower cost equivalent medications when available, and clinical trials.16–18 Pharmacoequity should also involve equity in access to interventions that aim to deprescribe PIMs, or avoid them altogether. Three recent studies illustrate the state of the current literature on disparities in PIM prescribing in the United States. Hwang and colleagues evaluated 4927 adults aged 66–90 in the Atherosclerosis Risk in Communities Study and found that patients with low cumulative socioeconomic status had 1.66 higher odds (95% confidence interval [CI] 1.02–2.71) of PIM use compared with people with high cumulative socioeconomic status. Cumulative socioeconomic status was defined as a combination of area deprivation index, household income, and education level. Lim and Jung used data from the 2012–2015 Medical Expenditure Panel Surveys to evaluate potentially inappropriate psychotropic medication prescriptions among 7616 older adults, finding that prescription rates were higher among Hispanic patients compared with non-Hispanic White and non-Hispanic Black patients (odds ratio [OR] 1.22, 95% CI 1.02–1.46). Jungo and colleagues evaluated PIM prescriptions among older adults with multiple chronic conditions who were treated at seven Massachusetts hospitals. In 2014, PIMs were more likely to be prescribed to female patients (OR 1.27 vs. male, 95% CI 1.25–1.30) and Hispanic patients (OR 1.41 vs. non-Hispanic, 95% CI 1.27– 1.56). PIM prescribing was less likely among Black patients (OR 0.87 vs. White, 95% CI 0.83–0.91). While all of these analyses adjusted for patient comorbidities, it is important to note that medication indication is rarely recorded in prescription data. It can thus be difficult to accurately determine from claims or electronic health record data whether a PIM prescription was appropriate or inappropriate. As clinicians and care managers attempt to reduce use of benzodiazepines, opioids, and other PIMs, they This editorial comments on the article by Niznik et al. in this issue. Received: 2 March 2023 Accepted: 19 March 2023