Effect of patient-facility language discordance on potentially inappropriate prescribing of antipsychotics in long-term care

Context: Appropriate use of medication is a key indicator of the quality of care provided in long-term care (LTC).

Objective: To determine whether resident-facility language concordance/discordance is associated with the odds of potentially inappropriate prescribing of antipsychotics (PIP-AP) in LTC.

Study Design and Analysis: A population-based, retrospective cohort study. The association between linguistic factors and PIP-AP was assessed using adjusted multivariable logistic regression analysis.

Dataset: Health administrative databases housed at ICES, Ontario’s data steward

Population Studied: Long-term care (LTC) residents in Ontario, Canada from 2010 to 2019

Intervention/Instrument: Use of antipsychotic medications, assessed using STOPP-START criteria

Outcome Measures: We obtained resident language from standardized resident assessments, and derived facility language by determining the proportion of residents belonging to each linguistic group within individual LTC homes. Using linked administrative databases, we identified all instances of PIP-AP according to the STOPP-START criteria, which have previously been shown to predict adverse clinical events such as ED visits and hospitalizations. Residents were followed for 1 year or to date of death, whichever occurred first.

Results: We identified 198,729 LTC residents consisting of 162,814 Anglophones (81.9%), 6,230 Francophones (3.1%), and 29,685 Allophones (14.9%). The odds of PIP-AP were higher for both Francophones (aOR 1.15, 95% CI 1.08–1.23) and Allophones (aOR 1.11, 95% CI 1.08–1.15) when compared to Anglophones. When compared to English LTC homes, French LTC homes had greater odds of PIP-AP (aOR 1.12, 95% CI 1.05–1.20), while Allophone homes had lower odds of PIP-AP (aOR 0.82, 95% CI 0.77–0.86). Residents living in language-discordant LTC homes had higher odds of PIP-AP when compared to LTC residents living in language-concordant LTC homes (aOR 1.07, 95% CI 1.04–1.10).

Conclusions: This study identified linguistic factors related to the odds of potentially inappropriate prescribing of antipsychotics (PIP-AP) in LTC, suggesting that the linguistic environment may have an impact on the quality of care provided to residents.