Changes in the initiation of antipsychotics and trazodone over time: a cohort study of new admissions to nursing homes in Ontario, Canada

Objectives — To investigate whether trazodone is being initiated in lieu of antipsychotics following antipsychotic reduction efforts, this study described changes in medication initiation over time.

Methods — We conducted a retrospective cohort study of new admissions to nursing homes in Ontario, Canada between April 2010 and December 2019 using health administrative data (N=61,068). The initiation of antipsychotic and trazodone use was compared by year of admission using discrete time survival analysis and stratified by history of dementia.

Results — Relative to residents admitted in 2014, antipsychotic initiation significantly decreased in later years (e.g., 2017 admission year hazard odds ratio [HOR₂₀₁₇]=0.72 [95% confidence interval (95%CI)=0.62-0.82]) while trazodone initiation modestly increased (e.g., HOR₂₀₁₇=1.09 [95%CI=0.98-1.21]). The relative increase in trazodone initiation was larger among residents with dementia (e.g., HOR_2017Dem =1.22 [95%CI=1.07-1.39]).

Conclusions — Differences in which medications were started following nursing home admission were observed and suggest trazodone may be initiated in lieu of antipsychotics.