Are primary care virtual visits associated with higher emergency department use? A cross-sectional analysis from Ontario, Canada

ImportanceThe COVID-19 pandemic has resulted in increased use of virtual care, however, few studies have looked at the association between virtual primary care visits and other healthcare use. ObjectiveTo determine whether there was an association between a high proportion of virtual visits in primary care and more emergency department visits DesignA cross-sectional study, using routinely collected data SettingOntario, Canada ParticipantsOntario residents alive on March 31st 2021 and family physicians with at least 1 visit claim between February and October 2021. ExposureFamily physicians stratified by the percentage of total visits that were virtual (phone or video) between February and October 2021 Main outcome(s) and measure(s)We calculated the emergency department visit rate for each stratum of family physician virtual care use. We used multivariable logistic regression models to understand the relative rate of patient emergency department use after stratifying for rurality and adjusting first for patient characteristics and then the 2019 emergency department visit rate. ResultsWe analyzed data for 15,155 family physicians and 12,951,063 Ontarians attached to these physicians. The mean number of emergency department visits was highest among patients whose physicians provided only in-person care (470.3 {+/-} 1918.8 per 1,000) and was lowest among physicians who provided >80 to <100% care virtually (242.0 {+/-} 800.3 per 1,000). After adjustment for patient characteristics patients seen by physicians with >20% of visits delivered virtually had lower rates of emergency department visits compared to patients of physicians who provided >0%-20% virtually (e.g. >80 to <100% vs >0%-20% virtual visits in Big Cities, Relative Rate (RR) 0.80 [95%CI 0.76-0.83]). This trend held across all rurality strata and after adjustment for 2019 emergency department visit rates. In urban areas, there was a gradient whereby physicians providing the highest level of virtual care had the lowest emergency department visit rates. Conclusions and RelevancePhysicians who provided a high proportion of care virtually did not have higher emergency department visits than those who provided the lowest levels of virtual care. Our findings refute hypotheses that emergency department use is being driven by family physicians providing more care virtually. Key pointsO_ST_ABSQuestionC_ST_ABSDo family physicians who provide more care virtually have higher emergency department visit rates among their patient panel? FindingsIn this cross-sectional study from Ontario, Canada, we examined data from February to October 2021 for 12,951,063 patients attached to 15,155 family doctors and found that physicians who provided a high proportion of virtual care did not have higher emergency department visits than those who provided the lowest levels of virtual care. This finding remained true after adjusting for patient characteristics. MeaningOur findings refute hypotheses that emergency department use is being driven by family physicians providing more care virtually.