Table 2.

Effects of asynchronous e-visits on clinical outcomes, quality, utilization, and costs.

Outcome type and citations		Specific measures	Impact
Clinical outcomes
	Hawes et al [44]	Abnormal international normalized ratio	E-visits were associated with lower abnormal international normalized ratios than in-person visits (P<.05).
	Hawes et al [44]	HbA1ca	E-visits were associated with lower HbA1c values than in-person visits (P<.001).
	Hawes et al [44]	Amount of diabetic patients with controlled blood pressure	Compared to the preimplementation period, more diabetic patients were observed with controlled blood pressure in the postimplementation period (P<.001).
	Hawes et al [44]	Amount of diabetic patients with HbA1c levels of less than 8%	Compared to the preimplementation period, more diabetic patients with HbA1c levels of less than 8% were observed in the postimplementation period (P<.0001).
	Hawes et al [44]	Amount of diabetic patients with HbA1c levels of less than 7%	Compared to the preimplementation period, more diabetic patients with HbA1c levels of less than 7% were observed in the postimplementation period (P<.001).
	Levine et al [45]	Systolic blood pressure	Equivalent outcomes
	Watson et al [38]	Total inflammatory lesion counts	Equivalent outcomes
	Watson et al [38]	Frontal inflammatory lesion counts	Equivalent outcomes
	Watson et al [38]	Leeds score	Equivalent outcomes
	Penza et al [47,51]	Mortality rate	Penza et al reported only descriptive statistics in both studies, so it is unclear if there are differences in mortality rates between e-visits and in-person visits.
	Murray et al [50]; Penza et al [51]	Hospitalizations	Murray et al [50] and Penza et al [51] reported only descriptive statistics, so it is unclear if there are differences in the number of related hospitalizations between e-visits and in-person visits.
	Murray et al [50]	Antibiotic retreatment rate	Equivalent outcomes
	Yokose et al [31]	Proportion of patients serum urate levels of less than 6.0 mg/dL	E-visits had greater proportions of patients with optimal control of serum urate levels when compared to in-person visits (P<.01).
Quality of care
	Hawes et al [44]	Amount of diabetic patients receiving aspirin, if clinically indicated	Equivalent outcomes
	Hawes et al [44]	Amount of diabetic patients receiving moderate-intensity statins	Equivalent outcomes
	Hawes et al [44]	Amount of diabetic patients receiving high-intensity statins	Equivalent outcomes
	Heyworth et al [42]	Medication discrepancy discovery rate	It is unclear what the impact is on the rate of discovering medication discrepancies as no P value was reported.
	Mehrotra et al [24]	Order rate of diagnostic test	E-visits had a lower order rate of diagnostic tests when compared to in-person visits (P<.001).
	Mehrotra et al [24]	Order rate of preventive care services	E-visits had a lower order rate of preventive care services when compared to in-person visits (P<.01).
	Yokose et al [31]	Rate that serum urate levels were checked	E-visits had more frequent checks of serum urate levels when compared to in-person visits (P<.05).
	Murray et al [50]; Penza et al [46,51]; Mehrotra et al [24]; Courneya et al [40]	Antibiotic prescribing rate	Mehrotra et al [24] reported that the rate of prescribing antibiotics was higher during e-visits than in-person visits for sinusitis (P<.001) but not for e-visits for urinary tract infections. However, Penza et al [46] and Murray et al [50] saw equivalent outcomes. Penza et al [51] reported e-visits had lower antibiotic prescribing rates than in-person visits (P<.001). Courneya et al [40] also investigated the association but did not report a P value, so the impact on antibiotic prescribing rate is unclear.
Health care utilization
	Levine et al [45]	Overall primary care visit utilization	Equivalent outcomes
	Levine et al [45]	Overall specialist visit utilization	Equivalent outcomes
	Levine et al [45]	Overall emergency department utilization	Equivalent outcomes
	Levine et al [45]	Overall inpatient admissions	Equivalent outcomes
	Rajda et al [49]	Number of specialist procedures done 60- and 90-days after initial consultation	E-visits were associated with a lower number of specialist procedures performed 60 and 90 days after an initial consultation when compared to in-person visits (P<.01)
	Murray et al [50]; Penza et al [51]	30-day follow-up rate (planned and unplanned)	Equivalent outcomes
	Penza et al [46]; Pathipati et al [43]; Albert et al [39]; Player et al [48]; Adamson et al [19]	Rate of patients who need planned follow-up visits	Penza et al [46] reported that e-visits were associated with higher rates of planned follow-up visits than in-person visits (P<.001). Pathipathi et al [43], Albert et al [39], Player et al [48], and Adamson et al [19] reported only the proportion of e-visits that required follow-up visits, so it is unclear what the association of e-visit usage and rate of planned follow-up visits is.
	Penza et al [47]; Mehrotra et al [24]; Courneya et al [40]; North et al [41]; Hertzog et al [30]	Unexpected follow-up encounter rate after initial encounter	Hertzog et al [30] reported that e-visits were associated with higher unexpected follow-up rates when compared to in-person visits (P<.05). However, Mehrotra et al [24], Courneya et al [40], and North et al [41] found equivalent outcomes. Penza et al [47] reported only descriptive statistics, so it is unclear if there are differences in unexpected follow-up encounter rates between e-visits and in-person visits.
Health care costs
	Rajda et al [49]; Courneya et al [40]; Rohrer et al [37]	Treatment costs	Courneya et al [40] and Rajda et al [49] reported e-visits were associated with lower treatment costs (P<.001). Rohrer et al [37] reported a lower median of costs associated with e-visits than in-person visits (P<.01).

^aHbA_1c: hemoglobin A_1c.