Frequency of Modification of Pharmacological Treatment Is Equivalent for Virtual and In-Person Psychiatric Visits

Background: During the coronavirus pandemic there was a rapid adoption of telehealth services in psychiatry, which now accounts for 40% of all visits. There is a dearth of information about the relative efficacy of virtual and in-person psychiatric evaluations. Methods: We examined the rate of medication changes during virtual and in-person visits as a proxy for the equivalence of clinical decision-making. Results: A total of 280 visits among 173 patients were evaluated. The majority of these visits were telehealth (224, 80%). There were 96 medication changes among the telehealth visits (42.8%) and 21 among the in-person visits (37.5%) (z = −1.4, p = 0.16). Conclusion: Clinicians were equally as likely to order a medication change if they saw their patient virtually or in person. This suggests that remote assessments yielded similar conclusions to in-person assessments.


Introduction
In December 2019, visitors to Huanan and Yangchahu markets in Wuhan and racoon dogs being sold there developed an illness that was subsequently determined to be caused by the severe acute respiratory syndrome-related coronavirus. 1 In the spring of 2020, the World Health Organization declared that the coronavirus 2019 (CoViD-19) outbreak was a pandemic. 2 As the rate of CoViD-19 infections, severe illnesses, and deaths increased, many countries, including the United States, began declaring shutdowns and lockdowns. 3 These shutdowns included nonemergency medical services. At that time only a fraction of states allowed parity of telehealth and in-person visits, 3 and its use was predominantly limited to rural settings and about 8% of medical provider visits. 4 By the spring of 2021, about one quarter of Americans were receiving telehealth services, 5 a 200% increase.
Even before the CoViD-19 lockdown, there had been a call to digitalize psychiatric outpatient care. 6 Nonetheless, psychiatrists had been concerned that use of  virtual settings would lead to deterioration of the doctor-patient relationship, which many believe is central to the practice of psychiatry. [7][8][9] Pre-CoViD-19 studies comparing in-person care with telepsychiatry generally showed that remote treatment was comparable or superior to in-person treatment. 10 Patient satisfaction was generally good, 10 but some patients felt a lesser connection to their clinician. 11 Provider satisfaction is more mixed with concerns about quality of services, patient perceptions, and technological issues. 10 Throughout CoViD-19 restrictions, clinician satisfaction with telepsychiatry effected adoption rates when an option was available. 12,13 However, participants in pre-CoViD-19 studies chose to be involved in a telehealth study. Patients and clinicians through the CoViD-19 shutdowns had telehealth ''forced'' upon them. That difference is a potentially important variable.
We performed a quality assurance study examining the objective outcome of medication intervention comparing individuals who attended the same clinic remotely or in person. The study was performed after the shutdown restrictions were slowly being lifted in 2021, and some patients were transitioning back to in-person visits.

Methods
The study was a quality assessment of visits at the University of Louisville outpatient psychiatric service. With the switch to virtual visits we needed to examine a quality measure to ensure that we were providing adequate care to our patients. We examined the rate of medication changes during virtual and in-person visits as a proxy for the equivalence of clinical decisionmaking because this method was utilized previously. 14 All patients were ‡18 years. We identified times when patients came to two consecutive visits with the same provider and the same primary diagnosis and evaluated treatment changes on the second visit. We documented when there was a change in prescribed medication or dosage but excluded ''as needed'' (PRN or Pro Re Nata) medications. A test evaluating differences in proportions was used. 15 Because this was a quality assessment study, it did not require evaluation or approval by the Human Subjects Protections Program of the University of Louisville.

Results
The study period spanned all of 2021. A total of 280 visits among 173 patients were evaluated (Table 1). There were some 20 providers in the clinic, all of whom did both in-person and virtual visits; we corrected for this by only examining patients who came to the same provider at least twice in a row. The majority of these visits were telehealth (224, 80%) ( Table 1). There were 115 women (66.5%), 57 men (33%), and 1 unspecified gender. The age range was 19-89 years (Median age 43 years; mean age 45.4 years; interquartile range 24 years); the majority of patients (98 or 56.6%) were under the age of 50 years.

Discussion
We performed this study because there is a dearth of information regarding the relative efficacy of telehealth visits when patients do not choose that modality. We examined medication change as a proxy for clinical decision-making by psychiatrists in an academic outpatient clinic. Previous study had suggested that this was a reasonable measure of clinical decision-making. 14 In our quality assurance examination, clinicians, were equally likely to act on their collected clinical data by ordering a medication change if they saw their patient virtually or in person. This suggests that remote assessments yielded similar conclusions to in-person assessments, even in patients who do not voluntarily participate in a telehealth study.
It is important to note that this study does not address the quality of care provided or health care outcomes. We did not measure patient satisfaction and did not collect any correlates of clinical outcomes. We utilized medication change as a measure because it is driven by patient complaints or clinician observations during patients' presentations. 16 Thus, it reflected the quality of the transfer of information during the session. The lack of significant difference suggests that the quality of information transfer is equivalent, or nearly so, in the two forms of evaluation.
Telepsychiatry has grown considerably. During the height of the CoViD-19 restrictions, 40% of all outpatient visits were provided virtually by mental health providers, compared with only 11% by other providers. 17 Even as CoViD-19 restrictions have been largely removed, and telepsychiatry continues to be highly utilized, accounting for 36% of all outpatient visits and 39% of all telehealth services. 17 There are clear limitations to our study design. As noted earlier, we did not examine the actual quality of outcome. Furthermore, this was not a randomized study. Nearly all patients seen virtually did not have a choice of how the evaluation would be done, but all patients seen in person chose that format. Furthermore, it is possible that patients who would not do well with virtual visits simply dropped out of treatment during CoViD-19 restrictions and were not studied in our population.
Telepsychiatry will continue to be a growing presence in future health care. Additional studies need to be done to confirm the equivalence of remote versus in-person outcomes. Nonetheless, the early results of our study and the current literature suggest that using telehealth for mental health conditions is a reasonable option.