COVID-19 and the promise of digital psychiatry

COVID-19 and the promise of digital psychiatry

David Karol and Patrick Triplett

 The practice of C-L psychiatry requires an ability to adapt to evolving circumstances. The COVID-19 pandemic is challenging C-L psychiatrists to care for patients in ways we have scarcely seen before.  In the earliest phase of the pandemic in the US, shortages of available personal protective equipment limited the frequency or even advisability of in-person psychiatric consultation in a variety of settings, particularly if practical electronic alternatives were available. Some hospitals had invested in the infrastructure for telepsychiatry and likely found the transition at this earliest phase less problematic. Many others had not. Through Listservs, special interest groups, and other means of communication, C-L psychiatrists shared their experiences, good and bad, with various electronic means of caring for patients.

The realities of digital technology have not always lived up to their lofty promises, at least not yet. In inpatient settings, challenges have included limited access to devices, the limited availability of personnel to assist with holding and operating the device (as when patients are too confused or too weak to do so themselves), and difficulty communicating due to physical barriers, such as plastic coverings over the speaker and microphone. Complications in the outpatient setting have been similar and include lack of internet access or appropriate devices, poor or failed connections, or insufficient technical savvy to navigate telehealth. Despite these frustrations, the increasing recognition of neuropsychiatric complications of COVID-19 and the mental health impact of social and physical distancing has led to the realization that we must find ways to return to in-person evaluations safely, often incorporating some degree of electronic support.

On the one hand, the adoption of telehealth by clinicians, patients, and third-party payors has enhanced our ability to provide patient assessments and care. On the other hand, telehealth has limitations and introduces unique considerations. In the end, we learned the essential nature of our services, the power of physical presence, and our need to share the sacred space with our patients. The pandemic has also reinforced what we already knew—that C-L psychiatrists must be ready for anything.

These latest adaptations in delivering psychiatric care in the age of COVID-19, including the use of technology to connect with patients and colleagues, share many of the aspirations of proactive C-L psychiatry, which aim to reduce barriers to care in a systematic fashion and to support staff on hospital units.

In addition to patients with known and untreated psychiatric conditions on COVID units, case reports of patients with severe agitation, psychosis and other concerning symptoms and behaviors have emerged.  Whereas the neuropsychiatric symptoms unique to COVID-19 infection remain challenging to predict, identifying patients at risk for agitation and aggression due to psychiatric conditions is often far less difficult. The technology we have come to embrace (or, reluctantly allowed into our lives, for some!) can help us in systematic screening and early clinical intervention, two of the cardinal principles of team-based proactive C-L. The electronic health record is a powerful tool currently being studied as a means to improve accuracy and outcomes related to these two principles. For instance, Finn et al demonstrated the feasibility of generating an automated daily report from the electronic health record that includes diagnoses, orders, and nursing care plans to guide decisions regarding proactive psychiatric consultation.1 Additionally, the team at the University of Pennsylvania has been pioneering machine-learning approaches to identify patients with acute psychiatric needs. Similar work on machine learning has sought to understand delirium misdiagnosis in the hospital,2 to predict risk of developing delirium in patients without known cognitive impairment,3 and to predict hospitalization following psychiatric crisis.4

Our field’s increasing comfort with technology and the growing opportunities for easy, secure communication are certain to further the other two cardinal principles of team-based C-L: the interdisciplinary model and integration with primary teams. Throughout the pandemic, many have experienced the pleasant surprise that virtual meetings can be just as productive as, or even more productive than, traditional face-to-face meetings. Having regular meetings is much easier with the acceptability of video or phone conferencing that can be done from almost anywhere (including some team members’ homes), and this essential communication seems less burdensome when travel time, waiting for late arrivers, etc., is less disruptive. In addition, these communications can be more impromptu – in between patients, even – allowing for more frequent and focused connections that supplement regularly scheduled communication. One expects this will lead to improved staff morale and building of mental health competence as ease of access to skilled team members increases.

The timing is right. Whereas telehealth has been available as an option for decades, the COVID-19 pandemic proved that digital technology is essential to deliver psychiatric care when traditional face-to-face evaluation is not possible or poses undue risks. Team-based proactive approaches to C-L psychiatry offer unique value to our field, to our ability to allocate mental health resources more accurately, and to enhance the healthcare our most vulnerable patients receive. Proactive C-L psychiatrists have the opportunity to lead, teach, and inspire others as we learn from our pandemic experiences to shape the future of C-L psychiatry. 


  1. Finn CT, Thakur D, Shea KM, Riblet NBV, Lee B, Heng G, Scott R, Gardner T, Randlett S, LaRock T, Siriwardana N, Green G, Torrey W. Electronic Medical Record Reporting Enhances Proactive Psychiatric Consultation. Psychosomatics 2018; 59:561-566.
  2. Hercus C, Hudaib A. Delirium misdiagnosis risk in psychiatry: a machine learning-logistic regression predictive algorithm. BMC Health Services Research 2020; 20:151.
  3. Wong A, Young AT, Liang AS, Gonzales R, Douglas VC, Hadley D. Development and validation of an electronic health record-based machine learning model to estimate delirium risk in newly hospitalized patients without known cognitive impairment. JAMA Netw Open 2018; 1(4):e181018.
  4. Blankers M, van der Post LFM, Dekker JJM. Predicting hospitalization following psychiatric crisis care using machine learning. BMC Med Inform Decis Mak 2020; 20(1):332.