Telepsychiatry adoption across hospitals in the United States: a cross-sectional study

Over 46 million Americans experienced a mental illness [1]; however, less than half (42.9%) of these individuals received mental health services in a 12-month window, partially due to stigma and limited access to care [1]. Every year, millions of emergency department (ED) visits in the United States (US) involve mental illness and substance use disorders [2]. Recent data from California indicate that nearly 30% patients seen in a ED had a prior mental health diagnosis [3]. The number of ED visits for primarily mental health reasons has increased markedly over past two decades [4]. For US adults under age 45, mental illness is the top reason for hospitalization [5]. In 2016, 7.7 million hospitalizations, accounting for 21.7% of national hospital stays, were attributable to mental and substance use disorders [6]. Linkage to and continuity of mental health care is a challenge, with persistent high suicide rates for mental health patients following hospital discharge [7].

Access to psychiatric care in the US is inequitable across communities due in part to an uneven distribution of psychiatrists, psychologists, and other mental health professionals [8]. A shortage of psychiatrists is more likely to occur in non-metropolitan counties than metropolitan counties; in 2015, nearly 70% of counties in the West North Central Census Division lacked a licensed psychiatrist—a rate over tenfold higher than the New England Census Division where only 6% of counties had no psychiatrist [8]. Amid provider shortages, the national number of psychiatric beds declined from 0.31 to 0.21 beds per 1000 population from 2000 to 2016 [9]. Limited access to psychiatric inpatient care may result in EDs being overcrowded with patients with psychiatric conditions [10], creating challenges for patient safety and increasing health care costs [11, 12]. Meanwhile, lack of or loss to follow-up after psychiatric inpatient discharge can also lead to non-adherence to medications [13], readmission among high-risk patients, and even suicidality [14]. Specifically, the period between psychiatric discharge and follow-up is a particularly critical time for suicidal risk, with high rates of suicides occurring within one week of hospital discharge for psychiatric patients [15]. However, in 2015, only 55.8% of hospitals delivered timely care following psychiatric hospitalizations; psychiatric specialty hospitals (52.9%) and publicly owned hospitals (52.3%) had lower follow-up rates within 30 days post discharge than their non-specialty (57.0%) and private hospital counterparts (59.2%), respectively [16].

Mental health advocates have promoted telepsychiatry as one possible solution to the shortage of mental health professionals [17‐19]. Within a hospital setting, telepsychiatry may enable providers to complete virtual psychiatric evaluations, provide teletherapy, communicate briefly and check-in with patients, and offer patient education [20, 21]. In addition, telepsychiatry may overcome some barriers to care by enabling patients to access mental health services from a private, trusted location. Telepsychiatry may also be an effective way to decrease ED visits for non-life-threatening mental health conditions and to ensure continuity of care for patients after being discharged from inpatient care. These virtual care systems have been proven to be well received by patients, associated with decreased admissions or readmissions to psychiatric hospitals, and cost effective [22]. In the wake of a pandemic like the coronavirus disease 2019 (COVID-19), telepsychiatry becomes a critical tool to cope with the growing needs for mental health care, while allowing for social distancing practices [23, 24].

While telehealth adoption has steadily increased nationwide, telepsychiatry has lagged behind [17, 18]. From 2010 to 2017, the proportion of hospitals with any telehealth capacity doubled from 35 to 76% in the US [25]. Despite well-documented acceptance of telepsychiatry by patients and clinicians and higher perceived efficacy than standard care by patient populations [26‐28], only about 20% of hospital-based EDs had telepsychiatry as of 2016 [29]. In 2013, less than 1% of rural Medicare beneficiaries reported ever having utilized telehealth for mental health services [30]. Telepsychiatry use was also uncommon among commercially insured populations, with only 0.5 telepsychiatry visits per 1000 members per quarter in 2017 [31]. Even among mental health facilities, only about 30% offered telepsychiatry by 2017 [32]. Full implementation of telepsychiatry takes months, even for well-resourced hospitals [21]. As community mental health services have been reduced over the past decades [33], hospital-based services play an increased role in preventing and managing psychiatric crises, especially for patients with severe mental health disorders or being discharged from inpatient care. Assessing the distribution of telepsychiatry adoption across hospital settings is an essential first step to understand the availability of services and to identify areas and settings where gaps exist.

Prior work has demonstrated geographic variations in the availability of mental health resources across counties nationwide [34‐36], and variations in telepsychiatry adoption in mental health facilities [32]. However, to the best of our knowledge, no study has examined geographic distribution of telepsychiatry adoption across US hospitals. Much remains unknown about characteristics of hospitals that choose to adopt telepsychiatry, which is vital to identify where the shortages of telepsychiatry are and where to enact policies to support more rapid telepsychiatry adoption. Therefore, this study aimed to assess the geographic distribution of telepsychiatry across hospitals in the US and investigate characteristics associated with telepsychiatry adoption.

Using nationwide hospital data, this study explored geographic variations in telepsychiatry adoption across US hospitals in 2017, and findings can inform efforts to improve access to psychiatric care and reduce persistent geographic disparities in mental health [17‐19]. Our data indicate that less than one in six (15.8%) hospitals had telepsychiatry as of 2017. This suggests that substantial challenges remain for increasing access to psychiatric services across the US. Although the majority of rural residents live in mental health shortage areas, telepsychiatry was not routinely being used to deliver psychiatric services in these areas. Hospitals in rural noncore areas were far less likely to have adopted telepsychiatry – with only 8% of rural noncore hospitals having telepsychiatry in 2017. Telepsychiatry adoption varied significantly by both hospital- and county-level characteristics, including provision of outpatient psychiatric services, system affiliation, hospital bed size, ownership, ratio of Medicaid inpatient days to total inpatient days, and designation as mental health shortage areas.

Although telehealth has long been advocated as a tool to improve access to care and to facilitate the transition from hospital-based care to community-based care [44], we find that telepsychiatry adoption by hospitals remains very limited. More importantly, our study reveals that hospitals in counties with more psychiatrists did not have higher telepsychiatry adoption rates. Clinical resources that are clustered in certain geographic areas may have little benefit for individuals residing outside of those areas without purposeful, targeted efforts to expand access. This finding may be due to the absence of incentives, a lack of hospital buy-in, and/or limited education or training for psychiatrists to provide telepsychiatry in the hospital settings [27, 45]. In addition, many patients may face barriers to engaging in such services, particularly due to the persistent rural-urban disparities in high speed Internet access [46].

It is concerning that hospitals in rural counties have lower rates of telepsychiatry adoption, especially hospitals in noncore areas. This finding is somewhat in contrast to previous research that showed that mental health facilities in rural noncore areas have greater rates of telepsychiatry than their urban counterparts [32]; however, it is possible that disconnect exists between rural hospitals and rural mental health centers and that administrators of local hospitals have not been sufficiently motivated to expand telepsychiatry. One prior study has revealed that around 40% of individuals who died by suicide had received care within 30 days of their suicide [47]. In addition, poor continuity of care and lack of follow-up for individuals discharged from psychiatric inpatient settings are major issues [7, 48]. Providing follow-up after psychiatric hospitalization discharge has proven useful to reduce risk of non-adherence to medication and suicide [13, 14]. Telepsychiatry may facilitate timely delivery of follow-up care after discharge and make it easier to support patients’ adherence to treatment [16, 17].

Bridging these gaps calls for a wider availability of telepsychiatry to improve continuity of care. Our study reveals that hospitals with inpatient psychiatric services but without outpatient psychiatric services did not report greater adoption of telepsychiatry than their counterparts. This may be related to the per diem prospective payment system for inpatient psychiatric facilities and insufficient payment for outpatient services [49, 50]. Also, telepsychiatry adoption rates vary tenfold by facility operation, with 80% of federal hospitals but only 16% of non-federal public and 8.6% of private non-profit hospitals reporting telepsychiatry. This result is likely due to the significant progress that has been made by the Veteran Affairs system in promoting telehealth [51]. To ensure access to psychiatric care for all, federal and state policymakers should expand the types of providers eligible to receive reimbursements for both live video and remote patient monitoring for patients in need.

Our study demonstrates that hospitals with a greater ratio of Medicaid inpatient days to total inpatient days were more likely to have telepsychiatry. This suggests that federal Medicaid policies could possibly promote telepsychiatry adoption in these hospitals; surprisingly, profit margins were not independent factors associated with telepsychiatry use, even though investing in telehealth systems is perceived as a way to increase the competitive advantage of a hospital [52]. This might be related to decreasing trends in average reimbursement for telepsychiatry [53]. In 2018, over 10 states still did not have parity legislation in place for private insurance coverage of telehealth [54]; these telepsychiatry disparities are likely historically rooted, in part, in regulation and reimbursement policies. Policies to improve access to care through expanded telehealth are evolving quickly as a result of the COVID-19 pandemic; however, the public and private funding sources that will be needed for expanded telehealth remain unclear [55].

With increasing demands for psychiatric services, our findings on lower rates of telepsychiatry adoption in counties designated as mental health professional shortage areas raise concerns about access to care for residents in these already low resource areas. Lack of telepsychiatry adoption in vulnerable communities is likely compounded by the limited supply of mental health professionals to begin with. Without purposeful state and federal efforts to address the inequitable distribution of mental health resources, disparities in access to care are likely to persist. These results call for allocating telepsychiatry funding based on local mental health care need. Otherwise, residents in these counties will be less likely to have access to evidence-based treatments for mental health disorders, and the health disparities affecting the rural US are likely to persist or even worsen.

This study has some limitations. First, the AHA Annual Survey asked about hospital-wide use of telepsychiatry via a single item without querying the extent of use or scope of services offered. About 25% of hospitals did not respond to the item on telepsychiatry. Assuming that hospitals without any telehealth tend not to respond to telehealth questions, the current national rate of telepsychiatry adoption may be overestimated. Second, our cross-sectional data did not allow us to make causal inferences, and we had no data about local psychiatric care needs. Third, our study focused on the telepsychiatry adoption at hospital settings, which include EDs, as well as inpatient and outpatient psychiatric services. We did not include mental health facilities in our analysis though they often provide a broad range of services [56]. Finally, this study documented telepsychiatry availability prior to COVID-19, which might have experienced uptick due to temporary waivers by the Centers for Medicare and Medicaid Services on originating sites for telehealth and the ability of healthcare professionals to prescribe remotely during the COVID-19 pandemic [57]. Future research is warranted to study how these policy waivers impacted telepsychiatry availabilities when nationwide hospital data on telepsychiatry during 2020 are made available.

This study is the first to examine the national geographic distribution of telepsychiatry adoption across US hospitals and the hospital characteristics that were associated with adoption. Significant regional and rural-urban disparities of hospital-based telepsychiatry adoption exist. Understanding the distribution of telepsychiatry adoptions and associated factors is vital to enact targeted policies to improve access to hospital-based inpatient and outpatient psychiatric care for those in need. This study found that factors related to both hospital capacity and external environments were important predictors of telepsychiatry adoption. Our results suggest that rural, isolated, small, and freestanding hospitals face disproportionate difficulties in adopting telepsychiatry. Given well-documented benefits of telepsychiatry, policies and enhanced resources are needed to ensure necessary infrastructure in small and less-resourced hospitals to ensure access to telepsychiatric care among residents, especially those in mental health professional shortage areas.