Economic evaluation and costs of telepsychiatry programmes: A systematic review

Abstract

Objective:

Telepsychiatry involves use of telecommunications technology to deliver psychiatric care and offers promise to reduce costs and increase access to mental health services. This systematic review examined cost reporting of telepsychiatry programmes for mental healthcare.

Methods:

We systematically searched electronic databases for studies reporting costs, including economic evaluations such as cost-effectiveness analyses, or costs of developing telepsychiatry programmes for clinical care of mental disorders. Included studies enrolled participants with mental disorders and involved telepsychiatry for depression, anxiety disorders, serious mental illnesses including schizophrenia spectrum disorders and bipolar disorder, post-traumatic stress disorder, dementia or epilepsy.

Results:

Twenty-six unique studies met inclusion criteria (17,967 participants), with most targeting depression (n = 7; 27%), general mental disorders and screening (n = 7; 27%), child mental health (n = 4; 15%) and geriatric mental health (n = 4; 15%). Nearly all studies (n = 25; 96%) compared telepsychiatry programme costs with either standard in-person consultation or usual care, with 15 (60%) reporting that telepsychiatry programmes were less expensive, and 8 (32%) showing telepsychiatry programmes were more expensive. Three studies reported cost-effectiveness analyses, favouring telepsychiatry programmes, but at highly elevated cost-effectiveness thresholds. Few studies reported costs of developing or delivering telepsychiatry programmes.

Conclusion:

Costs of telepsychiatry programmes varied widely, with substantial heterogeneity in how costs were defined and reported. Some programmes cost less than in-person services while others cost more. Therefore, rigorous cost-effectiveness studies following established standards in economic evaluation are needed to inform implementation and sustainability of these programmes in health systems.

Introduction

Mental disorders are a significant cause of disease burden, attributable to nearly a third of years lived with disability¹ and roughly 8 million deaths every year.² Most health systems are currently unable to address this increasing need for mental healthcare, largely due to a shortage and uneven distribution of resources.³ Key challenges to delivering adequate mental healthcare include shortages in mental health clinicians, difficulty accessing services, and fragmentation of care,⁴ as well as stigma-related barriers to seeking care.^5,6 In many countries, mental disorders are also associated with low utilisation of services and poor access to treatment.⁷

To make psychiatric services more widely available, health systems have increasingly turned to technology-based solutions as potentially sustainable and feasible methods for care of mental disorders.⁸ Telepsychiatry, the area of telehealth involving the use of telecommunications technology such as webcams, telephone, and videoconferencing for delivery of psychiatric care and assessment at a distance has gained prominence for supporting mental health services.^9,10 Telepsychiatry is defined as the ‘delivery of health care and exchange of health care information for purposes of providing psychiatric services across distances’ and encompasses both synchronous and asynchronous methods of communication.⁸ Synchronous communication refers to real-time communication, typically using videoconferencing between patient and provider. Alternatively, asynchronous communication refers to audio- or video recording patient encounters (typically with community providers), and then forwarding the recordings to speciality providers for assessment. This is also known as the ‘store-and-forward’ method.¹¹ Applications of telepsychiatry include screening, diagnosis and assessment, individual and group psychotherapy and medication management.¹¹ There is also growing interest in telepsychiatry in remote areas, allowing psychiatrists to reach patients and support primary care providers in rural settings.^9,12

Mounting evidence shows the clinical effectiveness of telepsychiatry for psychiatric assessment and treatment delivery.^13–17 Researchers have demonstrated that video-based telepsychiatry can produce similar clinical outcomes to face-to-face care,¹⁶ while telepsychiatry can lower the stigma associated with visiting mental health clinics and increase access to services.^18,19 Patient and clinician satisfaction with telepsychiatry is also generally high,²⁰ particularly in rural and underserved settings.²¹ However, there remains uncertainty about the costs of these programmes,^22,23 and specifically whether telepsychiatry is cost-effective.⁹ A myriad of expenses are associated with implementing telepsychiatry infrastructure, including equipment costs, system operation and maintenance, lines for information transmission, transportation to telepsychiatry site, and administrative costs among others.²⁴

To inform health systems as they seek to implement telepsychiatry for addressing gaps in care for mental disorders, a more complete understanding of the costs of these programmes is required. The objective of this systematic review was thus to examine the literature on cost reporting of telepsychiatry programmes for mental healthcare, including screening, assessment and psychiatric care for persons with various mental disorders. Our primary goal was to synthesise evidence reporting broadly on costs, and economic evaluations including cost-effectiveness or costs of developing or delivering telepsychiatry programmes for clinical care of mental disorders.

Methods

Search strategy

We searched electronic databases from 1 January 2000 to 16 March 2018: PubMed (MEDLINE), EMBASE, Cochrane Central, PsychINFO (EBSCOhost), CINAHL, EconLit, Web of Science, Health Economic Evaluations Database, National Health Service Economic Evaluation Database, Cost-Effectiveness Analysis Registry, Research Papers in Economics (RcPEc), and European Network of Health Economic Evaluation Database (EURONHEED). We selected 1 January 2000 as the start date because several systematic reviews found no published cost evaluation studies of digital technology for mental health before 2000,^25–29 and specifically, no economic evaluations of telepsychiatry programmes.^18,19,22,23

We combined search terms for ‘mental disorders’ and ‘technology, telepsychiatry or remote technology’ and ‘costs, cost-effectiveness or economic analysis/evaluation’. Each term was entered as a key word and corresponding medical subject heading (MeSH) term. MeSH terms refer to the United States Library of Medicine’s controlled vocabulary thesaurus and are used for indexing articles in Medline. Combining MeSH terms with general free text search terms identifies as many relevant records as possible.³⁰ No language limits were applied. Table 1 lists the complete search strategy used in Medline. We also searched reference lists of included studies, prior systematic reviews, and Google Scholar to identify additional relevant studies.

Table 1.

Search strategy used in Medline.

search	Search terms
#1  (Mental disorders)	‘serious mental illness’ OR ‘serious and persistent mental illness’ OR ‘severe mental illness’ OR ‘mental illness’ OR ‘mental health’ OR ‘mental disorder’ OR ‘schizophrenia’ OR ‘bipolar disorder’ OR ‘schizoaffective disorder’ OR ‘major depressive disorder’ OR ‘depression’ OR ‘anxiety’ OR ‘affective disorder’ OR ‘psychotic disorders’ OR ‘psychosis’ OR ‘post-traumatic stress disorder’ OR ‘PTSD’ OR ‘stress disorder’ OR ‘antipsychotics’ OR ‘dementia’ OR ‘epilepsy’ OR ‘mental disorders’[MeSH terms] OR ‘anxiety disorders’[MeSH] OR ‘stress disorders, traumatic’[MeSH] OR ‘mood disorders’[MeSH] OR ‘depressive disorder’[MeSH] OR ‘schizophrenia and disorders with psychotic features’[MeSH] OR ‘psychotic disorders’[MeSH] OR ‘dementia’[MeSH] OR ‘epilepsy’[MeSH]
#2  (Substance-use disorders)	‘alcohol’ OR ‘substance use’ OR ‘substance use disorder’ OR ‘substance related disorder’ OR ‘alcohol’ OR ‘alcohol use disorder’ OR ‘alcoholism’ OR ‘amphetamine’ OR ‘cocaine’ OR ‘crack cocaine’ OR ‘marijuana’ OR ‘cannabis’ OR ‘opioid’ OR ‘heroin’ OR ‘morphine’ OR ‘street drugs’ OR ‘drug abuse’ OR ‘illicit drug’ OR ‘recreational drugs’ OR ‘substance-related disorders’[MeSH] OR ‘alcohol-related disorders’[MeSH] OR ‘amphetamine-related disorders’[MeSH] OR‘cocaine-related disorders’[MeSH] OR ‘marijuana abuse’[MeSH] OR ‘opioid-related disorders’[MeSH] OR ‘street drugs’[MeSH] OR ‘crack cocaine’[MeSH]
#3	#1 OR #2
#4  (Digital technology)	‘telemetry’ OR ‘telemedicine’ OR ‘telepsychiatry’ OR ‘telehealth’ OR ‘telecare’ OR ‘Tele mental health’ OR ‘connected health’ OR ‘internet’ OR ‘internet health’ OR ‘web browser’ OR ‘website’ OR ‘web-based’ OR ‘social media’ OR ‘Facebook’ OR ‘mobile health’ OR ‘mobile technology’ OR ‘mobile phone’ OR ‘cellular phone’ OR ‘cellphone’ OR ‘smartphone’ OR ‘text message’ OR ‘text messaging’ OR ‘wireless technology’ OR ‘remote sensing technology’ OR ‘remote consultation’ OR ‘big data’ OR ‘telemetry’[MeSH] OR ‘telemedicine’[MeSH] OR ‘internet’[MeSH] OR ‘web browser’[MeSH] OR ‘social media’[MeSH] OR ‘cellular phone’[MeSH] OR ‘text messaging’[MeSH] OR ‘wireless technology’[MeSH] OR ‘remote sensing technology’[MeSH]
#5  (Costs, cost-effectiveness, and economic evaluation)	‘cost’ OR ‘cost-effectiveness’ OR ‘cost-benefit’ OR ‘cost-minimisation’ OR ‘cost-utility’ OR ‘economic’ OR ‘finance’ OR ‘pricing’ OR ‘spending’ OR ‘expenditure’ OR ‘costs and cost analysis’[MeSH] OR ‘cost-benefit analysis’[MeSH] OR ‘economics’[MeSH] OR ‘fees and charges’[MeSH]
#6  (Final search)	#3 AND #4 AND #5

MeSH indicates Medical Subject Heading.

Study selection criteria

We included studies that met the following eligibility criteria: (1) pertained to delivery of care, including screening, assessment or treatment of mental disorders, defined broadly as depression, anxiety disorders, serious mental illnesses including schizophrenia spectrum disorders and bipolar disorder, post-traumatic stress disorder, dementia and epilepsy; (2) evaluated use of telepsychiatry for clinical care of mental disorders; (3) recruited participants; and (4) reported telepsychiatry programme costs. Only studies involving telepsychiatry programmes, including synchronous and asynchronous approaches, were included. Telepsychiatry programmes could support various clinical services, including diagnosis, screening, treatment, ongoing support, medication management or follow-up.

We required that studies included primary data collected from patients receiving mental healthcare. All types of studies describing costs associated with development, evaluation or delivery of telepsychiatry were considered. This included experimental or observational studies, naturalistic studies or pilot/feasibility studies. Costs must be reported and were defined as either costs of development or delivery of the telepsychiatry programme. Costs also included out-of-pocket expenses and individual costs for patients, and costs for healthcare providers. Our goal was to include studies that objectively reported costs associated with development and delivery of telepsychiatry for mental disorders, and economic evaluations including cost-effectiveness analyses of these programmes. Commentaries, opinion pieces, editorials or perspective articles were not included.

Data extraction and analysis

After searching the databases, all entries were imported into reference management software. Duplicate entries were removed. Two researchers independently screened the titles and abstracts of the remaining entries to identify potentially relevant studies. Relevant studies were defined as entries that pertained to mental disorders, involved use of telepsychiatry, and mentioned cost measurement or cost as an outcome. English language abstracts were obtained for all relevant studies. When both researchers agreed that citations met these eligibility criteria, full-text articles were retrieved. Full-text articles were also retrieved for any citations for which eligibility could not be determined. For any foreign language articles, Google Translate online software was used to facilitate full-text review and data extraction. A third researcher resolved any disagreement about inclusion through discussion and consensus, and reviewed entries that were excluded as per the eligibility criteria to confirm exclusion. The two researchers independently conducted full-text review and coding of potentially relevant articles, and the third researcher reviewed the final list of full-text articles that met the inclusion criteria.

The two researchers extracted the following data from included studies: year, country of origin, participant characteristics (e.g. mental illness diagnosis, sample size, gender), study design and duration of follow-up, telepsychiatry programme (e.g. brief summary of the programme), and cost outcomes (e.g. development costs, delivery costs and cost-effectiveness). These details were summarised in a table, and were reviewed by all authors to confirm accuracy in data extraction. In some cases, results from a single study were published as multiple manuscripts, such as reporting of secondary outcomes. We were careful to avoid over counting studies, though secondary analyses from included studies were reviewed to supplement data extraction. We present a narrative summary of the included studies.

Results

Our database search yielded 6069 articles after removal of duplicates, of which 318 were relevant and required full-text review (Figure 1). In total, 26 unique studies met the inclusion criteria, with one secondary analysis to supplement data extraction (Table 2). The studies were from the United States (n = 15; 58%), Canada (n = 4; 15%), Australia (n = 2; 8%) and one study each from Colombia, Germany, Hong Kong, Israel and Spain. Two studies were published in languages other than English.^31,32 The studies enrolled a total of N = 17,967 participants, ranging from 15 to 10,879 participants. Most studies focused on depression (n = 7; 27%) or general mental disorders and screening (n = 7; 27%), followed by child mental health (n = 4; 15%), geriatric mental health (n = 4; 15%), post-traumatic stress disorder (n = 2; 8%), suicidal ideation (n = 1; 4%) and epilepsy (n = 1; 4%).

Figure 1.

Flow diagram of studies included in the review.

Table 2.

Cost evaluations of telepsychiatry for remote consultations.

Study, year, country	Study design and duration	Sample description	Digital intervention description	Cost outcome	Intervention costs
Ahmed et al., 2008, Canada (44)	Pilot RCT; 8 months	41 patients with epilepsy; 45% female	Telemedicine video follow-up visits with epilepsy physician covering seizure and medication profile, neurological exam, and discussions of treatment options and other issues. Compared to in-person follow-up visits.	Total costs for one telemedicine visit were CAD$498.85 (consisting of travel and lost work time for patients ($35.85) and videoconferencing costs ($463)), and $466 for one conventional care visit (consisting of travel, parking, accommodation, and lost work time for patients). This excluded medical personnel costs, which were assumed to be the same across both alternatives. When excluding videoconferencing equipment costs, cost savings were $430 for patients for each telemedicine visit; patients were satisfied with the telemedicine services.	Costs of the videoconferencing equipment for each visit were CAD $463.
Barrera-Valencia et al., 2017, Colombia (31)	RCT; 6-month follow-up	106 male prisoners with depression	Asynchronous telepsychiatry (store and forward) where prisoners see primary care physician, and clinical information sent electronically to a remote psychiatrist for diagnosis and treatment plan. Compared with synchronous telepsychiatry where prisoners see a psychiatrist for consultation via videoconferencing.	Both telepsychiatry models contributed to reduction in depressive symptoms among prisoners (p < 0.001). Asynchronous model showed greater decrease in symptoms compared with synchronous model (p = 0.01); costs were significantly lower for asynchronous model compared to synchronous model (p < 0.001).	Not reported.
Bounthavong et al., 2018, United States (34)	RCT; 3-month follow-up	121 veterans and military personnel with depression; 18% female	Home-based tele-behavioural healthcare using videoconferencing to deliver speciality mental healthcare to patients via laptop or desktop computer, webcam, and secure and encrypted software. Includes one session each week for 8 weeks. Compared with in-person treatment.	Tele-behavioural health cost US$51,652 more than in-person treatment (when accounting for the videoconferencing equipment), and produced 0.004 greater QALYs. ICER = $14,434,503 per QALY.	Delivery costs of tele-behavioural health were US$71,974 ($1161 per patient) compared with $20,322 ($344 per patient) for in-person treatment.
Butler and Yellowlees, 2012, United States (50)	Retrospective pilot cost-analysis	125 patients with non-urgent psychiatric problems seen in a primary care clinic	Asynchronous telepsychiatry using web-based software in which patient interviews with primary care physicians were videotaped. Electronic medical records and videotaped interviews were reviewed by remote psychiatrists for assessment and treatment plan. Compared with synchronous telepsychiatry using videoconferencing and in-person consultation.	Marginal costs for asynchronous telepsychiatry, synchronous telepsychiatry, and in person models were US $68.18, $107.50 and $96.36, respectively. Asynchronous telemedicine became the most cost-effective model beyond 249 sessions.	Asynchronous telepsychiatry costs were US $7000, which included development, equipment and provider training costs. Synchronous telepsychiatry fixed costs were $20,000 due to equipment, software and training.
Egede et al., 2017, United States (46)	RCT; 12-month follow-up	241 veterans with major depression; 2.5% female	Telepsychology therapy delivered by videoconferencing technology in patients’ homes. Eight weeks, weekly 60-min sessions of behavioural activation for depression. Compared with in-person delivery of same sessions.	Increasing costs observed over time for both groups. No differences between groups. Telepsychology did not contribute to greater costs compared with in-person therapy.	Not reported.
Elford et al., 2001, Canada (35)	Descriptive, naturalistic study; 3 months	30 patients for child psychiatry assessment; 30% female	Telepsychiatry involving videoconferencing system for child psychiatry assessments. Costs compared with travelling to see a psychiatrist for in-person services.	Telepsychiatry service cost CAD$419 per patient compared with $428 per patient for in-person service. Patients were satisfied with the telepsychiatry service.	Telepsychiatry costs were CAD$419.17 per patient, including equipment, phone line installation, monthly line fees and personnel time.
Garzón-Maldonado et al., 2017, Spain (45)	Prospective observational study; 12 months	97 patients with Alzheimer’s disease and their caregivers; 65% female in patients, 81% female in caregivers	Telephone assistance system where primary caregivers can call hospital liaison nurse during working hours, then nurse notifies neurologist and neurologist provides telephone assistance to caregiver within 24 h. Phone calls used for information and support. Compared to costs of in-person clinical visits.	Mean costs of intervention were €3.74 per consultation compared with €86.87 for in-person consultations. Significant savings for telephone intervention of €80.50 (SD = €27.07) per session (p < 0.001). Caregivers reported high satisfaction with the system.	Mean costs of intervention were €3.74 per consultation accounting only for telephone costs, not including personnel or equipment.
Gerlach-Reinholz et al., 2017, Germany (32)	Quasi-experimental; 12-month follow-up	3172 patients with depression; 69% female	Telephone coaching delivered by medical specialists or health care professionals with experience in psychiatry. Patients and their relatives receive telephone support with help seeking, self-help, drug compliance, response to symptoms, and coping with social isolation. Compared to matched treatment as usual control.	At 12 months, telephone coaching showed significant reduction in total healthcare costs (p = 0.0015). Investment costs of €256,683.42 resulted in total savings to the statutory health insurance fund of €415,532 amounting to a return on investment of 1.62.	Over 12 months, total costs of the intervention group were €2332 and the control group were €2626.
Hull and Mahan, 2017, United States (51)	Pre-post pilot study; 12–16 weeks	57 patients with general mental disorders; 67% female	Talkspace web-based text messaging platform allows licensed therapists to deliver treatment by SMS asynchronously. Costs compared with known costs of in-person therapy.	Significant reduction in life distress after 3.86 months of text therapy (p < 0.001). Text therapy is lower cost (41.2%) than traditional therapy.	Cost for 3.86 months of Talkspace text therapy was US$386 compared with traditional therapy (roughly 15 sessions) at $937.35.
Jones et al., 2012, United States (33)	Descriptive, naturalistic study; 12 months	10,879 soldiers returning from active duty complete mental health screening at three sites	Different telehealth technology was used at each site for mental health screening, including videoconference and webcam. Costs compared to in-person screening.	Costs per soldier were US $57.29 for telehealth screening compared with $49.30 for in-person screening, representing a 16% increase in costs. Soldiers had preference for in-person over telehealth screening.	Cost per soldier for telehealth screening was US $57.29, accounting for mental health personnel, additional staff and travel expenses. Does not include costs of equipment, technological support, or maintenance.
Jones et al., 2014, United States (62)	Pilot RCT; 8–12 weeks	15 children with disruptive behaviour disorders and their family; 47% female	Technology-Enhanced Helping the Noncompliant Child (TE-HNC) is a smartphone-enhanced version of the clinic-based behavioural parent-training programme (HNC) to promote improved behaviour and healthier family functioning for young children. Technology enhances clinic-based services with skills videos, brief daily surveys, text message reminders, video recording home practice and mid-week video calls. Compared with HNC alone.	No statistically significant differences in outcomes between groups. TE-HNC yielded larger effect sizes than HNC for engagement outcomes. Both groups yielded clinically significant improvements in disruptive behaviour. TE-HNC families required fewer sessions (mean = 8; range = 7–10) compared with HNC families (mean- = 10; range = 7–12).	Total delivery costs for TE-HNC were US$478 per family, and for HNC were $491 per family, excluding start-up costs. TE-HNC start-up costs (with inclusion of smartphone cost) were $671 per enrolled family and HNC start-up costs were $10 per enrolled family.
Jong, 2004, Canada (47)	Descriptive, naturalistic study; 1 year	71 patients with suicidal ideation	Telepsychiatry for suicide risk assessment where a local nurse/counsellor uses a videoconferencing system to connect with a remote physician. Costs compared to transferring patients for in-person assessment.	Videoconferencing costs CAD$206 per consultation compared with $1745 for in-person assessment. Cost savings for the health system were $104,088. Satisfaction with telepsychiatry was high for patients, doctors, mental health workers and nurses.	Cost for videoconferencing assessment was CAD$206 including equipment and personnel (i.e. booking fee, online charges, nurse/counsellor, telehealth coordinator and equipment and software).
Kennedy and Yellowlees, 2000, Australia (36)	Pre-post pilot study; 12-month follow-up	124 patients with a mental disorder (32 used videoconferencing to receive mental health services)	Telepsychiatry with psychiatrists providing consultation using videoconferencing. Costs compared to visiting psychiatrist and mental health team.	Cost per consultation was AUD$145 for videoconferencing, $162 for a visiting psychiatrist and $326 for the visiting public mental health team. Combination of videoconferencing and visiting psychiatry services was $307 per consultation. No difference in clinical outcomes from using videoconferencing.	Cost per telepsychiatry consultation was AUD $145 for videoconferencing only.
Modai et al., 2006, Israel (37)	Quasi-experimental; 12-month follow-up	81 mental health patients in remote communities; 41 % female	Telepsychiatry delivered by videoconferencing. Compared to in-person care.	Costs of telepsychiatry sessions were 32% higher compared with costs of control.	Delivery costs of telepsychiatry were US $94.40 per l-h session, including salaries, running costs, phone expenses and equipment.
Morland et al., 2013, United States (52)	RCT; 6 months	74 male veterans with PTSD	Clinical group video teleconferencing focused on anger management among veterans with PTSD. Therapist led sessions remotely on a large screen at the local VA clinic. Compared to in-person group therapy at the local VA clinic.	Average costs of the teleconferencing intervention were US$79 (95% CI $73, $84) compared with $792 (95% CI $727, $856) for the in-person group therapy, a significant difference (p < 0.01). Consistent cost reduction for teleconferencing across clinical outcomes.	Mean cost per participant was US$79 (95% CI $73, $84) for video teleconferencing sessions, accounting for per session costs for personnel, participants, travel and equipment.
Neufeld and Case, 2013, United States (48)	Descriptive, naturalistic study; 2 years	350 patients with various mental illnesses; 55% female	Telemental health services delivered by advanced practice nurses by videoconferencing, supplemented with walk-in clinics. Compared to in-person care.	Telemedicine contributed to more rapid access to services, reduced wait times for follow-up, and increased cost-efficiency measured as scheduled time converted to billable time compared with in-person care.	Not reported.
Painter et al., 2017, United States (38)	RCT; 12-month follow-up	265 veterans with PTSD; 10% female	Telemedicine-based collaborative care for PTSD. Community-based outpatient clinics supported by off-site PTSD care team (includes remote nurse care manager, pharmacist, psychologist and psychiatrist) using electronic health record system, telephone, and interactive video to review medications, deliver cognitive processing therapy, supervise the team and conduct psychiatric consultations. Compared to usual care.	Total costs were US $11,512 for the telemedicine intervention compared with $9,544 for usual care. Intervention group showed significantly greater improvements in PTSD, depression severity, and QALYs measured with QWB and SF-12 compared with usual care. ICER = $185,565 per QALY.	Telemedicine intervention cost US$2,029 per patient per year, which primarily included personnel costs.
Pyne et al., 2010, United States (39) Secondary outcomes: Fortney et al., 2011 (63)	RCT; 12-month follow-up	395 patients with depression from VA community-based outpatient clinics; 8% female	Telemedicine-based collaborative stepped-care model involving primary care providers, off-site tele-psychiatrists, off-site depression nurse care manager and off-site clinical pharmacist. Combination of videoconferencing and telephone-delivered consultations, enhanced with decision support software. Compared to usual care.	No difference in depression outcomes between groups. Significant intervention effects on SF-12 QALYs (p = 0.04) and lower expenditures (p < 0.001) compared with usual care. ICER = US$85,634 per QALY.	Development costs accounted for in the delivery costs. Delivery costs of the intervention were US$794 per patient, including cost of patient education pamphlets, care provider education, equipment, development and personnel.
Rabinowitz et al., 2010, United States (49)	Descriptive, naturalistic study; 5 years and 8 months	106 nursing home residents with various mental illnesses; 59% female	Telepsychiatry consisting of psychiatrists delivering care to nursing home residents via videoconferencing. Compared to in-person care.	Telemedicine reduced travel distance and time. Physician cost savings estimates for additional travel time ranged from US$84,347 to $253,040 and patient-transport personnel cost savings estimates ranged from $33,739 to $67,477. Patients, families, and staff were satisfied with the videoconferencing.	Telepsychiatry intervention costs were US $8,485 per site, including the cost of videoconference unit, support hardware and staff, and service contract. Additional cost for line fees per call.
Ruskin et al., 2004, United States (64)	RCT; 26-week follow-up	119 veterans with depression; 12% female	Telepsychiatry remote treatment consisting of eight sessions with a psychiatrist over 6 months. Consisted of use of personal-computer-based videoconferencing with mounted camera. Compared to in-person treatment.	Depression improved in both groups, no significant difference. The costs were US$86.16 for a telepsychiatry session and $63.25 for an in-person treatment session (t = 3.2, p< 0.001). Difference was not significant if psychiatrists’ travel costs to remote clinics more than 22 mi away were considered.	Delivery costs were US $86.16 for a telepsychiatry session, including equipment, access charges, maintenance and personnel time.
Simon et al., 2009, United States (54)	3-arm RCT; 24 months	600 primary care patients with depression; 74% female	Telephone care management intervention with five outreach calls or personalised mailings by care managers for monitoring and improving antidepressant adherence, and supporting care coordination and follow-up in primary care. Compared to telephone care management plus eight sessions of telephone-based cognitive behavioural psychotherapy with up to four additional calls for reinforcement. Compared to continued usual care.	Compared with usual care control group, telephone care management led to a gain of 29 depression-free days (95% CI: −6 to 63) and a US$676 increase in outpatient health care costs (95% CI: $596 lower to $1974 higher), without net benefit even if depression-free day was valued at $20. Telephone care management plus psychotherapy led to a gain of 46 depression-free days (95% CI: 12 to 80) and a $397 increase in outpatient costs (95% CI: $882 lower to $1725 higher), with positive net benefit when a depression-free day is valued at >$9.	Cost of US$98 per participant for telephone care management and $444 for telephone care management plus psychotherapy. Includes personnel time, support/facilities/overhead, training, supervision and infrastructure.
Simpson et al., 2001, Canada (40)	Descriptive, naturalistic study; 24 months	379 patients with range of mental disorders	Telepsychiatry service in rural areas using a videoconferencing system. Costs compared to visiting psychiatrist costs.	Health professional satisfaction scores of the telepsychiatry service were high. Telepsychiatry costs were CAD$140 per consultation and travelling psychiatrist costs $630 per consultation. Break-even point for telepsychiatry was 224 consultations/year. Telepsychiatry costs equal travelling psychiatrist costs at 348 consultations/year.	Cost for setting up telepsychiatry, including equipment, installation and line charges was CAD$169,800 per year. Total fixed costs for telepsychiatry consultations were $109,587.
Smith et al., 2007, Australia (41)	Descriptive, naturalistic study; 30 months	184 child and youth mental health patients in rural areas	e-Child and Youth Mental Health Service videoconferencing consultations. Costs compared were with patients travelling for in-person consultations with the same specialists, and also with a visiting outreach psychiatry service.	For 606 patient consultations, videoconferencing resulted in savings for the health service of AUD$420,000 compared with patient travel, and $70,000 compared with a visiting outreach service. Videoconferencing was the lowest cost option for child psychiatry services when the workload exceeded 131 consultations (compared with patient travel) and 379 consultations for the visiting psychiatrist outreach service.	Cost of videoconferencing was AUD$230,753 over 30 months, including equipment, line access, call charges and personnel.
Spaulding et al., 2010, United States (42)	Descriptive, naturalistic study; 6 months	132 patients for child psychiatry services	Telepsychiatry services using videoconference equipment and Internet connectivity for children and adolescents. Costs compared with travelling to see a psychiatrist for in-person services.	Telepsychiatry cost US $169 per consultation. Mean cost savings to patients and families was $138 (SD = $22) per consult. Overall savings for patients over 6 months was $35,369 in reduced travel costs. Costs of telepsychiatry to the health system were $31 per consult after subtracting savings for patients.	Cost per telepsychiatry consultation was US $169, including office space, administrative costs, line charges, personnel costs and equipment. Not including start-up costs.
Tang et al., 2001, Hong Kong (43)	Pre-post pilot study; 11 months	45 elderly care home patients mostly with dementia (67%) or other psychiatric conditions; 71% female	Psychogeriatric team provided monthly assessments of patients via teleconferencing. Costs compared to face-to-face on-site psychiatrist visits.	Cost of teleconsultation was HK$91.81, and was 13.2% lower than on-site visits with the psychiatrist ($105.78). Videoconferencing was feasible and acceptable to staff and patients.	Delivery cost per teleconsultation was HK $91.81, included set-up and system maintenance, and physician time.
Wray et al., 2010, United States (53)	RCT; 6-month and 12-month follow-up	158 caregivers of veterans with dementia	Telehealth Education Programme (TEP) consisting of 10 weekly 1-hr group telephone sessions with education, coping strategies and support for caregivers of veterans with dementia. Social workers and nurse dementia care managers lead the sessions. Compared to usual care.	At 6-months, TEP showed significant mean overall cost savings of US $2,768 compared with usual care (p = 0.039). Cost savings not maintained at 12 months.	From baseline to 6 months, per-patient costs were: US$7008 (SD = $9226) for TEP and $8831 (SD = $13,246) for control group. From 6–12 months, per-patient costs were: $6783 (SD = $7767) for TEP and $5648 (SD = $6353) for control group. Includes inpatient, nursing home and outpatient costs.

RCT: randomised controlled trial; QWB: quality of well-being scale; SF-12: Short form 12-item; ICER: incremental cost-effectiveness ratio.

Note: The percentages of female participants are not available in the studies where gender is not reported.

The telepsychiatry programmes mostly consisted of direct ‘synchronous’ connection between patients and mental health specialists or other clinicians using videoconferencing software or webcam (n = 20; 77%). The videoconferencing visits covered mental health screening,³³ psychiatric consultations,^34–43 neurological assessment,^44,45 provision of therapy,⁴⁶ medication management,^38,44 suicide risk assessment,⁴⁷ general mental health services delivered by non-physician providers,^39,48 psychiatric services for rural nursing home residents⁴⁹ and routine follow-up assessments.⁴⁴ Asynchronous approaches were used in primary care settings involving videotaping psychiatric histories for non-urgent patients in rural areas,⁵⁰ and sending clinical information electronically to a remote psychiatrist for diagnosis and treatment plan.³¹ In another study using asynchronous telepsychiatry, licensed therapists delivered psychotherapy to patients with various mental disorders via text messages.⁵¹ Four studies used videoconferencing or telephone to deliver therapy, education, coaching, or support sessions to patients or their caregivers.^{32,46,52–54}

Telepsychiatry programme costs

Table 2 summarises cost outcomes and intervention costs. Nearly all studies (n = 25; 96%) compared telepsychiatry costs with either standard in-person consultation or usual care. Among these studies, 15 (60%) reported that telepsychiatry programmes were less expensive than standard in-person or usual care, due to savings in travel time and out-of-pocket expenses and lost work time for patients and their families. An additional 8 (33%) studies showed that telepsychiatry programmes were more expensive, which appeared to be primarily due to the costs of the videoconferencing equipment, while one study did not report any differences in costs between telepsychiatry and standard in-person or usual care. The per session costs of telepsychiatry programmes varied widely, ranging from as low as e3.74 (US$4.38 in 2019) for the technology costs per neurology consultation for Alzheimer’s care in Spain (not including neurologist costs),⁴⁵ to CAD $498.85 (US$430.85 in 2019) for epilepsy consultation in Canada, covering only the costs of the videoconferencing equipment and not medical personnel.⁴⁴ For studies involving longer term delivery of telepsychiatry consisting of multiple sessions or clinician encounters, the costs over a 12-month period for stepped-care models for veterans in the United States were US$794 per patient ($930 in 2019) for community depression care³⁹ and $2,029 per patient ($2,116 in 2019) for specialised PTSD care.³⁸ Figure 2 illustrates the range in costs for individual telepsychiatry consultations and the per-patient costs for programmes consisting of multiple sessions. Two studies found that costs were lower for asynchronous compared to synchronous telepsychiatry.^31,50 In one study, few details were reported about the specific cost components that contributed to this difference,³¹ while the other attributed the lower cost of asynchronous telepsychiatry to a reduction in equipment costs and substitution of low-cost providers for specialists.⁵⁰

Figure 2.

Cost for each telepsychiatry programme by type of technology^a: (a) cost per telepsychiatry session (single consultation/session); (b) cost per patient (multiple telepsychiatry sessions.

^aAll amounts reflected in 2019 US dollars.

Cost-effectiveness of telepsychiatry

Three studies, all from the United States, reported cost-effectiveness analyses.^34,38,39 One assessed the cost-effectiveness of a telepsychiatry programme for veterans and military personnel with depression, and found that the telepsychiatry programme produced only 0.004 greater quality-adjusted life years (QALYs) compared with in-person treatment at considerably higher cost.³⁴ The resulting incremental cost-effectiveness ratio (ICER) was US$14,434,503 per QALY ($14,688,886 per QALY in 2019) gained relative to in-person care.³⁴ In a study of a telemedicine-based collaborative care programme for PTSD in veterans, the costs of the telepsychiatry programme were higher than usual care, and the ICER was $185,565 per QALY (US$193,542 per QALY in 2019) gained relative to usual care.³⁸ In another study among veterans with depression, a telemedicine-based stepped collaborative care programme with telepsychiatry consultation resulted in significantly higher expenditures, and the ICER was $85,634 per QALY (US$100,401 per QALY in 2019) gained relative to usual care.³⁹

Discussion

In the 26 studies that met inclusion criteria in this systematic review, the reporting of costs of telepsychiatry programmes varied widely. We found substantial heterogeneity in the way costs were defined across studies, such as whether development costs or equipment costs were included, or whether cost calculations accounted for medical personnel or patient costs such as travel time, out-of-pocket expenses, or lost employment. Several studies omitted key details about telepsychiatry programme costs. Given the high variation in cost reporting, it was not possible to conduct a quantitative synthesis of cost findings. While our review suggests that some telepsychiatry programmes appear to be less costly that traditional in-person services, there remains uncertainty in the literature regarding economic evaluation of telepsychiatry.

This variation in costs can be attributed to several factors. The studies employed a variety of designs and methods to measure and analyse costs, making it difficult to compare costs across studies.⁵⁵ Furthermore, the wide range in costs may be due to different types of equipment and platforms used, as reflected in Figure 2 where telephone or teleconference appears to cost less than videoconferencing per telepsychiatry session. Another potential driver of the wide heterogeneity of costs could be the inclusion of studies from multiple different countries and settings, as costs can vary from country to country.⁵⁶ For example, costs of healthcare delivery are substantially higher in the United States relative to other settings, such as in Europe, Canada or Australia. Differences in programme structure can also contribute to different cost outcomes, as some studies used telepsychiatry consultation to allow patients to remain in their home environment, thereby curbing costs associated with travel for both the patient and provider. Furthermore, telepsychiatry may contribute to greater cost savings when used in rural settings,²¹ where there are fewer speciality mental health providers and where burden of travel for patients or clinicians is greater compared with urban settings.

Consistent with prior studies,⁵⁵ we found that most telepsychiatry programmes appeared to cost less compared with in-person care. This was achieved through reduction in travel expenses and time, which also potentially reduced daily salary loss or other out-of-pocket expenses incurred by patients who would otherwise miss work and other responsibilities to attend in-person care.¹² Telepsychiatry may also reduce costs of patient transfers between departments, follow-up assessments, medical space usage, and redundant testing.⁵⁷

Some studies reported higher costs for telepsychiatry programmes. The high costs often include upfront costs, such as the substantial one-time costs of equipment purchases and set-up. Given the short time frame of many included studies, there may not have been sufficient time for the cost-saving of telepsychiatry to offset the high upfront costs. In addition, many telepsychiatry programmes rely heavily on consultations with speciality mental health providers, who are in short supply across most settings. The relatively high salaries of these providers can contribute to high costs of telepsychiatry programmes. We found several studies reporting costs of telepsychiatry programmes extending beyond speciality consultations, including delivery of psychological sessions,⁴⁶ group therapy sessions,⁵² remote coaching support,³² telephone care management and cognitive behavioural therapy for depression,⁵⁴ and weekly education sessions for caregivers of veterans with dementia,⁵³ all of which reported favourable cost outcomes relative to traditional in-person services or usual care. These findings contribute to increasing research to extend the use of telepsychiatry to cover other aspects of mental healthcare delivery.²¹

Among the three cost-effectiveness studies identified in our review, the telepsychiatry interventions emerged as cost-effective compared with standard in-person or usual care.^34,38,39 However, the ICERs were very large, exceeding what are typically considered standard willingness-to-pay thresholds.⁵⁸ Telepsychiatry programmes have been in use for a longer period of time relative to other emerging digital technologies for mental health, such as smartphone apps, wearable devices and sensing technology.²⁹ Therefore, it is some-what surprising that, despite the large number of studies showing the clinical benefits of telepsychiatry programmes, there have been so few rigorous economic evaluations and assessments of cost-effectiveness. The three cost-effectiveness analyses were published in 2010,³⁹ 2017³⁸ and 2018,³⁴ indicating that the field continues to lag behind in the economic evaluation of telepsychiatry. Most of the studies included in this review represent ‘cost studies’ (reporting costs) rather than true ‘economic evaluations’ (analysing costs), a concern that has been previously highlighted in the literature on emerging healthcare technologies.^18,56 Another challenge was the selective reporting of development and equipment costs, such as installation and upfront investment in the programmes, which makes it difficult to draw from these findings to inform implementation or delivery of these programmes in other settings. The lack of consistent reporting of costs has similarly been highlighted in prior reviews questioning the quality of cost data in telepsychiatry.^9,18,21,59 Future economic evaluation studies of telepsychiatry programmes should aim to adhere to best practices in quality economic outcome reporting,⁶⁰ as recommended by the Cochrane Collaboration.²⁵

Several limitations warrant consideration. First, given that many studies were preliminary and involved pilot or feasibility evaluations, used varying study designs including naturalistic evaluations, and reported heterogeneous cost outcomes, we did not conduct a formal assessment of risk of bias and study quality, and it was not possible to conduct a meta-analysis of quantitative findings and economic outcomes. Furthermore, the Cochrane Collaborative states that there is not a recommended approach for pooling combined estimates of cost data extracted from multiple economic evaluations.⁶¹ The studies employed telepsychiatry for various mental disorders across diverse settings and countries, which likely contributed to variation in the costs of these programmes, making it difficult to draw conclusions for any single setting and limiting the generalisability of these findings.¹⁸ It is also possible that there may have been publication bias, where studies showing that telepsychiatry programmes are highly costly or potentially less cost-effective compared with in-person or usual care were not published, and were not included in the literature search. This concern has similarly been raised in prior systematic reviews of telehealth programmes.²³ We did not identify any studies on the costs of telepsychiatry programmes for supporting clinical supervision or continuing education for healthcare providers at remote sites, both valuable applications of telepsychiatry programmes. This could be due to our inclusion criteria, though it may also reflect a gap in the literature, where cost benefits of remote supervision or training for mental healthcare providers have not been examined.

Conclusion

This review summarises recent studies reporting costs of telepsychiatry programmes. Some studies appear to demonstrate that telepsychiatry is less costly compared with conventional in-person or usual care, though there is limited evidence that these approaches are cost-effective. Given the high variability in cost reporting, inconsistent inclusion of cost components, and range of study designs, sample sizes, and study durations, there remains continued need for robust economic analysis of telepsychiatry. Future efforts should seek to determine in what settings and for what disorders the different applications of telepsychiatry are most cost-effective to inform implementation and sustainability of these programmes in health systems.