Cost issues with telepsychiatry in the United States

Abstract

Videoconferencing has increased patient access to psychiatric care by linking specialists at academic or regional health centres with primary health care professionals in shortage areas (Hilty et al, 1999, 2002). Preliminary studies have demonstrated positive outcomes and user satisfaction (Hilty et al, 2002). Information is still being sought regarding costs because of a paucity of clinical outcome studies, cost data and randomised trials.

The overall effectiveness of telepsychiatry has recently been evaluated (Hilty et al, 2004a). Ideally, effectiveness should be considered in terms of the patient, the provider of services, the programme receiving services, the community receiving services and society at large. Telepsychiatry appears effective with regard to access to care, quality of care (in terms of outcomes, reliability, satisfaction and comparison with in-person care), education and the empowerment of rural communities. It may be premature to claim it is cost-effective because of a lack of data. This article reviews the cost data, discusses issues that affect costs and makes recommendations to reduce costs.

Methods

A comprehensive review of the telepsychiatric literature was conducted in the Medline, PubMed, PsychInfo, EMBASE, Science Citation Index, Social Sciences Citation Index and Telemedicine Information Exchange databases (1965 to July 2003). Key words included ‘telepsychiatry’, ‘telemedicine’, ‘videoconferencing’ and ‘costs’. Article titles and abstracts were reviewed by the first author and references were reviewed for additional potential articles.

Results

Currently, there are over 50 telepsychiatry programmes in the USA and another 14 in Canada (Hilty et al, 2004b). Nearly all telepsychiatry services use dial-up integrated service digital network (ISDN) or T1 lines, and transmit at 128–512 kbit/s (there is typically a 0.3 s audio and video delay at the lower end of this range). Satellite transmission is eight times as costly and almost always involves a 0.5– 1.0 s delay in communication between parties.

Telepsychiatry works well in a number of regards. Most studies have shown it to be diagnostically reliable compared with in-person care for a wide range of diagnoses for adults, children and geriatric populations (Hilty et al, 2004b). It appears to be generally acceptable to patients. Telepsychiatry appears to allow the building of relationships, with clear advantages over telephone consultation and few disadvantages compared with in-person care. It may improve outcomes; for example, in a study by Nesbitt et al of specialty consultation to primary care providers, including telepsychiatry, specialists changed the diagnosis in 91% of cases and recommended medication changes in 57%. Subsequently, 56% of patients showed clinical improvement (further details available from TSN on request).

Formal studies of cost-effectiveness are limited because:

• the scope of data collection is often limited

• cross-sectional rather than longitudinal measurement is done

• data have not been collected in a systematic, controlled, prospective fashion.

One meta-analysis found that only 38 of 551 articles contained any quantifiable cost data (Whitten et al, 2000). Ideally, direct and indirect costs should be collected for patients, clinics, providers and society at large. Many detailed guidelines have been published with recommendations for the assessment of cost-effectiveness measures in health provider systems (Weinstein et al, 1996; Hailey & Crowe, 2000; Hilty et al, 2004b).

Costs of telepsychiatry services

Direct costs include equipment, installation of lines and other supplies. Fixed costs include the rental cost of lines, salaries and wages, as well as administrative expenses. Variable costs include data transmission costs, fees for service, and maintenance and upgrades of equipment. Costs are dependent on utilisation; for example, a break-even cost analysis is used in Alberta, Canada (Hilty et al, 2004b). In addition, the cost of telepsychiatry may seem high compared with that of ‘usual care’ in remote rural communities, where ‘usual care’ is often no care at all.

Telepsychiatry appears effective with regard to access to care, quality of care, education and the empowerment of rural communities. It may be premature to claim it is cost-effective because of a lack of data.

Telepsychiatry is in general less expensive for patients than a conventional consultation, largely because it reduces both travel and time away from work. Studies have been inconsistent in their estimation of whether telepsychiatry services are less expensive, as expensive or more expensive than outreach services that involve personal contact with the patient. Telepsychiatry, though, appears to be cost-effective in terms of reducing the numbers of patient transfers (e.g. Alessi, 1999) and hospital use (e.g. Lyketsos et al, 2001). Communities have been able to treat their patients locally and they have therefore retained money that would otherwise have been lost to suburban centres upon referral (Dimand et al, 2004).

The ongoing costs of maintaining telepsychiatry services have been a major problem throughout the United States. Start-up grants generally pay for technology, but not for staff coordination and long-term psychiatric (physician) service. Insurance or third-party payers have agreed to fund physician time in most regards, although they often require preliminary educational and administrative interventions. County mental health systems often deny telepsychiatry claims, partly in order to keep costs low but also because the services are provided outside of their system, in the medical sector. This is a problem because patients receive 60% of their mental health services from the medical sector, which they generally prefer because it generates less stigma and gives patients the ability to maintain their relationship with the primary care provider, and because of what is widely perceived to be inadequate care in the mental health sector.

The ongoing costs of maintaining telepsychiatry services have been a major problem throughout the United States. Start-up grants generally pay for technology, but not for staff coordination and long-term psychiatric service.

Federal programmes have been established with high specialist reimbursement for some rural patients, but telepsychiatry services do not qualify. This is because telepsychiatry consultations are viewed as being provided ‘outside’ the designated clinics.

When rural agencies have funds available, contracts with academic consultation–liaison services have proved successful in terms of patient outcomes. The use of residents with faculty supervision appears to reduce costs and provides them with a meaningful learning experience. Consultation–liaison services benefit from an expansion in the scope of their work to the out-patient sector and improved reimbursement (e.g. salary and benefits, as reimbursement for an in-patient medical centre consultation is limited) (Bourgeois et al, 2003).

Discussion

Little information is available about the cost-effectiveness and cost–benefit of telepsychiatry programmes; and data need to be collected in a standard, prospective, preferably longitudinal fashion. However, cost-effectiveness could be improved by use of a consultation–liaison model, whereby the telepsychiatrist evaluates the patient and makes recommendations for management by the primary care provider, who thereby gains skills that could benefit patients and the community setting. This educational role of telepsychiatry is especially important for the primary care providers of rural communities, in which 20% of the US population lives.

Little information is available about the cost-effectiveness and cost–benefit of telepsychiatry programmes.