Annotated Bibliograpgy

PSYCHIATRIC SERVICES ♦ http://psychservices.psychiatryonline.org ♦ July 2003 Vol. 54 No. 7997766

Telepsychiatry has been in exis-tence for more than 40 years,yet the issue of whether it is worth the cost, or whether it even pays for itself, remains controversial. Whitten and associates (1) recently concluded, after an extensive review of the literature, that “there is no good evidence that telemedicine is a cost-effective means of delivering health care.” We reviewed the litera- ture on the cost of telepsychiatry to determine whether telepsychiatry is worth the cost.

The basic components of telepsy- chiatry costs can be classified as di- rect costs—including the cost of equipment, lines for information

transmission, operation of the telepsychiatry system, supplies, main- tenance, and the salary of the telecommunications coordinator— and indirect costs, including trans- portation of patients and clinicians to the telepsychiatry site and adminis- trative overhead costs. Hidden costs include training individuals in the use of the equipment, maintaining dupli- cate records at several sites, transmit- ting clinical information between sites, and allocating space for equip- ment. In this article we note discus- sions of component costs in the stud- ies reviewed. We summarize findings of the studies about costs, derived both from actual services delivered

and from theoretical calculations, and about cost comparisons between telepsychiatry and in-person psychi- atric services. The methods and limi- tations of each study are also noted.

The cost-related terms used in the articles reviewed—for example, cost, cost-effectiveness, and cost-benefit— are defined elsewhere (2–5).

Methods Published studies were identified through English-language searches of MEDLINE and PsycINFO databas- es from 1956 through 2002 using the terms “telepsychiatry,” “telemedicine + psychiatry,” “teleconferencing + psychiatry,” “cost,” “cost analysis,” “cost-benefit,” “cost-effectiveness,” and “cost-consequences matrix.” Studies were also found in bibliogra- phies provided by the authors of two recent literature reviews (6,7). The studies we found were grouped by their method of looking at cost and were tabulated accordingly.

Results Although more than 380 articles re- lating to telepsychiatry were found, the literature search generated only 12 articles that focused specifically on the cost of telepsychiatry; ten of those were peer reviewed. The articles ap- peared from 1995 through 2002. The projects described in the studies were located in the United States, Canada, Australia, and Hong Kong, and one study included programs around the world. Table 1 summarizes the 12 studies reviewed.

Cost-feasibility Two articles used a cost-feasibility method, calculating costs theoretical- ly, with no actual service delivered (8,9). Preston (8), in 1995, reported on an assessment of the potential sav- ings of a rural telemedicine project

A Review of the Costs of Telepsychiatry SStteevveenn EE.. HHyylleerr,, MM..DD.. DDiinnuu PP.. GGaanngguurree,, MM..DD..

The authors are affiliated with the department of psychiatry at Columbia University in New York City. Dr. Hyler is also with the department of psychiatry at New York State Psychiatric Institute, and Dr. Gangure is also with St. Luke’s–Roosevelt Hospital Center in New York City. Send correspondence to Dr. Hyler at Box 130, NYSPI, 1051 Riverside Drive, New York, New York 10032 (e-mail, [email protected]).

Objectives: The issue of whether telepsychiatry is worth the cost or whether it pays for itself is controversial. This study investigated this question by reviewing telepsychiatry literature that focused on cost. Methods: Approximately 380 studies on telepsychiatry published from 1956 through 2002 were identified through MEDLINE, PsycINFO, and cross-referenced bibliographies. Of these, 12 studies with samples of more than ten persons or programs focused specifically on the cost of telepsychiatry. Results: The methods of examining cost used in the 12 studies were cost-feasibility, cost surveys, direct comparison of costs of telepsychiatry and in-person psychiatry, and cost analysis. It was con- cluded that in seven of the studies reported, telepsychiatry was worth the cost. One study reported that telepsychiatry was not financially vi- able. Three studies of cost-effectiveness reported the break-even num- ber of consultations, the number that make telepsychiatry comparable in cost to in-person psychiatry. One review concluded that the lack of a clear business plan contributed to the difficulty of determining whether any of the programs was cost-effective. Conclusions: Telepsychiatry can be cost-effective in selected settings and can be financially viable if used beyond the break-even point in relation to the cost of providing in-per- son psychiatric services. Whether governmental or private health agen- cies value telepsychiatry enough to assume its cost is a question that re- mains to be answered. (Psychiatric Services 54:976–980, 2003)

PSYCHIATRIC SERVICES ♦ http://psychservices.psychiatryonline.org ♦ July 2003 Vol. 54 No. 7 997777

TTaabbllee 11

Summary of 12 studies focusing on issues of cost in telepsychiatry (TP)

Type and study Sample and location Method Results and comments Study limitations

Cost-feasibility Preston Rural outpatients, rural Estimation followed by Estimation of 2.7 years for Simple comparison of ag-

(1995) (8) inpatients, and juvenile calculation of costs and positive return on initial gregated costs with aggre- offenders; Texas savings for 2 years only investment for TP gated savings; extrapolation

by estimation. Theoretical; no actual TP

Werner and Rural outpatients at a Cost comparison, University-based TP less Theoretical; no actual TP Anderson community mental CMHC-based TP and expensive; TP not finan- (1998) (9) health center (CMHC) university-based TP cially viable for rural

and another site; Michigan outpatients

Cost surveys Whitten et 16 TP programs with Survey of TP programs Lack of business No systematic data

al. (2000) >100 visits per year; strategic plan for most collection (10) worldwide programs; only one finan-

cially viable program

Simpson et Rural outpatients; Survey of outpatients In-person psychiatry (IP) Data derived only from al. (2001) Alberta, Canada in a major center more ex- patients’ qualitative re- (11) pensive for rural patients sponses

than TP (work time missed, child care expense)

Direct cost comparisons

Trott and Rural outpatients; Cost comparison, TP considerably less expen- Did not specify cost-effec- Blignault Queensland, Australia TP and IP sive than IP for nonclinical tiveness of using TP system (1998) (12) activities

Tang et al. Nursing home Cost comparison, TP 13 percent less expensive Pilot study; assumed setup (2001) geriatric patients; TP and IP than IP and maintenance cost shar- (13) Hong Kong ing by university depart-

ments by time used; hidden costs not included

Naval tele- Military; U.S.S. George Cost comparison, TP much less Qualitative data only medicine Washington and U.S. TP and IP expensive than IP study (1998) Naval Hospital, Bethesda, (14)a Maryland

Cost analysis Simpson et al. Rural outpatients; Cost analysis Break-even point in com- Estimates used for

(2001) (15) Alberta, Canada paring TP with IP varies determining IP cost with equipment cost

Doze et al. Rural outpatients; Cost analysis, Break-even point in com- Pilot study; telephone line (1999) Alberta, Canada sensitivity analyses paring TP with IP varies charges uncertain; no (17) with equipment cost procedure manual

Alessi et al.a Inmates at Event-based cost and TP much less expensive Only abstract published (1999) (18) a prison; Michigan cost-effectiveness than IP, especially with cost-

analysis effectiveness technique

Kennedy and Rural outpatients; Cost-effectiveness TP not always cost- Pilot study; did not specify Yellowlees Queensland, Australia analysis effective cost-effectiveness of using (2000) (19) TP system for nonclinical

activities

Hailey et al. Outpatients; Cost analysis from Patients: TP less expensive Weighting for TP’s indirect (1999) (20) Alberta, Canada three perspectives: than or the same as IP. influences on cost described

patients, payers, Payers: break-even point only theoretically (for ex- and society determined by volume of ample, travel stress, changes

use. Society: break-even in duration of treatment, point determined by sum work time lost) of two preceding perspec- tives; TP less expensive than IP

a Published as non-peer-reviewed report

that included a telepsychiatry compo- nent. The authors estimated that a positive return on investment would take approximately 2.7 years. Werner and Anderson (9), in 1998, reported their study of the cost-feasibility of implementing a telepsychiatry system called university-based telepsychiatry to link psychiatrists at Michigan State University with patients at a rural community mental health center (CMHC). The authors compared this type with one called CMHC-based telepsychiatry that used a link be- tween the CMHC and another rural site. The university-based system was found to be less expensive, chiefly be- cause of the preexistence at the uni- versity of integrated services digital network (ISDN) lines. The authors concluded that although “telepsychia- try is technologically feasible, it is pragmatically difficult, and not eco- nomically supportable in providing services to remote rural areas at this time.”

Cost surveys Two articles used surveys as a way to subjectively probe cost without de- termining it objectively (10,11). During 1998 and 1999, Whitten and colleagues (10) surveyed 16 U.S. and international programs that had each conducted at least 100 consultations by telepsychiatry. Only one program reported that it was self-sustaining through revenues from service deliv- ery. The authors identified a need for better financial planning for de- livering telepsychiatry services cost- effectively. Simpson and colleagues (11) reported in 2001 that the avail- ability of telepsychiatry led to cost savings for patients who would oth- erwise have had to travel and thus lose time at work and pay child care expenses.

Direct comparisons of costs Three studies directly compared the costs of telepsychiatry with those of in-person psychiatry (12–14). Trott and Blignault (12), in 1998, reported their calculations of travel savings ob- tained through the use of telepsychia- try compared with the same level of service provided in person over one year. The child and adolescent tele- consultation component proved to be

the highest generator of savings. These savings persisted even after the capital costs associated with the es- tablishment of the telepsychiatry sys- tem and the costs of the telepsychia- try calls were taken into account. No maintenance and equipment upgrad- ing costs were considered by Trott and Blignault. Tang and associates (13), part of the psychogeriatric team of the Chinese University of Hong Kong, reported in 2001 on telepsychi- atry for nursing home geriatric pa- tients. The authors found that the cost of telepsychiatry was 13.2 per- cent lower than that of an in-person visit when the setup and maintenance costs were shared by various depart-

ments of the university according to the proportion of time they used the system. In 1998 The U.S. Navy re- ported on a telepsychiatry project that connected the aircraft carrier U.S.S. George Washington with the U.S. Naval Hospital in Bethesda, Maryland (14). It was determined that telepsychiatry reduced the costs of psychiatric intervention both di- rectly, by cutting the high cost of pa- tient evacuation by air from the sea, and indirectly, by ensuring a rapid in- tervention that maintained the psy- chological balance and the functional capacity of both patient and col- leagues on board.

Cost analysis Cost analysis is an objective, more so- phisticated method of looking at cost- effectiveness (15–19). A cost analysis “examines what costs are associated with a particular [proposed] project and what may be done about those costs in the future” (16).

In a follow-up to the article by Simpson and colleagues (11) dis- cussed above, the same authors point- ed out in 2001 that the high cost of many in-person mental health servic- es could prohibit their delivery, ulti- mately decreasing the revenue of the health care provider (15). Similar re- sults were found by Doze and associ- ates (17), who reported in 1999 that telepsychiatry was more expensive than in-person psychiatry at a low vol- ume of service but less expensive at a higher volume. As part of their study, the authors analyzed the degree to which economic viability would change with the values of variables. For example, in considering the break-even point—the point at which telepsychiatry is comparable in cost to in-person psychiatry—the authors found that a reduction of 10 percent in equipment cost reduced the break- even point from 396 to 368 telepsy- chiatry consultations a year. Doze and associates concluded that in some scenarios the use of telepsychiatry might justify costs that remained be- low the break-even point and that cost analysis should not be the only factor considered by health service decision makers.

Alessi and colleagues (18) reported in 1999 on two economic analyses of a prison telepsychiatry service. The first analysis was event based and fo- cused on the costs of transportation. The second analysis measured cost- effectiveness by the time spent by health and prison professionals. A comparison of these two techniques demonstrated substantial cost savings through telepsychiatry, especially when the cost-effectiveness analysis was used.

Kennedy and Yellowlees (19) re- ported in 2000 that cost-effectiveness could not be determined solely by ex- amining the financial cost but also re- quired examining health outcomes, utilization, accessibility, quality, and needs for such services in the specific

PSYCHIATRIC SERVICES ♦ http://psychservices.psychiatryonline.org ♦ July 2003 Vol. 54 No. 7997788

The

cost of

telepsychiatry

must also be considered

in light of whether it would

draw funding away

from other

services.

population studied. These authors re- ported that a community-based telepsychiatry program was not neces- sarily cost-effective for all consumers, general practitioners, psychiatrists, and public mental health services. Hailey and associates (20) published a cost analysis in 1999 comparing telepsychiatry and in-person psychia- try and presented results from three different perspectives. For patients, telepsychiatry was less expensive than in-person psychiatry because of re- duced travel costs. For third-party payers, telepsychiatry was initially more expensive than in-person psy- chiatry, but an increase in the volume of use resulted in telepsychiatry’s be- coming less expensive. From the soci- etal perspective, an approximate sum of the other two perspectives, telepsy- chiatry was less expensive.

Discussion We concluded that in seven of the 12 studies dealing with the cost of telepsychiatry, telepsychiatry was worth the cost (8,11–15,18). One study determined that telepsychiatry was not financially viable for rural outpa- tients (9). Three studies of cost-effec- tiveness reported on the break-even number of consultations—the point at which the cost of telepsychiatry be- comes equivalent to that of in-person psychiatry (15,17,20). One review sur- veying 16 telepsychiatry programs concluded that the lack of a clear busi- ness plan contributed to the difficulty of determining whether any of the programs was cost-effective (10).

Limitations Limitations of this review are the small number of studies available, their weak methodologies, the lack of explicitly presented sources of fund- ing, the lack of consistency in presen- tation of costs, and the noncompara- bility of the cost factors across the 12 studies, as can be seen in Table 1. In addition, several studies were at least five years old, meaning that their findings and conclusions could now be different because of the rapid pace of technological change. Finally, in reports of most of the studies, the au- thors appeared to have a vested inter- est in the success of the telepsychiatry program at their institution; thus the

conclusions may not be objective. A formal meta-analysis of the cost

of telepsychiatry would require at least several studies with independ- ent reviewers, random assignment, matched controls, objective outcome measures, and a comprehensive analysis of the influences of sources of funding and indirect cost-related is- sues on costs. Not addressed by most of the studies reviewed were issues of outcome and efficacy, which still need to be studied.

Is telepsychiatry worth the cost? One question in determining whether telepsychiatry is worth the cost is “Does telepsychiatry cost more or less than in-person psychiatry?” The an- swer is that it depends on many fac- tors, including the price of the equip- ment and the transmission costs, whether the equipment cost is borne exclusively by the telepsychiatry pro- gram or shared by other programs (for example, other specialties or ad- ministrative programs), the cost of technical support, how far the treat- ing psychiatrist travels to conduct in- person treatment compared with the cost of support staff for telepsychiatry at the site where the patient is locat- ed, the volume of cases treated, and the reimbursement rate.

Last, the answer depends on the party paying the costs. For patients, telepsychiatry can be less expensive in that it requires less travel time to see specialists. On the other hand, in- surance companies might be con- cerned that their costs will increase as a result of an increase in the use of services made possible by telepsychi- atry. From the perspective of the health care provider, the break-even analysis—which considers the vol- ume of use needed to equalize the to- tal costs for the two types of service (15,17,20)—shows that at higher vol- umes, telepsychiatry is less costly. Analysis from a societal perspective, in which fixed and variable costs per patient for each alternative are calcu- lated, takes into account costs in- curred by the patient as well as the health care provider (20).

A cost-consequences matrix in- cludes costs for and benefits to spe- cialists, referring physicians, health care professionals, patients and their

families, and health care administra- tors and funders (21); consideration is then given to providing appropriate weightings for intangible benefits in association with those that have mon- etary valuations. With telepsychiatry, access to certain services might in- crease appreciably, with benefits to the health of a population but at addi- tional cost.

Funding issues The majority of telepsychiatry pro- grams worldwide still remain grant funded. Many of them face the immi- nent step of finding ongoing revenue streams to sustain them. These pro- grams must show that they are cost- effective if they are going to survive (22). Rapid changes in technology, such as decreases in equipment and transmission costs and the increased reliability of equipment, as well as the sharing of expenses between disci- plines, can be expected to continue to change the cost-benefit equation in the direction of decreased cost and increased benefit (6,23,24).

An important economic issue in the United States in geriatric settings is the status of Medicare reimburse- ment for telemedicine services (25). Because insurance companies often mirror Medicare reimbursement practice, such practice has implica- tions beyond the geriatric population. The Centers for Medicare and Medic- aid Services, formerly the Health Care Financing Administration (HCFA), published rules and regulations in No- vember 1998 for Medicare reim- bursement of telemedicine services; more recent legislation approved sig- nificant modifications in these original rules, beginning with October 2001. A “referring clinician” is no longer “medically necessary” at the patient site. The requirement that the reim- bursement be split with the referring clinician has also been dropped. The ability to bill for “teleconsultation” continues to be restricted geographi- cally but has been broadened to in- clude Medicare beneficiaries residing in rural areas with shortages of health professionals, counties that are not in- cluded in a metropolitan statistical area, and agencies participating in federal telemedicine demonstration projects (25).

PSYCHIATRIC SERVICES ♦ http://psychservices.psychiatryonline.org ♦ July 2003 Vol. 54 No. 7 997799

PSYCHIATRIC SERVICES ♦ http://psychservices.psychiatryonline.org ♦ July 2003 Vol. 54 No. 7998800

The cost of telepsychiatry must also be considered in light of whether it would draw funding away from other services. Because the budget for many communities is fixed, the allo- cation of money for telepsychiatry may result in less money for other en- deavors. Decisions will need to be made by local administrators about whether telepsychiatry provides enough “bang for the buck” com- pared with delivering other services.

Conclusions From a review of the recent literature on the cost of telepsychiatry—even given the limitations of many of the studies—we conclude that telepsy- chiatry can be cost-effective in select- ed settings. However, there is no as- surance that any governmental or pri- vate health care agency will be willing to assume the cost. The cost of telepsychiatry should be considered in relation to how it contributes to im- proving the health of the population through access to information and communication and how it changes the types of interaction between providers themselves and between providers and their patients (26).

Telepsychiatry’s ultimate survival will depend on its finding its niche. Future telepsychiatry might be part of a hybrid—distinct from current health care systems. In one possible model of care, initial comprehensive evaluations could be conducted in person and routine follow-up visits through telepsychiatry. When the al- ternative to telepsychiatry is no psy- chiatry, whether psychiatry is worth the cost will depend on the value placed on delivering psychiatric serv- ices at all. ♦

References

1. Whitten PS, Mair FS, Haycox A, et al: Sys- tematic review of cost effectiveness studies of telemedicine interventions. BMJ 324: 1434–1437, 2002

2. Berger ML, Teutsch SM: Understanding cost-effectiveness analysis. Preventive Medicine in Managed Care 1:51–58, 2000

3. Gold MR, Siegel JE, Russell LB, et al (eds): Cost-Effectiveness in Health and Medi- cine. New York, Oxford University Press, 1996

4. Garber AM, Weinstein MC, Torrance GW, et al: Theoretical foundations of cost-effec- tiveness analysis, in Cost-Effectiveness in Health and Medicine. Edited by Gold MR,

Siegel JE, Russell LB, et al. New York, Ox- ford University Press, 1996

5. Eddy DM: Benefit language: criteria that will improve quality while reducing costs. JAMA 275:650–657, 1996

6. Baer L, Elford L, Cukor P: Telepsychiatry at forty: what have we learned? Harvard Review of Psychiatry 5:7–17, 1997

7. Frueh BC, Deitsch SE, Santos AB, et al: Procedural and methodological issues in telepsychiatry research and program devel- opment. Psychiatric Services 51:1522– 1527, 2000

8. Preston J: Texas Telemedicine Project: a vi- ability study. Telemedicine Journal 1:125–132, 1995

9. Werner A, Anderson LE: Rural telepsychi- atry is economically unsupportable: the Concorde crashes in a cornfield. Psychi- atric Services 49:1287–1290, 1998

10. Whitten P, Zaylor C, Kingsley C: An analy- sis of telepsychiatry programs from an orga- nizational perspective. Cyberpsychology and Behavior 3:911–916, 2000

11. Simpson J, Doze S, Urness D, et al: Telepsychiatry as a routine service: the per- spective of the patient. Journal of Telemed- icine and Telecare 7:155–160, 2001

12. Trott P, Blignault I: Cost evaluation of a telepsychiatry service in northern Queens- land. Journal of Telemedicine and Telecare 4(suppl 1):66–68, 1998

13. Tang WK, Chiu H, Woo J, et al: Telepsychi- atry in psychogeriatric service: a pilot study. International Journal of Geriatric Psychia- try 16:88–93, 2001

14. Naval telemedicine: fleet lines up to pro- vide live teleconsulting; telepsychiatry saves time and money on the high seas. Telemedicine and Virtual Reality 3(4):40, 47, 1998

15. Simpson J, Doze S, Urness D, et al: Evalu- ation of a routine telepsychiatry service. Journal of Telemedicine and Telecare 7: 90–98, 2001

16. Shin JK, Siegel JG: Modern Cost Manage- ment and Analysis, 2nd ed. Hauppauge, NY, Barron’s Educational Series, 2000

17. Doze S, Simpson J, Hailey D, et al: Evalua- tion of a telepsychiatry pilot project. Jour- nal of Telemedicine and Telecare 5:38–46, 1999

18. Alessi NE, Rome L, Bennett J, et al: Cost- effectiveness analysis in forensic telepsychi- atry prisoner involuntary treatment evalua- tions (abstract). Telemedicine Journal 5:17, 1999

19. Kennedy C, Yellowlees P: A community- based approach to evaluation of health out- comes and costs for telepsychiatry in a rural population: preliminary results. Journal of Telemedicine and Telecare 6(suppl 1):155– 157, 2000

20. Hailey D, Jacobs P, Simpson J, et al: An as- sessment framework for telemedicine ap- plications. Journal of Telemedicine and Telecare 5:162–170, 1999

21. McIntosh E, Cairns J: A framework for the economic evaluation of telemedicine. Jour- nal of Telemedicine and Telecare 3:132– 139, 1997

22. Darkins A: Program management of tele- mental health care services. Journal of Geriatric Psychiatry and Neurology 14:80–87, 2001

23. May C, Gask L, Atkinson T, et al: Resisting and promoting new technologies in clinical practice: the case of telepsychiatry. Social Science and Medicine 52:1889–1901, 2001

24. Bashshur RL: Telemedicine effects: cost, quality, and access. Journal of Medical Sys- tems 19:81–91, 1995

25. Jones BN 3rd, Ruskin PE: Telemedicine and geriatric psychiatry: directions for fu- ture research and policy. Journal of Geri- atric Psychiatry and Neurology 14:59–62, 2001

26. Bashshur RL, Sanders JH, Shannon GW: Telemedicine: Theory and Practice. Springfield, Ill, Charles C Thomas, 1997

CCoorrrreeccttiioonn

The article “Mental Health Services Received by De- pressed Persons Who Visited General Practitioners and Family Doctors,” by Wang et al., in the June 2003 issue (pages 878–883) contained an error in Table 1 (page 880). The asterisked footnote should read p>.05, not p<.05. The corrected version of the table is available on the journal’s Web site at http://psychservices.psychiatry online.org.