Original Research Cost Effectiveness, Safety, and Satisfaction with Video Telepsychiatry Versus Face-To-Face Care in Ambulatory Settings

TELEMEDICINE AND e-HEALTH Volume 12, Number 5, 2006 © Mary Ann Liebert, Inc.

Original Research Cost Effectiveness, Safety, and Satisfaction with Video Telepsychiatry versus Face-to-Face Care in Ambulatory Settings

ILAN MODAI, M.D., M.H.A,1,2 MAHMOUD JABARIN, M.D.,1 RENA KURS, B.A.,3 PERETZ BARAK, M.D.,1 ILAN HANAN, M.D.,1 and LUDMILA KITAIN, M.D.1

ABSTRACT

Videoconference telepsychiatry provides an alternative for the psychiatric treatment of mental health patients who reside in remote communities. The objective of this study was to compare institutional ambulatory and hospitalization costs, treatment adherence, patient and physician satisfaction, and treatment safety between mental healthcare via videoconferencing and care pro- vided in person. Data collected for 1 year of telepsychiatry treatment was compared to that of the preceding year and a matched comparison group. Twenty-nine patients from Or Akiva and 20 patients from Reut Hostel in Hadera who met the inclusion criteria agreed to participate; 24 and 15 patients, respectively, completed the study. Forty-two matched patients, who continued face-to-face interviews, comprised the comparison group. Drop-out patients and those who did not consent to telepsychiatry treatment were not involved. During the year of telepsychiatry treatment, patients and physicians were satisfied and treatment was safe. However, 1 hour of telepsychiatry treatment was more expensive than face-to-face care, and a tendency of increased hospitalizations was noted. Adherence ratios before and during telepsychiatry treatment were similar, but were twice as high versus the comparison group. The limited sample size precludes the drawing of definite conclusions, and further studies involving a larger study population and longer duration of investigation is warranted.

INTRODUCTION included questionnaires for the evaluation of telepsychiatry3 and more recent studies dem- EW TELECOMMUNICATION TECHNOLOGIES onstrated patient and physician satisfaction.4–6 Nhave already influenced the spatial and In their systematic review of patient satisfac- temporal relationships between health profes- tion with telehealth care Williams et al.7 re- sionals and patients.1 Studies have shown that ported levels of satisfaction greater than 80% telepsychiatry offers a viable alternative for ser- and frequent reports of 100% satisfaction. Stud- vice delivery to patients who live in distant ies from the late 1990s found that diagnosis and rural areas.2 The 1980s introduced studies that treatment via telepsychiatry was as reliable and

1Shaar-Menashe Mental Health Center, Mobile Post Hefer, Israel. 2Ruth and Bruce Rappaport Faculty of Medicine, Technion, Haifa, Israel. 3Lev-Hasharon Mental Health Center, Netanya, Israel.

515 516 MODAI ET AL. efficient as face-to-face interviews, with high SMMHC. Transportation costs and travel-asso- interrater reliability in the diagnosis of people ciated hardships of OA patients account in part suffering from obsessive-compulsive disor- for the occasional difficulties in adherence to der,8 anxiety and depressive disorders,1 schizo- follow-up care. phrenia,9 and paranoid schizophrenia with We sought to evaluate the cost of ambula- no exacerbation of delusional symptoms.3 tory care and hospitalization, treatment adher- Telepsychiatry has also been successfully used ence, clinical safety, and patients’ and thera- for treatment of posttraumatic stress disorder10 pists’ satisfaction in two VCTP settings in and for the treatment of depression.11 comparison to face-to-face care. More recent studies placed the emphasis on actual geographic distance between the exam- iners and the patients and found that patients expressed equal satisfaction and ability to de- MATERIALS AND METHODS velop a relationship in both in-person and Subjects telepsychiatric interviews.1,12 Since 1960, the use of telepsychiatry has Patients were over 18 years of age, spoke He- proven beneficial in solving the problem of lack brew, and had at least 1-year of face-to-face of adequate psychiatric services in remote com- treatment before the initiation of the study. One munities. Many projects of telepsychiatric ser- hundred four patients from the towns of OA vices have been described in the literature. The and Reut Hostel in Hd sought help from most expansive project was undertaken in SMMHC ambulatory clinic during the trial pe- southern Australia, and began in 1994.13 riod. Twenty-nine patients from OA and 20 pa- Telepsychiatry was implemented to provide a tients from Reut Hostel in Hd who met the in- solution for a broad spectrum of psychiatric clusion criteria agreed to replace face-to-face services for community-based clients scattered interviews with VCTP and participated in the across 1 million square kilometers. More than study; 24 and 15 patients, respectively, com- 2,000 consultations–emergency services and li- pleted the study. Forty-two matched patients aison services for inpatients, and follow-up vis- who did not participate in VCTP treatment its were administered during the first 4 years comprised the comparison group. Five patients of the project. from OA and 5 from Hd dropped out. In OA, Freuh et al.14 conducted a survey of telepsy- 2 withdrew consent, 2 relocated, and 1 discon- chiatry that dealt mainly with administration tinued after hospitalization. In Hd, 1 patient of psychiatric services in geographically distant committed suicide while hospitalized in a gen- communities and reported a lack of empirical eral hospital and 4 relocated. Patients who did studies of telepsychiatry (i.e., clinical outcome not meet inclusion criteria for participation in studies and cost analysis). In a comprehensive the study and patients who dropped out from systematic review of the literature of the as- VCTP treatment were not included in the sessment of telemedicine, Roine et al.15 con- analyses. All participants met International Sta- cluded that while relatively convincing evi- tistical Classification of Diseases and Related Health dence of effectiveness was found for Problems (ICD-10) diagnostic criteria for major telepsychiatry, evidence regarding the effec- psychiatric disorders. The Institutional Review tiveness and cost-effectiveness of telemedicine Board (IRB) of SMMHC approved the study. is still limited. The participants retained the option to return We opened the first two satellite videocon- to face-to-face care. ference telepsychiatry (VCTP) clinics in Israel, Demographics of the VCTP and comparison in Or Akiva (OA) and Hadera (Hd), located 20 groups are presented in Table 1. The groups and 15 km, respectively, from the Shaar- matched for age, gender, education, family sta- Menashe Mental Health Center (SMMHC) am- tus, and diagnoses. The social counselor from bulatory clinic. Prior to the study, the treating OA and the social worker from Reut Hostel psychiatrist traveled to Hd to see patients. OA who served as case managers accompanied the patients came to the ambulatory clinic at patients during their sessions. VIDEO TELEPSYCHIATRY VS. FACE-TO-FACE CARE 517

TABLE 1. DEMOGRAPHICS OF VIDEOCONFERENCE TELEPSYCHIATRY (n 39) AND COMPARISON (n 42) GROUPS

VCTP group Comparison group Statistics

Gender M 26 M 22 2 1.709 F 13 F 20 df 1 p 0.191 Age (y) 44.64 9.48 46.21 13.60 df 79 t 0.607 p 0.546 Family status Bachelors 17 Bachelors 15 2 0.607 Married 13 Married 17 df 3 Divorced 8Divorced 9 p 0.895 Widowed 1 Widowed 1 Education (y) 10.15 3.34 10.54 7.07 df 79 t 0.324 p 0.747 Diagnosis Schizophrenia 33 Schizophrenia 27 2 6.030 AD 2 AD 3 df 4 Anxiety 2 Anxiety 2 p 0.197 Dementia 1 Dementia 5 Organic PD 5

VCTP, videoconference telepsychiatry; AD, affective disorders; PD, personality disorders.

Instruments tocol (IP) version 3 for IP and Integrated Ser- vices Digital Network (ISDN) (VCON Ltd., Safety was monitored using the Brief Psy- Ra’anana, Israel). ISDN communication lines chiatric Rating Scale (BPRS)16 and the Clinical connected the main clinic to the two satellites Global Impression Scale (CGI)17 at baseline and and 29-inch television monitors were used for every 3 months thereafter. viewing. Satisfaction was measured at the 3-, 6-, 9-, and 12-month visits using the Patient Satisfac- tion Questionnaire (PSQ)-completed by pa- Statistical analyses tients and Therapist Satisfaction Questionnaire Analysis of variance ANOVA with re- (TSQ)-completed by therapists, modifications peated measures was used for the compari- of the Missouri Telehealth Network18 satisfac- son of BPRS, CGI, and satisfaction question- tion questionnaire. Answers for both questionnaires, two-tailed t test was used to compare naires ranged from 1, (not satisfied) to 5 (ex- duration of hospitalizations. The adherence tremely satisfied). The PSQ included the ratio was the mean number of visits/mean following items: general impression, the ability number of missed visits. t Test or 2 was to contact the therapist, the ability of the ther- used for the comparison of demographic apist to diagnose, feeling of comfort, the level data. of discomfort and nervousness, accessibility, the quality of treatment, similarity to face-to- face interview, the quality of equipment, will- RESULTS ingness to use again and the sense of therapist’s presence. TSQ included similar items aside Results are presented in Tables 2, 3, and 4. from “the sense of comfort” and “nervous- Adherence ratios were twice as high in the ness.” VCTP group as in the comparison group but similar within the groups. Mean hospitaliza- Videoconference telepsychiatry materials tion days were not significantly increased (2.49; VCTP facilities were set up near the town 8.28; p 0.05) during VCTP as well as in the center of OA and in Reut Hostel located in Hd. parallel periods of the matched comparison The equipment used was Falcon Internet Pro- group (12.45; 28.59; p 0.05). 518 MODAI ET AL.

TABLE 2. PATIENTS’ ADHERENCE RATIO AND NUMBER OF HOSPITALIZATION DAYS FOR ONE YEAR PRIOR TO AND FOR THE YEAR-LONG DURATION OF VIDEOCONFERENCE TELEPSYCHIATRY TREATMENT COMPARED TO FACE-TO-FACE TREATMENT

VCTP group (n 39) Comparison group (n 42)

Before During Before During

Adherence ratioa 11.05/2.15 5.13 12.33/2.66 4.63 16.74/8.38 1.99 15.33/6.52 2.35 No of hospitalization days 2.49 10.33b,c 8.28 24.29b,d 12.45 32.20d,e 28.59 101.15c,e (Mean SD)

aMean number of visits/mean number of missed visits. bp 0.19. cp 0.38 dp 0.062. ep 0.270. SD, standard deviation.

Average institutional costs for hospitaliza- ing interview, an item not generally included tion were 223.7% higher during VCTP treat- in satisfaction scales also showed significant ment and 132.5% during the parallel periods of improvement in patients’ and therapists’ eval- the matched comparison groups. The costs of uations during the study (df 3,172; F 5.81, an hour-long session were 32% higher during p 0.001; df 3,172; F 15.93, p 0.0001, re- VCTP treatment. When travel expenses of OA spectively). patients for the year preceding VCTP were in- As presented in Table 4, VCTP treatment was cluded in calculations, VCTP hour costs were clinically safe; BPRS scores significantly de- higher by only 10.6%. creased during the study and CGI scores re- As expected, patients and therapists were mained stable. generally satisfied (df 3,172; F 9.38, p 0.0001; df 3,172; F 12.69, p 0.0001, respectively). All PSQ final scores significantly DISCUSSION improved (p 0.05) and were above 4 except treatment quality (df 3,172; F 2.69, p Institutional costs (hospitalizations and ses- 0.05), satisfaction with equipment (df 3,172; sions) are higher during VCTP. Ruskin et al.11 F 0.74, p 0.05) and nervousness (df 3,172; found that the cost of remote treatment was F 1.47, p 0.05), which remained stable. In equal to that of in-person treatment if the psy- addition, all TSQ final scores were above 4 and chiatrist had to travel 35.2 km. In a more dis- significantly improved during the study (df tant facility, VCTP treatment was even less ex- 3,172; p 0.0001). The sense of presence dur- pensive than in-person treatment. In the

TABLE 3. COMPARISON OF COSTS FOR ONE YEAR PRIOR TO AND FOR THE YEAR-LONG DURATION OF VIDEOCONFERENCE TELEPSYCHIATRY TREATMENT

Telepsychiatry (n 39) Comparison group (n 42)

A year before A year during A year before A year during

Institutional operating costs $71.50 $94.40 $70.90 $70.90 per houra Total operating costs per hourb $85.30 $94.40 $84.70 $84.70 Mean cost of hospitalization $372.50 $1,205.80 $1,842.20 $4,282.80 per patient

aSalaries, psychiatrist travel expenses, running costs, phone expenses, equipment depreciation (10-year basis). bPlus patients travel expenses. VIDEO TELEPSYCHIATRY VS. FACE-TO-FACE CARE 519

TABLE 4. MEAN ( SD) SCORES OF BPRS AND CGI AT EACH TIME POINT OF THIRTY-NINE PATIENTS DURING TELEPSYCHIATRY TREATMENT

Baseline 3 months 6 months 9 months Year Statistics

BPRS 32.04 5.68 30.67 5.74 28.93 5.99 27.69 6.12 26.18 6.37 df 4,222; F 5.90, p 0.0005 CGIs 4.31 0.81 4.3 0.81 4.33 0.87 4.27 0.87 4.08 0.87 df 4,220; F 0.60, p 0.66 CGIc — 3.83 0.59 3.71 0.62 3.71 0.62 3.36 0.62 df 3,172; f 3.98, p 0.83

SD, standard deviation; BPRS, Brief Psychiatric Rating Scale; CGI, Clinical Global Impression Scale. present study, the distance between the main were more meticulous about their appearance, clinic and satellites was less than 35.2 km (15 and second, the interviews were more orga- to 20 km), nevertheless, costs during VCTP nized, with patient and doctor more courteous treatment were increased most probably as a in that they did not interrupt one another. result of the fact that travel expenses affected This study had several limitations. The fact the therapist in Hd and the patients in OA. that the comparison group included patients Hospitalization rates and costs increased that did not consent to VCTP treatment might (not significantly) during VCTP treatment and account for inherent differences between the in the parallel periods within the comparison comparison and the VCTP groups. The small group. Because only a few patients were hos- number of hospitalized patients and study pitalized and the study groups were relatively groups might account for the statistically in- small, results were statistically not significant significant differences in hospitalization days and should be considered with caution. between the investigated groups. The cost anal- Adherence ratios were twice as high in the ysis is limited to the presented procedures; out- VCTP group as a whole than those of the com- reach and in-person clinics, items included in parison group but did not change within groups. calculations, and the setting of a governmental The differences in adherence ratios between institution, and might differ when other pro- VCTP and comparison groups might arise from cedures or psychiatric settings such as private methodology; it might be that the compliance of institutions are evaluated. those patients who agreed to participate in VCTP was better than those who refused. VCTP in the treatment of major psychiatric CONCLUSION disorders is known to be safe and both therapists and patients have expressed satisfac- Patients and physicians are satisfied with tion.1,4–6,9,11,12 VCTP, treatment is safe and effective, and ad- Ruskin et al.11 reported that telepsychiatry herence to treatment remains stable. However, was effective in the ambulatory treatment of operational costs of VCTP may be higher than 119 depressed outpatients as measured by de- face-to-face care, and a tendency of increased pression, anxiety, global, clinical and health hospitalization costs was noted. The limited sam- scales. Medication adherence and patient satis- ple size precludes the drawing of definite con- faction were also positive. clusions, and further studies involving a larger In the present study, both therapists and pa- study population and longer duration of inves- tients experienced “presence,” a fact that may tigation is warranted. account for the ability to establish positive relationships between them,1,12 and overcome detachment that might be created while speak- ACKNOWLEDGMENTS ing to a faceless electronic device. Two unique phenomena associated with The research was made possible with a grant VCTP sessions were observed. First, patients from the Galil Center for Telemedicine and 520 MODAI ET AL.

Medical Informatics, Technion, Haifa. The au- 10. Deitsch SE, Frueh BC, Santos AB. Telepsychiatry for thors acknowledge the assistance of Mrs. Iris post-traumatic stress disorder. J Telemed Telecare Cohen in data collection. 2000;6:184–186. 11. Ruskin PE, Silver-Aylaian M, Kling MA, Reed SA, Brad- ham DD, Hebel JR, Barrett D, Knowles F 3rd, Hauser P. Treatment outcomes in depression: Comparison of REFERENCES remote treatment through telepsychiatry to in-person treatment. Am J Psychiatry 2004;161:1471–1476. 1. May C, Gask L, Atkinson T, Ellis N, Mair F, Esmail 12. Rohland BM. Telepsychiatry in the heartland: If we A. Resisting and promoting new technologies in clin- build it, will they come? Community Ment Health J ical practice: The case of telepsychiatry. Soc Sci Med 2001;37:449–459. 2001;52:1889–1901. 13. Kavanagh S, Hawker F. The fall and rise of the South 2. Aas IH. Telemedical work and cooperation. J Telemed Australian telepsychiatry network. J Telemed Telecare Telecare 2001;7:212–218. 2001;7(Suppl 2):41–43. 3. Dongier M, Tempier R, Lalinec-Michaud M, Meunier 14. Frueh BC, Deitsch SE, Santos AB, Gold PB, Johnson D. Telepsychiatry: Psychiatric consultation through MR, Meisler N, Magruder KM, Ballenger JC. Proce- two-way television: A controlled study. Can J Psychi- dural and methodological issues in telepsychiatry re- atry 1986;31:32–34. search and program development. Psychiatr Serv 4. Stevens A, Doidge N, Goldbloom D, Voore P, 2000;51:1522–1527. Farewell J. Pilot study of televideo psychiatric as- 15. Roine R, Ohinmaa A, Hailey D. Assessing telemedi- sessments in an underserviced community. Am J Psy- cine: S systematic review of the literature. CMAJ chiatry 1999;156:783–785. 2001;165:765–771. 5. Simpson J, Doze S, Urness D, Hailey D, Jacobs P. 16. Overall JE, Gorham DR. The brief psychiatric rating Telepsychiatry as a routine service—The perspective scale (BPRS). Psychol Rep 1962;10:799–812. of the patient. J Telemed Telecare 2001;7:155–160. 17. Guy W. Clinical Global Impression (CGI). National 6. Simpson J, Doze S, Urness D, Hailey D, Jacobs P. Institute of Mental Health. In: ECDEU assessment man- Evaluation of a routine telepsychiatry service. J ual for psychopharmacology. Rockville, MD: 1976:217– Telemed Telecare 2001;7:90–98. 222, 1976. 7. Williams TL, May CR, Esmail A. Limitations of pa- 18. The Lewin Group Inc. Missouri Telehealth Network, tient satisfaction studies in telehealth care: A system- 2003. www./telehealth.muhealth.org/eval, Last accessed atic review of the literature. Telemed E-Health December 28, 2004. 2001;7:293–316. 8. Baer L, Cukor P, Jenike MA, Leahy L, O’Laughlen J, Coyle JT.. Pilot studies of telemedicine for patients with obsessive-compulsive disorder. Am J Psychiatry Address reprint requests to: 1995;152:1383–1385. Professor Ilan Modai 9. Yoshino A, Shigemura J, Kobayashi Y, Nomura S, Shaar-Menashe Mental Health Center Shishikura K, Deor R, Wakisaka H, Kamata S, Ashida Mobile Post Hefer, 38814 H. Telepsychiatry: Assessment of televideo psychi- Israel atric interview reliability with present- and next-gen- eration Internet infrastructures. Acta Psychiatr Scand 2001;104:223–226. E-mail: [email protected]