Objective: The authors conducted a review and meta-analysis of the literature comparing telepsychiatry with “in-person” psychiatric assessments.Method: Approximately 380 studies on telepsychiatry published between 1956 and 2002 were identified using MEDLINE, PsycINFO, and cross-referenced bibliographies. Of these, 14 studies with an N>10 compared telepsychiatry with in-person psychiatry (I-P) using objective assessment instruments or satisfaction instruments. Three of these studies compared high bandwidth (HB) with low bandwidth (LB) telepsychiatry.Results: Fourteen studies of 500 patients met inclusion criteria and were included in the meta-analysis. Telepsychiatry was found to be similar to I-P for the studies using objective assessments. Effect sizes were on average quite small, suggesting no difference between telepsychiatry and I-P. Bandwidth was found to be a significant moderator. Three moderators were tested, effect sizes remained largely heterogeneous, and further analyses are needed to determine the direction of effect. There was no difference between I-P and telepsychiatry between the HB and LB groups, although there are anecdotal data suggesting that H B was slightly superior for assessments requiring detailed observation of subjects.Conclusion: Out of a large telepsychiatry literature published over the past 40+ years, only a handful of studies have attempted to compare telepsychiatry with I-P directly using standardized assessment instruments that permit meaningful comparisons. However, in those studies, the current meta-analysis concludes there is no difference in accuracy or satisfaction between the two modalities. Over the next few years, we expect telepsychiatry to replace I-P in certain research and clinical situations.

S: 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-person psychiatric services. Whether governmental or private health agencies value telepsychiatry enough to assume its cost is a question that remains to be answered.

Telepsychiatry in the 21st century poses a wide range of legal and ethical challenges. The authors review issues related to licensure, credentialing, privacy, security, confidentiality, informed consent, and professional liability in the use of telepsychiatry services and illustrate the discussion with hypothetical clinical vignettes. It is clear that there will be a need in the immediate future to create legal instruments as well as formal professional ethical guidelines for the practice of telepsychiatry.

Numerous studies have assessed the acute efficacy of antidepressants, including selective serotonin reuptake inhibitors, in treating dysthymic disorder; however, escitalopram, the S-enantiomer of citalopram, has not been studied. Thirty-six outpatients with Structured Clinical Interview for DSM-III-R-diagnosed dysthymic disorder, aged 23-65 years (mean±SD=44.7±11 years), were randomly assigned to double-blind escitalopram (maximum dose 20 mg/day) versus placebo for 12 weeks. Inclusion criteria included age 18-65 years and Hamilton Depression Rating Scale (HDRS) score≥12. We hypothesized that escitalopram would be superior to placebo in the HDRS-24 item total score at week 12. We also hypothesized the superiority of escitalopram over placebo for secondary measures, including the percentage of participants classified as responders and remitters, as well as social functioning (Social Adjustment Scale), clinical global impression-improvement, Global Assessment of Functioning Scale. Participants' baseline HDRS-24 averaged 23.4±5.9. Final HDRS-24 scores at last observation carried forward did not differ significantly between escitalopram-treated (mean±SD=10.88±5.83) and placebo-treated individuals (mean±SD=16.4±6.34) (F=2.82, degrees of freedom=1,32, P=0.10). Significant differences favoring active medication were found on the Social Adjustment Scale and the Clinical Global Impression Severity and Global Assessment of Functioning Scale, but not in the percentages of responders or remitters. A larger study sample or higher escitalopram dose may show more significant placebo-medication differences.