COVID-19 has changed the field of psychotherapy overnight from in-person to “virtual,” remote teletherapy. This shift will likely have lasting effects on psychotherapy practice. What had been primarily an adjunctive therapeutic approach for patients in geographically isolated settings who lacked access to in-person care is suddenly, thanks to mandated “social distancing,” the standard mental health care intervention. Teletherapy can use traditional telephones, smartphones and therapy-related apps, Internet video calls on platforms that are compliant with Health Insurance Portability and Accountability Act rules (e.g., Zoom), or online computer–mediated treatment programs. There seems to be no question that patient access to clinicians is better than no access, and telephone or video therapy seems undoubtedly preferable to more detached media such as texting for a human encounter.
As psychotherapy researchers and therapists, we all had prior experience in conducting telepsychotherapy, but in recent weeks we have rapidly gained a more nuanced appreciation of its strengths and weaknesses through ongoing practice. A brief literature review unearthed several reviews of studies (
1–
10) but relatively little discussion of the trade-offs of teletherapy relative to in-person treatment. We first summarize the empirical background, then focus on the technical approach, clinical benefits, and difficulties that teletherapy presents. We restrict our focus to telephone and video therapies, which are both the basis of our own experience and the direction in which the field has overwhelmingly shifted since the COVID-19 outbreak.
Evidence Base
Although our focus in this article is on clinical issues in delivering teletherapy, we would be remiss to ignore its empirical support. Whereas more than a thousand in-person psychotherapy trials have been published for depression alone (
11), the newer field of teletherapy is less developed (
12). Much reported “teletherapy” is not tele-
psychotherapy. Many teletherapy studies reported in reviews (
2–
4,
7) are not traditional efficacy or effectiveness trials, but instead document heterogeneous outcomes, such as reduced travel time for rural patients, psychoeducation, or addition of a video to standard care to reduce depressive symptoms (
3). Hilty et al. noted that before 2013 there had been little assessment of treatment effectiveness in the usual sense (
3).
Outcome data are generally encouraging but sparse, and come with caveats. For example, in 2016 Leach and Christensen (
1) reviewed 14 studies spanning a range of psychiatric disorders, treatment approaches, and selected populations. They describe mostly positive outcomes of telephone therapy, generally compared with no treatment or with treatment as usual. The latter, in the fractured U.S. mental health care system, often amounts to nocebo and hence is a weak comparator (
13). While some of the 14 studies are small pilot trials, others treated hundreds of patients (
1), underscoring the potential reach of teletherapy.
A few rigorous trials have demonstrated the effectiveness of telephone cognitive-behavioral therapy (CBT) (
14) and telephone interpersonal psychotherapy (
15,
16) in reducing depressive symptoms. Simon and colleagues (
14) found that eight sessions of telephone CBT enhanced usual care for primary care patients (N=600) who were starting antidepressant pharmacotherapy. Heckman et al. (
15) found that nine sessions of telephone interpersonal psychotherapy produced long-term reductions in depressive symptoms, more than usual care, for depressed rural patients living with HIV (N=132), and Dennis et al. (
16) showed that 12 sessions of nurse-delivered telephone interpersonal psychotherapy reduced depressive symptoms more than usual care in postpartum mothers (N=241). In a 2008 meta-analysis, Mohr and colleagues (
5) found that telephone therapy significantly reduced depressive symptoms across 12 trials (mean effect size, d=0.81 in pre- to posttreatment reduction in symptoms), with a low mean attrition rate of 7.6%. Eight of these were CBT trials. Mohr et al. do not report mean depressive symptom severity.
Evidence for the efficacy of videotherapy, which has now taken the field by storm, is less robust than that for telephone therapy. Berryhill et al. (
8) reviewed 33 videotherapy depression studies, which in fact comprised mixed psychiatric diagnoses, with only nine focusing primarily on depression; 14 of the studies were randomized, and sample sizes ranged from 1 to 243, with median effect sizes ranging from medium to very large for reducing depressive symptoms. Treatment types and comparisons varied: CBT was the most common modality, and studies tended to find equivalent outcomes for in-person and video CBT. The quality (
17) of studies appears highly variable.
For telephone-administered group therapy, the limited number of randomized controlled trials, their mixed findings, and their disproportionate focus on people living with HIV preclude conclusions about general efficacy. In perhaps the largest group teletherapy trial, Heckman and colleagues (
18) assigned 361 HIV-infected older adults to 12 weekly sessions of telephone-administered supportive-expressive group therapy, 12 weekly sessions of telephone-administered coping effectiveness group training, or a standard-of-care control condition. The supportive-expressive group therapy participants reported significantly fewer depressive symptoms at 8-month follow-up compared with the coping effectiveness group training and standard-of-care participants. Heckman and Carlson (
19) tested the depression efficacy of an 8-week telephone coping group compared with a telephone information-support group or standard care. Neither telephone group intervention was found to reduce depressive symptoms.
One non-HIV randomized controlled trial compared telephone-administered to in-person group cognitive-behavioral stress management for 100 patients with chronic fatigue syndrome (
20). While both patient groups reported significantly reduced perceived stress, the effect size was large for in-person treatment and medium for telephone-administered treatment. While telephone-administered group therapies can potentially reach large numbers of patients with relative convenience and cost benefits, the approach presents many challenges. This medium requires speakers to identify themselves by name and to indicate when they have finished speaking. Heckman’s telephone-administered supportive-expressive group therapy patients had 8% overall attrition, but on average attended only 7.4 of 12 sessions (62%) and the telephone-administered coping effectiveness group training patients attended 6.4 of 12 sessions (53%), despite numerous e-mail and telephone appointment reminder messages (
18). Perhaps patients felt that the remote group telephone format would obscure their skipping sessions.
Group videotherapy is still less developed. Scattered trials have been published (
9,
10). A 2019 review (
10) found six randomized trials of differing approaches and disorders, which suggested feasibility and roughly comparable outcomes between in-person and tele-group treatments. The reviewers noted, however, that the studies were insufficiently powered to test noninferiority (
10).
Beyond the limited empirical data, thinly spread across therapies and diagnoses, stand broader concerns. Caution about using remote therapy has affected research study design. Because many therapists (and institutional review boards) have felt less comfortable treating high-risk patients remotely than in-person, most teletherapy outcome trials have limited the symptom severity of enrolled patients. Less severely symptomatic patients have a greater chance of response to any treatment, including placebo (
21). Furthermore, teletherapy studies may have selected for patients and therapists who prefer teletherapy. Thus the extant research has limited generalizability to the present situation: selection of less ill, tele-friendly patients may imply more generous outcomes than apply to sicker, not invariably tele-comfortable patients whom therapists are now perforce treating remotely.
Probably because of the same caution about remote care, teletherapy has often been studied as an adjunct to or augmentation of treatment as usual. This raises the question of its efficacy when it becomes treatment as usual. Overall, the available research presents a fragile foundation for the broad treatment edifice of telepsychotherapy and naturally occurring public health experiment it must now support. Moreover, insurance plans, when they have reimbursed remote therapy at all, have arbitrarily tended to pay only for synchronous videotherapy, even though no compelling evidence indicates its superiority to (audio-only) telephone remote therapy.
Doing Remote Psychotherapy
Many clinicians may currently care less about the research base of teletherapy than about how to adopt and adjust to this new medium. The great strength of remote therapy is that it expands access: the great majority of Americans have access to a telephone or computer, a claim teletherapy advocates have touted (
22,
23). Yet recent reports suggest that many at-risk populations, including poorer and elderly Americans, lack high-speed Internet access (
24–
26). At least one of our formerly homeless patients declined to continue therapy even by telephone because sessions would have cost him precious billed minutes. Several others lacked any private space to speak away from difficult family members.
In reaching patients, however, remote therapy requires important adjustments, on several levels. As therapists and supervisors, we sense great differences treating patients with psychotherapy by webcam rather than in person. Some of this may reflect the sudden, drastic switch to videotherapy and may change as therapists adjust over time (
27).
Setting
Many clinics that have practiced telepsychiatry, including ours (
28), had required at least one initial in-person visit to evaluate the patient and develop a therapeutic alliance before continuing treatment remotely. That is no longer practical or safe: now treatment is distanced from the start. This may subtly alter the therapeutic relationship. Nor can one offer a tearful patient a tissue as a mute acknowledgment of pain. Patients who began therapy in-person and had anticipated that it would continue in that format can find the adjustment to remote therapy “weird” (as several said) and discomfiting, although they soon seem to settle in. Therapists may, too.
Maintaining a consistent intimate focus is more difficult. The patient is no longer in the room but on a screen (or a telephone). Instead of two human beings fully engaging in a common space, one meets an image of a patient on a computer screen (or a disembodied voice) surrounded by too many distracting stimuli. Although studies indicate that good therapeutic alliance and psychotherapeutic common factors can be established in remote therapies (
29–
31), they may reflect selective, enthusiastic therapist and patient samples.
Distractions abound. The usual instruction to patients is to find a private, quiet space where they are unlikely to be overheard or interrupted, but that is not always possible, particularly for less privileged patients under lockdown. People and pets walk in. Outside noises distract. Even if they do not, the screens themselves teem with diversions. Because the computer volume is on (sometimes set very loud) to allow therapist-patient interchange, the frequent ping of arriving e-mail occurs at both venues. We have seen patients scanning the screen as if reading an e-mail, rather than making eye contact. Eye contact itself is tricky: if the patient is addressing the computer camera lens, making virtual eye contact, he or she may not be looking at your image; and vice versa. Thus the patient’s gaze may be misleading. Your own image on your screen is an anomalous presence: you or your patient may be looking at yourselves rather than each other.
There is the visual risk of viewing each other as talking heads rather than having the office experience of seeing one another in full view, permitting assessment of nonverbal behaviors. Proximity to the keyboard raises the temptation to check e-mail. Maneuvers (
Table 1) such as distancing yourselves from the camera allow a purer focus on the patient. These maneuvers may reduce unhelpful sensory stimuli. On the other hand, too much distance from the microphone can hurt sound quality, and headphones can be obtrusive. The issues video treatment raises suggest that telephone therapy might present fewer distractions, although at the therapeutic cost of nonverbal cues, particularly for those many patients unable to easily express their feelings in words.
Remote therapy grants the therapist revealing glimpses of a patient’s home and life. This may include meeting pets and babies, and seeing elements in the surroundings that the patient might not think to mention. One patient appeared in her childhood bedroom, where a devout religious icon hung on the wall. She had mentioned having been raised in a “kind of religious, Catholic” home, but the camera brought her mother’s consuming devout piety, and graphic reminders of the strictures this imposes on the patient’s life, into sharp relief. Other patients could only find private space in a bathroom, on the stairs of their apartment building, or outside in a park. Most patients seem not to mind allowing their therapists into their homes, although those preoccupied with their outward appearance, hoarders ashamed of their household interiors, and some mistrustful patients with social anxiety disorder or posttraumatic stress disorder (PTSD) have requested telephone rather than video sessions. Anecdotal conversations with colleagues suggest that more than half of their patients prefer standard telephones to videophones for remote therapy purposes. One patient personalized her screen background to surround herself with a family portrait. A therapist noted that some patients have been conditioned to feel a work ethic in front of their computer screen, and seem more relaxed on the telephone. Just as good therapists offer patients informed consent and a choice of treatment modality, remote therapists might offer patients a choice of treatment medium: i.e., telephone or video.
For their part, some therapists feel odd about treating patients from the therapists’ own personal spaces, such as a bedroom, to which a crowded house may confine them. It may be important to precheck the camera frame so as to avoid unwanted, inadvertent self-disclosure of personal home details. Whereas many therapists do not miss commuting to work, they note in retrospect that it provided time to decompress and think through the progress of treatments before rejoining private or family life. That transitional buffer may no longer exist when one works from home. It may help to allot time to reflect before and after treatment sessions to ease the shift to domestic life. During the COVID-19 crisis, however, many people lack extra time as details of domestic life have become more burdensome (
32).
Therapists worry about behavioral hazards of remotely treated patients whom they may never have met in person. Therapists have limited control over outpatients presenting in their offices, but distance magnifies concerns about suicide and violence risk.
Physical Discomfort
We all find remote psychotherapy physically and psychically more exhausting than the in-person variety. There are several apparent reasons. It is harder to stay focused on and more difficult to read the patient’s cues across the medium. Sitting before a screen constricts physical movement, including the subconscious mirroring movements in which patients and therapists sharing a space engage. Therapists feel rigidly locked before the camera, tensing different muscles. Multiple consecutive seated sessions can feel like a long-haul airline flight.
Transmission
Technical difficulties can impede communication or interrupt treatment sessions: difficulty connecting, frozen screens, “unstable Internet connection” warnings, garbled or delayed audio, poor lighting, dropped calls. Videotherapy has turned out to have confidentiality risks (
33). Time spent countering these inefficiencies means less time for engagement in therapy.
Emotional Distancing
There is a loss of affective nuance on telephone or screen, a factor that seems to bother therapists more than patients. We believe this affective diminution makes the experience less emotionally vibrant, particularly for patients with the psychological tendency to dissociate. Media separation makes it hard to gauge nonverbal behavioral subtleties (
34), such as when a patient with PTSD may be dissociating. A pause on the telephone can mean (too) many things. Although research has found that tele-exposure therapy benefits patients (
6,
35), it seems easier for patients to avoid exposure at geographic and interpersonal distance. In affect-focused psychotherapies, distance impedes emotional engagement with the therapist in the moment, which is key to the process of change.
Patients who participate in teletherapy in the familiar “safety” of their home, particularly those with anxiety, panic, and agoraphobia, may underreport symptoms likely to be present (or activated) when presenting in clinical venues. Thus physical remoteness appears to aggravate these patients’ avoidance of uncomfortable affects and experiences.
Recommendations
Teletherapy thus presents difficulties. In response, we suggest some partial remedies (
Table 1).
The Pandemic
It would deny reality to pretend that current teletherapy is therapy as usual. The COVID-19 pandemic is a world crisis, and it tends to aggravate underlying anxiety in at least three ways: 1) by evoking appropriate fears of contagion, which may rapidly merge into panic attacks (anxiety as signal versus symptom [
36]); 2) by disrupting the comfortable structure and rhythm of the patient’s (and therapist’s) work and life schedule, often including where and with whom they are living, and sources of income and relaxation; and 3) through physical distancing, which stretches attachment bonds and risks loss of social support (
37). Truncated public transport and bans on longer-distance travel have sometimes made what once were trivial distances between patients and significant others impassable. As the pandemic has persisted, initial panic in the general population seems to be giving way to frustration, despondency, and depression for many, with concern that suicide risk may be increasing (
38). As in previous pandemics (
39) and disasters (
40), highly exposed groups such as medical teams, first responders, and the bereaved may well present with lingering PTSD and complicated grief.
The pandemic not only evokes new symptoms but functions as a Rorschach test, magnifying aspects of patients’ ongoing inner struggles. People respond to crisis in varying, idiosyncratic ways. Embarrassing behavioral differences between severely agoraphobic and social phobic patients and “normal people” have temporarily shrunk. Some depressed patients have become more depressed, whereas others say that the crisis has led them to downgrade their previous concerns—that COVID-19 has been de-catastrophizing, as it were. A severely symptomatic veteran with PTSD who had been in productive, exploratory, trauma-focused psychodynamic psychotherapy (
41), became distanced, repeatedly telling his therapist that all symptoms were “just the same,” that “nothing’s new really” once his treatment switched to videotherapy. He evidently said this in response to an unspoken or unconscious urge to protect her from his rage-filled fantasies and recurrent dreams after having learned early in the pandemic that she required isolation (earlier than other VA employees required it) because of an immune-compromising illness. This patient began to improve, and to use therapy more productively, only after the therapist pointed out this seeming attempt to protect her. The therapist had uncharacteristically avoided making this observation for several stymied weeks because of her own concerns about being less available to the patient than she would have been in person.
Recommendations
Therapists should acknowledge the crisis, and perhaps that teletherapy is a limited substitute for more direct contact. They can attempt to maintain the helpful structure of therapy by maintaining regular sessions and treatment approach. They should encourage patients not to let the physical distancing of “social distancing” impede their existing relationships and cost them protective social support (
37). Many relationships can be preserved either by remote means (Skype, Zoom, or Facetime) or by masked, six-feet-apart, “socially distanced” in-person walks.
Discussion
Several modalities of teletherapy can preserve the crucially important link of psychotherapy in a highly anxiety-provoking, socially burdened time of quarantine, physical distancing, and deep emotional need and despair. Teletherapy has some empirical backing, but the outcome literature is very limited relative to that of in-person treatment and has unclear generalizability to its broad current use among a wide range of patients with various serious psychiatric problems. Telepsychotherapy offers access and convenience during a time of unprecedented crisis, at the cost of important elements of in-person treatment. This is an early report among what will doubtless be many commentaries and studies on the post-COVID-19 world. At least in the United States, COVID-19 has changed the long-standing requirement that therapists see patients in person. This may well change the face of psychotherapy and enhance the ongoing use of teletherapy, whether or not it is an optimal approach. Our experience to date suggests that in-person psychotherapy, now necessarily in limbo, has many advantages over remote treatment and should eventually return.
Is videotherapy preferable to telephone therapy? The limited empirical research does not support this. Vision, being the dominant human sense, may have prejudiced insurance reimbursement requirements for visual patient tele-contact (when insurance has paid for videotherapy at all [
42]). Video has obvious advantages over audio for group therapy, but it may provide more distractions than a simple telephone call for some or many patients in individual therapy. Such preferences, which can affect treatment outcome (
43,
44), deserve study. Telephone therapy may also provide broader access to economically disadvantaged patients (
24–
26). Perhaps insurance should reimburse both media: teletherapy research sparked by COVID-19 could reveal what has always been an arbitrary insurance requirement to be an unnecessary one.
COVID-19 will eventually be contained, and the world will resume some new form of normalcy. Nonetheless, COVID-19 may continue to have ongoing effects on social closeness and on how (remotely) psychotherapy is practiced. The wave of viral contagion may pass, only to be followed by a wave of psychopathology (
45,
46). Inasmuch as previous, far more contained disasters have raised the incidence of anxiety and mood disorders, PTSD, and substance use, COVID-19 likely will as well.
This report has limitations. It relies on a small sample of therapist observations. It omits treatment of patients with severe mental illness, such as psychotic disorders, who are not the patients we primarily treat. A survey of two psychiatric Listservs elicited sometimes contradictory clinical anecdotes: some patients with schizophrenia miss telephone appointments; a few preferred telephone to video to diminish the intensity of sessions, and one patient only sends text messages. Further reports on treating this population (
47) and on tele-psychopharmacology are needed. Nor can this team of adult therapists provide data or impressions on conducting telepsychotherapy with children or adolescents, with or without parental involvement.
Future Directions
The scattered, relatively sparse research literature, which has largely addressed selective populations (e.g., rural HIV-positive or veteran patients), does not suffice to support the extent to which remote therapy, and particularly videotherapy, is now being used. Hopefully the National Institute of Mental Health, which in recent years has moved away from clinical research funding in its pursuit of neuroscience (
48), will recognize the need for immediate, rigorous clinical research on the use of a range of teletherapies for the broader range of psychiatric patients. Whether different types of therapy (e.g., affect-focused versus exposure-focused [
49]) have differing online benefits is unknown. Because no treatment benefits all patients, we recommend testing the benefits and limitations of a range of remote psychotherapies: CBT, interpersonal psychotherapy, psychodynamic psychotherapy, and perhaps others. Research on teletherapy dissemination and implementation science will be important to scale up and rapidly transition evidence-based teletherapies from the research arena to community settings.
Another historical complication of teletherapy has been that in the United States, each state required therapist licensure in the patient’s state of residence. This legal requirement greatly impeded a key strength of remote therapy, namely, its broad and relatively inexpensive geographic reach. In response to the pandemic, the federal government in March 2020 relaxed this requirement, allowing therapists to treat patients across state lines (
50). The authors hope that this freedom to cross state boundaries will continue after the pandemic passes.