Online Therapy in Dubai: What the Research Actually Says — A Clinical Reading of Five Papers

In Dubai and across the UAE, online therapy has shifted from a workaround to a first-choice format for many people. Expats working long hours, clients managing cross-emirate commutes, couples whose schedules rarely line up. That makes the question “is online therapy as good as in-person?” more than academic here. It shapes a decision people make in real life, often under time pressure and without much guidance.

The question has also become one of those that gets answered too quickly.

Yes, says one camp. The research supports it. Here’s a meta-analysis.

It depends, says another. Nothing replaces the room.

Both responses tend to flatten what is actually a complex evidence base. When you read it carefully, it is more nuanced, more honest about its own limits, and more clinically useful than either position suggests.

So I read five papers. Not abstracts – the actual studies, methods sections and all. What follows is what I found:

A note before we start

Most of what gets shared about online therapy research is either “it works just as well” or “we can’t be sure.” Both are technically true. Neither is particularly useful on its own.

What I wanted to understand is how it works, for whom, under what conditions, and where the evidence genuinely runs out. That’s a different question – and a harder one to answer with a single study.

These five papers approach it from different angles. There are meta-analyses with different methodologies, a real-world intensive treatment comparison, a deep look at the therapeutic alliance, and the only couples-specific study I could find. Together they give a picture I think is worth having.

I wrote earlier, in an opening piece on online therapy for Dubai and UAE clients, about the broader benefits and limitations of the format. This article sits underneath that one, looking at the evidence in more detail.

Paper 1 – Lin, Heckman & Anderson (2022): No significant difference in randomised trials

The efficacy of synchronous teletherapy versus in-person therapy: A meta-analysis of randomized clinical trials. Clinical Psychology: Science and Practice, 29(2), 167–178. https://doi.org/10.1037/cps0000056

This is the paper with the strictest methodology: a meta-analysis of randomised controlled trials only. From an initial pool of 1,393 studies (after de-duplication), 20 satisfied the inclusion criteria. That’s a deliberately narrow pool. Randomised designs control for self-selection effects. People who choose online therapy may differ from those who choose in-person, which would skew any comparison.

The finding: no significant difference in treatment outcomes at post-treatment (g = −0.043) or 3–6 month follow-up (g = −0.045). Attrition rates were also equivalent overall (odds ratio = 1.006). Effect sizes that close to zero point to no meaningful difference between conditions.

The caveats the headline misses

Read past the headline and the paper is more careful than it first appears:

• Videoconferencing therapy showed higher dropout than telephone therapy.
• Trainee therapists had significantly worse client attrition in teletherapy than licensed therapists.

The authors write, in their own conclusion: “It is premature to conclude that teletherapy is as efficacious as in-person therapy across all conditions, symptoms, and patient populations.”That sentence appears in the same paper people most often cite as proof of equivalence. I find that worth noting.

The evidence base here is largely pre-pandemic. It skews toward anxiety, depression, and PTSD, and is almost entirely CBT. So it doesn’t tell us much about psychodynamic or person-centred approaches online. Likewise, the literature says little about children and adolescents. And it doesn’t speak to presentations outside those diagnostic categories.

What it does tell us clearly, is that for the conditions and modalities studied, teletherapy holds up against in-person in RCT conditions. That’s meaningful. It just isn’t unlimited.

Paper 2 — Fernandez et al. (2021): Where the picture gets more textured

Live psychotherapy by video versus in-person: A meta-analysis of efficacy and its relationship to types and targets of treatment. Clinical Psychology & Psychotherapy, 28(6), 1535–1549. https://doi.org/10.1002/cpp.2594

Where Lin et al. stayed narrow, Fernandez and colleagues cast a much wider net. 103 studies – 56 within-group and 47 between-group – covering over 5,000 participants. Video therapy only (no telephone). That makes the modality more homogenous than Lin et al., but the evidence base much larger and more varied.

The between-group finding mirrors Lin: near-zero difference between video therapy and in-person (g = 0.01). What Fernandez adds is the subgroup breakdown.

The subgroup breakdown

Within-group effects — how much did people improve during video therapy on its own terms — were large for anxiety (g = 0.99), depression (g = 1.29), PTSD (g = 1.00), and notably high for OCD (g = 1.83). CBT outperformed non-CBT approaches (g = 1.34 vs g = 0.66). These are encouraging findings, particularly for the presentations that make up the bulk of outpatient practice.

Then there is the eating disorders finding: g = 0.07. Strikingly weak, and based on a small number of studies. It doesn’t prove that online therapy doesn’t work for eating disorders – the evidence base is too limited for that conclusion. But it’s a signal. And it rarely gets discussed in the way that the depression and anxiety findings do.

Publication bias and heterogeneity

Two other things matter here. First, the authors examined publication bias. A known problem in this literature: studies with positive results are more likely to make it into print. They applied a sensitivity correction. That correction brought the within-group estimate from g = 0.99 down to g = 0.54. Still significant, still meaningful, but a more conservative estimate of what is actually happening.

Second, the heterogeneity across studies was very high (I² = 88–94%). In plain English: the 103 studies varied enormously in their findings. The overall average conceals a lot of variation underneath. Context – setting, population, approach, therapist experience – matters enormously, and a single average statistic can’t capture it. It’s all about the detail and the nuance.

Paper 3 — Bulkes et al. (2021): Does telehealth hold up at higher clinical intensity?

Comparing efficacy of telehealth to in-person mental health care in intensive-treatment-seeking adults. Journal of Psychiatric Research, 145, 347–352. https://doi.org/10.1016/j.jpsychires.2021.11.003

The first two papers were meta-analyses. This one is a real-world matched comparison. 2,384 adult patients at Rogers Behavioral Health System in the US, matched for age, sex, race, diagnosis, and level of care. 1,192 received in-person treatment before COVID; 1,192 received telehealth during COVID.

What makes this study important is where it sits on the intensity spectrum. These were not weekly outpatient sessions. Partial hospitalisation ran for 6 hours per day, 5 days per week. Intensive outpatient programmes ran for 3 hours per day, 5 days per week. The question being asked was whether telehealth can deliver at high clinical intensity. The answer was largely yes.

Equivalent outcomes

No significant differences in depression symptom reduction (p = .90) or quality of life improvement (p = .20) between groups. Both reduced depression symptoms by around 37%. Both improved quality of life by around 21%. Effect sizes were moderate to large across all conditions and levels of care.

The 2.8-day puzzle

The finding that puzzles me: telehealth patients in the partial hospitalisation programme stayed 2.8 days longer in treatment than in-person patients. The authors flag it; the study doesn’t explain it. It may reflect more cautious remote discharge decisions. It may reflect something about how clinical readiness presents differently through a screen. We don’t know. But it suggests that high-intensity remote treatment operates with its own particular logic. The evidence base for it is still relatively new.

A limit on who the research reached

Over 93% of participants in both groups were white. This is a gap in who the research reached, not a statement about any population. And it means the findings should be generalised with caution to the culturally diverse populations that make up much of contemporary urban practice. That includes here in Dubai.

Paper 4 — Aafjes-van Doorn et al. (2024): What happens to the therapeutic alliance online

The association between quality of therapeutic alliance and treatment outcomes in teletherapy: A systematic review and meta-analysis. Clinical Psychology Review, 110, 102430. https://doi.org/10.1016/j.cpr.2024.102430

This paper asks a different question entirely. Papers 1–3 looked at whether outcomes are equivalent. This one looks inside the therapeutic process and asks: does the quality of the therapeutic alliance predict outcomes in teletherapy, the way it does in in-person work?

The therapeutic alliance – broadly, the emotional bond between therapist and client, agreement on goals, and agreement on tasks – is one of the most consistently robust predictors of outcome in psychotherapy research. If it functions differently online, that matters regardless of whether headline outcomes are equivalent.

What the meta-analysis found

Across 31 studies and 34 independent samples (4,862 participants), the alliance–outcome association in teletherapy is real but small. Average weighted correlation r = 0.15, p = .001, 95% CI [0.07, 0.24]. Significant, but notably weaker than what we typically observe in the in-person literature, where the alliance–outcome correlation sits closer to r = 0.28.

Two moderators that matter

First, the association was stronger when alliance was measured later in treatment. Early alliance is less predictive online than in-person. That suggests it may take longer to build trust through a screen. Early ratings may not reflect where the relationship eventually goes.

Second, the association was stronger when outcomes were patient-reported rather than clinician-rated. What the patient experiences of the relationship matters most to what they get from it.

The qualitative picture from therapists

The qualitative picture from therapist reports is consistent and worth taking seriously. Working online requires more active effort to create emotional closeness. Sessions feel less emotionally deep. Therapists become more directive, avoid silences, and self-disclose more. Some find it harder to read their patients’ emotions.

There are also unexpected features of the online setting that cut the other way. Patients are in their own space rather than the therapist’s. That shifts the power dynamic. Home-based delivery gives the therapist direct access to the patient’s environment in ways that aren’t possible in a clinic. For some modalities and some patients, that access is clinically useful.

Not better or worse, but different

The framing I keep returning to comes from Aafjes-van Doorn’s own work: teletherapy is not better or worse, but different. It has its own clinical texture. The therapeutic process through a screen is not the same process compressed into a smaller frame. It calls for different skills, more deliberate relationship-building, and a different kind of therapist presence.

That distinction matters for training, supervision, and the standards we hold for online practice. Knowing how to do therapy well in-person does not automatically transfer.

Paper 5 — Bradford et al. (2024): The only couples-specific study in the field

Call me maybe? In-person vs. teletherapy outcomes among married couples. Psychotherapy Research, 34(5), 611–625. https://doi.org/10.1080/10503307.2023.2256465

All four previous papers focused on individual therapy. This is the only study in the series looking specifically at couples. 1,157 married clients, the largest couples-specific teletherapy study to date. The data draws on university training clinics and private practices.

Where Paper 4 looked at the strength of the alliance–outcome association, this paper looks at the rate at which the alliance itself develops. They are related questions, but distinct.

Equivalent outcomes, but slower alliance growth

The headline finding: equivalent outcomes overall. No significant difference in couple satisfaction between in-person and teletherapy groups. That is reassuring, and consistent with everything that came before.

The finding that’s harder to set aside: the therapeutic alliance developed at twice the rate in in-person sessions as in teletherapy. Not the eventual level – the pace of growth across sessions.

This matters clinically. If alliance builds at half the speed online, in time-limited work there may simply not be enough sessions for it to reach the depth it would in-person. In couples work, alliance must be built with two people at once. Each has their own relationship to the therapist. And the relationship between them is itself the focus of treatment. That slower development is not a small thing.

The sexual satisfaction puzzle

The sexual satisfaction findings were complicated. Outcomes varied not by modality but by clinic type. Clients in private practices improved. Training clinic clients showed decreases. Which suggests therapist experience, not the screen, may be driving some of what we assume to be a modality effect. A reminder of how entangled these variables are.

Limits worth naming

I want to be honest about the limits of this study too. Trainees featured prominently in the sample. We know from Lin et al. that trainees have worse outcomes in teletherapy than licensed practitioners. The sample was restricted to married couples. That doesn’t cover the full range of relationship structures or cultural contexts in which couples seek therapy. And the data was retrospective, not randomised.

Still: for couples weighing online therapy in Dubai or anywhere, this is the most substantive evidence we have on couples and teletherapy. And I think it supports being more rather than less deliberate about the conversation with a couple before they start. Not to discourage online work, but to name honestly what online couples therapy may require.

What five papers add up to

The evidence for online therapy is real and growing. For individual therapy addressing anxiety, depression, and trauma – particularly with structured approaches like CBT – the evidence base is probably strong enough to say that online delivery is a clinically legitimate choice, not a compromise.

But “the evidence is there” is the beginning of the clinical question, not the end of it.

Where to be more thoughtful

The therapeutic alliance functions differently online. It requires more deliberate effort. Beyond that, it builds more slowly. As a result, it is perhaps best assessed over time rather than early in treatment.

Trainee therapists face specific challenges in teletherapy that are not well addressed by training programmes designed for in-person work. Certain presentations – eating disorders, high-acuity cases, couples work – warrant more careful thought about fit than the headline equivalence findings suggest.

And the evidence base itself is largely white, largely Western, and largely pre-pandemic. That limits how confidently we can generalise to the populations I work with in Dubai and the UAE.

None of that undermines the case for online therapy. But it does mean the question worth asking is not “is online therapy as good?” but “is online therapy the right fit — for this person, with this presentation, at this point in their life?”

That’s where the research meets the consultation room.

Online therapy in Dubai – what this means in practice

What online therapy Dubai actually delivers

For clients in Dubai and across the UAE, the research points in a reasonably clear direction. Online therapy is a legitimate first-line option for common presentations – anxiety, depression, trauma – particularly when the work is well-structured. For many of the people I see here, it is also the only format that fits real life: time-zone-straddling work, travel, family life across continents, partners who are only occasionally in the same city.

The places to be more thoughtful are the ones the research flags. Higher-acuity presentations, work with trainees, couples therapy in time-limited frames, and eating disorders – these are not reasons to rule out online therapy. But they are reasons to have a more deliberate conversation about fit before starting. The cultural range of clients I see in Dubai also sits outside the populations represented in most of these studies. That is another argument for clinical care rather than confident generalisation.

In my own practice – online individual and couples therapy across Dubai and the UAE – II often combine in-person and online sessions across the course of a therapy process, drawing on what each format offers. I find that each modality brings a different kind of intimacy. In-person sessions let us share the same physical space and read each other’s full non-verbal communication. Online sessions, in contrast, bring me into your home, where you often feel safer and more comfortable. I get to meet the cats and dogs you’ve been telling me about, and to see how they come to comfort you when emotions surface. And on screen, the finer nuances of each other’s facial expressions are surprisingly visible.

How to book with me

If you are considering online therapy in Dubai and would like to talk through whether it is the right fit for what you’re bringing, bookings are made through Thrive Wellbeing Centre in Jumeirah Lake Towers, Dubai. I also see clients at Satori Health Centre. Initial consultations are available in both online and in-person formats depending on my availability. The decision between the two is one we make together.

 

Christi Gadd is a Clinical Psychologist and Neuropsychologist offering online therapy for individuals and couples across Dubai and the UAE through Thrive Wellbeing Centre (Jumeirah Lake Towers) and Satori Health Centre.