The research team, led by Mohammed A. Alhassan of Majmaah University in Saudi Arabia and spanning institutions from Riyadh to Liverpool, Saitama, Dire Dawa, and Toronto, followed the PRISMA 2020 reporting guidelines and prospectively registered their protocol with PROSPERO before screening a single record. They searched six databases and one clinical trial registry for studies published between January 2010 and September 2025. From 2,775 initial hits, 106 studies survived the inclusion criteria, forming the largest synthesis to date of severe mental illness epidemiology across the MENAT region.

Methodologically, the team employed a random-effects meta-analysis using restricted maximum likelihood estimation, a statistical approach designed to accommodate the reality that the underlying true prevalence almost certainly differs between studies drawn from different populations, settings, and decades. They stratified their analyses by population type, diagnostic approach, and study characteristics, assessed methodological quality with the Joanna Briggs Institute tool, and ran formal tests for publication bias alongside sensitivity analyses. This layered architecture matters, because in a field where estimates can swing by an order of magnitude depending on whether researchers surveyed a general community or recruited from a psychiatric clinic, the analytical choices determine whether the resulting numbers mean anything at all.

The headline figures are striking. In general-population samples, the pooled prevalence of severe mental illness came out at 7.3 percent, with a 95 percent confidence interval running from 5.0 to 9.5 percent. In clinical and high-risk samples, by contrast, the pooled estimate jumped to 31.1 percent, with a confidence interval of 23.9 to 38.2 percent. That fourfold gap between community and clinical settings is not a contradiction; it is exactly what one would expect when one set of studies samples everyone and the other deliberately oversamples people already in contact with mental health services or screened as being at elevated risk. But it underscores how much the answer depends on where and how you look.

Comorbidity emerged as perhaps the most clinically consequential finding. Across studies that reported co-occurring conditions in people with severe mental illness, the pooled prevalence of any comorbidity was 32.0 percent, with a confidence interval of 25.5 to 38.6 percent. Crucially, physical comorbidities, the metabolic, cardiovascular, and other somatic conditions that so often shadow serious psychiatric diagnoses, were more common than additional psychiatric comorbidities. This aligns with a well-documented global pattern in which people with severe mental illness die years, sometimes decades, earlier than the general population, largely from treatable physical diseases rather than from the psychiatric conditions themselves.

The authors are refreshingly candid about the limits of their own numbers. Between-study heterogeneity, quantified by the I-squared statistic, ranged from 96.1 percent to over 99 percent, figures that sit at the extreme end of what meta-analysts ever encounter. In practical terms, this means the variation between individual studies dwarfs the variation that would be expected from chance alone, and the pooled estimates should be read as broad summaries of a scattered literature rather than as precise regional prevalence figures. The team also detected small-study effects, the tendency for smaller investigations to report more extreme results, which can inflate pooled estimates and complicate interpretation further.

Why is the evidence base so heterogeneous? The answer lies partly in the region itself and partly in the nature of psychiatric epidemiology. Studies differed in whether they used structured diagnostic interviews or screening questionnaires, whether they were conducted in hospitals, primary care clinics, or communities, and which populations they targeted, from refugees and displaced people to university students to general adult samples. Diagnostic thresholds for what counts as severe major depressive disorder, in particular, vary considerably across instruments and cultures. Add to this the region’s exposure to conflict, displacement, and economic instability, all of which reshape both mental health need and the capacity to measure it, and the statistical turbulence becomes unsurprising.

Even with those caveats, the review delivers a clear public health signal. Severe mental illness in the MENAT region is common enough, and its physical comorbidity burden high enough, that it demands systematic attention from health planners. The finding that roughly one in three people with severe mental illness carries a significant comorbid condition argues for integrated care models in which psychiatric treatment is paired with routine screening and management of metabolic and cardiovascular risk, rather than the fragmented separation of mental and physical healthcare that still characterizes many health systems in the region and beyond.

The study also exposes a gap that only better primary research can close. The authors call explicitly for future epidemiological studies using standardised diagnostic approaches, which would allow estimates from different countries and settings to be compared and combined with far greater confidence. Until such studies exist, the 7.3 percent community figure and the 31.1 percent clinical figure should be treated as signposts rather than destinations, useful for signaling scale and urgency but not for calculating exact service requirements or budget allocations.

What makes this work resonate beyond its regional focus is the methodological honesty on display. In an era when meta-analyses are sometimes presented as definitive answers, this team demonstrates what rigorous synthesis looks like when the underlying evidence refuses to cooperate: register the protocol in advance, quantify the heterogeneity, test for publication bias, and tell readers plainly that the pooled numbers are summaries, not truths. For the MENAT region, the message is twofold. Severe mental illness represents a substantial and underappreciated burden, particularly among people already in clinical contact, and the physical health of psychiatric patients demands urgent integration into routine care. And for the global research community, it is a reminder that before prevalence can be measured precisely, the instruments, definitions, and sampling strategies used to measure it must first be made to speak the same language.

Subject of Research: Prevalence and comorbidities of severe mental illness in the Middle East, North Africa, and Türkiye region

Article Title: Severe mental illness in the MENAT region: systematic review and meta-analysis of prevalence and comorbidities

Article References: Alhassan, M. A., Alarabi, M. A., Albalawi, W. M., Nisar, A., Arachchige, O. K. T., & Kitaw, T. A. (2026). Severe mental illness in the MENAT region: systematic review and meta-analysis of prevalence and comorbidities. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08615-8

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08615-8

Keywords: severe mental illness, MENAT region, systematic review, meta-analysis, prevalence, comorbidity, schizophrenia, bipolar disorder, major depressive disorder, public health, psychiatric epidemiology, physical comorbidity