The researchers analysed cross-sectional data from 2,389 Qatari Biobank participants, split almost perfectly evenly between 1,192 males and 1,197 females. Rather than examining individual measurements in isolation, the team constructed five composite scores from standardised z-scores, each capturing a distinct health domain: muscular strength, blood pressure and adiposity, respiratory function, cognitive function, and a socio-environmental index combining education, income, tobacco exposure and night-shift work. This composite approach matters because single measurements can be noisy and domain-specific patterns can hide inside averages. By aggregating related variables into coherent phenotypes, the analysis could ask whether men and women differ across whole physiological systems rather than on isolated laboratory values.

The headline statistical result was unambiguous. A multivariate analysis of variance, which tests whether the entire profile of five scores differs between the sexes simultaneously, produced a Pillai’s Trace of 0.598 with F(5,667) = 198.5 and a p-value below 0.001. In practical terms, sex explained an enormous share of the variation across these combined health domains. But the study’s real contribution lies in what happened next, when the researchers decomposed that overall signal domain by domain, estimated effect sizes using Cohen’s d, and then re-estimated everything in multivariable models adjusting for age, education, income and smoking. The picture that emerged was one of starkly domain-specific differences, some of which are biology and some of which are policy targets.

The largest difference anywhere in the dataset was muscular strength, where males scored dramatically higher, with an effect size of d = 2.41 in the raw comparison. Even after adjustment for age, education, income and smoking, the gap remained at d = 2.28, an effect size so large it sits at the boundary of what social and behavioural science considers practically maximal. The authors interpret this, correctly, as consistent with established structural dimorphism: differences in muscle mass and body composition that are rooted in physiology, not in unequal access to resources or healthcare. Their conclusion is pointed and policy-relevant. Such differences should not be interpreted as disparities requiring intervention, and treating them as such would misdirect resources away from the gaps that are genuinely actionable.

The respiratory findings deliver the study’s most instructive methodological lesson. When lung function was measured as raw volumes, such as forced expiratory volume in one second and forced vital capacity, males showed a massive advantage of d = 2.05, seemingly comparable in magnitude to the strength difference. But the team had prespecified a critical alternative: respiratory function was also quantified using reference-equation z-scores, which standardise each individual’s lung volumes against expected values for their age, sex, height and ethnicity. Under this properly referenced metric, the male advantage collapsed from d = 2.05 to d = 0.30, and after multivariable adjustment it shrank further to d = 0.17. In other words, the enormous raw-volume gap was almost entirely an artefact of body size. Men have bigger lungs because they have bigger bodies, not because their respiratory health is better. Any surveillance system that tracks raw spirometry volumes across sexes without referencing would systematically misread anatomy as pathology.

The cardiometabolic domain tells the opposite story, and it is here that the study’s clearest warning for prevention policy emerges. Females showed a more favourable blood pressure and adiposity profile, with an effect size of d = -0.58 in the raw comparison that actually strengthened to d = -0.82 after adjustment. On its face, that looks like good news for women. But the researchers then tested whether this female advantage was uniform across the lifespan, and it was not. Below age 50, the female cardiometabolic advantage was d = -0.78, whereas at age 50 and above it narrowed to d = -0.37, a sex-by-age interaction with a p-value below 0.001. The pattern is consistent with a menopause-related narrowing: the well-documented cardiometabolic protection that women carry through their reproductive years attenuates after midlife, and in this Qatari cohort it attenuates substantially.

That finding has direct implications for how screening and prevention are timed in Gulf populations. If the female advantage in blood pressure and adiposity erodes after 50, then the midlife transition marks the point at which Qatari women’s cardiovascular risk profile begins converging toward that of men, and prevention programmes that assume women remain relatively protected deep into later life may miss the window in which intervention is most effective. The authors flag this menopause-related narrowing, together with the divergent exposure profiles discussed below, as the two findings with the clearest implications for population health policy. It is a reminder that sex differences in health are not static properties but trajectories that change across the life course, and that cross-sectional snapshots can conceal those dynamics unless age is explicitly modelled, as it was here through prespecified effect-modification analyses.

Cognitive function, by contrast, turned out to be largely a story about confounding. In the unadjusted comparison, males showed a small advantage on the cognitive composite, which drew on the Cambridge Neuropsychological Test Automated Battery. But after adjustment for age, education, income and smoking, the difference shrank to d = 0.18 with a p-value of 0.068, no longer statistically distinguishable from zero. The team also ran prespecified sensitivity analyses addressing selection into the cognitive subsample, component overlap between the composites, and internal consistency, which strengthens confidence that these attenuations reflect genuine statistical structure rather than analytical fragility. The lesson echoes the respiratory finding: apparent sex differences in a health domain can dissolve once socioeconomic and behavioural context is accounted for, which is precisely why the study distinguishes so carefully between dimorphism and disparity.

Perhaps the most socially revealing result concerns the socio-environmental composite, which showed essentially no net sex difference, d = -0.01, but only because two opposing gradients cancelled each other out. Males in the cohort had higher education and income, advantages that would push their composite score upward, but they also carried substantially greater exposure to tobacco and to night-shift work, risk factors that pushed it downward. Averaged together, men and women looked environmentally identical. Looked at component by component, they occupy strikingly different risk landscapes. For policy, this cancellation is itself the finding: a summary index that reports no sex gap would conceal the fact that Qatari men disproportionately bear the burden of smoking and shift-work exposure, both established drivers of cardiometabolic and respiratory disease, while women’s environmental profile differs in composition rather than magnitude. Targeted interventions, the authors argue, need to see these opposing gradients, not the net zero.

The study’s overall conclusion is that sex-specific health phenotypes in this Qatari cohort are domain-specific, and that only some of the observed differences represent actionable health inequity. Muscular strength gaps and raw lung-volume differences reflect normative physiological dimorphism and should be excluded from disparity metrics. The narrowing of women’s cardiometabolic advantage after midlife and the divergent socio-environmental exposures of men and women are the signals that prevention policy should act on. The authors are appropriately cautious about generalisability, noting that whether these patterns extend to other Middle Eastern populations requires comparative data, and the cross-sectional design cannot establish causal sequences within individuals. Still, for a region where sex-specific differences have often been overlooked in population health strategy, the analysis offers a template: measure domains comprehensively, reference measurements properly, adjust for social context, model age explicitly, and only then decide which differences between men and women are biology to be understood and which are inequities to be fixed.

Subject of Research: Sex-specific health phenotypes and their policy implications in a Qatari population cohort

Article Title: Sex-specific health phenotypes in a large qatari cohort: a cross-sectional analysis with implications for population health policy

Article References: Sex-specific health phenotypes in a large qatari cohort: a cross-sectional analysis with implications for population health policy. (n.d.). https://doi.org/10.1186/s12889-026-29801-z

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29801-z

Keywords: sex differences, sexual dimorphism, Qatar Biobank, health phenotypes, cardiometabolic risk, respiratory function, muscular strength, menopause, population health policy, cross-sectional study, tobacco exposure, cognitive function