Abstract
BACKGROUND: While tobacco smoking biologically elevates hemoglobin through chronic hypoxia, this study investigates an unexpected paradoxical reversal in conflict-affected Yemen, testing whether socioeconomic status may override biological pathways in extreme resource-limited settings. METHODS: We conducted a multi-center cross-sectional study of 600 Yemeni university students (aged 18-25). Data on smoking, socioeconomic proxies, and hematological parameters were collected. We employed multivariate logistic regression, causal mediation analysis with bootstrapping, and E-value sensitivity analysis. RESULTS: Non-smokers had substantially higher odds of abnormal hemoglobin (adjusted Odds Ratio [aOR] = 11.25, 95% CI: 3.45--36.70, p < 0.001) and abnormal Mean Corpuscular Hemoglobin Concentration (aOR = 3.41, 95% CI: 1.58--7.35, p = 0.002) compared to smokers. Mediation analysis suggested that 38% of smoking's total effect on hemoglobin was mediated through nutritional pathways (indirect effect = 0.24, 95% CI: 0.08--0.45). The paradoxical association was significantly stronger among students from lower socioeconomic backgrounds (interaction p = 0.012) and females. E-value analysis (E = 4.32) indicated that substantial unmeasured confounding would be needed to explain away this association. CONCLUSION: These findings suggest a "healthy smoker" paradox in a humanitarian crisis context, where smoking status may serve as a proxy for higher SES and better nutritional access. This is consistent with the hypothesis that fundamental social causes can reverse established biological risk associations in contexts of extreme deprivation.