Causal Inference in Non-Allergic Asthma Associated with Obesity: Application of the Bradford Hill Viewpoints to a Complex Epidemiological Association.

Background/Objective: Obesity has been linked to a distinct non-allergic, late-onset, neutrophilic asthma phenotype for more than two decades, yet whether obesity should be regarded as a cause of that phenotype rather than a coexisting comorbidity remains unresolved because randomised allocation to obesity is neither feasible nor ethical. This review asks whether the evidence accumulated between 1995 and 2026 supports a causal interpretation. Methods: A structured narrative synthesis was undertaken. PubMed/MEDLINE, Scopus and Web of Science were searched from January 1995 to June 2026 (the synthesis emphasises the 2020-2026 evidence while drawing on the full window for landmark studies) for cohort, cross-sectional, Mendelian randomisation (MR), interventional, mechanistic and pharmaco-epidemiological studies; landmark earlier work was retained where it remains the primary source for a given viewpoint. The retrieved evidence was mapped onto Sir Austin Bradford Hill's nine viewpoints (1965) and triangulated using E-values for unmeasured confounding, GRADE, directed acyclic graphs and MR. Results: All nine viewpoints were satisfied to varying degrees. Strength: effect estimates 1.4-6.8, E-values 4.8-13.1, MR summary risk ratio 1.05 per 1 kg/m2. Consistency: replication across four continents, both sexes and multiple study designs. Temporality: prospective cohort and life-course MR evidence. Biological gradient: body mass index dose-response for TNFα, IL-17A and Th17 frequency. Plausibility: a twelve-layer architecture from adipose dysfunction to bronchial epithelium. Coherence, experimental evidence (bariatric surgery, lifestyle weight loss, GLP-1 receptor agonists, murine and in vitro models) and analogy were also supported; specificity was met at the phenotype level. Conclusions: The cumulative evidence strongly supports a causal contribution of obesity to the development of the late-onset, non-allergic asthma phenotype while falling short of definitive proof (GRADE: moderate certainty), with implications for primary prevention, endotype-targeted therapy and longitudinal research. Phenotype-stratified MR remains the principal missing element of the causal argument.