Among 480,344 UK Biobank participants free of sepsis at baseline, current smokers faced a 42% higher adjusted risk of hospital-recorded sepsis than never smokers (AHR 1.42), rising to 1.57 for pulmonary sepsis and 1.36 for non-pulmonary sepsis. Former smokers who had quit for 30 or more years sat at an AHR of 0.73 relative to the comparison group. In a 50,564-person proteomics subset (2,923 Olink proteins), neutrophil degranulation and extracellular matrix pathways were activated in sepsis cases and statistically mediated 14-25% of the pulmonary and 25-70% of the non-pulmonary association.
Smoking's link to cancer and heart disease is well known; sepsis rarely appears on the list of reasons to quit, yet it is a leading cause of hospital death. The dose-response gradient and the time-dependent benefit of cessation strengthen the case, and the neutrophil signal is biologically plausible, since smoking primes neutrophils toward degranulation and tissue damage. Caution is warranted. This is observational, so residual confounding by alcohol, deprivation, diabetes and frailty is possible. The 30-year-quitter comparison is ambiguous as reported, and long-term quitters differ from continuing smokers in many ways. Sepsis was captured only through hospital admissions, missing community cases. Protein levels were measured once, and the very wide 25-70% mediation range for non-pulmonary sepsis signals unstable estimates, not a clean mechanism. UK Biobank volunteers are also healthier and less diverse than the general population. Overall, the finding is confirmatory but adds a concrete infection-related reason to quit, with benefits that keep accruing for decades.