Sepsis is a life-threatening condition characterized by organ dysfunction resulting from a dysregulated host response to infections [1]. Epidemiological studies indicate that approximately 48.9 million people worldwide were diagnosed with sepsis annually, leading to 11 million deaths and representing 19.7% of all global deaths [2]. In China, sepsis affects 20% of patients in intensive care units, with a 90-day mortality rate of 35.5% [3]. The substantial epidemiological burden underscores the critical importance of sepsis as a public health issue [4]. Despite decades of research, current therapies remain limited to organ support and antimicrobial interventions, without any targeted treatment available [5-7]. These limited therapeutic options highlight the urgent need for shifting focus toward modifiable risk factors for prevention. The World Health Organization 2020 Sepsis Report specifically highlights vulnerable populations and health care seekers, while outlining future research directions and priorities in sepsis epidemiology. Although sepsis manifests diversely and can be life-threatening, it remains preventable and reversible [4].

Host risk factors for sepsis are multifaceted, encompassing extremes of age, immunosuppression, and comorbidities [8,9]. Emerging evidence suggests that socioeconomic determinants (eg, education level and occupation) and metabolic disorders (eg, obesity and diabetes) may influence susceptibility to sepsis [10-16]. However, the existing observational studies report inconsistent conclusions. For instance, although lower socioeconomic status is broadly linked to higher sepsis incidence [10,11], another study paradoxically found no association using adjusted models [17]. Metabolic comorbidities further illustrate this ambiguity. Although obesity is widely implicated in sepsis risk [18], some observational studies found that individuals with obesity showed a lower incidence and fatality than did those with underweight among patients with sepsis. These discrepancies may stem from methodological limitations in observational research, such as residual confounding and reverse causality. For instance, sepsis-induced systemic inflammation could transiently alter metabolic markers, obscuring true causal directions.

The core idea of Mendelian randomization (MR) is to use genetic variants (typically single-nucleotide polymorphisms [SNPs]) as instrumental variables to infer whether a potential causal association exists between an exposure and a health outcome. Recent MR studies have explored individual factors such as obesity, diabetes mellitus, and lipid profiles in patients with sepsis [19-23]; however, critical gaps persist. First, lifestyle factors (eg, smoking and diet) remain understudied despite their clinical relevance. Second, the existing MR analyses of sepsis predominantly examine isolated risk factors, neglecting potential synergies between socioeconomic, metabolic, and behavioral determinants [8,9,24-28]. Third, the interaction between modifiable factors and genetic predisposition remains unexplored, limiting personalized preventive strategies.

To address these gaps, we conducted a comprehensive 2-sample MR analysis of 30 modifiable factors spanning lifestyle, socioeconomic status, and metabolic health. This approach not only circumvents confounding biases but also systematically evaluates multifactorial contributions to sepsis pathogenesis. Our findings aimed to clarify controversial associations in prior literature and prioritize actionable targets for intervention.

In this bidirectional MR study, we investigated the association of lifestyle, educational and psychological, and metabolic factors with the risk of sepsis. Genetic variants such as smoking initiation, smoking frequency, and waist circumference were associated with an increased risk of sepsis, whereas light physical activity, higher education, and high levels of HDL-c were causally associated with a lower risk of sepsis (Figure 3).

Genetic variants such as smoking initiation, smoking frequency, and waist circumference were associated with an increased risk of sepsis, whereas light physical activity, higher education, and high levels of HDL-c were causally associated with a lower risk of sepsis.

Owing to the modifiable nature of lifestyle, the impact of lifestyle on diseases with high mortality, such as sepsis, has received increasing attention. Numerous observational studies have described an association between smoking, alcohol consumption, and the risk of various infectious diseases [22,30-35]. However, establishing a causal link between lifestyle and sepsis remains challenging. This study aimed to explore the potential causal relationships between lifestyle factors and sepsis. The results showed that smoking initiation and a high frequency of smoking were associated with an increased risk of sepsis. Previous immunologic experiments reported that smoking has been associated with impaired phagocytic function and cytokine expression of T lymphocytes and macrophages in the respiratory tract or peripheral blood [36-38]. These studies collectively underscore the detrimental impact of smoking on the immune system. These findings provide a better understanding of the role of smoking in patients with sepsis. Therefore, healthy lifestyle modification is a preventive measure against sepsis; individuals should be advised to reduce smoking to reduce the risk of developing sepsis.

Physical exercise is a planned, purposeful action to maintain a healthy lifestyle and improve physical fitness (World Health Organization, 2010) [39]. Furthermore, the evidence suggests that physical exercise affects multiple organ systems under various conditions [40]. It has been found that inadequate exercise is associated with a doubled risk of sepsis-related mortality [40]. Given that the incidence of sepsis is increasing, these results suggest that exercising could be immediately effective in reducing disease incidence. Currently, no MR analyses of the association between physical activity and sepsis are available. Our analysis confirmed that light physical activity is associated with a lower risk of sepsis.

The socioeconomic status affects health through environmental exposure, health behaviors, and lifestyle and has been shown to be positively related to health [10,41]. Education stands as the most potent indicator of socioeconomic status, exerting influence over lifestyle choices and access to health resources [42]. Observational studies suggest an inverse association between educational attainment and sepsis risk. Wang et al [43] reported significantly higher sepsis risk among individuals with lower education (hazard ratio 1.88, 95% CI 1.54‐2.29), though without controlling for comorbidities and other confounders. Another cohort study adjusting for socioeconomic factors still demonstrated a higher risk of sepsis-related intensive care unit admission in moderately educated versus highly educated individuals [44]. It should be noted that these observational findings may be influenced by residual confounding and do not establish causality. This study found a negative correlation between genetically predicted educational levels and sepsis risk. Higher education may reduce sepsis risk through multiple pathways, including enhanced health literacy (eg, earlier recognition of infections), greater access to health care resources, and adoption of healthier behaviors (eg, smoking cessation and improved diet). These mechanisms could collectively mitigate exposure to risk factors and improve timely medical intervention.

In recent years, an increasing number of studies have focused on the impact of metabolic factors such as blood glucose, blood lipids, and obesity on various infectious diseases [45-49]. Previous studies have reported that high HDL-c levels are significantly associated with reduced sepsis risk and other sepsis-related outcomes [50,51]. Extremely low serum HDL-c levels (≤20 mg/dL) are associated with an increased risk of death, sepsis, and malignancy [52]. However, observational clinical and epidemiologic studies have potential problems such as confounding, reverse causation, and unaccounted comorbidities. An MR study found that lower HDL-c levels were significantly associated with an increased risk of sepsis and related outcomes in infected patients [53]. This is consistent with our findings. Our study found a genetic relationship between high HDL-c levels and low risk of sepsis. HDL-c might exert protective effects against sepsis via its anti-inflammatory and antioxidant properties, as well as its role in modulating endothelial function and neutralizing bacterial endotoxins [54-56].

In observational studies on the general population, a higher BMI has been associated with an increased incidence of mortality from bloodstream infections and sepsis [57-59]. This finding is confirmed in this study. We also found that BMI, arm fat mass, leg fat mass, whole body fat mass, waist circumference, and trunk fat mass were significantly associated with sepsis. However, using the multivariate factor analysis, only increased waist circumference was an independent risk factor for sepsis, which might be because other factors included the causes of increased waist circumference, suggesting that these factors might influence sepsis risk as a consequence of shared risk factor profiles.

This study had several limitations. First, the population limited to the European ancestry hampered the generalization of the findings to individuals of other ancestries, as genetic and environmental factors influencing sepsis risk may vary across ethnic groups (eg, differences in inflammatory responses or health care access). Further studies are required to assess the modifiable risk of sepsis in other races. Second, as with all MR studies, pleiotropy in an MR setting is challenging. Additionally, gene-environment interactions (eg, how lifestyle modifies genetic risk) and potential measurement errors in exposure variables were not fully addressed, which might have introduced bias. We performed various sensitivity analyses that made different assumptions regarding the underlying nature of pleiotropy. Most tests showed stable results. Future studies should incorporate multi-ancestry cohorts and explore gene-environment interplay to strengthen causal inference.

In conclusion, our combined MR analysis supports the causal roles of smoking, educational level, HDL-c level, physical condition, obesity, and physical activity in patients with sepsis. The findings of this study highlight actionable targets for sepsis prevention. For instance, smoking cessation programs, public health initiatives to promote physical activity, and interventions to improve HDL-c levels (eg, dietary modifications or pharmacotherapy) could be integrated into personalized preventive strategies for managing high-risk populations. Furthermore, educational campaigns targeting health literacy and early symptom recognition might reduce delays in seeking care, particularly among socioeconomically disadvantaged groups. Future research should prioritize replication in diverse populations, mechanistic studies to elucidate pathways (eg, HDL-c’s immunomodulatory effects), and translational interventions targeting high-risk subgroups.