By Betty Tran, MD, MSc

Assistant Professor of Medicine, Pulmonary and Critical Care Medicine, Rush University Medical Center, Chicago

Dr. Tran reports no financial relationships relevant to this field of study.

SYNOPSIS: Based on a large, national, prospective cohort study, lower neighborhood socioeconomic status was associated with a higher incidence of hospitalizations for infection (but not sepsis) at presentation.

SOURCE: Donnelly JP, Lakkur S, Judd SE, et al. Association of neighborhood socioeconomic status with risk of infection and sepsis. Clin Infect Dis 2018;66:1940-1947.

Community or neighborhood factors play a key role in determining outcomes in many diseases and overall health.1-3 In addition, other investigators have reported that lower neighborhood socioeconomic status (nSES) is associated with increased sepsis hospitalizations and sepsis-related mortality, although these studies were localized to a specific city/state or used imprecise measures of nSES.4-6

Using prospectively gathered data from a study called the REasons for Geographic and Racial Differences in Stroke (REGARDS), a national cohort of more than 30,000 community-dwelling adults > 45 years of age with intentional oversampling of blacks and individuals from the southeast United States, Donnelly et al sought to examine further the association between nSES and hospitalization for infection and sepsis. nSES was summarized by a score comprised of several factors including: percentage of adults who completed high school and college; percentage of participants who work in professional, executive, or managerial jobs; median household incomes; median home value; and percentage of households receiving rental fees, interest, or dividends.

Regarding outcomes, ED visits and hospital admissions for serious infection were identified by trained abstractors. These abstractors reviewed medical records to confirm the existence of infection on first presentation as one reason for admission, with additional adjudication with physician review (if needed). There was excellent interrater agreement for presence of serious infection (Kappa = 0.92).

Hospitalizations for sepsis were defined as infection with two or more sepsis-related organ failure assessment (SOFA) score points. Additionally, hospitalizations for infection with two or more systemic inflammatory response syndrome (SIRS) criteria and two or more “quick” SOFA (qSOFA) criteria also were identified. Ten-year infection incidence per 1,000 person-years was examined across quartiles of nSES, with adjustment for several confounding factors (chronic kidney disease, age, race, education, sex, chronic lung disease geographic region, stroke, hypertension, alcohol use, smoking, myocardial infarction, and high-sensitivity C-reactive protein). The effects of potential mediators on the associations also were evaluated, including: individual income, physical weakness (based on the physical composite score on the 12-Item Short-Form Health Survey), self-reported exhaustion and physical activity, obesity, and diabetes.

The authors included 26,604 participants in the analysis. They observed significant variation among the nSES components across nSES quartiles. Participants in the lowest nSES quartile were disproportionately smokers, female, black, and nonusers of alcohol. Additionally, such subjects were disproportionately from the Stroke Belt, earned less annual income than others, had attained less education, and were more likely to present with comorbidities, abnormal biomarker labs, and reduced functional status.

After adjusting for participant characteristics, Donnelly et al noted that infection incidence was 0.84-fold lower for the highest quartile vs. lowest quartile of nSES. However, after adjustment, the authors noted no association between quartiles of nSES and sepsis incidence as defined by SOFA or qSOFA scores. Donnelly et al found that comorbid diabetes, physical weakness, and participant income produced modest (at least 10%) indirect effects on the association between nSES and infection risk.

Overall, infection type was similar across nSES quartiles, with respiratory infections listed as the most common. Median length of stay was longer for participants in the lowest nSES quartile, but the authors found no significant differences in SOFA scores or percentages of in-hospital death among nSES quartiles.


Donnelly et al have expanded on prior efforts to explore the complex association between social determinants of health and risk of hospitalization for infection and sepsis. Compared to prior studies, the Donnelly et al investigation features several strengths. First, data were collected prospectively as part of a national cohort. Second, the definition of nSES used was much more granular, based on census block groups and incorporating multiple domains as opposed to a single proxy such as insurance status.6 Outcomes were measured more accurately based on medical abstraction and published criteria for sepsis (e.g., SOFA scores) as opposed to ICD-9 codes.4 Third, follow-up was fairly extensive at 10 years. Finally, the mediating role of several factors was examined, providing hypotheses for further explorative and intervention-based studies. For example, individual income was found to mediate the association between nSES and infection risk, suggesting factors such as social isolation, reduced food availability, and lack of transportation may be targeted areas that could reduce hospitalizations for infection.

Overall, no association was found between nSES and sepsis incidence, to the extent that hospitalizations for infection may portend future hospitalizations for sepsis. However, the findings from this study support the premise that improving sepsis outcomes might be achieved outside of focusing exclusively on inpatient care by way of improvements in decreasing healthcare disparities that result in infection hospitalizations. This could include social interventions such as improving safe transportation options as well as improved aggressive medical management of chronic diseases such as diabetes. As such, findings from this study are a helpful reminder for us to think outside the box in our attempts to improve sepsis outcomes.


  1. Schultz WM, Kelli HM, Lisko JC, et al. Socioeconomic status and cardiovascular outcomes. Challenges and interventions. Circulation 2018;137:2166-2178.
  2. Nobel L, Jesdale WM, Tjia J, et al. Neighborhood socioeconomic status predicts health after hospitalization for acute coronary syndromes: Findings from TRACE-CORE (Transitions, Risks, and Actions in Coronary Events-Center for Outcomes Research and Education). Med Care 2017;55:1008-1016.
  3. DeRouen MC, Schupp CW, Koo J, et al. Impact of individual and neighborhood factors on disparities in prostate cancer survival. Cancer Epidemiol 2018;53:1-11.
  4. Goodwin AJ, Nadig NR, McElligott JT, et al. The impact of place of residence on severe sepsis incidence and mortality. Chest 2016;150:829-836.
  5. Galiatsatos P, Brigham EP, Pietri J, et al. The effect of community socioeconomic status on sepsis-attributable mortality. J Crit Care 2018;46:129-133.
  6. O’Brien JM Jr, Lu B, Ali NA, et al. Insurance type and sepsis-associated hospitalizations and sepsis-associated mortality among US adults: A retrospective cohort study. Crit Care 2011;15:R130.