according to a study published in Critical Care Medicine.
Dr. Panagis Galiatsatos
The association between living in a disadvantaged neighborhood and 30-day readmission remained significant even after adjustment for “individual demographic variables, active tobacco use, length of index hospitalization, severity of acute and chronic morbidity, and place of initial discharge,” wrote Panagis Galiatsatos, MD, of Johns Hopkins University in Baltimore, and colleagues.
“Our findings suggest the need for interventions that emphasize neighborhood-level socioeconomic variables in addition to individual-level efforts in an effort to promote and achieve health equity for patients who survive a hospitalization due to sepsis,” the authors wrote. “With a third of our cohort rehospitalized with infections, and other studies emphasizing that the most common readmission diagnosis was infection, attention toward both anticipating and attenuating the risk of infection in sepsis survivors, especially among those who live in higher risk neighborhoods, must be a priority for the prevention of readmissions.”
Although she did not find the study results surprising, Eva DuGoff, PhD, a senior managing consultant with the Berkeley Research Group and a visiting assistant professor at University of Maryland School of Public Health, College Park, said in an interview that she was impressed with how clinically rigorous the analysis was, both in confirming an accurate sepsis diagnosis and in using the more refined measure of the Area Deprivation Index (ADI) to assess neighborhood disadvantage.
“I think it makes sense that people who have less means and are in neighborhoods with fewer resources would run into more issues and would need to return to the hospital, above and beyond the clinical risk factors, such as smoking and chronic conditions,” said Dr. DuGoff, who studies health disparities but was not involved in this study.
Shayla N.M. Durfey MD, ScM, a pediatric resident at Hasbro Children’s Hospital in Providence, R.I., said in an interview she was similarly unsurprised by the findings.
“People who live in disadvantaged neighborhoods may have less access to walking spaces, healthy food, and safe housing and more exposure to poor air quality, toxic stress, and violence – any of which can negatively impact health or recovery from illness through stress responses, nutritional deficiencies, or comorbidities, such as reactive airway disease, obesity, hypertension, and diabetes,” said Dr. Durfey, who studies health disparities but was not involved in this study. “Our research has found these neighborhood-level factors often matter above and beyond individual social determinants of health.”
Dr. Galiatsatos and associates conducted a retrospective study in Baltimore that compared readmission rates in 2017 at Johns Hopkins Bayview Medical Center among patients discharged after a hospitalization for sepsis, coded via ICD-10. They relied on the ADI to categorize the neighborhoods of patients’ residential addresses. The ADI rates various socioeconomic components, including income, education, employment, and housing characteristics, on a scale of 1-100 in geographic blocks, with higher score indicating a greater level of disadvantage.
Among 647 hospitalized patients with an ICD-10 code of sepsis who also met criteria for sepsis or septic shock per the Sepsis-3 definition, 17.9% were excluded from the analysis because they died or were transferred to hospice care. The other 531 patients had an average age of 61, and just under one-third (30.9%) were active smokers. Their average length of stay was 6.9 days, with a mean Charlson Comorbidity Index of 4.2 and a mean Sequential Organ Failure Assessment score of 4.9.
The average ADI for all the patients was 54.2, but the average score was 63 for the 22% of patients who were readmitted within 30 days of initial discharge, compared with an average 51.8 for patients not readmitted (P < .001).
Among those 117 readmitted, “39 patients had a reinfection, 68 had an exacerbation of their chronic conditions, and 10 were admitted for ‘concerning symptoms’ without a primary admitting diagnosis,” the investigators reported. Because “a third of our cohort was readmitted with an infection, it is possible that more disadvantaged neighborhoods created more challenges for a person’s immune system, which may be compromised after recovering from sepsis.”
Dr. DuGoff further noted that health literacy may be lower among people living in less advantaged neighborhoods.
“A number of studies suggest when patients leave the hospital, they’re not sure what they need to do. The language is complicated, and it’s hard to know what kind of medication to take when, and when you’re supposed to return to the doctor or the hospital,” Dr. DuGoff said. “Managing all of that can be pretty scary for people, particularly after a traumatic experience with sepsis at the hospital.”
Most patients had been discharged home (67.3%), but the 31.6% discharged to a skilled nursing facility had a greater likelihood of readmission, compared with those discharged home (P < .01); 1% were discharged to acute rehabilitation. The average length of stay during the index hospitalization was also greater for those readmitted (8.7 days) than for those not readmitted (6.4 days). The groups did not differ in terms of their acute organ dysfunction or severity of their comorbidities.
However, even after adjustment for these factors, “neighborhood disadvantage remained significantly associated with 30-day rehospitalization in patients who were discharged with sepsis,” the authors said. Specifically, each additional standard deviation greater in patients’ ADI was associated with increased risk of 30-day readmission (P < .001).
“Given that the ADI is a composite score, we cannot identify which component is the predominant driver of rehospitalizations for patients who survive sepsis,” the authors wrote. “However, all components that make up the index are intertwined, and policy efforts targeting one (i.e., unemployment) will likely impact others (i.e., housing).”
Dr. Durfey said that medical schools have not traditionally provided training related to management of social risk factors, although this is changing in more recent curricula. But the findings still have clinical relevance for practitioners.
“Certainly, the first step is awareness of where and how patients live and being mindful of how treatment plans may be impacted by social factors at both the individual and community levels,” Dr. Durfey said. “An important part of this is working in partnership with social workers and case managers. Importantly, clinicians can also partner with disadvantaged communities to advocate for improved conditions through policy change and act as expert witnesses to how neighborhood level factors impact health.”
Dr. DuGoff also wondered what implications these findings might have currently, with regards to COVID-19.
“People living in disadvantaged neighborhoods are already at higher risk for getting the disease, and this study raises really good questions about how we should be monitoring discharge now in anticipation of these types of issues,” she said.
The authors noted that their study is cross-sectional and cannot indicate causation, and the findings of a single urban institution may not be generalizable elsewhere. They also did not consider what interventions individual patients had during their index hospitalization that could have increased frailty.
The study did not note external funding. One coauthor of the study, Suchi Saria, PhD, reported receiving honoraria and travel reimbursement from two dozen biotechnology companies for keynotes and advisory board service; she also holds equity in Patient Ping and Bayesian Health. The other authors reported no industry disclosures. In addition to consulting for Berkeley Research Group, Dr. DuGoff has received a past honorarium from Zimmer Biomet. Dr. Durfey has no disclosures.
SOURCE: Galiatsatos P et al. Crit Care Med. 2020 Jun;48(6):808-14.