Abstract
Background:
According to the 2013 HCUP (Health Care Utilization Project) statistical brief, sepsis accounted for US$23.7 billion, or 6.2% of the aggregate costs for all hospitalizations. Sepsis is the second most common cause for hospitalizations, accounting for 3.6% of hospital stays. When considering the sepsis high cost burden, it is of paramount importance to identify the responsible factors. The main objective of this study is to evaluate impact of Non-Hodgkin Lymphoma (NHL) as a comorbidity either a history or as a present diagnosis of NHL, in patients admitted in the hospital with sepsis. Studies suggested increased risk of infections and mortality from bacteremia in patients with NHL in the past, (Shaikh et al, Cancer, 1987; Williams et al, Critical Care 2004 ) but recent data on health care cost burden associated with sepsis in NHL is limited.
Methods:
We analyzed the admissions for sepsis among adult patients using the Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample, between 2010 and 2014. Patients were stratified into two groups based on the status of NHL (sepsis with NHL and sepsis without NHL) using the ICD-9 CM diagnostic codes. Baseline characteristics including age, race, gender, Charlson comorbidity index, were described in the two groups, using descriptive statistics. A survey weighted multivariate regression analysis was used to adjust for confounders. STATA version 15 (College Station, TX) was used to perform statistical analysis.
Results
Between 2010 and 2014, a total of 4,501,621 admissions were identified with a primary diagnosis of sepsis. A total of 45,943 admissions had an underlying diagnosis of NHL. Mean age of patients hospitalized with sepsis and with NHL, was higher than patients without NHL (69.04 ±0.167 years; p<0.05 ), and 55% of the patients were males. Average hospital cost for each admission for patients with sepsis and NHL was ($78122.98±$1541.9;p<0.05), significantly higher than those without NHL. When adjusted for patient and hospital factors affecting population with NHL, such as age, race, gender, size of hospital, the average cost of a hospitalization was still significantly higher for patients with NHL. Similar results were found regarding outcomes, with a 32% increase in inpatient mortality for patients with sepsis and NHL (Odd ratio 1.32; p value=0.001)
Conclusion
Sepsis is responsible for a major healthcare cost burden in the US. Identification of signs of early sepsis and high-risk patients could support the design of early therapeutic strategies with the goal of decreasing negative outcomes and also decrease overall costs This study highlights NHL as an independent factor for mortality and healthcare costs for sepsis. It may help implement preventive strategies to avoid complications from infections and reduce cost burden associated with sepsis and NHL. Future studies are needed to focus on improving preventive strategies e.g vaccinations, early diagnosis and prompt treatment of infections to prevent progression to sepsis in patients with NHL.
No relevant conflicts of interest to declare.
