Outpatient antibiotic prescribing and demographic factors associated with state-level septicemia mortality rates in US adults

Background Rates of septicemia/sepsis mortality and hospitalization in the US have risen significantly during the recent years, and antibiotic use may contribute to those rates through various mechanisms. Methods We used multivariable linear regression to relate state-specific rates of outpatient prescribing overall for fluoroquinolones, penicillins, macrolides, and cephalosporins between 2013-2014 to state-specific rates of septicemia mortality (ICD-10 codes A40-41 present as either the underlying or contributing causes of death) in each of the following age groups of adults: (18-49y, 50-64y, 65-74y, 75-84y, 85+y) between 2013-2014, adjusting for median household income, average annual temperature, age-specific percentages of state residents who (i) lived below the poverty level, (ii) were African American, (iii) lacked health insurance (in non-elderly age groups), and random effects associated with the different US Health and Human Services (HHS) regions. Results Rates of penicillin prescribing were positively associated with septicemia mortality rates in the analyses for persons aged 65-74y, 75-84y and over 85y. Rates of cephalosporin prescribing were positively associated with septicemia mortality rates in the analyses for persons aged 18-49y and 65-74y. Rates of fluoroquinolone prescribing were positively associated with septicemia mortality rates in the analyses for persons aged 18-49y. Percent African Americans in a given age group was positively associated with septicemia mortality rates in the analyses for age groups over 65y, and 18-49y. Percent of residents in a given age group living below the poverty level was positively associated with septicemia mortality rates in the analysis for persons aged 65-74y. Conclusions Our results suggest that rates of penicillin prescribing are associated with rates of septicemia mortality in older US adults, while rates of cephalosporin prescribing are associated with rates of septicemia mortality in persons aged 18-49y and 65-74y, and rates of fluoroquinolone prescribing are associated with rates of septicemia mortality in persons aged 18-49y. Further studies are needed to better understand the potential effect of antibiotic replacement in the treatment of different syndromes, such as replacement of fluoroquinolones by other antibiotics, possibly penicillins and cephalosporins following the recent US FDA guidelines on restriction of fluoroquinolone use, on the rates of sepsis mortality.