University of Arizona College of Medicine Phoenix, AZ
Jiali Ling, MS1, Alec Zamarripa, MS2, Saroja D. Geetha, MBBS1, Tyson Amundsen, MD3, Ricardo Badillo, MD4, Nirav Thosani, MD, MHA4, Sushovan Guha, MD, PhD4, Bijun S. Kannadath, MBBS, MS1, Michael B. Fallon, MD1; 1University of Arizona College of Medicine, Phoenix, AZ; 2University of Arizona College of Medicine, Sacramento, CA; 3Baylor Scott & White Health, Temple, TX; 4University of Texas Health Science Center, Houston, TX
Introduction: In this study we aim to look at the effects of hospital parameters such as location, teaching status, bed size and ownership on incidence and outcomes of Clostridium difficile infection (CDI) using the National Inpatient Sample (NIS) database. Methods: All adult hospitalizations (age ≥ 18 years) between 2012 and 2017 were evaluated. Admissions with CDI were identified based on the appropriate ICD9 (00845) and ICD10 (A047, A0471, A0472) codes. Incidence and mortality rates with respect to hospital facilities such as hospital location, bed size, teaching facilities and ownership were analyzed after applying recommended weights using Stata Statistical Software and Python. Results: In this study, 181,132,460 records were analyzed from 2012-2017. A total of 2,089,130 admissions had CDI. Of those, urban hospitals had 1,906,150 (91.24%) and rural hospitals had 182,980 (8.76%) CDI admissions. Further, urban teaching hospitals had 1,282,380 (67.28%) and urban non-teaching hospitals had 623,770 (32.72%) CDI admissions. Though mortality rates in both rural and urban hospitals showed an overall decline from 2012-2017, urban teaching hospitals had a higher mortality rate than urban non-teaching [90,890 (68.92%) vs 40,995 (31.08%) deaths]. CDI incidence was higher in large bedded hospitals, trending from 1.19% to 1.10% (2012-2017) and lowest in small bedded hospitals, 1.07% to 1%. Mortality rates were also higher in large bedded hospitals with a significant improvement seen in medium bedded hospitals after 2013. At the end of study, private non-profit hospitals had a higher CDI incidence (1.10%) than government non-federal (1.04 %) and private investor owned hospitals (0.87%). Mortality rates in government non-federal hospitals peaked in 2013 (7.74%) but there appears to be an increasing trend in 2016 (not statistically significant).
Discussion: From 2012-2017, urban teaching hospitals had higher CDI incidence and mortality rates compared to urban non-teaching and rural hospitals. Furthermore, large bedded hospitals had a higher CDI incidence and mortality rates and private, non-profit hospitals saw higher incidence rates without a statistically significant change in mortality rates. Our results could be explained by the larger number of complex, often tertiary referral, cases seen at teaching hospitals and the larger census encountered in large hospital systems, compared to smaller, privatized, non-teaching hospitals. Further studies are necessary to evaluate other confounding factors.
Table 1: CDI incidence & mortality rates from 2012-2017, based on hospital location & teaching facilities, bed size and ownership: results from NIS database (2012 - 2017)
Figure 1: Incidence of CDI based on hospital location & teaching status
Figure 2: Mortality rates in patients with CDI based on hospital location and teaching
Disclosures: Jiali Ling indicated no relevant financial relationships. Alec Zamarripa indicated no relevant financial relationships. Saroja Geetha indicated no relevant financial relationships. Tyson Amundsen indicated no relevant financial relationships. Ricardo Badillo indicated no relevant financial relationships. Nirav Thosani indicated no relevant financial relationships. Sushovan Guha indicated no relevant financial relationships. Bijun Kannadath indicated no relevant financial relationships. Michael Fallon indicated no relevant financial relationships.