Gram Type Infection-Specific Sepsis Identification Using Machine Learning

The focus of this study will be to conduct a prospective, randomized controlled trial (RCT) at Cape Regional Medical Center (CRMC), Oroville Hospital (OH), and UCSF Medical Center (UCSF) in which a Gram type infection-specific algorithm will be applied to EHR data for the detection of severe sepsis. For patients determined to have a high risk of severe sepsis, the algorithm will generate automated voice, telephone notification to nursing staff at CRMC, OH, and UCSF. The algorithm's performance will be measured by analysis of the primary endpoint, time to antibiotic administration. The secondary endpoint will be reduction in the administration of unnecessary antibiotics, which includes reductions in secondary antibiotics and reductions in total time on antibiotics.

The experimental arm will involve patients monitored by the Gram type infection-customized version of InSight.

The control arm will involve patients treated with the regular diagnosis and treatment protocol for gram-type infection, where fluid cultures are run to determine infection type.

The InSight algorithm which draws information from a patient's electronic health record (EHR) to predict the onset of severe sepsis, and in this study will be customized to differentiate between various Gram-type infections.

Change in time period between diagnosis of Gram infection and administration of antibiotics to treat infection

Inclusion Criteria: All adults above age 18 who are a member of one of the three subpopulations studied in this trial (patients with Gram-positive infection, patients with Gram-negative infection, and patients with mixed Gram-positive and Gram-negative infection) are eligible to participate in the study. Exclusion Criteria: Under age 18 No record of Gram infection

Calvert J, Mao Q, Hoffman JL, Jay M, Desautels T, Mohamadlou H, Chettipally U, Das R. Using electronic health record collected clinical variables to predict medical intensive care unit mortality. Ann Med Surg (Lond). 2016 Sep 6;11:52-57. doi: 10.1016/j.amsu.2016.09.002. eCollection 2016 Nov.

Shimabukuro DW, Barton CW, Feldman MD, Mataraso SJ, Das R. Effect of a machine learning-based severe sepsis prediction algorithm on patient survival and hospital length of stay: a randomised clinical trial. BMJ Open Respir Res. 2017 Nov 9;4(1):e000234. doi: 10.1136/bmjresp-2017-000234. eCollection 2017.

Mao Q, Jay M, Hoffman JL, Calvert J, Barton C, Shimabukuro D, Shieh L, Chettipally U, Fletcher G, Kerem Y, Zhou Y, Das R. Multicentre validation of a sepsis prediction algorithm using only vital sign data in the emergency department, general ward and ICU. BMJ Open. 2018 Jan 26;8(1):e017833. doi: 10.1136/bmjopen-2017-017833.