Skip to contents
Hojjat Salmasian, MD, PhD, MPH
Instructor, Harvard Medical School

Brigham and Women's Hospital
Department of Medicine
75 Francis Street
Boston, MA 02115

Research Narrative:

My research area of interest is the use of data science and informatics, to measure and improve quality and safety of health care. I have extensive research background in design and evaluation of automated solutions implemented through the Electronic Health Record (EHR) system to reduce inappropriate ordering. Besides my education as an MD, I have also been trained in epidemiology and biostatistics and received my PhD in biomedical informatics from Columbia University. The combination of skills and experience I have acquired through my education has allowed me to significantly contribute to several relevant projects aimed at improving the quality of care, care coordination, and identifying and reducing errors in the delivery of health care.

Before joining BWH and HMS, I worked at NewYork-Presbyterian Hospital (NYPH) and was on the faculty of biomedical informatics at Columbia University. My main research focus in this period was on hospital acquired infections (particularly C. diff), and order entry safety (particularly on measuring and reducing wrong-patient errors). In collaboration with colleagues in the division of gastroenterology at Columbia University, I studied the risk factors for C. diff infection; in a temporospectral analysis, we showed that patients who are admitted into beds that were previously occupied by another patient who receive high dose antibiotics, but did not manifest C. diff infection, are more likely to present this infection. I led the team that programmed the interface in the EHR to allow the display of patients’ photograph, and rolled out a randomized controlled trial to measure the impact of displaying photos to providers on reducing their rate of wrong-patient errors; the study is funded by the AHRQ and is ongoing. Prior to that, I was part of a study in which we showed the introduction of a patient verification alert reduced wrong-patient order entry by 30% immediately, and by 25% after two years.