Heat map for data visualization in infection control epidemiology: An application describing the relationship between hospital-acquired infections, Simplified Acute Physiological Score II, and length of stay in adult intensive care units

• Published in: American journal of infection control Vol. 45; no. 7; pp. 746 - 749
• Main Authors: Ohannessian, Robin, MPH, Bénet, Thomas, MD, PhD, Argaud, Laurent, MD, PhD, Guérin, Claude, MD, PhD, Guichon, Céline, MD, Piriou, Vincent, MD, PhD, Rimmelé, Thomas, MD, PhD, Girard, Raphaele, MD, Gerbier-Colomban, Solweig, MD, PhD, Vanhems, Philippe, MD, PhD
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.07.2017; Elsevier
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Automatic Data Processing; Bacteriology; Cross Infection - epidemiology; Cross Infection - pathology; Data visualization; France; Heat map; Hospital-acquired infections; Humans; Immunology; Infection Control; Infection Control - methods; Infectious Disease; Intensive care unit; Intensive Care Units; Length of Stay; Life Sciences; Microbiology and Parasitology; Middle Aged; SAPSII; Severity of Illness Index; Surveillance; Virology; Young Adult; Surveillance; Hospital-acquired infections; Data visualization; SAPSII; Intensive care unit; Heat map; Length of stay
• ISSN: 0196-6553; 1527-3296
• DOI: 10.1016/j.ajic.2017.02.012

Background Hospital-acquired infections (HAIs) in intensive care units (ICUs) are associated with increased length of stay (LOS). The objective of this study was to graphically describe by heat mapping LOS of patients hospitalized in ICUs related to the occurrence of HAI and severity at admission measured by the Simplified Acute Physiological Score II (SAPSII). Methods Adult patients hospitalized in ICUs of Lyon University Hospitals (France) were included in an active standardized surveillance study of HAI from January 1, 1995-December 31, 2012. Surveillance included adult patients aged ≥18 years hospitalized ≥2 days. Patient follow-up ended at ICU discharge or death. LOS was calculated in days from differences between dates of entry and discharge from ICUs. HAIs recorded were pneumonia, bacteremia, and urinary tract infection. The heat map was designed with a spreadsheet software. Results A total of 34,694 patients were analyzed. Among infected patients, 72.3% had 1 infected site (IS), 23% had 2 ISs, and 4.7% had 3 ISs. Median LOS was 24 days in infected patients (20.4 days among patients with 1 IS, 34.2 days among patients with 2 ISs, and 45.3 days among patients with 3 ISs) and 5 days in noninfected patients ( P  < .001). Two groups of multi-infected patients with long LOSs were identified with the heat map. Conclusions The heat map facilitated easy-to-implement semi-quantitative visualization of increasing LOS through the SAPSIIs and number of ISs.