Volume 8, Issue 5 p. 248-253
Original Research

Beyond statistical prediction: qualitative evaluation of the mechanisms by which pediatric early warning scores impact patient safety

Christopher P. Bonafide MD, MSCE

Corresponding Author

Christopher P. Bonafide MD, MSCE

Division of General Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Center for Pediatric Clinical Effectiveness, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania

Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, Pennsylvania

Address for correspondence and reprint requests: Christopher P. Bonafide, MD, MSCE, Division of General Pediatrics, The Children's Hospital of Philadelphia, 34th St and Civic Center Blvd, Room 12NW80, Philadelphia, PA 19104; Telephone: 267-426-2901; Fax: 215-590-2180; E-mail: [email protected]Search for more papers by this author
Kathryn E. Roberts RN, MSN, CNS, CCRN, CCNS

Kathryn E. Roberts RN, MSN, CNS, CCRN, CCNS

Department of Nursing, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Search for more papers by this author
Christine M. Weirich MPH

Christine M. Weirich MPH

Division of General Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Search for more papers by this author
Breah Paciotti MPH

Breah Paciotti MPH

Mixed Methods Research Lab, University of Pennsylvania, Philadelphia, Pennsylvania

Search for more papers by this author
Kathleen M. Tibbetts MS

Kathleen M. Tibbetts MS

Division of General Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

GfK Healthcare, Blue Bell, Pennsylvania

Search for more papers by this author
Ron Keren MD, MPH

Ron Keren MD, MPH

Division of General Pediatrics, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Center for Pediatric Clinical Effectiveness, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania

Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania

Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, Pennsylvania

Search for more papers by this author
Frances K. Barg PhD

Frances K. Barg PhD

Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania

Mixed Methods Research Lab, University of Pennsylvania, Philadelphia, Pennsylvania

Search for more papers by this author
John H. Holmes PhD

John H. Holmes PhD

Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania

Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, Pennsylvania

Search for more papers by this author
First published: 13 March 2013
Citations: 43

Abstract

BACKGROUND

Early warning scores (EWSs) assign points to clinical observations and generate scores to help clinicians identify deteriorating patients. Despite marginal predictive accuracy in retrospective datasets and a paucity of studies prospectively evaluating their clinical effectiveness, pediatric EWSs are commonly used.

OBJECTIVE

To identify mechanisms beyond their statistical ability to predict deterioration by which physicians and nurses use EWSs to support their decision making.

DESIGN

Qualitative study.

SETTING

A children's hospital with a rapid response system.

PARTICIPANTS

Physicians and nurses who recently cared for patients with false-positive and false-negative EWSs (score failures).

INTERVENTION

Semistructured interviews.

MEASUREMENTS

Themes identified through grounded theory analysis.

RESULTS

Four themes emerged among the 57 subjects interviewed: (1) The EWS facilitates safety by alerting physicians and nurses to concerning changes and prompting them to think critically about deterioration. (2) The EWS provides less-experienced nurses with vital sign reference ranges. (3) The EWS serves as evidence that empowers nurses to overcome barriers to escalating care. (4) In stable patients, those with baseline abnormal physiology, and those experiencing neurologic deterioration, the EWS may not be helpful.

CONCLUSIONS

Although pediatric EWSs have marginal performance when applied to datasets, clinicians who recently experienced score failures still considered them valuable to identify deterioration and transcend hierarchical barriers. Combining an EWS with a clinician's judgment may result in a system better equipped to respond to deterioration than retrospective data analyses alone would suggest. Future research should seek to evaluate the clinical effectiveness of EWSs in real-world settings. Journal of Hospital Medicine 2013;8:248–253. © 2013 Society of Hospital Medicine