The morbidity and mortality conference in PICUs in the United States: A national survey

Objective: Determine how many morbidity and mortality conferences in PICUs across the United States conform to key elements of medical incident analysis.

Design: Web-based cross-sectional survey open from March to September 2013. Setting: Seventy-five PICUs with regular morbidity and mortality conferences in the United States identified by cross-referencing publicly available databases. Participants: Multidisciplinary PICU staff who attend the PICU morbidity and mortality conference. Interventions: None. Measurements and Main Results: Eighty-four identified PICUs of 206 PICUs contacted had at least one respondent, with a 40.8% PICUlevel response rate. The PICUs had a mean of three respondents (sd, 2.5; range, 1-13), with 45 of 84 (54%) having greater than one respondent. There were 305 total respondents to the survey, of which 220 of 300 (73%) self-identified as attending physicians and 47 of 300 (16%) as fellows. Four PICUs with only one respondent were excluded due to poor question response rates. Forty-nine of eightythree PICUs (59%) had fellowship training programs. Five of eightythree PICUs (6%) had no regular morbidity and mortality conference. Among 75 PICUs with regular morbidity and mortality conferences, morbidity and mortality conference process and structure characteristics varied widely. Among PICUs with greater than one respondent, when asked about morbidity and mortality conference conformity to each of the three key elements of medical incident analysis, 62-68% had intra-PICU disagreement among respondents. Fifteen of thirtyseven PICUs with greater than one respondent (41%) had intra-PICU disagreement on all three key elements. Conclusions: Morbidity and mortality conferences varied widely in structure and process across PICUs in the United States. There was marked disagreement as to whether the morbidity and mortality conference conforms to key elements of medical incident analysis, which might itself be revealing a lack of morbidity and mortality conference structure and consistency. Future research is needed to identify barriers to the use of the morbidity and mortality conference as a patient safety improvement tool and to test strategies for effective implementation linked to improved patient outcomes.