A Pragmatic Method for Identification of Long-Stay Patients in the PICU

To develop a pragmatic method of identifying long-stay patients (LSPs) in the PICU.

We surveyed 40 expert stakeholders in 14 Canadian PICUs between February 2015 and March 2015 to identify key factors to use for defining LSPs in the PICU. We then describe a pragmatic method using these factors to analyze 523 admissions to an academic, tertiary-care PICU from February 1, 2015, to January 31, 2016.

The overall response rate was 70% (28 of 40). Of respondents, 75% (21of 28) stated that it was important to define LSPs and identified present and future resource consumption (18 of 21 [86%] and 16 of 21 [76%], respectively) as the key reasons for defining LSPs. Respondents valued a definition that was consistent and ranked a percentile cutoff as the preferred analytic method for defining LSPs. Of respondents, 86% (24 of 28) though the LSP definition should include factors other than length of stay. We developed a surrogate marker for LSPs using mechanical ventilation and presence of a central venous catheter in our sample population to compare to varying percentile cutoffs. We identified 108 patients at the 80th percentile as LSPs who used 67% of total bed days and had a median length of stay of 11.3 days.

We present a pragmatic method for the retrospective identification of LSPs in the PICU that incorporates unit- and/or patient-specific characteristics. The next steps would be to validate this method using other patient and/or unit characteristics in different PICUs and over time.