An Empirical Derivation of the Optimal Time Interval for Defining ICU Readmissions

Abstract

Background ICU readmission rates are commonly viewed as indicators of ICU quality. However, definitions of ICU readmissions vary, and it is unknown which, if any, readmissions are associated with ICU quality. Objective Empirically derive the optimal interval between ICU discharge and readmission for purposes of considering ICU readmission as an ICU quality indicator. Research Design Retrospective cohort study Subjects 214,692 patients discharged from 157 U.S. ICUs participating in the Project IMPACT database, 2001–2008. Measures We graphically examined how patient characteristics and ICU discharge circumstances (e.g., ICU census) were related to the odds of ICU readmissions as the allowable interval between ICU discharge and readmission was lengthened. We defined the optimal interval by identifying inflection points where these relationships changed significantly and permanently. Results 2,242 patients (1.0%) were readmitted to the ICU within 24 hours; 9062 (4.2%) within 7 days. Patient characteristics exhibited stronger associations with readmissions after intervals greater than 48–60 hours. By contrast, ICU discharge circumstances and ICU interventions (e.g. mechanical ventilation) exhibited weaker relationships as intervals lengthened, with inflection points at 30 to 48 hours. Due to the predominance of afternoon readmissions regardless of time of discharge, using intervals defined by full calendar days rather than fixed numbers of hours produced more valid results. Discussion It remains uncertain whether ICU readmission is a valid quality indicator. However, having established two full calendar days (not 48 hours) following ICU discharge as the optimal interval for measuring ICU readmissions, this study will facilitate future research designed to determine its validity.

Assessing the Utility of ICU Readmissions as a Quality Metric

“…Th e complete model building strategy may be viewed in e-Appendix 1 . 18,19,25,34,35 Th e Mortality Prediction Model, version III (MPM 0 -III) was used to model severity of illness at ICU admission. 39 Because MPM 0 -III is not valid in patients admitted for coronary conditions, cardiac surgery, or burns, they were excluded from the analysis.…”

“…IMPACT data have been shown to be highly reliable, and ICUs are largely representative of US ICUs. 32 As in prior studies, 18,19,25,[34][35][36] we used a specially prepared version of this dataset in which ICU and hospital admission and discharge dates and times were available for all patients; more detailed patient data were collected on a random sample of 50% to 100% of patients.…”

“…Readmissions to other ICUs may be more likely to be elective 35 ; readmissions after longer intervals are less likely to be related to ICU discharge circumstances. 19 ICU mortality included patients dying during the first ICU admission 37 ; postdischarge mortality included all in-hospital deaths aft er fi rst ICU discharge, including deaths occurring during subsequent ICU admissions. 25 We examined the time period from April 1, 2001, to December 31, 2007 ( Fig 1 ) .…”

“…12,18,19,35 ICU readmissions included patients readmitted within two calendar days to the ICU from which they were originally discharged. Readmissions to other ICUs may be more likely to be elective 35 ; readmissions after longer intervals are less likely to be related to ICU discharge circumstances.…”

“…[17][18][19] Additionally, the performance of ICUs on ICU readmissions is not correlated with performance on other measures of ICU quality, such as mortality or process measures. 3,18,[20][21][22][23][24] Th ese latter observations suggest that ICU readmissions may result from practice variations residing outside the causal pathway to other poor outcomes.…”