Siegel, Lianne

IMPORTANCE: Provider staffing models for ICUs are generally based on pragmatic necessities and historical norms at individual institutions. A better understanding of the role that provider staffing models play in determining patient outcomes and optimizing use of ICU resources is needed. OBJECTIVES: To explore the impact of transitioning from a low- to high-intensity intensivist staffing model on patient outcomes and unit composition. DESIGN, SETTING, AND PARTICIPANTS: This was a prospective observational before-and-after study of adult ICU patients admitted to a single community hospital ICU before (October 2016–May 2017) and after (June 2017–November 2017) the transition to a high-intensity ICU staffing model. MAIN OUTCOMES AND MEASURES: The primary outcome was 30-day all-cause mortality. Secondary outcomes included in-hospital mortality, ICU length of stay (LOS), and unit composition characteristics including type (e.g., medical, surgical) and purpose (ICU-specific intervention vs close monitoring only) of admission. RESULTS: For the primary outcome, 1,219 subjects were included (779 low-intensity, 440 high-intensity). In multivariable analysis, the transition to a high-intensity staffing model was not associated with a decrease in 30-day (odds ratio [OR], 0.90; 95% CI, 0.61–1.34; p = 0.62) or in-hospital (OR, 0.89; 95% CI, 0.57–1.38; p = 0.60) mortality, nor ICU LOS. However, the proportion of patients admitted to the ICU without an ICU-specific need did decrease under the high-intensity staffing model (27.2% low-intensity to 17.5% high-intensity; p < 0.001). CONCLUSIONS AND RELEVANCE: Multivariable analysis showed no association between transition to a high-intensity ICU staffing model and mortality or LOS outcomes; however, the proportion of patients admitted without an ICU-specific need decreased under the high-intensity model. Further research is needed to determine whether a high-intensity staffing model may lead to more efficient ICU bed usage.

Often clinicians are interested in determining whether a subject's measurement falls within a normal range, defined as a range of values of a continuous outcome which contains some proportion (eg, 95%) of measurements from a healthy population. Several studies in the biomedical field have estimated reference ranges based on a meta‐analysis of multiple studies with healthy individuals. However, the literature currently gives no guidance about how to estimate the reference range of a new subject in such settings. Instead, meta‐analyses of such normative range studies typically report the pooled mean as a reference value, which does not incorporate natural variation across healthy individuals in different studies. We present three approaches to calculating the normal reference range of a subject from a meta‐analysis of normally or lognormally distributed outcomes: a frequentist random effects model, a Bayesian random effects model, and an empirical approach. We present the results of a simulation study demonstrating that the methods perform well under a variety of scenarios, though users should be cautious when the number of studies is small and between‐study heterogeneity is large. Finally, we apply these methods to two examples: pediatric time spent awake after sleep onset and frontal subjective postural vertical measurements.