Introduction: Occupied bed census at midnight is typically the parameter used to estimate overall inpatient hospital unit census. For intensive care units (ICU’s), administrative decisions concerning optimal unit size and staffing are typically based on this sole data point. However, the census of an ICU at other time periods, or at peak levels of activity, might more accurately reflect the true needs and characteristics of the ICU. We compared hospital census at midnight with other time periods during the day in a Pediatric ICU (PICU), to assess whether midnight census accurately reflects peak ICU census. Methods: A retrospective review of census data was conducted from August 1, 2012 through July 31, 2013, for a 24-bed PICU located in a children’s hospital. In addition to the census at midnight, the PICU census for 7AM, 4PM, and 7PM were captured. The mean census for all four time periods was compared. In addition, the mean peak census, derived from these time points, was determined and compared with the midnight census. It should be noted that the PICU cannot expand to a census greater than 24 patients. Data are presented as mean ± S.E.M., and analyzed using the two-sample T-Test and ANOVA (Minitab). Results: For the 365 days studied, the mean PICU census at midnight (21.04 ± 0.12) was compared to the mean census at 7 AM (21.66 ± 0.16), 4 PM (21.14 ± 0.12), and at 7 PM (20.67 ± 0.12), and the daily peak census (22.29 ± 0.09). The peak census was significantly higher than the census at midnight as well as the other three time points (p<0.001 for each). The peak census occurred at midnight 40.3% of all days; the census at midnight was the lowest for the day 44.1% of all days. Conclusions: The mean census of a PICU at midnight, as well as other preset time points during the day, do not accurately reflect the peak utilization of ICU beds. In fact, the census at midnight is the lowest census for the PICU nearly half the time. The use of midnight or other preselected time points, as opposed to peak patient census, underestimates the number of patients in the ICU, which can result in inadequate staffing levels, insufficient bed allocation, delays in patient care, and bottlenecks in patient throughput.
© 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins