predictable for the next shift. However, as the environment becomes less stable as a result of patient turnover (patient discharges and new admissions) and changes in patient status, projections become less accurate. To the extent that input (patient admissions) to a unit is not predictable, PCS predictions will be less accurate (Seago, 2002).
Historically, hospitals have predicted patient volume—and thereby staffing levels—based on a daily census, typically taken at midnight. A midnight census, however, underestimates care requirements. The actual number of patients cared for by nurses during a 24-hour period is actually the total of four patient types—those on the unit for the full 24-hour period, patient admissions, patient discharges, and patients admitted and discharged on the same day (often referred to as “observation-only” patients). The midnight census fails to capture two of these four—discharges prior to midnight and observation-only patients (Lawrenz, 1992). The latter patients often are in need of care because of an outpatient surgical or radiological procedure and frequently require the same level of care as other inpatients. Moreover, admissions and discharges are well known to be high-activity, time-demanding processes. Thus when hospitals base predictions on projected patient volume as indicated by the midnight census alone, they fail to accurately measure the true level of patient volume—and nurses’ workload (Budreau et al., 1999; Jacobson et al., 1999).
A 1997 study was conducted at a large midwestern medical center to determine the difference between the midnight census and the actual number of patients receiving nursing care in one unit for a 24-hour period. The midnight census counted 23 patients on the unit, while the unit had actually cared for 35 patients during the day. This study also examined periods of peak activity and found that, contrary to historical patterns of peak activity, the evening shift in today’s environment is just as busy as the day shift in terms of total hours of care required. This situation is attributed to several changes brought about by today’s health care environment, such as the fact that a late discharge maximizes the number of hours a patient is in the hospital while avoiding incurring an additional “day” charged at midnight. Late discharges also occur because family members often prefer to pick up the patient after their workday, when they will be home to monitor the patient more closely. Because discharges are late, admissions are also late, awaiting the availability of a bed (Jacobson et al., 1999).
Staffing principles that can help compensate for these problems As of 2002, five states had laws requiring hospitals to develop and follow internal nurse staffing plans (as opposed to adhering to nurse staffing ratios) (ANA, 2002). A number of nursing organizations have put forth guidance for HCOs to follow when developing such staffing plans. This guidance often speaks to issues broader than patient safety, including nurses’ “degree of