nipulation. While nursing staff can consciously or subconsciously manipulate patient classification levels (and thereby project needs for greater staffing—a phenomenon referred to as “acuity creep”), managers can also influence staffing need projections through their selection of the staffing allowances for the various acuity levels (Norrish and Rundall, 2001).
PCSs are time-consuming. Most PCSs require nurses to check off activities, treatments, and procedures according to their frequency of occurrence for each patient several times a day. A survey of California hospitals, for example, revealed that three-fourths of hospitals must complete 36 items for their PCS, while half must complete 20 items; one PCS was found to contain 200 items (Seago, 2002).
As a result of the above concerns, researchers studying patient staffing and PCSs note “widespread distrust” of virtually all these tools (Spetz et al., 2000), and the AHRQ evidence report Making Health Care Safer: A Critical Analysis of Patient Safety Practices (Seago, 2001:427) concludes that “although PCSs are used for multiple purposes, they are an inadequate tool for determining unit staffing on a daily or shift basis. In addition, there are numerous patient classification systems and most are specific to one hospital or one nursing unit. The validity and reliability of PCSs are inconsistent and the systems cannot be compared with one another.” Nonetheless, a number of states mandate the use of PCSs. Five states (Texas, Oregon, Kentucky, Nevada, and Virginia) require hospitals to develop and implement nurse staffing plans, methodologies, or systems (ANA, 2002). The California nurse staffing legislation described above requires hospitals to adhere to both nurse staffing ratios and the results of a hospital-selected PCS, whichever is higher. Some speculate that hospitals will have incentives to readjust their PCS staffing factors to predict staffing levels no higher than the ratios mandated by law (Seago, 2002).
For this reason, many researchers, hospital executives, and policy analysts call for more reliable and valid measures of patient acuity (Reed et al., 1998) or the use of approaches other than PCSs to determine nurse staffing in relation to current patient needs (Kovner et al., 2000). Some urge the development of a formula approach to determining nurse staffing levels that would take into account multiple variables in addition to patient acuity, including RN staff expertise; work intensity; unit physical layout; and availability of NAs, other support staff, and physicians (Seago, 2002).
Failure of methods for predicting patient volume to keep pace with changes in hospital admission practices Compounding the above problems with predicting workload on the basis of patient acuity are problems in predicting daily and hourly patient volume. As discussed above, PCSs document the acuity level of patients at a point in time. If patient volume and acuity are assumed to be stable, acuity, workload, and theoretically staffing are