mum Nurse Staffing Ratios in Nursing Homes—Phase II Final Report.
Address staffing levels for nurse assistants, who provide the majority of patient care.
With respect to requiring the presence of at least one RN in each nursing home at all times, two previous IOM studies made this same recommendation to achieve better patient outcomes (IOM, 1996, 2001b). This committee additionally calls attention to this minimal staffing requirement as essential to patient safety.
With respect to the recommendation that DHHS specify staffing standards in regulations that would increase with the number of patients and be based on the findings and recommendations of the Phase II DHHS report to Congress on the appropriateness of minimum staffing ratios in nursing homes (CMS, 2001), the committee notes that the thresholds identified in that study above which no further benefit from staffing ratios could be identified are above the staffing levels of 75–90 percent of facilities, depending on the type of staff. However, a minimum standard set by DHHS need not approach the threshold level above which there is no further benefit. In fact, such a standard would go beyond the expectation for a minimum, which is intended to identify situations in which facilities unequivocally place residents at an unacceptable level of risk. The challenge is that there is no absolute minimum level of risk for untoward events that is considered acceptable.
If every single resident in a nursing home experienced an avoidable untoward event, that would clearly be considered unacceptable. But there is no absolute rate of pressure ulcers, weight loss, or hospitalization for infection that is considered unacceptable. Even if one were assured that every event counted was due to a care error (i.e., avoidable), one would still tolerate some errors and would have to choose an acceptable rate. Thus, it is most defensible to set an unacceptable rate based on relative standards using the distribution of rates across facilities and identifying outliers. In the Phase II CMS study, the worst 10 percent of facilities is used as the relative standard, recognizing that an argument could be made for other standards. Facilities in the worst decile, however, were generally three or more times more likely than those at the mean to have untoward events. The study does not propose a specific minimum standard for RNs, licensed nurses, and NAs because agreement must first be reached about what is an unacceptable level of risk. However, data exist from this national study with which to determine the staffing levels for each type of staff that are associated with any level of risk for untoward events.
The committee believes it is feasible to establish a minimum staffing level for each type of staff based on the consensus of experts about unac-