Why adverse outcomes are not "par for the course." (How to be a Quality Standout: Adventures in Nursing Homes CQI)

Over the years a mindset has grown in some nursing homes that such resident outcomes as pressure sores, disruptive behavior or verbal aggression are "to be expected," at least to a degree, considering that their populations tend to be frail and elderly. Another possible contributor to this philosophy has been the influx in recent years of mentally ill elderly admitted to nursing homes due to "deinstitutionalization" (or, as I call it, transinstitutionalization). While it would be inaccurate to say that adverse outcomes of any frequency are considered to be "acceptable," they are not uncommonly looked upon as par for the course.

Now, however, there is evidence that the opposite is true -- that through careful attention to good-quality care, their incidence can be reduced virtually to nil. Recently my colleagues and I published a study comparing the "best" and the "worst" of 30 nursing homes ranked for prevalence of pressure sores.(*) The "best" had a prevalence of zero in residents who had been in the facility at least six months; the "worst" had an prevalence of 17.2% for such residents. Similarly, the best facility -- we called it "Facility A" --had a prevalence of only 2% of aggressive or disruptive behavior and a conversion to such behavior after six months of 0%. By contrast, the worst -- Facility B -- had prevalences of 15.6% and 32% for aggressive and disruptive behavior, respectively, and six-month conversion to such behavior of 9% and 11 %.

Decline in activities of daily living (ADLs) over six months showed similar discrepancies between facilities A and B: 2.9% decline vs. 47.8% for eating ADLs; 7.7% vs. 59.4% for mobility ADLs; 8% vs. 55.6% for transfer ADLs; and 9.5% vs. 61.5% for toileting ADLs. Furthermore, the nutritional status of residents in Facility A was superior to that of those in Facility B.

Despite these marked differences in outcomes, both facilities were comparable in terms of the populations they served. Residents were generally similar in age, sex, length of stay, degree of dependency, frequency of dementia, level of nursing care, and other severity measures. The most significant difference was that Facility B had more schizophrenics among the resident population, 28% vs. Facility A's 12%.

What explained the differences? It is true that the types of facility were quite different: Facility A was a 60-bed, rural-based, free-standing nursing home; Facility B was a 280-bed teaching facility based in an urban medical center. This in itself, however, was obviously not explanatory. Methods of resident care could have been compared, but the study was not designed to investigate these (those studies are underway). Nevertheless, other potentially significant differences did emerge.

Facility A had a 48% higher ratio of medical personnel to residents than did Facility B, a 49% higher ratio of nursing personnel to residents, a 54% lower staff turnover rate, and 37% higher expenditures per bed per day. In partial explanation, it is worth noting that a nursing facility based at a tertiary, teaching hospital, such as Facility B, must compete for budgetary funds with many high-tech acute medical and surgical services and technologies, as well as physician training requirements, and these may take priority over the long-term care unit. Even so, the parallel of discrepancies in financial and management support with discrepancies in clinical outcome is instructive.

Would the higher proportion of schizophrenics in Facility B have had some bearing on its relatively negative outcomes? This may have indeed adversely impacted the facility's environment -- and it is well-known that the frail elderly are highly sensitive to their environments. Disruptive behavior is contagious, and it probably doesn't take a great deal of it to become self-perpetuating. Certainly the environmental difference between the two facilities was quite noticeable, even on a brief visit.

This raises questions about the wisdom of depending as heavily as we do on nursing homes to look after the chronically mentally ill, who as a practical matter are often mixed in with the other residents. Perhaps the Preadmission Screening and Annual Resident Review (PASARR) regulations will provide some relief, though they themselves raise many questions as currently implemented -- a worthy subject, no doubt, for separate investigation.

In any event, the difference in schizophrenic population was not sufficient to account for all the differences in behavioral disruption, which remained significant for the organic brain syndrome and non-neuropsychiatric subgroups in both nursing homes.

We and other investigators are in the process of studying quality of care issues in greater depth, and my guess is that we will have a firmer grasp of the factors important to this within a few years. For now, however, we can see that the "irreducable" adverse outcomes expected of nursing homes (pressure sores, loss of ADLs, weight loss and behavioral disturbances) are anything but "irreducable". They can be reduced to virtually nil with sufficient support and resources provided by nursing home management.

* Rudman D. Mattson DE, Alverno L, Richardson TJ, Rudman IW. Comparison of Clinical Indicators in Two Nursing Homes. J Am Geri Soc 1993;41:1317-1325.

Daniel Rudman, MD, is Professor of Medicine at the Medical College of Wisconsin, Milwaukee.