| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
GUEST EDITORIAL |
Department of Medicine, Division of Geriatrics, University of California–Los Angeles, California. UCLA and Jewish Home for the Aging Borun Center for Gerontological Research, Reseda, California. Veterans Administration Greater Los Angeles Healthcare System, Sepulveda Geriatric Research, Education, and Clinical Center, Sepulveda, California.
The report by Kane and colleagues (1) in this Journal issue makes an important contribution to our understanding of how to measure the quality of a resident's nursing home (NH) experience. Their finding that 60% of residents, in a study using (and here's the key phrase) highly inclusive selection criteria, provided useable information about their quality of life should spur change in our approach to assessing quality in NHs.
Their study supports the argument that the current quality assessment process should bolster efforts to obtain information directly from NH residents, partly because no one is better positioned to comment on quality than residents themselves, but also because their reports will provide balance to the extensive information currently obtained from staff reports. As it now stands, the state and federal survey process for evaluating NH home care is biased against resident assessments of the care they receive. Although some NH residents are interviewed about their care during on-site survey visits, Kane and colleagues point out that many more residents could be interviewed than is currently the case, and they present guidelines for identifying how many residents can be interviewed based on cognitive functioning criteria. The required Minimum Data Set (MDS) for NHs, which includes resident assessment data, instructs staff to interview residents about such conditions as pain and depression before making categorical judgments about residents' status. Thus, these staff ratings on the MDS are presumably based on what the residents said and/or their behavior. Unfortunately, evidence that both pain and depression are significantly underestimated by MDS data suggests that resident interviews are oftentimes skipped in the MDS assessment process (2,3).
One possible reason that NH home staff and even federal and state survey staff fail to interview more residents about their mood and other conditions is that they do not believe cognitively impaired residents can provide useful information. The Kane study shows that this stereotyped belief is simply inaccurate and calls into question the heavy reliance on MDS data within the survey process, which emphasizes staff ratings of resident behavior over direct responses from residents. The only defensible argument for this indirect approach is that there are not enough NH residents who can provide reliable interview responses to a standardized protocol; hence, subjective judgments by staff are unavoidable. Kane and colleagues attack the validity of this assumption, as have several other researchers. Prior studies have shown that many NH residents are capable of completing standardized pain interview protocols (4). A surprisingly large number of residents can also provide accurate information about the kind of care they prefer [e.g., the number of walking assists they prefer each day] and receive [e.g., the number of walking assists they actually received on a given day (5)]. These findings argue for the increased use of standardized interview protocols along with explicit criteria for selecting residents for interview in the quality assessment process.
These interview and selection protocols would improve the science of NH quality assessment in several important ways. First, the protocols would allow for measurement of care dimensions that currently are not addressed in the MDS (e.g., quality of life domains and residents' reports of the kind of care they actually received); resident selection criteria that is based on cognitive performance information would prevent both NH and quality assurance personnel from subjectively excluding residents from interviews. Second, the direct measurement of residents' perceptions of care and quality would help improve the accuracy of medical record and MDS data, both of which currently represent the major sources of quality information. Other studies have shown that both medical record and MDS data can be blatantly inaccurate or at least at odds with other measures of care quality (e.g., direct observation or resident interview data). For example, a recent study found no relationship between medical record documentation of resident repositioning, which indicated that 95% of the residents were repositioned every 2 hours, and data generated from electronic, thigh-worn monitors that continuously measured each resident's position and movement, which showed that 78% of the residents were not moved or repositioned by staff for at least 3 hours (6). Other studies have reported discrepancies between medical record documentation of toileting assistance and oral food and fluid intake and either direct observations by research staff or resident reports of these care processes (7,8). The collection of quality of care information independent of the medical record and the MDS is an important accuracy check that may prevent erroneous conclusions derived from medical record information alone. The collection of these independent data might even motivate NH home staff to document care processes more accurately.
Third, resident reports about the care they receive and their quality of life could be used to evaluate improvements in NH performance, thus reducing the need to use controversial "risk adjustments" to MDS-based quality indicators. A strong argument can be made that quality-of-life domains (e.g., feeling safe) and resident reports of the kind of care they receive and prefer (e.g., walking assistance) are less subject to a resident's health status than are clinical care domains. In addition, NH staff should be able to exert a greater influence on resident quality of life than on such conditions as weight loss or pressure ulcers, which in some cases may be unavoidable even when care is provided at the highest level. In short, a quality improvement process that draws on resident reports would enable NH care providers to affect quality outcome measures by, for example, providing walking assistance to ambulatory residents who, regardless of their physical or mental condition, prefer such assistance. In this example, the effect on the outcome measure would be independent of whether the walking assistance resulted in improved health (it may well not) or a higher score on MDS-based quality indicators (again, it may not).
Several feasibility issues must be addressed before implementing a quality assessment process that more actively and directly measures resident perceptions of care. Foremost among these issues are labor intensity and resident response burden. The Kane interview requires approximately 60 minutes to complete and does not examine some aspects of care that should be addressed in a comprehensive quality assessment. It is preferable that each resident interview require less than 1 hour if a sufficient number of residents are to be assessed during a short on-site visit for the purpose of forming a defensible judgement about a NH's quality of care. These cost-related feasibility issues, however, can be readily delineated and resolved. Experts likely could identify and agree on what should be included in a comprehensive interview protocol and, using objective criteria from the literature, predict how many residents should and can be interviewed. Methodologies could then be developed so that the protocol can be used within the cost constraints of the quality assessment process. These restraints will vary by facility, and probably will be less burdensome to in-house quality improvement programs than to the federal or state survey process.
In developing interview protocols, attention must be paid to concerns about reduced expectations among residents or acquiescent response biases. In the Kane study, these problems may be evidenced by the relatively high average scores for most of the domains (above three on a four-point scale). Many individuals would be surprised by these relatively high ratings in light of their own experiences with and knowledge of NH care quality; they might suspect that these high scores reflect a reluctance among residents to complain. Two recent studies have discussed acquiescent response pressures among NH residents and have identified interview methodologies that appear to be less sensitive to these pressures (5,9).
The key point is that, with a relatively small effort, we can overcome barriers to using the standardized interview protocol developed by Kane and colleagues and other research groups and thereby improve the quality assessment process in NHs. These barriers should not hinder our efforts to immediately improve quality measurement by more actively involving NH residents in the process and by systematically identifying residents for interviews based on objective criteria. The Kane study presents sound evidence that many residents with cognitive impairment can still provide useful information about their care and quality of life in the NH. This study, along with two reports by Simmons and Schnelle, helps delineate specific guidelines for selecting residents to interview (10,11). The question now is not, "Should we improve the quality assessment process by interviewing more residents?" but rather, "How do we do interview more residents within the cost constraints of the quality assessment process?"
Received November 20, 2002
Accepted January 8, 2003
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  P. D. Sloane, S. Zimmerman, C. S. Williams, P. S. Reed, K. S. Gill, and J. S. Preisser Evaluating the Quality of Life of Long-Term Care Residents With Dementia Gerontologist, October 1, 2005; 45(suppl_1): 37 - 49. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| All GSA journals | The Gerontologist |
| Journals of Gerontology Series B: Psychological Sciences and Social Sciences | |
