Quality of Care or Quality of Life--What Really Tells the Story?

by Charles Crecelius, MD, PhD, CMD
Past President, Missouri Association of Long-Term Care Physicians,
Medical Director, Delmar Gardens,
St. Louis, MO

Quality indicators, quality measures, and nursing home "report cards" are just a few of the many "tools" and "ratings" used to evaluate care in nursing facilities. All have been in the news lately because of the Centers for Medicare and Medicaid Services' release of nursing home data to the public in November 2002.

But resident care and quality of life (QOL) are elusive quantities to accurately assess, much less communicate to the general public. Such care is often broken down into concepts of disease states, tasks, and outcomes--none of which necessarily makes sense to nonclinicians. This is where the perceived ideal of a small facility with a well groomed, quiet grandmother rocking in her chair, engaged in simple pastimes, often runs head-long into a nightmarish vision of a thin, child-like, babbling, demented resident confined to a gerichair. Yet, the reality is that QOL could be equally good in both these situations!

Some might say that clinical outcome measures would surely tell us that the first situation is one in which staff are clearly superior. Yet, those familiar with long-term care know that the grandmother may be quiet for fear of retribution by intimidating staff, and that the simple activities she's engaged in may not be of her choosing and may, in fact, be profoundly boring to her.

By contrast, the demented resident may be thin due to advance directives affirming the right to refuse feeding tubes in the face of severe dysphagia; the gerichair may be the least restrictive environment she can function in; and her occasional agitation may be the result of staff's efforts to minimize psychopharmacologic drugs.

Current Measures Fall Short

Quality of care currently is typified by collection of resident data via the Minimum Data Set to arrive at quality indicators and quality measures; simply put, it is concerned with the structure, processes, and outcomes of clinical matters. Structure refers to staffing variables and facility-resident mix; process refers to methods and utilization of services; and outcomes relate the health status of residents to processes used (or not used). These measures are centered on geriatric diseases and syndromes; by default, their relative absence or presence denotes quality. Despite the fact that undesirable outcomes are often the result of complex situations and conditions, they nevertheless may end up being equated with poor care.

CMS recently had the validity and reliability of quality indicators and measures studied and, not surprisingly, some fell short of the mark.¹ Behavior symptoms, weight loss, and antipsychotic use are among the assessments that don't accurately represent care processes or, presumably, quality.

Certain measurements, such as prevalence of urinary catheters, infections, and pressure ulcers, do appear to accurately reflect clinical care upon analysis. However, while in theory the important part is the process by which these clinical parameters are managed over time, most indicators are prevalence based, and may not reflect efforts made to deal with a complex, frail population. Without longitudinal tracking, it is impossible know if many measures are changing for better or worse. Furthermore, virtually none of the quality indicators or measures represent a positive aspect of resident life. The public can access abuse and neglect citations, but how can they find out about kindness and extraordinary care?

Certainly, QOL is more than the sum of the diseases, syndromes, and adverse events that beset our elders. The structure, processes, and outcomes of clinical care still do not define all that transpires in nursing homes--nor do they solely define quality. Many subjective parameters have been cited as important by residents or their families, and these can be used to help define quality as it relates to the individual in a particular situation (see box at right).

The subjective nature of QOL assessments means that there needs to be more assessor-resident interactions when filling out standard forms that the MDS requires. Ideally, trained personnel should do such assessments via resident interviews. If this is not possible, for example because of cognitive impairment, then direct observation of positive and negative aspects of the resident's daily life in the facility needs to be assessed.

Several QOL measures have been reported for cognitively intact nursing home residents. For example, REAL (Real Experiences and Assessment of Life), developed under an NIH grant at Vital Research in Los Angeles, CA, measures via yes-no questions six factors: help and assistance; communication with staff; autonomy and choice; companionship; safety and security; and food and environment.² The Confidence Satisfaction Survey³ uses many similar items as REAL. Although administered by facility staff, it requires interpretation by the developer, Life Services Network of Illinois in Chicago. However, neither assessments are standardized or validated by large group studies.

Determining QOL for residents with dementia has been done by both direct interview and indirect measures using observation, interview, or caregiver reports. For example, Rabins and colleagues have reported a caregiver-based QOL instrument with five domains: social interaction; awareness of self; feeling and mood; enjoyment of activities; and response to environment.⁴

Brod and colleagues have developed a direct resident interview instrument with five domains: aesthetics; positive affect; negative affect; self-esteem; and feelings of belonging.⁵ Although inter-rater reliability has been demonstrated for some of these tools, there are no larger validation studies.

At least one provider has attempted to determine quality using "in-house" indicators, derived from quality indicators, quality assurance projects, and staffing-related considerations.⁶ However, there were no QOL indices, and the only consumer-related measure, family satisfaction, was dropped from analysis. Many nursing homes gather data regarding family perceptions, but how many relay them to staff, track them systematically, or publish them? How can any measure really represent quality without considering the life of the resident?

QOL & Risk Taking

The Institute of Medicine's recent report on long-term care quality underscores the importance of QOL, noting as one of nine general principles, "measures of the quality of long-term care should incorporate its many dimensions, especially quality of life."⁷ Rosalie Kane, PhD, one of the report's authors, described 11 domains for use in nursing home QOL assessments that encompass non-medical outcomes experienced by the individual, rather than structures or processes⁸ (see box at right).

Implicit in the concept of QOL is the consideration of risk taking--i.e., a clinical adverse event may be acceptable if QOL is significantly improved. An advanced Alzheimer's resident with poor oral intake and impaired swallowing is normally not an appropriate candidate for artificial feeding. The resulting poor nutrition and subsequent debility will risk decubitus ulcers, loss of ADLs, and weight loss, among other care-related quality issues.

But are these measures always the most important? Isn't it important to determine whether residents are allowed to walk in the facility, despite the risk of falls? Are they given their favorite foods, even if the textures aren't strictly "the best"? Can we measure whether residents are treated with respect, provided with appropriate activities, or feel secure--even if, to accomplish this, it means risking medical complications?

Measuring quality at end of life is especially important, since about one-fourth of all Americans die in nursing homes. One-third of new nursing home admissions will die within a year. It is naive to not expect bad outcomes. We can try to risk adjust measures so that they are not influenced by dying patients--but excluding such patients still does not give us any information about the equally important QOL considerations in declining residents. How do we measure their experiences in a meaningful way?

One worthwhile response to these dilemmas has been the addition of pain management assessment in quality measures. However, even important clinical symptoms such as pain should not necessarily take precedence over QOL issues, which may be more responsive to the nursing home staff's efforts and interventions than a possibly innate clinical characteristic.

Quality indicators and measures are important clinical guides, but cannot, in and of themselves, define the nursing home experience. More research is needed to improve the standardization and utility of QOL measures. Providers, regulators, and consumers deserve accurate methods of assessing the entire nursing home experience. This will only happen when quality of life, as well as quality of care, can be meaningfully presented.

back to top