Providing Structured Opportunities for Nursing Home Residents to Choose Community Care

The assumption of the inevitability of nursing home placement is convenient for facility personnel, attending professionals, and even family members. Families in particular may resist revisiting a decision that is often agonizing in the first place. But the article by Christy M. Nishita and colleagues in this issue refutes the contention that few elderly long-stay nursing home residents would choose to return to the community. The authors approached the topic head-onFin a small, carefully designed study, they asked the residents or their proxy spokespersons a series of direct questions. They found that 23% of respondents believed that the resident could make a successful transition to the community and that 46% expressed a preference to return to that community. After the residents received information about housing and services in the community, the percentage saying that the resident could make a successful transition went up to 33%, suggesting that the receipt of structured information changed choices. A test–retest interrater reliability study showed that the responses on this screening questionnaire were stable. The authors also compared the responses they received from the direct approach with responses on the single item in Section Q of the Minimum Data Set (MDS) 2.0, which asks the MDS assessor to judge whether the resident wishes to be discharged from the nursing home. The MDS responses were incongruent with those of the investigators, and the MDS results identified fewer who wanted to leave than did the direct interview. Using the inferred MDS data in lieu of actually asking the residents about their preferences for leaving is a pity. Research has shown that residents whose recent MDS cognitive scores are poor can respond reliably to questions about their preferences. The results have implications for assessment of residents. As the authors point out, the MDS is flexible about how assessors gather the information to complete it. For Question Q, the MDS manual even provides a paternalistic nudge by advising against disturbing a long-stay resident or raising unrealistic expectations and suggesting a subtle approach with leading questions such as ‘‘It’s been about 1 year that we’ve known each other. How are things going for you here at (name of nursing home)?’’ Nobody who takes a data collection instrument seriously could endorse this approach to eliciting a preference for remaining at the nursing home or moving out. Research neophytes are taught to ask their questions clearly, directly, and consistently, without skipping questions or rewording them. The MDS is not a questionnaire per se but rather a data collection tool for recording clinical judgments, and this example illustrates biases inherent in unscripted assessments. An analogy can be made to the way physicians and others historically approached eliciting end-of-life preferences by using oblique and indirect questions rather than openly engaging the subject, a phenomenon also noted when professionals assess suicidal ideation, sexuality, or incontinence in older people. Information is improved if professionals use standardized assessments such as have been developed for assessing preferences at the end of life or even a single clear item. In the case of suicidal ideation, the question ‘‘How often have you thought about killing yourself?,’’ was posed to a large group of nursing home residents without unduly upsetting anyone and eliciting information that about 10% of residents think about it all the time. One cannot assess a possibly sensitive topic through avoidance. Moreover, residents typically welcome open discussions of their fears and concerns. If a resident is actively and deeply upset about being in a nursing home, assessors need not worry that their question will remind the resident of this misery. The problem upsetting the resident is not the asessment question, but the fact of being in the nursing home and his/her experiences there. The MDS 3.0, now in the testing phase, includes new sections asked directly of residents, including a direct depression measure. These changes are a major advance, although it would be important not to disqualify too many respondents from direct interviews based on judgments about their cognitive abilities. In general, professional assessors worry too much about getting biased information because of inaccurate responses from older people who may deny their problems or fail to recall the information but too little about biases from their own omissions of questions and inferences about the answers. Over the last 5 years, advocates for community care have become aware of Section Q, a previously little-analyzed component of the MDS. They began counting and publicizing the responses in a simple state-by-state report on the proportion of residents wanting to move out. As often occurs with unanalyzed items in long assessment batteries, quality of the data was poor and the item often left blank. Despite that, the percentages of all residents whose MDS indicates a wish to move out is more than 10% in most states, and sometimes up to one-quarter of the responses are affirmative. Some states have made aggregate Section Q results available according to facility and have shared actual data with transition counselors who act as DOI: 10.1111/j.1532-5415.2007.01560.x