Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website

2006 HSR&D National Meeting Abstract


3089 — Feasibility of Including Self-Reported Preferences in VA Nursing Home Resident Assessments

Author List:
Housen P. (VA Greater Los Angeles HSR&D Center of Excellence)
Shannon GR (VA Greater Los Angeles HSR&D Center of Excellence)
Simon B (VA Greater Los Angeles HSR&D Center of Excellence)
Cadogan MP (VA Greater Los Angeles HSR&D Center of Excellence)
Orlando M (RAND CORP.)
Sohn L (VA Greater Los Angeles HSR&D Center of Excellence)
Buchanan J (Harvard University)
Jones M (RAND CORP.)
Saliba D (VA Greater Los Angeles HSR&D Center of Excellence)

Objectives:
Improving care quality is a high priority for the VA National Nursing Home (NH) Service. A primary focus of this effort is fostering a culture shift toward resident-centered care. In order to develop resident-centered care plans, it is important to systematically ascertain residents’ preferences for day-to-day care and activities. High prevalence of cognitive impairment in NHs might impede efforts to determine these preferences. This study examines the feasibility of assessing resident preferences across a full range of cognitive abilities.

Methods:
Based on recommendations from an expert panel, researchers developed an instrument soliciting resident self-reported preferences and their relative importance. In all, 24 items were identified in an environmental scan, refined in cognitive interviews, and mapped onto seven Quality of Life (QoL) domains. We examined response and completion rates of residents in two VA NHs in order to evaluate instrument performance. Cognitive ability was assessed using the Cognitive Performance Scale (CPS), a measure of cognitive status highly correlated with the Mini-Mental Status Examination.

Results:
The sample was average age 74.2 (SD=10.6) and 33% non-white. Of 143 interviews initiated, 83% were completed without interruption. Item by item response rates for 117 residents ranged from 94%-100%. Mean interview duration was 8 minutes (SD=6). The remaining 25 interviews were truncated after respondents were nonresponsive in two domains. Completion rates varied by cognitive ability. Approximately 94% of cognitively intact participants (CPS<2, n=95) and 78% of those with mild to moderate impairment (CPS=2 or 3, n=27) completed the interview, compared to 32% with moderate to very severe impairment (CPS>3, n=19). Items also varied in test-retest reliability.

Implications:
High completion rates and short average administration time indicate obtaining preference information directly from NH residents with mild to moderate cognitive impairment is feasible. Results suggest the 24-item instrument is problematic for only the most cognitively impaired residents.

Impacts:
Incorporating preferences in care planning may improve NH resident QoL by increasing perceptions of control and thus improving physical and mental outcomes. Next steps should test inclusion of these items in standardized assessments already scheduled in NHs, and development of approaches to aid staff in incorporating resident preferences into care plans.