• end of life;
  • instrument development


The purpose of this study was to develop and validate an instrument, the Preferences for Care near the End of Life (PCEOL) scale. Following a literature review and using the domain-referenced approach and a test grid, a large pool of items was generated to reflect five dimensions of preferences for care near the end of life. These were reviewed for relevance and clarity by an expert panel. A convenience sample of 198 adults was secured to provide data for the study of the psychometric properties of the scale. Data screening and item analysis resulted in a final sample of 43 items. A principal factor analysis (PFA) resulted in an interpretable, meaningful five-factor solution. Reliability estimates (internal consistency) for the factors on the multidimensional instrument ranged from .68 to .91. Retest stability estimates for the PCEOL showed correlations of .80 to .94 for factor scores over a 2-week interval. © 2001 John Wiley & Sons, Inc. Res Nurs Health 24:298–306, 2001