Developing a Scale for Home-Visit Nurses to Start End-of-life Discussions with Cancer Patients

Abstract

ContextHome-visit nurses find it challenging to determine the appropriate time to initiate end-of-life discussions with cancer patients. ObjectiveThis study aimed to develop the Timing of End-of-Life Discussions (T-EOLD) scale to help home-visit nurses determine the appropriate time to initiate end-of-life discussions with cancer patients and to test its reliability and validity. MethodsThe scale items were developed based on qualitative data extracted from interviews, literature reviews, and expert panel discussions. We conducted a preliminary study involving 93 home-visit nurses and evaluated the construct validity, consistency, and test-retest reliability of the scale. Finally, using a sample of 234 home-visit nurses, we conducted the primary study and assessed the construct validity and scale consistency. ResultsA total of 41 items were initially developed. Floor effect, item-total correlation, good-poor, and exploratory factor analysis in the preliminary and primary studies yielded a three-factor, 16-item model. The model's goodness-of-fit was CFI = 0.94, GFI = 0.90, AGFI = 0.87, and RMSEA = 0.06. Cronbach's alpha for the overall scale was 0.91. ConclusionsThe reliability and validity of the T-EOLD is acceptable, as it is an appropriate scale that home-visit nurses can use to determine the time to initiate end-of-life discussions with cancer patients. However, further study is required to examine T-EOLD's clinical utility, both nationally and internationally.