EOL care. The questionnaire then was piloted with 28
experienced oncology nurses from three hospitals in
one western state. The final questionnaire contained
68 items, including 50 Likert-type items, 4 open-ended
questions, and 14 demographic questions. Data reporting
oncology nurses’ perceptions of EOL care were
published (Beckstrand et al., 2009).
Participants were asked to rate obstacle and supportive
behavior items on two criteria, size and frequency
of occurrence. Items were rated on a size scale
ranging from 0 (not an obstacle or supportive behavior)
to 5 (an extremely large obstacle or supportive
behavior) and a frequency scale ranging from 0 (never
occurs) to 5 (always occurs). After the questionnaires
were returned, results were entered into SPSS®,
version 18.0. Mean scores for obstacle items and
supportive behavior items on size and frequency of
occurrence then were calculated. The size mean score
and frequency mean score for each individual item
were multiplied to yield an impact score (Sawatzky,
1996) for each obstacle and supportive behavior item.
Items then were ranked from highest to lowest impact
score to determine which obstacle and supportive
behavior items were perceived to have the greatest
impact.