Abstract| Volume 14, ISSUE 4, SUPPLEMENT , S35, April 2013

Discriminative sensitivity of the KnowPain-12 pain management survey

      Healthcare provider knowledge, attitude and performance surveys are commonly used yet little is known about the validity and reliability of these instruments. The KnowPain-50 has published psychometric properties1 showing correlations with clinical behaviors, and appears to distinguish between physicians with different levels of pain management expertise, but is burdensome to complete. The purpose of this study was to investigate the reliability and validity of a reduced number of items (KnowPain-12). Phase I used a modified Delphi approach with a group of pain experts to determine face validity and reduce the KnowPain-50 to 12 questions (2 items per original subscale). Phase II used a convenience sample of pain specialists and health professionals generated from public directories. Between April 4 and September 16, 2012, 846 respondents completed the survey. Respondents included RNs (34%), physicians (23%), advanced practice nurses (14%) and other allied health professionals and students. Twenty-six percent of the total sample self-identified as “pain specialist.” Pain specialists more often selected the most correct response to all knowledge assessment items than those who did not identify as a pain specialist, with the exception of one item predicting outcomes from back surgery. The Cronbach’s Alpha for the KnowPain-12 was alpha =.67. Pain specialists scored significantly better (p<.0001) on the Know Pain-12 when compared to general health professionals. The KnowPain-12 appears to have adequate ability to discriminate between pain specialists versus others (AUC=.75). In summary, the KnowPain-12 is a brief survey that may be a useful measure the effectiveness of pain management education programs. Further work is needed to determine sensitivity to change. (1. Harris JM et al, Pain Medicine, 2007.) This work was supported in part by the National Institute of Nursing Research Grant # R01NR012450 and the National Cancer Institute grant #R42 CA141875.