2012 HSR&D/QUERI National Conference Abstract
3012 — Implementation of Person-Centered Care Practices in VHA CLCs: Reliability of a Person-Centered Care Staff Survey
Sullivan JL, Meterko M, and Parker VA, COLMR, Boston University; Adjognon O, Stolzmann K, and Baker E, COLMR; Ballah K, HCFE;
The research objective of this study is to test and revise a staff assessment of person-centered care (PCC) within the Veterans Health Administration (VHA) Community Living Center (CLC) setting. Although VHA has undertaken the culture change process to move toward a PCC model, it is unclear how and to what extent CLCs are implementing PCC model practices. Further study is necessary to understand the variation in implementation of PCC practices across VHA CLCs.
Starting with measures of PCC initially developed through a Better Jobs Better Care (BJBC) study, we conducted two focus groups with CLC staff to assess applicability to the VHA setting. We then 1) modified the questionnaire based on respondent feedback, 2) administered the revised PCC survey via internet to 958 staff at eight VHA CLCs, and 3) examined the psychometric properties of the PCC instrument using multitrait analysis.
Based on focus group findings, items on the survey were all applicable to the VHA setting and only minor modifications were made to the questions. Following the revisions, we tested the psychometric properties of the 50-item BJBC PCC instrument using data obtained from 8 VHA CLCs. The response rate was 35%. The pattern of convergent and divergent correlations present strong evidence of the reliability and validity of the hypothesized scales. Cronbach’s alpha coefficients ranged from 0.84-0.92 and were comparable to those reported outside the VHA setting.
We observed excellent evidence for discriminant validity (90%), convergent validity (100%), and reliability (all scales demonstrated alpha levels exceeding the criterion for group comparisons). Thus, our results support the broader use of the BJBC survey within VHA. To boost response rates in future studies, we would advocate adding a paper mode of survey administration.
It is possible to measure key domains of PCC in a psychometrically sound manner using the BJBC survey. This instrument may be used in future studies to better understand the impact PCC has on staff and resident outcomes in VHA. The PCC survey is valuable because it is a staff assessment and measures some PCC domains omitted by the Artifacts Tool.