Development and psychometric characteristics of the pediatric inpatient experience survey (PIES)

作者：Ziniel Sonja I^*; Connor Jean A; Graham Dionne; Kupiec Jennifer Koch; Rauscher Nina A; Growdon Amanda S; Berger Anne; Jenkins Kathy J; Harris Sion Kim

来源：International Journal for Quality in Health Care, 2016, 28(2): 191-199.

DOI：10.1093/intqhc/mzv118

• pediatric inpatient experience
• validity
• reliability
• quality of care
• family-centered care

To study the psychometric properties of the Pediatric Inpatient Experience Survey (PIES), a mail and phone survey for parent reporting of family-centered aspects of inpatient care experiences.
Two waves of cross-sectional survey data were collected by mail and phone in 2009 to design a measurement instrument with good psychometric characteristics. Additional cross-sectional data from a mail administration in 2011 confirmed the measurement domains.
Free-standing pediatric hospital in the northeastern USA.
A convenience sample of English-speaking parents of hospitalized children, stratified by patient type (medical versus surgical) and previous stays at this hospital (yes versus no), constituted the instrument design phase. Four hundred and seventy-nine (63%) of those approached agreed to participate and were randomly assigned to mail or phone survey administration. Four hundred and one of these respondents completed the first wave of the survey and 354 respondents completed the second wave. A shortened instrument was mailed to parents randomly selected from patient discharge records. Data from 929 parents (response rate: 36.2%) were used for confirmatory analysis of the created measurement domains.
The main outcome measures of this psychometric validation study were individual item performance, test-retest reliability, internal consistency, and construct validity.
The resulting survey includes 61 items with 35 rating items with satisfactory test-retest reliability loading on eight domains. The factor structure was supported by Cronbach's alpha and confirmatory factor analysis. The survey supported construct validity in distinguishing between medical versus surgical and first time versus previous hospital stay groups known to differ with regard to satisfaction. Comparing mail and phone administrations, differences in scores were exacerbated in domain scores and showed the need for mode adjustment.
PIES shows satisfactory test-retest reliability, internal consistency, and construct validity. A new domain measuring emotional connectedness to staff and the hospital is highly correlated with overall satisfaction.

• 出版日期2016-4-1