Dimensions analysis of the Hospital Survey on Patient Safety Culture questionnaire in Iran: Psychometric properties

Abstract

Aim: The present study aimed to investigate the possible alternative factorial structure of the patient safety culture model in Iran. Methods: This study was performed based on data collected by Hospital Survey on Patient Safety Culture (HSOPSC) questionnaire from 420 staff in four hospitals. Internal consistency reliability and construct validity were evaluated by Cronbach's alpha and correlation analysis. Exploratory factor analysis (EFA), confirmatory factor analysis (CFA) and structural equation modeling (SEM) were used to investigate the possible alternative factorial structure, examine and confirm the obtained structure, alternatively. Kaiser-Meyer-Olkin measure and Bartlett test were calculated to determine the factor ability of sample and fit of the factor analysis, alternatively. SPSS and AMOS version 25 were used. Results: EFA identified 12 dimensions which one dimension has been also created from a new question. Distribution of

Psychometric evaluation of the Persian version of the Hospital Survey on Patient Safety Culture Questionnaire (HSOPSC) in neonatal intensive care units: a methodological study

“…Also, the sample size and data obtained from relatively small sample sizes in different studies can explain this lack of sameness in factor structures compared to the original version [ 15 ]. The discrepancy in the results is explained according to the various psychometric analysis methods used [ 18 , 23 ]. However, what matters is that the psychometric evaluation be done in such a way that the results can be trusted.…”

“…Regarding the dimensions of Organizational learning-continuous improvement and Teamwork within/across hospital units , Cronbach’s alpha coefficient was low, as in the Bulgarian version [ 38 ]. The dimension “ Organizational learning-continuous improvement ” is reported as unacceptable or low in several studies [ 14 , 18 , 49 ]. In addition, in many of the previous studies, internal consistency has been lower than in the original questionnaire in most areas [ 12 , 44 ], which could be due to the differences in the number of items in these dimensions.…”

“…Among these, can mention the studies conducted in Japan and Taiwan, in which Cronbach’s alpha coefficients were 0.47–0.88 and 0.26–2.83, respectively [ 50 ]. In other psychometric evaluations performed in Iran, the alpha coefficients obtained for the dimensions of the questionnaire were unacceptable and questionable [ 15 , 18 ]. Regarding the test-retest results, this questionnaire has proper stability.…”

“…In another study in 2020, although the 12-dminesion model fit the data, the distribution of the items in the dimensions (except two dimensions) was reported to be different from the original version. The results of internal consistency were reported to be unacceptable for all the dimensions of the questionnaire, except for the dimension frequency of events reported [ 18 ]. The results of these Iranian studies are challenging and make it difficult to use the results.…”

“…In addition, several studies have shown the need to study the psychometric properties of this tool in different parts of health care [ 14 , 18 ], and emphasized the evaluation of the extent and the level of safety culture in order to improve the level of patient safety and provide optimal care [ 10 , 24 ].…”