Evaluate construct validity of the Revised American Pain Society Patient Outcome Questionnaire in gynecological postoperative patients using confirmatory factor analysis

Abstract

Background: The Revised American Pain Society Patient Outcome Questionnaire (APS-POQ-R) evaluates the patient-reported quality of pain management in adults. A validated APS-POQ-R is pivotal to guide effective pain management with better patient satisfaction. Previous studies revealed that subscales of "patients' perception of pain management" were unstable cross-culturally. This study aims to evaluate the construct validity of the APS-POQ-R in gynecological postoperative patients with a multi-cultural background using confirmatory factor analysis to allow comparisons among different a priori models at the latent factor level.

Methods: Patients aged 18 years old or above and who were scheduled for gynecology surgery were selected. Three different models with a combination of latent factors were based on a priori hypotheses from previous studies. The root-mean-squared error of approximation, comparative fit index, Tucker-Lewis Index, Chi-squared test, and change in Chi-squared statistic given a change in degrees of freedom between models were used to assess the model fit to the present data.

Results: A total of 302 patients completed the questionnaire. The five-factor model which was based on Gordon's study has an acceptable fit for the data and was superior when compared to the one-factor baseline model. Although the four-factor model, which originated from Botti's study, also demonstrates a good model fit, the "perception of care" construct was excluded in this model. The "perception of care" construct is conceptually important as patient-centered care has become the focus of quality improvement of pain service.

Conclusions: The APS-POQ-R is easy to administer and is useful for quality evaluation in postoperative pain management. The present study demonstrates that a five-factor structure of the APS-POQ-R is the best fitting model in our patient sample. The results of this study provide further evidence to support the use of APS-POQ-R as a measurement tool for pain management evaluation in acute postoperative patients with a multi-cultural background.

Keywords: Construct validity; Factor analysis; Pain management; Patient management; Postoperative pain.

Fig. 1

Flowchart of patient identification and recruitment processes according to STROBE guidelines

Fig. 2

One-factor model (Model I) as a base model

Fig. 3

Five-factor model (Model II)

Fig. 4

Four-factor model (Model III)

Fig. 5

Three-factor model (Model IV)