Design and validation of an instrument to evaluate Person-Centered care in health services

Abstract

Background: The concept of quality in health care has evolved, placing greater importance on the patient's needs, culture, and social context, as well as their participation in clinical decision-making, as highlighted by Mead and Bower's Person-Centered Care Model. The aim of the present study was to design and validate an instrument to assess the extent to which healthcare services provided by PEMEX (Petróleos Mexicanos) offer person-centered care according to user perceptions.

Methods: The first phase comprised the development of 57 items based on the analysis of responses from an open-ended questionnaire administered to 30 users of Pemex healthcare services. This questionnaire was designed considering the four factors of the person-centered care model, however, the high correlation between the 4 factors (i.e., r ≥ .80) indicated an overfactoring effect and consequently an increase in the risk of overfitting. Therefore, an exhaustive analysis of the instrument was performed, starting with the review of the individual behavior of each item, and carrying out exploratory and confirmatory factor analysis. Using a sample of 330 individuals, an exploratory factor analysis was perfomed. Afterward, a confirmatory factorial analysis was carried out with 335 participants. Finally, a new confirmatory factorial analysis included 130 participants due to the refinements made in the previous phase. Internal consistency was assessed using Cronbach's α and McDonald's ω at every phase.

Results: The exploratory factor analysis retained 35 items in a single factor that accounted for 49% of the variance with an internal consistency of Cronbach's α and McDonald's ω = 0.97. Because the factorial structure by confirmatory factorial analysis was unsatisfactory, the initial model was refined, leading to the retention of 11 items and a final model with adjustment index of χ2 = 127.53, χ2/gl = 2.89, RMSEA = 0.07, IC RMSEA 0.06 to 0.09, TLI = 0.95 and CFI = 0.96, with an internal consistency of Cronbach's α and McDonald's ω = 0.93. Due to the refinements, a new confirmatory factorial analysis was conducted with suitable goodness-of-fit criteria in most items (χ2 = 151.44, χ2/gl = 3.43, RMSEA = 0.13, IC RMSEA 0.11 to 0.16, TLI = 0.93 and CFI = 0.94), resulting in a Cronbach's α and McDonald's ω = 0.98.

Conclusions: The instrument exhibits suitable psychometric properties to be employed to measure the degree to which medical care is patient centered. This instrument represents a strategy for promoting an innovative healthcare model.

Keywords: Confirmatory factorial analysis; Exploratory factorial analysis; Instrument design; Instrument validation; Person-centered care.

Fig. 1

Design and validation of the Person-Centered Care Instrument

Fig. 2

Factorial structure of the instrument to evaluate Person-Centered Care in Health Services. Values shown on the lines that join the factor with the reactants represent the factor loadings and those values placed on one side of the rectangle that represents the reactants are the R2. Hair et al. [36] argues that standardized factor loadings should ideally be 0.7, which implies a variance of 49%, that is, R2 = 0.49