The objective of this work was to validate the content of a scale formulated in Spanish for older adults in Mexico, with the aim of comprehensively measuring the access of this population group to outpatient primary-care services. To this end, we carried out a methodological content-validity study in four stages: (1) construction of the scale; (2) evaluation of item legibility; (3) quantitative content evaluation by two groups of judges selected by convenience: participant-judges including older adults with adequate reading comprehension, surveyed in person (n = 23), and expert-judges comprised of researchers specialized in the fields of health services, psychometrics and aging, surveyed online (n = 7); and (4) collection of qualitative feedback from several of the participant-judges (older adults, n = 4). The content was validated both by sequentially examining the level of consensus in the responses of both groups of judges, using the Tastle and Wierman method, and by calculating Aiken’s Validity Coefficient with a 90% confidence interval. The scale contained 65 items pertaining to 10 dimensions of two major constructs: accessibility (n = 39) and personal abilities (n = 26). Five items were eliminated in accordance with the minimum-consensus criterion (0.5). This is the first psychometric scale to be developed in Mexico with the view of integrating the characteristics of health-care services and the abilities of the older adults in a single questionnaire designed to measure the access of this population group to outpatient primary-care services.

This study evaluates the psychometric properties of the Access of Older Adults to Outpatient Primary-Care Health Services Scale (AOAOPHSS), in research conducted among 707 Mexican older adults selected by convenience from 14 rural and one urban locations. The AOAOPHSS explores 10 dimensions of two integrated subscales: Accessibility and Personal Abilities. Data analysis was performed in five phases. First, potentially biased responses were identified. Second, the response efficiency of the items and their association with external variables were evaluated. Third, the basic properties of the scores for the subscales’ dimensions of the AOAOPHSS were identified using non-parametric Mokken Scaling Analysis (MSA). Fourth, the Structural Equation Modeling methodology was used to identify the properties of the internal structure of the latent construct. Finally, reliability and internal consistency were evaluated at both score and item levels. The following findings emerged. 13 items with inefficient response options were removed, and 24 were retained using the MSA. The latent structure of the latter was defined based on 21 items of five Accessibility Subscale dimensions. Its internal consistency reliability ranged between 0.67 and 0.81 (omega coefficients) and between 0.61 and 0.78 (alpha coefficients). Accordingly, this paper discusses the overall implications of using the Accessibility Subscale.