S Haddad1, P Fournier, L Potvin. 1. Groupe de Recherche Interdisciplinaire en Santé, Université de Montréal, Québec, Canada. haddad@ere.umontreal.ca
Abstract
INTRODUCTION: Methodologies that have been developed and validated in accordance with accepted scientific standards are needed to monitor and assess the quality of primary health care in developing countries. OBJECTIVE: To present the results of reliability and validity testing of a new instrument of measurement intended to document the user's opinion on the quality of primary health care services. METHODS: The 20-item scale includes three subscales related to health care delivery personnel and facilities. There were 241 people in one city and two villages in Upper Guinea who responded to the questionnaire. An item analysis preceded the test of psychometric properties of the three subscales and of the total score. Reliability was estimated by analyses of internal consistency and the Cronbach's alpha coefficient. A variety of statistical procedures were used to test factorial validity, trait validity (convergent and discriminant) and nomological validity. RESULTS: The reliability of the subscales ranged from 0.71 to 0.88. The validity analyses supported the initial dimensionality and suggested good construct validity. CONCLUSION: Results confirm the value of the use of the scale developed and highlight the need to take into account the diversity of how quality is perceived by lay people in developing countries. It is suggested that the process of formalization of this type of measurement scale be pursued.
INTRODUCTION: Methodologies that have been developed and validated in accordance with accepted scientific standards are needed to monitor and assess the quality of primary health care in developing countries. OBJECTIVE: To present the results of reliability and validity testing of a new instrument of measurement intended to document the user's opinion on the quality of primary health care services. METHODS: The 20-item scale includes three subscales related to health care delivery personnel and facilities. There were 241 people in one city and two villages in Upper Guinea who responded to the questionnaire. An item analysis preceded the test of psychometric properties of the three subscales and of the total score. Reliability was estimated by analyses of internal consistency and the Cronbach's alpha coefficient. A variety of statistical procedures were used to test factorial validity, trait validity (convergent and discriminant) and nomological validity. RESULTS: The reliability of the subscales ranged from 0.71 to 0.88. The validity analyses supported the initial dimensionality and suggested good construct validity. CONCLUSION: Results confirm the value of the use of the scale developed and highlight the need to take into account the diversity of how quality is perceived by lay people in developing countries. It is suggested that the process of formalization of this type of measurement scale be pursued.
Authors: Emily Dansereau; Felix Masiye; Emmanuela Gakidou; Samuel H Masters; Roy Burstein; Santosh Kumar Journal: BMJ Open Date: 2015-12-30 Impact factor: 2.692