While significant progress has been made towards improving health outcomes in low-resource settings, unacceptably high maternal mortality remains a problem. Efforts to improve maternal mortality in low-resource settings did not yield intended results. One hypothesized reason for insufficient maternal mortality progress is poor interpersonal quality of care during childbirth at health facilities. Qualitative studies support the assumptions of quality of care frameworks that connect structural inputs (e.g. drugs and supplies, equipment, human resources) to interpersonal quality. However, there is no quantitative evidence for this relationship. Further, although maternal health researchers developed quantitative tools to measure interpersonal quality of care, the construct is mainly operationalized as a single, bipolar dimension, measured as respectful maternity care (good care) or disrespect and abuse (poor care). To address these limitations, this dissertation used an epidemiologic perspective to test the underlying assumptions of quality of care frameworks and to create a robust measure of interpersonal quality of care. This dissertation consists of three parts: an empirical study to test the hypothesis that structural inputs have a positive effect on interpersonal quality of care; a systematic review of the literature of instruments measuring the construct of interpersonal quality of care and their reliability, validity, and dimensionality; and an empirical study to assess the dimensionality and construct validity of the Maternal Health Interpersonal Quality Scale, a measure of interpersonal quality of care.
The first empirical study did not find meaningful associations between HIV structural inputs and maternal health structural inputs and interpersonal quality of care during childbirth. These results do not support the assumptions of quality of care frameworks nor qualitative evidence linking structural inputs and interpersonal quality of care. The systematic review suggested that the construct of interpersonal quality of care is not well-defined, that few instruments met psychometric standards for adequate reliability and validity, and that studies that assessed the instruments were generally of poor quality. The second empirical study found that interpersonal quality of care formed a two-dimensional, correlated structure, with one dimension measuring respectful maternity care and one dimension measuring disrespect and abuse. Overall, this dissertation used an epidemiologic lens to address an implementation problem in maternal health. While there is a need to improve interpersonal quality of care during childbirth, in order to impact change and to avoid implementation failure, it is imperative to ensure interventions have a strong evidence base and to use validated measures of the construct.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/D8GB3M29 |
| Date | January 2018 |
| Creators | Kujawski, Stephanie Allison |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0022 seconds