Using Integrative Mixed Methods to Examine Type 2 Diabetes and Inform Prevention Interventions

Introduction: Type 2 diabetes (T2D) disproportionately affects Latinos, who have exhibited a higher relative risk of developing T2D when compared with non-Hispanic whites (WHO, 2013). Although many gaps in the literature persist regarding the reasons for this health disparity among Latinos, it is known that dietary and exercise behaviors are central to effectively reducing the risks of T2D.  One of the major challenges in the prevention and control of T2D involves understanding the cognitive, behavioral, environmental, and cultural factors that operate as risk or as protective factors (which can motivate risk reducing behaviors). To examine such factors in a culturally-sensitive manner, the Integrative Mixed Methods (IMM) methodology was used to generate probing responses to these questions (Castro et al., 2010).

 Methods: This presentation will provide detailed information on how IMM can be utilized to design and implement a study that uses “focus questions” to generate detailed text narratives that can be used to study deep-structure responses.  The sample consists of 31 adult Latinos and Latinas from the Maricopa County Diabetes Registry, who were at risk for developing T2D (based on their fasting blood glucose levels). The IMM methodology will be described as generating: (a) response codes, answers to a focus question, (b) their organization into thematic categories, as identified by two independent coders, and with final identification via a roundtable analysis, and (c) how these thematic categories (categorical data) can be converted to numeric thematic variables.  Specific illustrative examples will be given for the application of this methodology in the analysis of motivational factors for diet and exercise, as examined for each of two focus question domains: (a) the perceived causes of diabetes, and (b) sources of familial social support.     

 Results: The use of “in vivo” coding using Atlas.ti will be demonstrated, for yielding thematic categories and thematic variables as generated for the two noted domains.  From preliminary analyses of these text narratives, emerging themes for beliefs about causes of T2D and sources of familial support will be discussed.   

 Conclusions: This presentation will show how emergent thematic variables can be generated in a rigorous manner, and then used to conduct IMM data analyses that can test specific hypotheses, thus providing more potent conclusions about developing T2D. The use of this IMM methodology can thus generate more detailed and in-depth contextualized results that can inform the development of culturally-sensitive prevention interventions for T2D.