EXAMPLES OF MIXED METHODS APPROACHES IN THE INTERVENTION DEVELOPMENT PIPELINE
In this section, we describe examples of studies employing mixed methods along the intervention development pipeline—intervention development, intervention evaluation in efficacy and effectiveness trials, implementation in diverse community and service settings, and understanding results. Mixed methods can augment an RCT or intervention design by gathering exploratory data before, during, or after a trial (Creswell & Plano Clark, 2011). Mixed methods studies provide insight by giving study participants a chance to describe how an intervention might be modified before full deployment, or to explain outcomes observed of a trial (Farquhar et al., 2011).
Barg and colleagues (2006) examined how patient notions of depression differed from or were similar to standard definitions. Older adults were administered a structured interview assessing symptoms of depression (“quantitative methods”), but also were asked to provide the investigators with ideas about depression and its treatment in their own words in semistructured interviews (a form of a qualitative method that uses a set of open-ended questions to gather ideas from the interviewee; Barg et al., 2006). What emerged was a concept of depression that included loneliness as a prominent component, with evidence drawn from both semistructured interviews and quantitative assessment. These findings have led to strategies that employ community members to deliver interventions to isolated older adults, to be more in line with conceptions of depression held by older adults.
Nastasi and colleagues (2007) used a multiphased mixed methods design for the Sri Lanka Mental Health Promotion Project, conducted in the Central Province of Sri Lanka. The group used formative research to identify and refine individual and cultural constructs and variables that explain or predict mental health, violent behavior, and academic achievement among Sri Lankan youths and teachers. Researchers conducted focus group interviews with students and teachers, individual interviews with school administrators, and observed participants in schools to identify culture-specific definitions of mental health constructs, such as “stress.” On the basis of the theories generated with a combination of interviews and school- based observations, the researchers developed scales—using factor analysis—that assessed culturally specific values of adolescents’ competency and prominent stressors. Those highly targeted psychological measures for the Sri Lankan context were administered to students across six schools and were quantitatively analyzed. This formative research phase led to the design of a mental health promotion program that used a RCT to test the effectiveness of an intervention that used culturally appropriate coping strategies and peer support activities. During the program implementation, researchers collected information on program acceptability, cultural relevance and social validity, integrity, and immediate impact of the intervention. The quantitative and qualitative results consistently found that girls, but not boys, became more aware of feelings of distress and limited helpfulness of social support for situations over which they had little control. Girls also had a heightened sense of responsibility in problem solving of complex family problems. Teachers found new roles in contributing to students’ social and emotional development, and students sought emotional support from them outside of the immediate intervention settings.