Abstract B17: Measures to assess narrative influences on cancer prevention behaviors

Abstract

Abstract B17 Background Narrative approaches to health promotion in cancer prevention have recently gained much attention. Cultural grounding to produce narrative-based messages is a well-suited method of health promotion to address health disparities, but cognitive and emotional mechanisms that purportedly work through narratives to change behavior are rarely specified nor directly measured. A model for assessing narrative characteristics and effects on intention to change behaviors (derived from narrative theory and entertainment-education) was proposed and tested in pilot studies of a community-based, storytelling intervention for Latinos. Method Seven sessions with health information embedded in narratives taught by experienced Promotoras de Salud (who assisted in developing story content) were pilot-tested prior to full study implementation. Novella-style stories featured a group of neighbors with various interpersonal challenges, and incorporating colorectal cancer prevention (dietary, physical activity, and screening) messages. Participants evaluated each story presented and answered questions pre- and post-intervention regarding intention to modify health behaviors. Scale items to assess factors theoretically predicted to change behavior were created; data reduction and regression modeling were used to examine model elements. Results Fifty-five Latinos completed the pilot study and assessed stories in each session; mean age 39.6, mean education level, 9th grade; 92% prefer Spanish; median annual income $15-25K. Factor analysis on story characteristics produced two scales, Cultural (α = .755) and Emotional Content (α =.817). Factor analysis on mediators expected to predict behavior yielded Identification (α = .925), Engagement (α = .976) and Attention (α = .966). Model testing indicated significant relationships between some of the story characteristic measures and mediators: Cultural and Emotional Content of stories was related to Identification with stories and characters, and Cultural Content was related to Attention (p < .01 for all). Mediators showed moderate relationships to health behavior change intentions, with significant findings for Identification predicting intentions to increase physical activity (p < .05) and Engagement predicting intentions for dietary change (p < .01). Conclusions Development and preliminary tests of measures to assess narrative characteristics and factors expected to mediate behavior change show promise for theory building. Story characteristics measured as cultural and emotional content provided predictive value for mediators shown in previous research to be important for persuasion and behavior change, Identification and Engagement (otherwise known as “transportation” into the story). Further, Identification and Engagement preliminarily appear to provide predictive power for cancer prevention behavioral intention. Implications The model and corresponding measures tested in this pilot study provide a starting point for evaluating narrative interventions prior to intervention. As our larger community study progresses and additional data are collected on attitudinal and behavioral responses to narratives, a more detailed model will be specified and tested (including social proliferation effects) for theory-building to understand more about narrative effects on behavior change. Funding: NCI Citation Information: Cancer Prev Res 2008;1(7 Suppl):B17.