University of Michigan Center for Health Communications Research
- Project 1
A multi-phased dynamic experimental process to screen and refine the effects of psychosocial and communication factors influencing smoking cessation and movement through stages of change
- Project 2
Development a tailored fruit and vegetable intervention for African American adults using three psychosocial constructs, which to date have been the subject of little empirical examination: cultural factors, ethnic identity, and motivational predisposition
- Project 3
A multi-phased experimental process to explore methods of communicating risk regarding tamoxifen prophylaxis to women at high risk for breast cancer
The purpose of the University of Michigan Center for Health Communications Research (UMCHCR) will develop an efficient, theory-driven model for generating tailored health behavior interventions that is generalizable across health behaviors and sociodemographic populations. The research conducted by the Center will advance the evidence base, methodologies, technologies, and conceptual frameworks relevant to developing and implementing tailored web- and print-based cancer prevention and control materials. The proposed UMCHCR will support three research projects. Project 1 will aim to facilitate smoking cessation; Project 2 will focus on promoting fruit and vegetable intake among African American adults; and Project 3 will develop a decision aid to help women decide whether to undergo tamoxifen prophylaxis for breast cancer prevention. All projects will employ a resolution IV fractional factorial design to determine the potential active ingredients of tailoring, including, but not limited to, communication factors such as message content, message framing, message source, and graphical presentation; individual factors such as culture and socio-demographics; and psychometric factors such as motivation and self-efficacy.
The UMCHCR will collaborate with the Cancer Research Network and three of its members: Group Health Cooperative of Seattle, Henry Ford Health System of Detroit, and Kaiser Permanente of Atlanta to recruit and enroll research subjects. The Center will also employ many internationally recognized consultants to serve as committee staff, advisory board members, and co-investigators. Further, the Center will support a number of developmental projects that will contribute to the methodological knowledge relevant to enhancing intervention delivery and data collection in the UMCHCR's projects. All UMCHCR research will be supported by a set of core resources including an Administration Core, a Biostatistics Core, a Theory and Measurement Core, a Tailoring Technology Core, and a Recruitment and Data Collection Core. Each core will interact with the projects and other cores, creating both a significant economy of scale and a conceptual and technological synergy unavailable through the sum of individual projects. The Center will also place great emphasis on career development opportunities by hosting pre-and post-doctoral fellowship, cancer communication seminars, and a two-week tailoring training institute.
Project 1 of the UMCHCR uses a multi-phased dynamic experimental process to screen and refine the effects of psychosocial and communication factors influencing smoking cessation and movement through stages of change.
Phase I (Screening): A fractional factorial design will be used to screen and identify the factors influencing cessation and stage movement from a broad spectrum of potential active components of a web-based cessation induction intervention. These factors include message content (tailored versus untailored outcome expectations and tailored versus untailored efficacy expectations), message framing (gain versus loss), dose schedule (one large message versus six smaller messages over six weeks), use of testimonials (present versus absent), and message source (health maintenance organization versus primary care provider). The six factors of our screening experiment will be examined across a set of individual characteristics, including stage of change (which will be block randomized in the design), baseline level of motivation and self-efficacy, nicotine dependence, need for cognition, health locus of control, attitude toward the subject's PCP, attitude toward the subject's HMO, and socio-demographic and health status characteristics. The primary outcomes will be smoking cessation and movement through stages of change six months following baseline assessment.
Phase II (Refining): Follow-up experiments will refine and specify how the active factors identified in Phase I effect cessation and stage movement and develop optimal treatment combinations for cessation induction. Specifically, investigators will explore the optimal dose of important factors, more accurately determine the interactive relationship between factors and how these factors interact with participant characteristics, and further decompose the active factors that influence stage movement. The final product of Phase II will be a refined set of principles for tailoring anti-smoking communications to the specific needs and interests of individual smokers. Participants in Phases I and II will be drawn from two health maintenance organizations associated with NCI's Cancer Research Network (CRN). The collaboration between the Center and the CRN will allow for testing leading-edge web-based technology in a real-world environment that has the infrastructure for both evaluating and disseminating population-based cancer prevention and control programs.
Project 2 of the UMCHCR will focus on developing a tailored fruit and vegetable (F & V) intervention for African American (AA) adults using three psychosocial constructs, which to date have been the subject of little empirical examination: cultural factors, ethnic identity, and motivational predisposition. This project will employ several features common to the other UMCHCR projects, including a fractional factorial design and recruitment through NCI’s Cancer Research Network. The 5-year project has two primary goals: In Phase I, investigators will identify key elements of cultural and motivational communications for increasing F & V intake among a sample of 1000 AA adults. In Phase II, investigators will develop an individually-tailored F & V print intervention for AAs, based on the findings of Phase I, and examine how best to maximize their delivery and efficacy. Specific aims of Phase I include:
- Developing conceptually distinct intervention messages for modifying F & V intake for AAs, derived from three domains of cultural tailoring: surface structure, deep structure, and culturally-based
- Developing conceptually distinct intervention messages for modifying F & V intake for AAs based on three main domains of motivation: extrinsic (introjected), intrinsic (autonomous); and values-based
- Using a fractional factorial design, identifying from the cultural and motivational factors noted in aims 1 and 2, the most powerful messages and/or combination of messages that impact F & V intake at six month follow-up
- Identifying person characteristics that modify the impact of these messages, most notably ethnic identity and motivational predisposition.
During Phase II of the study, investigators will refine their understanding of how to maximize the delivery and impact of individually-tailored cultural and motivational communications. Specifically, using a different sample of 1000 AAs, the primary aims of Phase II are to:
- Develop an individually-tailored, print F & V intervention using conceptually distinct cultural and motivational messages
- Explore the optimal "dose" of tailored cultural and motivational messages
- Explore the optimal "source" for the tailored messages, such as peer, professional, or religious leader
- Explore possible person characteristics that may interact with message, dose, and source factors, for example baseline motivation, religiosity, age, and health status. Participants will be AA adult members recruited from HMOs that are part of the NCI-CRN: Henry Ford Health System and Kaiser Permanente of Georgia.
Decision aids, consisting of multiple components including risk messages and probabilistic information, have been developed to help patients make informed decisions. However, little research has investigated which formats are best used for presenting risk messages and statistical information. Project 3 proposes a multi-phased experimental process to explore methods of communicating risk regarding tamoxifen prophylaxis to women at high risk for breast cancer. The project will also test which methods of risk communication work best for different people. This will allow investigators to have unprecedented data about the strengths and weaknesses of several risk communication methods, as well as valuable data about whether future research should focus on tailoring risk communication information. Phase I will screen for the effect of risk messages, graphical formats, and individual characteristics using a fractional factorial design in order to identify promising ways to communicate risk information. UMCHCR will test a variety of risk messages, such as presenting absolute and relative risk reduction, number needed to treat, and message framing. It will also test a series of graphical formats, such as pictographs, line graphs, and risk ladders, to represent the risk of breast cancer. Randomized subjects based on self-reported numeracy will be blocked, to see whether this individual characteristic interacts with risk messages and graphical formats. The effects of other individual characteristics will also be studied. Primary outcomes measures will focus on knowledge, subjective perception of tamoxifen prophylaxis, and satisfaction with the decision aid. Phase II will expand the exploration of the most promising risk messages and graphical formats found in Phase I. Investigators will design experiments that either refine
- our understanding of how risk messages and/or graphical formats influence comprehension and decision making or
- our ability to explore interactions between risk messages, graphical formats, and individual characteristics.
Experimental methods will be similar to Phase I methods. Subjects for both phases will be randomly drawn from a sampling frame of members of two health maintenance organizations associated with NCI's Cancer Research Network. This population provides a representative sample with broad ethnic, age, health status, and geographic representation.