Professional Documents
Culture Documents
Jjnlcoms 184
Jjnlcoms 184
ORIGINAL ARTICLE
doi:10.1111/j.1460-2466.2006.00289.x
Brennan, 1999; Kreuter & Skinner, 2000; Kreuter, Strecher, & Glassman, 1999; Rimer
& Glassman, 1998). To date, researchers have usually inferred these needs based on
an individual’s responses to assessment items measuring behavioral determinants.
Ideally, the concept of ‘‘needs’’ might be expanded to also include specific message
those chapters that applied to their particular stage of readiness to quit. Clear
Horizons, developed by Rimer et al. (1994), used a similar approach and was
significantly more effective than generic smoking cessation materials for smokers
aged 50 and older.
information in the amount, type, and through channels of delivery preferred by the
individual, thus potentially reducing barriers to exposure of individuals to commu-
nication interventions. Such an approach then could increase attention, lead to
subsequent yielding, and, ultimately, enhance the likelihood of behavior change.
To date, most THC interventions have focused primarily on the first of these
approaches, although there are examples of other levels of tailoring. By providing
content of interest or concern to specific individuals, readers should be motivated
to pay closer attention to the information, process it more carefully, and be more
likely to use it to make decisions and take actions to improve health (Kreuter
et al., 1999). Most but not all THC studies show that THCs indeed lead to these
expected outcomes (e.g., Brug et al., 1999; Campbell et al., 1994; Kreuter & Strecher,
1996; Prochaska et al., 1993; Rimer et al., 2002; Skinner et al., 1994). Still needed are
studies that explicate the pathway that leads from exposure variables to behavior
change. Newer methods of analysis, like structural equation modeling, may aid in
identifying the communication and behavioral impacts of THCs. Moreover, labora-
tory studies could be used to advantage in this regard.
This explanation of tailoring effects—that behavior change occurs through
increasing motivation to process information—is consistent with Petty and
Cacioppo’s Elaboration Likelihood Model (ELM; Petty & Cacioppo, 1981). ELM
asserts that under certain conditions like elevated motivation and ability, people are
active information processors—considering messages carefully, relating them to
other information they have encountered, and comparing them to their own past
experiences (Cacioppo, Harkins, & Petty, 1981; Petty, 2006; Petty, Cacioppo,
Strathman, & Priester, 1994). One condition under which people are motivated
to process information actively is when they perceive the information to be per-
sonally relevant.
Fazio’s (Fazio & Towles-Schwen, 1999; Fazio, Powell, & Williams, 1989) MODE
model, an integrative dual-process framework of the attitude–behavior relationship,
expands upon ideas introduced in ELM (Petty, Haugtvedt, & Smith, 1995) and is
complementary to theories of behavior change, such as the Theory of Reasoned
= Intervention
= Behavioral state or outcome
Figure 2 Using tailored health communications along the behavioral pathway to colorectal
cancer screening.
Yet, one could use different approaches to message design for different types of
people in different behavioral readiness states.
Weinstein’s Precaution Adoption Process Model (Weinstein & Sandman, 2002)
proposes that there are distinct phases to taking precautions about one’s health.
Individuals must first be aware of a problem, engaged in thinking about it, deciding
whether or not to take action, taking action, and finally maintaining any change that
was made. Communication objectives would be different for individuals in each
phase, as are the theories or change mechanisms that would inform communication
development for individuals in each phase. If a person is unaware of CRCS, it may be
important to capture his or her attention, perhaps using a compelling narrative or
image to encourage the person to think about CRCS. If he or she is aware of CRCS
but not particularly engaged in the topic, communication strategies would aim to get
the person to think about it and apply it to his or her situation. The person might be
given data about his or her risk and then encouraged to think about how a diagnosis
of colorectal cancer could affect his or her live. This may be particularly effective for
people with high family risk. If he or she was thinking about CRCS but had not yet
decided whether to be screened, theories of persuasion and attitude should apply.
A loss-framed message could highlight the consequences of not being screened,
using individualized information about the person to make the case more salient and
compelling and to facilitate behavior change (Salovey, Schneider, & Bailey, 1999).
Tailored lists of an individual’s reasons for getting screened (pros) and reasons for
not getting screened (cons) might help them evaluate the issues and move toward
action. Such approaches are consistent with the Transtheoretical Model of Change
(Prochaska et al., 2002) in helping people change the balance of pros and cons and
also with ELM in encouraging more effortful processing (Petty et al., 1995). Tailored
telephone calls have been used to help people make informed decisions about
2004). Even overt claims of relevance (e.g., This information is designed just for you
based on the answers you provided) could be manipulated experimentally, an
approach that has been shown to increase perceived relevance among recipients
(Burnkrant & Unnava, 1989). Skinner et al. (2002) varied graphics and messages
on physical activity in a community sample of adults. Moreover, there has been little
attention to evaluating the quality of messages or of the overall communication
product.
The cost-effectiveness of tailoring (vs. not) and of different doses of tailoring is
on each item—the experience of the booklet seems to have been different for the two
intervention groups and was significantly more favorable for the TP 1 TC group.
These data are not definitive in explaining how the intervention exerted its effect.
However, process items can be useful in understanding how women responded to
Future directions
The first generation of research on THC focused appropriately on the basic question
of whether tailored materials were more effective than nontailored materials. It is
now time to progress to a next generation of tailoring studies to explore how and
under what conditions tailoring works and how its effects can be optimized. Inte-
gration of a persuasion and message effects perspective will help in both types of
inquiry. More attention should be paid to issues, such as the appropriate control
group for THC studies, and to creating standards and metrics so that THC studies
can be compared at the message level.
When developing any health communication, tailored or not, certain operational
tasks must be completed (Kreuter & Wray, 2003). At the simplest level, these tasks
could include choosing credible sources, developing a message strategy, defining the
appropriate sources of data, and determining the settings and/or channels for opti-
mal communication delivery. Although each of these decisions presents an oppor-
tunity to tailor health communications, it should not be assumed that THCs are
always the ideal or even necessary approach to intervention. To date, tailoring has
been used primarily as a message strategy, providing specific content to individuals
based upon information from or about them, typically responding to a theory-driven
assessment of behavioral precursors. There is considerable need and opportunity to
explore a much wider range of tailoring strategies, formats, effects, and mechanisms
for effectiveness. Collaborations between behavioral and communication scientists
may be important in refining understanding of message effects within tailoring
paradigms.
Specifically, this paper identifies multiple intermediary points on the behavioral
pathway and suggests that different tailoring variables and strategies and even dif-
ferent message formats might be better suited for different points along the way. This
assumption should be empirically tested, as should questions about the extent to
which and mechanisms through which different intermediary variables contribute to
the end goal of behavior change, and whether some message strategies and individual
data are more useful than others in influencing different points along the behavioral
pathway. Moreover, to date, most printed health THC strategies have used
researcher-defined data as the basis for developing THCs. Using a combination of
quantitative and qualitative data, we generally decide what interventions people
should receive. Computerized online interventions offer more opportunities for
collaborative, real-time design. Products could range from those that are expert
designed to those that are user designed, with combinations in between. Although
some interventions have offered people choices about the amount of information
they could receive on various topics (e.g., Skinner et al., 2002) and similar options,
we are aware of no research that has assessed the efficacy of user-designed versus
expert-designed THCs.
Acknowledgments
This research was supported in part by the National Cancer Institute’s (NCI) Center
of Excellence in Cancer Communication Research program (CA-095815-02) to
M.W.K. and National Institutes of Health (NIH) Grant 1 R01 CA105786-01 and
NCI Grant 5-P30-CA16086-28 to B.K.R. The authors thank Linda Kastleman at the
University of North Carolina and Chris Casey, Danielle Davis, and Keri Jupka at
Saint Louis University for assistance in the preparation of the manuscript.
Fazio, R. H., & Towles-Schwen, T. (1999). The MODE model of attitude-behavior processes.
In S. Chaiken & Y. Trope (Eds.), Dual-process models in social psychology (pp. 97–116).
New York: Guilford.
Glasgow, R. E., Goldstein, M. G., Ockene, J. K., & Pronk, N. P. (2004). Translating what we
Rakowski, W., Lipkus, I. M., Clark, M. A., Rimer, B. K., Ehrich, B., Lyna, P. R., et al. (2003).
Comparing a reminder letter, tailored stepped-care, and self-choice for repeat
mammography. American Journal of Preventive Medicine, 25, 308–314.
Rimer, B. K., & Glassman, B. (1998). Tailoring communications for primary care settings.
Weinstein, N. D., & Sandman, P. M. (2002). The Precaution adoption process model. In
K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education:
Theory, research and practice (pp. 121–143). San Francisco: Jossey-Bass.
Williams-Piehota, P., Pizarr, J., Schneider, T. R., Mowad, L., & Salovey, P. (2005). Matching