Center of Excellence in Cancer Communication Research

Framing Genetic Risk



The explosion of knowledge in the field of molecular genetics has raised the possibility of understanding the role of genetic factors in disease and of tailoring effective interventions and targeted prevention efforts. The Framing Genetic Risk project is concerned with communicating genetic risk in a way that empowers people rather than inviting feelings of fatalism. There is good evidence for a genetic basis for breast and prostate cancers, some types of colon cancer and for genetic variants associated with complex behavioral traits such as substance abuse. Accompanying these developments in science are two parallel developments in the public sphere: (1) very strong interest in genetics by the news media and (2) the possibility that the public will infer incorrectly that genetic susceptibility to disease or to lifestyles means they have lost control of their lives. The consequence of these trends could be an increased sense of fatalism rather than empowerment. How can information about genetic influences on disease and behavior be presented to empower recipients and minimize their sense of fatalism? To answer these questions, the project is describing the frames used by the news media to present information on genetics, genetic risk, and genetic testing related to cancer and to behaviors that enhance cancer risk. Secondly, the project is assessing the impact of news stories about genetic influence on perceptions of risk, personal efficacy, and intentions to engage in healthy behaviors especially in vulnerable populations. Finally, the project explores ways to present information that will enhance people’s perceived efficacy in the face of genetic information about risk.

Effects Studies: Genetic

Annenberg National Health Survey: Genetics Items

A. Leader, J.N. Cappella

Started in January of 2005, the Annenberg National Health Survey is a monthly survey of health and health communication issues. The core of the survey includes questions about exposure to health information, health behavior and knowledge, attitudes and health policy. Part of the core is dedicated to genetic information, in particular, knowledge about genetics, exposure to information about genetics in the media, perceived influence of genes on health and the frequency of genetic testing. Data on these questions over time will allow for a trend analysis of genetic knowledge, reports of exposure to information about genetics, the salience of genetic influence and a comparison to media coverage of genetics as well as interrupted time series in the event of a major news event pertinent to genetics.


back to top

Effects of Genetic News Reports on Perceived Disease Controllability, Responsibility, and Helping Behavior

S. Jeong, J.N. Cappella

This study tests the effects of genetic information in news media. Based on the theory of reasoned action and attribution model, the study posits weight-loss related attitudes, self-efficacy, intentions and perception about others with obesity as the outcomes. The study results suggest that genetic information in news messages influences the audience’s attitudes toward one’s own weight loss, one’s genetic inference related to obesity and perception about others’ control over obesity. Genetic inference was both (a) positively associated with intentions to try to lose weight and (b) negatively related to intentions mediated by efficacy.


back to top

Effects of Prescription Drug Ads with Genetic Cues on Priming Health-Related Beliefs, Efficacy and Behavioral Intention

M. Shim, J.N. Cappella

Some direct-to-consumer (DTC) advertising of prescription drugs has explicitly linked a genetic or hereditary component to the health conditions that are to be treated with the advertised drugs. Two contrasting hypotheses can be proposed concerning the effects of genetic cues on self-efficacy and behavioral intention: on one hand, genetic information about diseases in DTC advertising may generate a sense of fatalism and thereby reduce self-efficacy and intention to engage in healthy lifestyles; on the other hand, this information may be thought of as another risk factor one should deal with, facilitating efficacy and intention.

This study tests these alternative hypotheses, conducting a simple two-factor (genetic vs. non-genetic), between-subject experiment (N= 395). In the genetic condition, three prescription drug ads about heart disease, osteoporosis and high cholesterol employing genetic cues were shown; whereas the ads were slightly edited to exclude these genetic cues in the non-genetic condition. Results suggest 1) family history and ad believability moderated the effects of genetic cues on inference of genetic susceptibility, such that family history had a strong association with genetic inference in the non-genetic condition, whereas such association became less strong when participants disbelieved the ads in the genetic condition, 2) genetic cues had a main, positive effect on behavioral intention to engage in healthy lifestyles, and 3) genetic cues had some weak effects on the depression of self-efficacy in a limited context. Overall these results support the second hypothesis that genetic cues affect personal sense of risk and thus facilitate intention to be healthier.


back to top

Genetic Information in News: Deterministic and Probabilistic Frames about Genetic Susceptibilities to Smoking Addiction and Disease

J.N. Cappella, C. Lerman, Y. Kang

Print news stories about genetics convey information to the public that is relevant to their assumptions about their own genetic predispositions. The objective of this study is to assess the effects of two types of information in news stories about genetics: deterministic and probabilistic frames and efficacy information. The context is genetic susceptibility to smoking addiction and efficacy to assist during quitting attempts. Outcomes include inference of genetic susceptibility to addiction, personal efficacy to quit, control over smoking and intentions to quit. Subjects were 201 young adult smokers recruited in a national random-digit-dialing sample. They were surveyed on the telephone in a randomized experiment embedded in a survey about cigarette smoking practices. In the framing condition, respondents heard either deterministic or probabilistic frames for genetic influence. Crossed with these is an efficacy or no-efficacy condition. Framing and efficacy interacted in their impact on inference of genetic susceptibility such that the no efficacy condition is associated with stronger inferences in the deterministic frame. The inference of genetic susceptibility mediates personal efficacy and control over smoking, which in turn are associated with the intention to quit.


back to top

News about Genetics and Smoking: Priming, Family Smoking History, and News Story Believability on Inferences of Genetic Susceptibility to Tobacco Addiction

J.N. Cappella, C. Lerman, A. Romantan, L. Baruh

Print news stories about genetics convey information to the public. This study assesses the effects of priming a belief in genetic susceptibility to smoking addiction on smokers’ inferences about their own susceptibility to smoking addiction, their efficacy to quit smoking and their intention to get a genetic test for addiction susceptibility. Respondents were 450 young adult smokers surveyed on the telephone in a randomized experiment embedded in a questionnaire about cigarette smoking practices. In the priming condition, respondents heard a news story about genes for smoking addiction. In the unprimed condition, respondents heard a news story concerning the gender of the offspring of smokers. Priming with the genetics news story did not affect respondents’ inferences about personal genetic susceptibility to smoking addiction. However, those who found the news story believable and had a strong family history of smoking were more likely to infer a greater personal genetic susceptibility.
back to top

Scaling Deterministic and Probabilistic Verb Phrases as Action Frames in Propositions about Genetic Influence

J. Weiner, J.N. Cappella

The purpose of this study is to assess participants’ subjective perceptions of different verb phrases (Q-sort methodology). The findings of this project will help researchers understand how individuals categorize and discriminate between deterministic and probabilistic verb forms. In addition, this study will set the stage for a follow-up study on judgments about genetic framing sentences in various contexts.


back to top

The Effects of Genetic Framing Statements on Perceptions of Determinism-Probabilism: Factorial Manipulation of Frames and Contexts

J. Weiner, J.N. Cappella

This study is one in a series of studies that will ultimately test framing of news stories. This project will use the verb forms generated from the Q-sort and assess them in various contexts. Results from this study will be used to develop effective news stories that vary by frame (i.e., deterministic versus probabilistic).


back to top

The Emotional Bases for Quitting Smoking: Anticipated and Concurrent Emotions as Predictors of Intention and Behavior

J.N. Cappella, A. Romantan, F. Patterson, C. Lerman

Extending the Integrated Model of Behavior Change, this study tested the hypothesis that anticipated emotions about quitting smoking would correlate with intentions to quit smoking, and that intentions to quit smoking would, in turn, predict quitting behavior one year later. Baseline (n=450) and 1-year follow-up (n=201) data were collected from a population-based sample of young adults, aged 18-25, who had smoked >1 cigarette in the past month. Anticipated emotions about quitting the correlated with quitting intention (p<.001) over and above other attitudinal, experiential and social normative predictors, while subsequent quitting behavior was predicted by baseline quitting intentions (OR= 1.79; p< .01) and smoking history (OR= .19; <.001). These data suggest that emotional appeals may be a useful appeal in smoking cessation programs.


back to top

Using Exemplars in News Accounts about Genetic Influences on Smoking Addiction to Deliver Efficacy Information

A. Klein, A. Leader, J.N. Cappella

In this experiment we manipulate the presence and content of exemplars, which are personalized stories about the experience of individuals, to examine their ability to deliver efficacy information in news media. The goal of the experiment is to determine whether smokers perceive smoking cessation information to be more efficacious when delivered inside or outside of an exemplar. Results from this study will be useful in developing effective messages about smoking cessation and other health topics.


back to top

Content Analysis of Genetic Influence

CAS: A Method for Computer-Assisted Human Content Analysis

D. Schreiber, M. Giorno, L. Humphreys, T. Falcone & J. N. Cappella

In this project we have developed CAS (Content Analysis System), a web-based interface, that enables human coders to both read and code articles on-screen using a series of simple, pull-down menus. The goal of this application is to ensure the reliability and consistency of results by streamlining the decision-making process and using familiar web-based forms. We further expect that this project will serve as a template for future efforts in computer-assisted human content analysis. Please visit the CAS Website for more information at http://toolbox.asc.upenn.edu/cas/ or click here.


back to top

Computer-Assisted Content Analysis of Genetic News Frames

L. Humphreys, T. Falcone, M. Ghiselli, D. Mittermaier & J.N. Cappella

A content analysis of news stories (i.e., broadcast and print) from 1997 to 2007 will map the public information environment for how genetic information is framed. To prepare for the analysis, we ensure the reliability of coders by employing a consensus-coded standard set of 105 articles for training and reliability purposes. Future coders will be trained against the coding results obtained from the standard. Instead of trying to achieve intercoder reliability, coding reliability for each coder will be measured against this standard, representative set of consensus-coded articles. This will ensure consistent coder training throughout the study as well as consistent reliability measures.


back to top

Computerized Analysis of Exemplars Appearing in News Stories about Genetic Influence

A. Klein, L. Humphreys, J.N. Cappella

Preliminary content analyses of news about genetics indicate that many stories contain human interest accounts, what we call exemplars. We are coding these exemplars for occurrence, vividness and inclusion of efficacy information. This study complements the human content analysis of exemplars with computerized content using James Pennebaker’s linguistics analysis software (LIWC system). This project applies Pennebaker’s LIWC software to an archive of exemplars, determining the extent to which the exemplars meet criteria for texts with the potential for positive health consequences.


back to top

Computerized Content Analysis of Genetic Framing Statements using Automap

Chul-joo Lee, D. Schreiber, J.N. Cappella

This project explores recent developments in computerized content analysis, or semantic network analysis, to determine whether framing of genetic influence can be inferred. Carley and Diesner’s (2004) Automap software combines a variety of text processing algorithms with network analysis tools to extract the mental models in the text and identify information about the framing of genetics as a factor in disease outcomes. We follow the Gene-Framing Phrase-Outcome link and are trying to develop an Automap thesaurus that can efficiently classify words and phrases into corresponding categories of gene, framing verbs and outcome. Currently, we are testing the validity and reliability of our preliminary thesaurus against human coding on a sample of 105 gold-standard news articles.


back to top

Social Support and Linguistic Analysis Software

A. Klein, L. Humphreys

This study utilizes Pennebaker’s linguistic analysis software (LIWC) to examine breast cancer support messages. According to Burleson’s Social Support Hierarchy, person-centered messages explicitly acknowledge and elaborate the feelings of the distressed other, and they are therefore perceived as being more helpful, supportive and involved. They also reflect the highly emotional language that has been associated with positive effects of insightful disclosure (based on Pennebaker’s linguistic analyses).


back to top

Validating the Effectiveness of Comments in On-line Breast Cancer Support Groups: An Analogue Study

A. Klein, J.N. Cappella

This study looks to validate whether messages classified as person-centered by Burleson’s scheme are perceived as socially supportive by recipients. We obtained the messages, which were previously classified as more or less person-centered according to the hierarchy developed by Brant Burleson, from online breast cancer support groups. Since perceived social support has been linked to improved health outcomes for cancer patients, these online groups may be a valuable source of support for patients and their families. This study will examine whether messages of different levels of person-centeredness will produce differential levels of perceived social support.


back to top

Effects Studies: Anti-Smoking PSAs

A Measure of Argument Strength for Anti-drug PSAs Targeted to Adolescents: Reliability and Validity

X. Zhao, A. Strasser, J. N. Cappella, M. Fishbein & D. Barrett

The purpose of this research is to develop a measurement scale that can be used to assess the strength of the arguments presented by PSAs. This scale should both incorporate the strengths of the traditional thought-based measure and at the same time overcome its limitations.


back to top

Evaluation of Anti-Marijuana Ads: The Impact of Message Sensation Value and Argument Strength on Ad Effectiveness

Y. Kang, J.N. Cappella

This study analyzed 60 anti-drug ads, which were evaluated in a series of studies by adolescents. The data were aggregate-level data in the sense that adolescents’ evaluations of ad effectiveness, argument strength, etc. were averaged for each ad. Message sensation value (MSV: audio, visual, and content characteristics of video advertisements) was coded for each ad. High MSV ads, by definition, should attract attention. It is hence possible that ads with high MSV value can distract adolescents’ attention away from argument processing, resulting in a tempered ability to appreciate strong arguments. This study tested the hypothesis that MSV may moderate the impact of argument quality on ad effectiveness, such that strong arguments would be evaluated as more effective than weak arguments in low MSV ads, whereas no such difference may exist for high MSV ads. Significant interaction effects were found between MSV and argument quality on three different measures of ad effectiveness, suggesting that higher levels of MSV may be distracting.


back to top

Physiological Response to Anti-Drug PSAs: Message Sensation Value and Message Effectiveness

Y. Kang, J.N. Cappella, A. Strasser, M. Fishbein & C. Lerman

Message sensation value (MSV) refers to audio, visual, and content characteristics of video advertisements that might appeal to high sensation-seekers. High MSV ads should draw attention of all audience members and especially the attention of high sensation-seekers. They should also elicit higher levels of arousal. This study tested whether high MSV ads do have these hypothesized effects using both physiological measures (orienting response, skin conductance and facial electromyography) and self-reported measures (perceived ad effectiveness, perceptions of ad arousal value and valence). High MSV ads were shown to have a direct impact on physiological arousal measured with skin conductance response. High MSV ads also seem to interfere with adolescents’ ability to differentiate strong versus weak arguments of the ads, reflected through their corrugator responses while viewing the ads. Conflicting results are found between participant’s self reports and physiological reactions. Although participants rate ads in the high-effectiveness condition as less effective than ads in the low-effectiveness condition, they indeed show more positive and fewer negative facial reactions while viewing ads in the high-effectiveness condition. This effect is especially evident for low-MSV ads and among high-risk adolescents. Thus psycho-physiological measures may provide an alternative and, perhaps, more sensitive way to detect deeper-level ad effects.


back to top

Pilot Testing the Visual and Verbal Redundancy of Anti-Smoking PSAs with a Between Subject Design

A. Leader, J. N. Cappella, C. Lerman, K. White

The purpose of this study is to assess whether perceived overlap exists between visual and auditory messages in anti-smoking Public Service Announcements (PSAs). This study utilized a Visual/Verbal Redundancy scale that was created and validated during a previous pilot study. Additionally, this study seeks to examine the impact that certain individual difference and demographic factors (i.e., sensation-seeking, need for cognition, smoking history, vulnerability to smoking and visual/verbal orientation) have on adult smokers' perceptions regarding how transportive (i.e., involving) different anti-smoking PSAs are, as well as the kinds of emotions they experience from watching these ads. It is theorized that the more redundant the visual and verbal messages in the advertisements are, any attention to the visual aspect of the ads enhances, rather than undermines, the effectiveness of the ad.


back to top