News Release

Paper: Don't rely on mixed messages to change health behaviors

Peer-Reviewed Publication

University of Illinois at Urbana-Champaign, News Bureau

Dolores Albarracin, University of Illinois at Urbana-Champaign

image: Self-improvement messages to lose weight, quit smoking or eat more fruits and vegetables can fall on deaf ears if the intervention message is mixed, says new research from U. of I. psychology professor Dolores Albarracin. view more 

Credit: Photo by L. Brian Stauffer

CHAMPAIGN, Ill. -- Self-improvement edicts to lose weight, quit smoking or eat more fruits and vegetables can fall on deaf ears if the intervention message is mixed, says new research co-written by a University of Illinois expert in social psychology.

When recommending multiple behavior changes, it's more effective if the messages are framed along similar polarities, according to a new paper from Dolores Albarracin, a professor of psychology at Illinois.

"There are all these programs to get people to execute multiple behavioral changes in the area of health, which is especially important because something like a healthy lifestyle is a conglomerate of behaviors," Albarracin said.

In other words, two actions or two inactions are more effective than a mix of positive and negative instructions.

"What we found is that it's better to combine them in a homogeneous way," said Albarracin, also a professor of business administration. "If you tell people to engage in two behaviors and the behaviors are a mix of actions and inactions, it's likely that recipients will adhere less. It's easier to frame the recommendations along the same axis of action or inaction. You don't want to work against yourself or at cross-purposes with your message."

Trying to get people to increase both their intake of fruits and their volume of exercise "goes together better in that both are increases in recommended action and may be remembered and executed more effectively," Albarracin said.

"Telling someone to increase exercise while decreasing their fat intake isn't nearly as effective, as shown in behavioral and clinical outcomes," she said. "It's better to say 'Increase exercise and increase the amount of vegetables you eat' than 'Increase exercise and decrease fat intake.'"

The paper, which will be published in the journal Health Psychology Review , is a meta-analysis of 150 research reports of interventions promoting multiple behavior domain change and measuring change at the most immediate follow-up.

The findings provide important insights on how to best combine recommendations when interventions target clusters of health behaviors, Albarracin said.

"The problem of how many behaviors we can combine is important, but equally critical is how to combine them," she said. "Previously, there wasn't much guidance on what the optimal combination is. There wasn't any sort of logic to it. This paper gives you a fairly robust answer because it looks at the literature as a whole."

The paper is relevant to health care organizations, health care marketing and any type of organization that is trying to change employee behavior.

"In terms of organizational behavior, if you're trying to train people to execute a few different behaviors, it would be better to frame it this way rather than mix and match," Albarracin said. "If you're trying to create employee health programs, for example, think of it as one package. You don't want to overload people with information, but you also don't want to have a message that's at cross-purposes with itself."

###

Albarracin's co-authors are Man Pui Sally Chan and Marta Durantini of the University of Illinois; Flor Sanchez of the Autonomous University of Madrid; and Kristina Wilson of the Duval County (Florida) Health Department.

The research was supported by the National Institutes of Health.

Editor's notes: To contact Dolores Albarracin, call 217-244-7019; email dalbarra@illinois.edu.

The paper "Combining action and inaction health recommendations" is available online.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.