WHICH COMMUNICATION STRATEGIES IMPACT THE EFFECTIVENESS OF MESSAGES ABOUT MEDICAL EVIDENCE?

Purpose:   To examine the comparative effectiveness of different strategies for communicating medical evidence to increase its use by consumers and health care providers.  We focused on four communication strategies:  (1) Message targeting; (2) Message tailoring; (3) Message framing; and (4) Use of narratives (stories). 

Methods:   In accordance with strict AHRQ Evidence-Based Practice Center standards, we conducted a systematic review of the literature.  Using a combination of medical subject headings and keywords, we searched MEDLINE® (from January 1, 2000 to March 15, 2013), the Cochrane Library, Cochrane Central Trials Registry, and PsychInfo and found 106 articles that met our initial exclusion and inclusion criteria.  Two reviewers independently reviewed titles, abstracts, and full-text articles.  Articles that passed this phase were abstracted into detailed evidence tables. Two reviewers then independently graded study quality using pre-specified criteria.  Articles that failed quality grading (i.e., poorly conducted studies) were excluded from the review. 

Results:  Seven studies met the inclusion criteria.  Four trials directly compared the effectiveness of targeting versus tailoring messages related to prevention behaviors.  The studies produced mixed results; two found no differential effect and one found targeting to be more effective than tailoring.  The fourth trial expected the combination of tailoring and targeting would be more effective than targeting alone in terms of the percent of women who were likely to get screened for breast cancer, but this was not the case.  In two additional trials, loss framed non-targeted messages were more persuasive than gain framed non-targeted messages or targeted messages.  One small trial examined the combination of framing and narratives and found that loss framed narrative messages were more persuasive in influencing women to get a mammogram than gain framed narrative messages.  However, the strength of the evidence of this small group of studies was limited. 

Conclusions:  Comparative studies fail to provide evidence that tailoring messages are more effective than targeting messages as many have hypothesized. The effect of framing information appears to be moderated by the format in which the evidence is presented. Loss-framed messages coupled with another strategy may show some promise for influencing outcomes. Larger, well-designed studies are needed to replicate these results and examine both head to head comparisons of single strategies as well as combined strategies.