Debra Sprague1,2, Donna L LaVallie1,3, Fredric M Wolf3, Clemma Jacobsen1,2, Kirsten Sayson2, Dedra Buchwald2. 1. Center for Clinical and Epidemiological Research (DS, DLL, CJ, DB) University of Washington, Seattle. 2. Department of Medicine (DS, CJ, KS, DB) University of Washington, Seattle. 3. Medical Education and Biomedical Informatics (DLL, FMW) University of Washington, Seattle.
Abstract
BACKGROUND: Presentation of risk information influences patients' ability to interpret health care options. Little is known about this relationship between risk presentation and interpretation among American Indians. METHODS: Three hundred American Indian employees on a western American Indian reservation were invited to complete an anonymous written survey. All surveys included a vignette presenting baseline risk information about a hypothetical cancer and possible benefits of 2 prevention plans. Risk interpretation was assessed by correct answers to 3 questions evaluating the risk reduction associated with the plans. Numeric information was the same in all surveys, but framing varied; half expressed prevention benefits in terms of relative risk reduction and half in terms of absolute risk reduction. All surveys used text to describe the benefits of the 2 plans, but half included a graphic image. Surveys were distributed randomly. Responses were analyzed using binary logistic regression with the robust variance estimator to account for clustering of outcomes within participant. RESULTS: Use of a graphic image was associated with higher odds of correctly answering 3 risk interpretation questions (odds ratio = 2.5, 95% confidence interval = 1.5-4.0, P < 0.001) compared to the text-only format. These findings were similar to those of previous studies carried out in the general population. Neither framing information as relative compared to absolute risk nor the interaction between graphic image and relative risk presentation was associated with risk interpretation. CONCLUSION: One type of graphic image was associated with increased understanding of risk in a small sample of American Indian adults. The authors recommend further investigation of the effectiveness of other types of graphic displays for conveying health risk information to this population.
BACKGROUND: Presentation of risk information influences patients' ability to interpret health care options. Little is known about this relationship between risk presentation and interpretation among American Indians. METHODS: Three hundred American Indian employees on a western American Indian reservation were invited to complete an anonymous written survey. All surveys included a vignette presenting baseline risk information about a hypothetical cancer and possible benefits of 2 prevention plans. Risk interpretation was assessed by correct answers to 3 questions evaluating the risk reduction associated with the plans. Numeric information was the same in all surveys, but framing varied; half expressed prevention benefits in terms of relative risk reduction and half in terms of absolute risk reduction. All surveys used text to describe the benefits of the 2 plans, but half included a graphic image. Surveys were distributed randomly. Responses were analyzed using binary logistic regression with the robust variance estimator to account for clustering of outcomes within participant. RESULTS: Use of a graphic image was associated with higher odds of correctly answering 3 risk interpretation questions (odds ratio = 2.5, 95% confidence interval = 1.5-4.0, P < 0.001) compared to the text-only format. These findings were similar to those of previous studies carried out in the general population. Neither framing information as relative compared to absolute risk nor the interaction between graphic image and relative risk presentation was associated with risk interpretation. CONCLUSION: One type of graphic image was associated with increased understanding of risk in a small sample of American Indian adults. The authors recommend further investigation of the effectiveness of other types of graphic displays for conveying health risk information to this population.
Authors: Laura S Larsson; Dorothy Champine; Dee Hoyt; Lillian Lin; Emily Salois; Sharon Silvas; Terri Weasel Tail; Matthew Williams Journal: Public Health Nurs Date: 2015-06-02 Impact factor: 1.462