Date Published: June 11, 2018
Publisher: Public Library of Science
Author(s): Katherine Ross, Justin Stoler, Nick Carcioppolo, Sabine Rohrmann.
Low numeracy may skew patient perceptions of information about cancer. This paper examines the relationship between self-reported measures of perceived numeracy and cancer knowledge, beliefs, and affect, using results from 3,052 respondents to the 2007 Health Information National Trends Survey (HINTS-3). Chi-squared tests were used to identify differences in responses between high- and low-numeracy groups using three measures of perceived numeracy. Multivariable logistic regression models were used to evaluate the association between the three perceived numeracy measures and cancer information overload, cancer fatalism, cancer prevention knowledge, and cancer worry. Respondents with low perceived numeracy as expressed by discomfort with medical statistics were more likely to report information overload, to display fatalistic attitudes towards cancer, to lack knowledge about cancer prevention, and to indicate that they worried about cancer more frequently. After controlling for sociodemographic characteristics, this measure of perceived numeracy remained significantly associated with information overload, fatalism, lower prevention knowledge, and worry. The other measures of perceived numeracy, which measured understanding and use of health statistics, were not associated with cancer perceptions. Our findings suggest that individuals with low perceived numeracy broadly differ from individuals with high perceived numeracy in their perceptions of cancer and cancer prevention. By improving our understanding of how perceived numeracy affects patient perceptions of cancer, health providers can improve educational strategies and targeted health messaging.
Health care providers and patients increasingly support shared decision making as the preferred alternative to paternalistic or autonomic approaches to medical decision-making . Patients need the ability to comprehend and utilize health information to meaningfully participate in their own care. This is referred to as “health literacy,” an essential facet of functional adult literacy which shapes personal health outcomes [2–4]. Numeracy is defined as “the ability to access, use, interpret, and communicate mathematical information and ideas, to engage in and manage mathematical demands of a range of situations in adult life” . Health numeracy includes individual skills necessary to understand and use quantitative health information, including reading and interpreting medical information, performing basic computations, and communicating orally about quantitative information .
The demographic characteristics of the 3,052 respondents analyzed from HINTS-3 are presented in Table 1. The weighted study sample was predominantly non-Hispanic white (68.8%) with a slight female majority (50.5%), with 76.7% completing at least a high school diploma, and 25.7% attaining a Bachelor’s degree or higher. Just 31.9% of participants had an annual household income of greater than $75,000, with 53.2% over $50,000. The mean age of the sample was 55.3 years old (standard deviation [SD] = 16.4 years), with a range of 18–97. The three perceived numeracy measures also were associated with many of the demographic characteristics. Respondents with low Nunderstand were more likely to be of ethnic/racial minority status (X2 = 18.3, P < .001), report lower household income (X2 = 18.3, P < .001), have lower educational attainment (X2 = 58.3, P < .001), and be more likely to prefer Spanish (X2 = 9.9, P = .002). Respondents with low Ncomfort were more likely to be female (X2 = 10.0, P = .002), be of ethnic/racial minority status (X2 = 11.1, P = .010), report lower household income (X2 = 49.4, P < .001), have lower educational attainment (X2 = 49.4, P < .001), and be more likely to prefer Spanish (X2 = 9.3, P = .002). Respondents with low Nuse were more likely to report lower household income (X2 = 10.4, P = .030), and have lower educational attainment (X2 = 10.2, P = .020). These results underscore the relationships between socio-demographic characteristics and perceived numeracy and the importance of controlling for these characteristics in a multi-variable modeling approach. This study examined the association between low perceived numeracy and responses to questions concerning cancer-related knowledge, beliefs, and affect in the 2007 HINTS-3 survey. Chi-square tests showed statistically significant differences between high- and low-numeracy group responses for Nunderstand and Ncomfort concerning information overload, fatalism, prevention knowledge, and worry about cancer. After controlling for sociodemographic factors, Ncomfort continued to have a significant positive association with information overload, fatalism, and worry, and a significant negative association with prevention knowledge. This is notable in light of the strong significant association between lower levels of education and these three constructs. This study explored the relationship between perceived numeracy and cancer-related perceptions measured by HINTS. Our findings suggest that individuals with low perceived numeracy differ from individuals with high perceived numeracy in perceptions of cancer and cancer prevention, which may increase avoidance of cancer prevention and screening behaviors. As the volume of quantitative information and health data available to consumers continues to rise, particularly through new channels such as mobile and wearable technologies, it becomes essential to understand how consumers interpret and utilize this information. This study has implications for public health professionals, health communication specialists, and health care providers interested in the production and effectiveness of targeted health messages given the natural population variation in numeracy. Future research should extend this work to explore the potential associations between low numeracy, health behaviors, and clinical outcomes. Source: http://doi.org/10.1371/journal.pone.0198992