The Internet offers many benefits to people seeking health information, such as the convenience of accessing information at any time, and the protection of viewing information anonymously; however, such information is unregulated and can be misinterpreted (Raine et al., 2000; Starcevic & Berle, 2013). Escalation, the observed increase in medical severity of search terms within a single search session, could occur. For example, escalation occurs when an initial search for “headache” leads to a later search for “brain tumor”. Researchers have recommended including incidence rates to reduce escalation; however, this phenomenon has yet to be tested empirically. The purpose of the current research is to investigate the effects of adding probability rates to Internet health search results. One-hundred-and-fifty undergraduates were randomly assigned to one of three presentation groups (control, pictorial, and numeric) where they evaluated four search results pages. Incidence rates were not displayed in the control whereas participants in the pictorial condition saw incidence rates displayed as bar graphs and those in the numeric condition saw incidence rates displayed as percentages. Escalation was evaluated using the severity and susceptibility measures from the Risk Behavior Diagnosis Scale. Severity was defined as the magnitude of harm expected from a threat or the significance or seriousness of a threat (Witte, Meyer, & Martell, 2001). The Risk Behavior Diagnosis Scale evaluated severity in three questions: (Symptom) is a serious threat; (Symptom) is harmful; and (Symptom) is a severe threat. Susceptibility is the likelihood that a specific person will experience a threat, the degree of vulnerability, or risk of experience a threat. In the current study, three questions were used to assess susceptibility: If I have (symptom), I am at risk for having (serious condition); It is likely that I have (serious condition) if I experience (symptom); (Symptom) is nothing to worry about. Four symptoms were evaluated and each symptom was paired with four conditions: two benign and two serious. Results indicated that participants believed symptoms were more severe after reviewing search result pages than before reviewing search result pages (p<.001); however, there were no display group differences in perceived severity. Participants also believed that they were the most susceptible to benign conditions when incidence rates were shown numerically, followed by pictorially, and the least susceptible when there were no incidence rates present. The numeric group was significantly higher than the control group (p=.002); however, there were no differences between the numeric and pictorial group, and between the control and pictorial group (p<.10). Similarly in the serious condition, the highest ratings were in the numeric group, followed by pictorial, and the lowest perceived susceptibility was in the control. Numeric was significantly higher than the control (p=.003) and pictorial (p=.028), but there was no difference between the pictorial group and the control (p<.10). Although susceptibility was higher when incidence rates were present for both benign and serious conditions, rates were higher for benign conditions than the serious conditions (p<.001) suggesting that people are not escalating. These results also fall in line with the statistical results shown on the search result’s page. Previous escalation studies suggested this phenomenon (Aiken et al., 2012; Starcevic & Berle, 2013; White & Horvitz, 2009), but it has not been tested up until this point. It would be beneficial to replicate the study with a more diverse population to obtain more generalizable results; however, the findings from this study could be helpful in understanding how patients comprehend healthcare information and could conceivably provide direction for how health care professionals distribute information to their patients.
展开▼
