An Initial Look at the Associations of a Variety of Health-Related Online Activities With Cancer Fatalism

ABSTRACT

It has been an important public health goal to remove cancer fatalism because of its negative influence on both cancer screening and preventive behaviors. The present study examines roles of the Internet, as an emerging, crucial source of cancer information, in the effect of education on cancer fatalism. Based on our secondary analysis of a nationally representative survey (i.e., Health Information National Trends Survey 4 Cycle 1), we found that people with low levels of education are less likely than their more educated counterparts to engage in online health information seeking that is negatively linked to fatalistic beliefs about cancer prevention. In addition, the effect of education on online health information seeking was detected only among people who trust online health information. The implications of these findings for cancer control and for research on the digital divide and communication inequalities are discussed.

Funding

This study was supported by the Institute of Communication Research at Seoul National University, by the CPRC program of MSIP/IITP (IITP-2015-H8201-15-1004), and by Grant #2015S1A58017994 from the National Research Foundation of Korea.

Notes

¹ Although a lack of self-efficacy is closely related to fatalism, these two concepts are not identical. Straughan and Seow (1998), for instance, showed that people with high levels of fatalism tend to exhibit low levels of self-efficacy. Likewise, Ramirez et al. (2013) suggested that cancer fatalism decreases self-efficacy or perceived behavioral control. Thus, we believe that cancer fatalism is a predictor of a lack of self-efficacy.

² People who do not trust online health information are less likely than those who do to seek health information on the Web. However, the associations between trust and media exposure are only modest (Tsfati & Cappella, 2003). In the current data set, the correlation between trust and online health information seeking is .21. In other words, many people are surfing the Web to seek health information to some extent, despite the fact that they do not trust online health information.

³ We could not conduct a factor analysis because two indicators for each latent variable is a statistical minimum (Kline, 2005). Muthén and Muthén (2009, p. 102), for example, contended that “Number of factors can be influenced by the number of variables per factor” and “similar number of variables per factor, at least four or five variables are recommended.” Health product purchase was measured with a single item. Moreover, social media use consists of three items.

⁴ It was not possible to differentiate effects among these sources. There was a strong association of attention across these sources. The average Pearson correlation was .51 and ranged as high as .83. Their multicollinearity indicates that their effects cannot be separately estimated, which reflects the tendency for those who paid attention to one source to pay attention to the others.

⁵ We tested alternative mediation model. That is, we examined whether cancer fatalism is a mediator between education and online health information seeking. This was not supported. To be more specific, education was negatively associated with cancer fatalism (b = −.045, SE = .007, p < .001). Cancer fatalism, however, was not negatively associated with online health information seeking (b = −.058, SE = .032, p = .082). The estimate for the indirect pathway was 0.003 (SE = .002). The asymmetric 95% CI revealed that the indirect pathway from education to online health information seeking via cancer fatalism is not statistically significant (95% CI = [0.00, .006]).

⁶ We generated the slopes to determine the associations of education with online health information seeking according to the levels of trust. For those who trust online health information, education was positively related to online health information seeking (ß = .29, p < .001). By contrast, among those who do not trust, education was not significantly related to online health information seeking (ß = −.10, p = .273).

Additional information

Funding

This study was supported by the Institute of Communication Research at Seoul National University, by the CPRC program of MSIP/IITP (IITP-2015-H8201-15-1004), and by Grant #2015S1A58017994 from the National Research Foundation of Korea.