This study demonstrates that many of the factors specified by the C-SHIP model are associated with subsequent online cancer information seeking behaviors among rural breast cancer patients. This supports the extension of the model’s applicability beyond demonstrated self-protective behavior for preventing cancer to also discerning the reasons why people afflicted with cancer may seek information to cope with their disease. While we approached these analyses with some uncertainties about the directional influences of some of these antecedent variables, our results clearly suggest that those who were in a worse condition regarding their perceived state of affairs were generally more likely to seek out more online cancer information. It is encouraging that those women who appeared to need information most were also among the most likely to seek it out.
One interesting highlight from these results is that the variables associated with the C-SHIP model appeared to have more frequent relationships with experiential as compared to didactic information seeking. This may be due to the fact that rural cancer patients are more likely to feel geographically isolated from attending face-to-face support groups and are less likely to have the opportunity to connect with other breast cancer patients facing similar issues (Curran & Church, 1999
). As such, there may be a strong need among rural cancer patients to access experiential information from other cancer patients on the Internet where it is easier to find than in their own communities.
While the C-SHIP model encompasses numerous antecedent psychological and cognitive variables related to information seeking behaviors among breast cancer patients, it should be noted that there are likely other important variables that influence the degree to which patients seek online cancer information including the cancer-related physical drivers that may contribute toward those psychosocial states. For example, fatigue resulting from cancer treatments has been found to cause significant distress among breast cancer patients (Boehmke, 2004
). While we controlled for some physical indicators such as cancer stage in our analyses, it is important to remember that some of these physical states may be precursors of information seeking behaviors in their own right, and future research should continue to explore these relationships.
It should also be noted that many caregivers and companions often play an active role in seeking out online health information for their friends and family afflicted with cancer (Basch, Thaler, Shi, Yakren & Schrag, 2004
). This current study says nothing of the antecedent characteristics of patients who rely on others to seek out information about their illness or the characteristics of caregivers who assume this role. Future research should explore these questions to better understand the needs of cancer patients and their caregivers to better inform education and support interventions to serve these populations.
Previous research indicates that many rural patients are medically underserved (e.g., Engelman et al., 2005
). They often do not have the same degree of access to comparable health services as breast cancer patients from more populated areas. In a recent, separate study examining some of the same rural breast cancer patients, we found that the use of high-quality online health information was associated with improvements in the doctor-patient relationship (Shaw, Han & Hawkins et al., 2007
). These women may seek out online cancer information to compensate for perceived inadequacies in the health education materials they receive from their healthcare providers and it may complement the information and advice they receive from their clinicians.
The current study also contributes to our understanding of why people turn to the Internet to cope with cancer, suggesting that patients may have very different reasons for accessing conceptually distinct types of services (e.g., seeking information vs. online social support) within an ICCS. Previous studies have demonstrated the efficacy of ICCSs in general (Gustafson et al., 2001
). However, there may be added benefits in examining system use at the service level as done in this study and determining the characteristics of patients who are more or less likely to use these conceptually distinct services. Such an approach is more likely to shed insights on the mechanisms underlying these effects. For instance, while the current study found that negative conditions related to patients’ perceived state of affairs were generally associated with higher levels of online cancer information seeking, another recently published study looking at a separate sample of breast cancer patients revealed different patterns indicating that patients with more positive appraisals of their state of affairs were more likely to participate in the online support groups (Shaw, Hawkins, & Arora et al., 2006
). While it is true that sharing information is part of what cancer patients do in online support groups (Shaw, McTavish, Hawkins, Gustafson & Pingree, 2000
), they also participate in support groups for a variety of other reasons such as openly and anonymously expressing their thoughts and feelings in ways that help them make sense of their cancer experience (Shaw, Hawkins, McTavish, Pingree & Gustafson, 2006
) and praying as a way to cope with their diagnosis (Shaw, Han & Kim et al., in press
). In other words, motivations for using each type of conceptually distinct service may be driven by different psychosocial factors and precursors for using these different services within an ICCS may vary significantly for a cancer patient – a phenomena that will be obscured if researchers solely examine use of ICCSs at the systems level as is common in many randomized clinical trials.
It is also worth reiterating that low income rural women were recruited to this study and provided a computer, Internet access and training to use the system, at no cost. Since a recent study provided evidence that the most common barrier to seeking online cancer information among rural cancer patients was a lack of Internet access (Helft, Eckles, Johnson-Calley & Daugherty, 2005
), and older, low income women in rural communities may have even less access to the Internet than many other segments of the population (Bell, Reddy, & Rainie, 2004
), this study provides promising evidence about how this population of breast cancer patients might utilize online cancer information when barriers to accessing the Internet are removed. From a public policy standpoint, our results suggest that underserved rural breast cancer patients will use online health information if provided the access. Moreover, results also indicate that the psychologically neediest rural patients are among the most likely to capitalize on this opportunity. One economical, sensible solution for bridging the digital divide to assist lower income breast cancer patients in rural communities may be to create a loan bank of computers that could be provided to patients for a specified period while coping with their treatments. The computers could then be re-circulated to new patients as needed.
Given that differing antecedent characteristics are associated with preferences for different forms and topics of information, it may also be possible that people benefit from exposure to different forms and topics of information as well. While previous research has examined the benefits of access and exposure to information services in more holistic terms (e.g., Gustafson et al., 2001
; Shaw, Han, & Baker et al., 2006
), future research should take a more granular approach examining who benefits from what types of information and topics to maximize the potential influence of ICCSs on patient outcomes.
While this study focused on an exclusively rural sample of breast cancer patients in Wisconsin, it should be noted that rural communities differ considerably across geographic regions of the country (Weinert & Hill, 2005
). Compared to national figures, this particular sample was mostly Caucasian, reflecting the ethnic profile of much of rural Wisconsin. Additionally, over 50% of these study participants had at least some college so this sample had higher education levels than might be found in some other rural areas. Future research should examine more diverse rural communities in different geographic locations, particularly regions that might have lower education and literacy levels as patients living in these areas might have the greatest need for information while facing a cancer diagnosis.
Another limitation of this study is that the variables selected to represent the C-SHIP model only looked at patients’ subjective perceptions of their state of affairs and healthcare services. As important as patients’ subjective experience may be related to influencing online cancer information seeking behaviors, it is important to note that barriers to information may be very real – and not just perceived. For example, short office visits (Sepucha, Belkora, Tripathy & Esserman, 2000
), as well as the fact that some healthcare professionals and institutions may not be supportive of questioning or information-seeking patients, could influence patients’ propensity to seek cancer information on the Internet (e.g., Helft, Hlubocky & Daugherty, 2003
). Future research should also examine these more objective, extrinsic measures related to the larger healthcare environment to complement the insights about more intrinsic variables generated in this current study.
Finally, this study looked at the entirety of online cancer information seeking behavior aggregated over four months, but previous research indicates that the timing of the desire for cancer information varies over time (Leydon et al., 2006). Future research should examine online information seeking behaviors on a longitudinal basis. The C-SHIP model successfully applied as a conceptual framework in the present study provides a valuable framework for such longitudinal investigations. The model’s emphasis on the interactions of personal and situational characteristics (Miller et al., 1996
) can help generate hypotheses regarding the use of cancer-relevant information over the course of time. Insights generated from such studies may inform more sophisticated tailoring frameworks that can provide patients with the most useful information when they need it most.