Finding health information online continues to help patients understand new symptoms. However, incomplete information or advice that errs on the site of caution can cause distress or anxiety and prompt visits to a doctor. Cyberchondria, a new phenomenon, is defined as excessive compulsive searching for health information online that leads to distress and increase utilization of healthcare services. Grounded theory guided the conception of a mixed method study to investigate cyberchondria and symptom appraisal within Emergency Departments. A questionnaire and interviews were used to collect data from patients visiting a local Emergency Room. Results showed 63.3% of respondents looked up their symptoms online before their visit, and exhibited higher levels of cyberchondria and health anxiety than those who did not (p .001, p .004), and that health information consulted online can impact the decision to seek out immediate care. Strategies are needed to improve and promote quality online sources to benefit both seekers and services.

Background. Lung cancer is the leading cause of cancer deaths worldwide. Low survival rates have been attributed to delays to diagnosis, and some patients report having symptoms for several months before presenting to health services. Strategies are needed to encourage timely help-seeking. The Web is increasingly used as a health information source. Aim. The aim of this thesis is to explore whether the Web plays a role in help-seeking behaviour of people with lung cancer prior to diagnosis, and how the Web can be utilised to encourage earlier presentation to health services for symptomatic people. Systematic review. To begin, I carried out a systematic review of the literature (N=34), which highlighted a scarcity of research on Web use for symptom appraisal among cancer populations. Mixed-methods study. I conducted a survey with recently diagnosed (6 months or less) lung cancer patients (N=113). Based on survey responses, I purposively selected a sub-sample of patients and their family/friends ("proxies") for semi-structured interviews (N=33). In the survey, 20.4% of participants reported they or proxies had researched their condition online before the diagnosis. Interview results suggest perceived impacts of online information on symptom appraisal, forming the decision to seek help, and on interactions with healthcare professionals. Intervention development and evaluation. Based on my findings, I developed a Web-based intervention. The intervention provides tailored information about lung cancer, and uses components based on the Theory of Planned Behaviour ("TPB-components") to encourage earlier help-seeking. This intervention was tested in an online feasibility study (N=130), and subsequently in an online randomised controlled trial (N=212) with a factorial design to test main and interaction effects of tailoring and TPB-components. The feasibility study and first trial helped identify methodological issues which were addressed in a second trial with a mixed factorial design. This trial (N=253) indicated that the self-reported likelihood of visiting a doctor increased significantly by 11.8% from before to after viewing study information (p < 0.001), but no effects of tailoring or TPB-components were found. When examining only those aged ≥50 years, who are at highest risk of lung cancer, those receiving tailored information reported an increase of 13.2% in likelihood of seeking help, compared to 3.2% in the untailored group (p=0.01). Participants aged 50+ receiving the TPB-component reported a larger increase (13.8%) than those who did not receive the TPB-component (5.2%), but this did not meet the significance criterion (p=0.054). Conclusions. According to patients' perceptions, the Web can impact on processes in the appraisal, help-seeking and diagnostic interval leading up to diagnosis. Presentation of information about symptoms and risk factors online can produce significant effects on self-reported likelihood of seeking help. These findings indicate that there is potential for the Web to be utlised in the endeavour to educate the public about symptoms and to promote earlier presentation to health services, but due to small effects and differential dropout in this study, further research is required.