Breast cancer on the Internet: the quality of Swedish breast cancer websites

Abstract

The aim of this study was to investigate the quality of Swedish-language breast cancer information available on the Internet. The questions explored were the extent and type of breast cancer information available, the coverage and correctness of that information, and whether the websites fulfilled the European Commission quality criteria for health-related websites. Three search engines were used to find websites containing medical information on breast cancer. An oncologist then evaluated the 29 relevant sites. Only seven of these were judged suitable for breast cancer patients. The coverage and correctness of the medical information varied considerably. None of the websites fulfilled the European Commission quality criteria. Therefore, considerable effort will be required before the Internet can serve as a valuable and up-to-date source of information on breast cancer for both professionals and laypersons. Our findings broadly match the results of earlier studies of English-language websites.

Introduction

The potential benefits of using the Internet as a source of information on breast cancer are obvious. For health professionals, it can be a valuable clinical tool allowing the exchange of information with colleagues and patients. The Internet can also be a valuable and constantly growing and updated source of the latest information both for professionals and for patients and other laypersons.¹

A diagnosis of breast cancer often leads women to seek information as a means of coping with their new situation. It has been shown that patients with cancer experience significant benefits when they receive information; for example, they conform more closely to the therapeutic regimens prescribed for them and enjoy better self-esteem.² Women with breast cancer also want a substantial degree of involvement in decisions about their medical treatment. The majority want to select their own medical treatment or at least be involved in its selection in collaboration with their physicians. However, most are not able to adopt the role they would prefer in decisions on their initial cancer treatment.³ Information on treatment options has a high priority for this patient group, albeit not as high as information on the chances of cure and the risk of disease spread.³ Furthermore, less than an hour of weekly Internet use to find information on breast health issues has been shown to be associated with greater social support and less loneliness among breast cancer patients.⁴

Health-related websites are now among the most frequently accessed on the Internet; current estimates indicate that there are now over 100 000 sites offering health-related information, and their number is growing rapidly.⁵ A study in Sweden on using the Internet for information on health issues found that 14% of those contacted had sought such information in 2002, and that 41% of these had looked for medical advice, 24% for health advice, and 20% for information on drug therapies.⁶ Two earlier Canadian studies found that as many as 40% of breast cancer patients used the Internet for medical information.7., 8. A person looking for breast cancer information on the Internet faces at least three problems: finding relevant websites, judging whether the information obtained is complete and accurate, and finally, finding whether the website is providing information from a trustworthy provider.

A common way of searching for information on the Internet is to use a search engine. However, search engines are only moderately efficient in locating information on a particular health topic. Although many links can be found for a specific search word, only a minority of such links lead to websites with relevant content.⁹ The number of links retrieved from different search engines also varies considerably.¹⁰

A second problem facing a person searching the Internet for breast cancer information is that of judging whether the information found is complete and accurate. For an oncologist this is a time-consuming, but relatively easy, task. For a layperson, the task can be much too difficult. In a study in which the researchers evaluated English-language information about Ewing sarcoma, because of the more manageable amount of information retrieved than in the case of cancer of the breast or prostate, four cases (6%) of clearly erroneous information were found. Websites reported survival rates as low as 5% and as high as 85%, when according to experts the actual survival rate is 70–75%.¹⁰ Clearly, this is information that can cause widespread devastation and confusion among patients and other laypersons. The situation with English-language information on breast cancer websites seems to be similar. One study found 12 sites (7%) that contained inaccurate medical statements.¹¹ A second study investigated whether English- and Spanish-language websites covered key consumer questions on four different conditions, one of which was breast cancer, and whether the information was accurate. They found that only half the topics that the expert panels thought were important for consumers had been given more than minimal coverage. This was particularly striking in the case of the Spanish-language sites. Even so, the situation was much better for breast cancer sites than for sites covering the other three conditions (childhood asthma, depression, and obesity). Sixty-three per cent of the elements on English-language sites were more than minimally covered and the information was completely accurate. The corresponding score for Spanish-language sites was only 39%. On both English and Spanish sites, the mean percentage of clinical elements covered that were rated as mostly incorrect was 0%.⁹

As mentioned earlier, a layperson is not qualified to judge whether the information obtained is sufficiently extensive and correct. One way of solving this problem is to evaluate whether the provider of the information is a trustworthy source. To make this possible, the website must display information on (for example) authorship of the medical content, sources, date of last update, ownership of the website, sponsorship, and advertising policies. Such quality criteria allow both consumers and professionals to make reasonable judgements and take informed decisions on whether or not what they are reading is credible, reasonable, and useful. Some examples of suggested or applied quality criteria are those of Silberg et al.¹² and the principles governing American Medical Association websites.¹³ The European Commission has also established a core set of quality criteria for health-related websites, which can be used, for example, as a basis in the development of user guides, voluntary codes of conducts, trustmarks, and accreditation systems.¹⁴ The quality criteria¹⁴ fall into six separate groups: transparency and honesty (display of site provider's name, purpose and objective of site, target audience, and sources of funding for the site), authority (display of sources for all information including name, credentials, and dates), privacy and data protection (display of privacy and data protection policy and system), updating of information (clear and regular updating of the site with date of last update), accountability (display of name of person responsible for the site and of responsible partnering and editorial policy), and accessibility (findability, searchability, readability, usability, etc.) (http://europa.eu.int/information_society/eeurope/ehealth/quality/index_en.htm). There are several other examples of quality criteria and instruments based on these criteria, which have been applied to produce awards, quality ratings or seals of approval. However, it is not known whether these—often incompletely developed—instruments measure what they claim to measure and whether they really do more good than harm.¹⁵

A few studies have evaluated quality criteria for English-language breast cancer websites. In one study¹⁶ it was found, for example, that the credentials of the site owners were identifiable only for one-third of the sites and that less than one-third of the sites clearly identified sources for information cited on the website and its updating. In another study,¹¹ the result was slightly better: more than half the websites considered displayed sources of information, and slightly under half of them provided information on updating. In an earlier study on Swedish health-related websites, none of the 35 websites evaluated was found to fulfil the European Commission's quality criteria.¹⁷

As the earlier research mentioned above shows, there is room for improvement in the quality of English-language breast cancer websites. Therefore, several authors recommend that in future more people working in the medical community should evaluate Internet information.1., 18. Physicians would then be in a position to give patients recommendations on the relevant websites.¹⁹ This is especially important, as the websites that are most popular among users are not necessarily of the highest quality.¹¹

To our knowledge, this is the first study to examine Swedish-language information on breast cancer on the Internet. The first question explored was that of the extent and type of breast cancer information that is available to anyone searching the Internet for such information in the Swedish language. The second question explored was the degree of coverage and the correctness of the available information. The third question explored was whether the available websites fulfilled the European Commission quality criteria for health-related websites.