Healthy Skepticism Library item: 20229

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: Magazine

Quality standards for health information on the Internet
SIM Quarterly 1997 Dec
http://www.eee.bham.ac.uk/sim/simq/issue3/editorial.html

Full text:

‘Quality’ is a word that means different things to different people. As a result it invites a variety of definitions, but it remains in essence a relative measure. Relative, that is, to a standard. The ‘quality’ of health information on the Internet has attracted a good deal of attention in recent months, and emerged as a strong theme at MEDNET 97. Several individuals and groups have taken up the challenge of defining ‘quality’ through the promotion of desirable standards. This is no easy task; most of us are highly capable of identifying something that is not quality – yet less readily are we able to discern resources of merit.

The ease with which anyone can become an information provider on the Internet [1] can be a weapon on the side of Light, but also on the side of Darkness. Self-publishing on the WWW, according to La Porte and Hibbitts, would allow medical scientists to report their findings without delay, do so for little cost, expand their audience, receive prompt feedback, and they would not need to relinquish control over copyright [2]. However, it also has the potential to populate the Web with misinformed opinion and unsubstantiated claims. Recent discussion has tended to focus on this ‘darker side’ of medical information on the Internet.

So how rife is the problem of misinformation? Little evidence of actual or possible harm has thus far been documented. Impicciatore and colleagues recently assessed the reliability of Web-based advice for the home management of feverish children, by comparison with published guidelines [3]. They found that only 4 out of 41 surveyed Web pages adhered closely to these guidelines. Rightly, they do point out that the Internet does not have a monopoly on misinformation. A wide variety of drugs are now on sale via the Web [4], and one fatality resulting from such a sale has already been reported [5]. Indeed, the World Health Organization set up a working group in May of this year to consider the problem of unsafe medicine sales via the Internet [6] (see also BHIA initiative on pharmaceutical advertizing, SIMQ Issue 1, June 1997).

In this context, it soon becomes clear why a growing number of voices are calling for ‘standards’, ‘regulation’, and ‘codes of conduct’. Consumers of health-care information need some form of protection from inadvertent misinformation no less than they do from intentional deception (such as the use of HTML frames to present content from other sites as their own). Information on the Web deserves special attention because the hallmarks of quality with which we are familiar from traditional publishing are often not so clear – brand name, quality of presentation, trust, clear credentials, and other markers that tend to give information ‘weight’.

What standards have been proposed, then, what themes do they share? Since quality issues were first broached in SIMQ Issue 1 (see Quality issues receive attention), there have been several new proposals and the subject has matured into one worthy of more than just a ‘brief’ mention. It is worth summarising the important work in this area to date.

Silberg et al point out that quality assessment should not focus solely on the accuracy of medical content [7]; factors such as effective use of technology (e.g. hyperlinking), attention to site navigation, the provision for feedback, and interactivity are also important measures. Based on a foundation of accountability, they go on to suggest rejection of information not adhering to four core standards: authorship (including mention of affiliations and credentials); attribution (appropriate referencing with attention to copyright); disclosure (listing site ownership/sponsorship and any potential conflicts of interest); and currency (content should be dated).

Wyatt considers the minimum criteria of Silberg et al only part of the answer, and advises consideration of other aspects of a site when evaluating its reliability [8]. These additional aspects comprise readability of material, the quality of links to other sites, types of media used to communicate material, accessibility of the site via search engines, user profile, ease of site navigation, and its impact on users, clinical practice, and patient outcomes. Noting that there have been few published trials concerning the effect of the WWW on clinical practice, Wyatt advocates the use of laboratory tests and field trials as part of this evaluation process. We need to work out ‘how to recruit suitable subjects, develop valid and reliable methods of measurement, and carry out many more rigorous evaluations’, he says. His paper stops short of specifying how this might be done.

The International Committee of Medical Journal Editors (known as the Vancouver Group) tackled the issue of Internet medical information standards at a meeting in May 1997 [9], agreeing on a minimum set of standards. The Group highlighted the importance of indicating authorship, affiliation, credentials, conflicts of interest, references and sources, copyright, and disclosing site ownership, sponsorship, advertising, and commercial funding. In particular, they insist that ‘any commercial content should be clearly identified as such’, and that ‘advertising banners for drugs should not appear on the same screen as editorial content’.

In follow-up to it’s ‘Quality Standards for Medical Publishing on the Web’ published in December 1996, the British Healthcare Internet Association has proposed a ‘Bill of Rights for the Internet’, dated November 7 1997. The Bill greatly simplifies the earlier recommendation, stating simply that ‘medical Webmasters shall build no site that does not include who they are; their credentials; honest and non-misleading information; accurate and up to date information; their authority and references; who is behind the site; whether I have to pay for anything; my rights and guarantees if I buy anything; domain of relevance and availability in my country; and how to get in touch.’

One of the more substantial efforts addressing quality criteria to date is that of Dr. Helga Rippen, Director of the Health Information Technology Institute at Mitretek Systems, and colleagues of the Health Summit Working Group. Their draft White Paper ‘Criteria for assessing the quality of health information on the Internet’ (Oct 14 1997) outlines the Group’s objective of developing an accurate and reliable tool that will allow health-care consumers to assess Internet health information quality. The White Paper was presented to the American Medical Informatics Association Internet Working Group at its 1997 Annual Spring Congress (May 28 – May 31, San Jose, California). In summary, the paper identified the following broad categories as needing assessment: credibility; content; disclosure; links; design; interactivity; and caveats (e.g. beware sound-alike names). Each of these groupings is further defined in the paper using various sub-categories. A small survey (n=28, 30/9/97), not without its own methodological problems, determined that respondents considered the ‘top 10’ specific criteria to be the information source itself; disclosure of associations; accuracy; currency; citation of original sources; a hierarchy of evidence (e.g. randomised-controlled trials carry more weight than a case report); relevance/utility (i.e. usefulness); use of disclaimers; content enhancement via hyperlinks to quality sites; and evidence of editorial review. Translating this work into an easy-to-use tool for consumers will be no easy task. Importantly, the paper emphases the need for ongoing testing and refinement of the tool once developed.

The HON Code of Conduct, from the Geneva-based Health On the Net Foundation, was one of the earliest initiatives. The current version (1.6, April 1997) lists eight Principles covering qualifications, the complimentary role that online information has to the traditional doctor-patient relationship, confidentiality, dating and hyperlinking of citations, an evidence-based approach to claims for products and services, contact information, disclosure of support or funding, and responsibilities in relation to advertising.

The Medical Matrix Code of Conduct, published in SIMQ Issue 2 (September 1997), was devised by the Medical Matrix Editorial Board as an interim guide until such time as a consensus standard is supported by appropriate authorities. Like the HON Code, medical information providers adhering to this standard are invited to display an emblem identifying their support for the Code’s principles. In summary, the Code highlights the importance of peer review (before and after publication), reference to evidence-based literature and traditional scientific methodology, accountability (a clear statement of authorship and affiliation), publication date, statement of any conflict of interest, recognition of copyright, acknowledgement of the limitations inherent to telemedical consultations, and the ‘moral duties’ of expanding the information-Web via the creative use of hyperlinks, assisting disabled access to information, and preventing the misuse of data collected upon Web site visitors.

The need to avert yet another ‘standards war’ on the World-Wide Web is obvious. One candidate for the origination of a consensus statement on quality standards – a kind of ‘RFC’ – is the Internet Healthcare Coalition (IHC). Seeded by discussion on the MEDWEBMASTERS mailing list, the IHC held their first Board meeting in September 1997. This international group includes representatives from the pharmaceutical industry, academia, government, publishing, and non-profit organisations. It was founded with the aim of fostering education ‘regarding use of technology to deliver healthcare information and services, guidance regarding quality of healthcare information, publicity regarding positive examples of use of technology to deliver healthcare information and services, advocacy of appropriate uses of technology for healthcare information and services, and research and development of frameworks to promote good-quality healthcare information’ [10].

In conclusion, the existing proposals appear very similar in scope; we seem to know what is required. None emerges a clear favourite for adoption, as each standard/code has its own particular merits. Self-regulation through the adoption of voluntary standards is a much more realistic goal than compulsory, external enforcement; the Internet simply does not lend itself to the latter. Standards, however, are probably only part of the overall solution. The Internet is already being used to help the public identify fraudulent claims concerning health care products and services e.g. the Quackwatch Web site. Many catalogues (such as CliniWeb, Medical Matrix, MedWeb, and OMNI) attempt to filter information by controlling quality not at source, but by positively influencing access to peer-reviewed resources meeting a certain standard. Another approach is to establish ‘controlled’ consumer health information repositories – an approach taken by the US Department of Health with Healthfinder, providing access to ‘vetted’ federal and non-profit health information sites. In the UK, the Health Education Board for Scotland (HEBS) has taken a similar approach. What ever measures may lie around the corner, as medical information providers on the Internet, we must place accountability above all else.

1 McKenzie B. Becoming an information provider. In: Medicine and the Internet: Introducing online resources and terminology 2nd ed. Oxford: OUP; 1997. p. 253-64.

2 La Porte, RE, Hibitts, B. Rights, wrongs, and journals in the age of cyberspace. BMJ 1996;313:1609-11.

3 Impicciatore, P, Pandolfini, C, Casella, N, Bonati, M. Reliability of health information for the public on the world wide web: systematic survey of advice on managing fever in children at home. BMJ 1997;314:1875-9.

4 The Lancet. The web of information inequality. The Lancet 1997;349:1781.

5 Weisbord SD, Soule JB, Kimmel PL. Poison online – acute renal failure caused by oil of wormwood purchased through the Internet [Brief Report]. NEJM 1997;337:825.

6 Thompson, C. Surfer, heal thyself. New Scientist 1997 Oct 4;2102:18-9.

7 Silberg, WM, Lundberg, GD, Musacchio, RA. Assessing, controlling, and assuring the quality of medical information on the Internet: Caveant lector et viewer – let the reader and viewer beware [Editorial]. JAMA 1997;277:1244-5.

8 Wyatt, JC. Commentary: measuring quality and impact of the world wide web. BMJ 1997;314:1879-81.

9 Horton, R. Sponsorship, authorship, and a tale of two media. The Lancet 1997 May 17;349:1411-2.

10 Mack, J. [MWM-L] News from IHC Board Meeting. MEDWEBMASTERS-L@LISTSERV.ACOR.ORG 1997 Sep 11.