OMNI Corner: Information or Hysteria, "Talking sensibly" in the Biomedical Field

Defining the subjects

Here at OMNI, our subject gateway essentially covers all of the health studies subjects in scope (basically medicine and health), and various biomedical aspects of the life sciences, such as molecular biology research, clinical psychology and sports medicines.

Of late, it has struck us that most of the national controversies that tend to fill the tabloids (and other newspapers) are sprung directly from these subject areas. In this section, we outline six examples most applicable to the UK; for countries such as the US, there would be a similar array of issues, including abortion as an issue of high public profile.

1. Dolly the sheep and cloning

Dolly entered the media spotlight in a blaze of glory a few years ago, when the Roslin Institute [roslin] announced the cloning of a sheep. Since then, speculation has increased on the ethical use of cloning, focusing especially on the ability to clone people. This issue covers both the life and health sciences, as well as impinging on agricultural areas of concern.

2. BSE

BSE (Bovine Spongiform Encephalopathy) [bse] is the major crisis to hit the animal farming industry in the UK in the last twenty years. The condition is found in cattle, and has been linked to the food products that they ingest. However, BSE is also heavily linked to the human condition of Creutzfeldt-Jakob Syndrome, therefore again making the wider issue of BSE something that straddles several areas within the health, medical and life sciences.

3. Genetically modified food

At the time of writing, the major public debate concerns the use of genetically modified ingredients, such as soya, in foods. This “debate” quickly spawned off a myriad of smaller debates, including:

• labelling of food
• ethics of inclusion of genetically modified ingredients
• effects on the environment of growing GM foods
• third world issues relating to GM foods
• medical implications from consuming GM foods
• testing and safety procedures for GM foods and ingredients

As is evident, the wider issue of GM foods covers many areas of the life and health sciences, as well as environmental and political issues [bio].

4. Use of cannabis for medicinal purposes

The issues regarding cannabis include whether it should be legally administered in cases where it assists someone with a serious medical condition, such as multiple schlerosis. There are periodic attempts to make use of cannabis legal in this way; thus, the issue is one that is frequently in the news [welsh]. This subject covers both the health and medical sciences, as well as biological research (why makes cannabis good?), and political and legal issues.

5. Euthanasia

Euthanasia is an issue which covers both the health studies areas, and the area of law. As well as being a controversial area, with an accompanying array of lobbying groups and emotive web sites, the law and common practise (not always the same) differ from one country to another.

6. Alternative/allied and complementary medicine

The issues surrounding alternative and allied medical practises are ones which seem to inspire more heated debate within medical circles than any other. At a recent OMNI seminar, a lively debate enraged between a speaker, who was justifying the need for a gateway to quality allied and complementary health resources, and one of the delegates, a GP with experience of dealing with patients who had been given incorrect alternative medical advice.

The tabloid and newspaper treatment of this area often verges towards the “hippy, josh stick and crystal magic” stereotype, wheras allied and complementary medicines cover a much wider field than most people think; herbalism, homeopathy, reflexology, chiropractise, and so forth. There is a growing movement of people, and increasing medical support and recognition, for many of the areas within this field [nih]. However, due to its nature, it is liable to be a controversial area for many years to come.

Defining the issues

Examination of the subject areas above draws out two particular issues:

• cross-subjectional interest. If a subject is controversial, for whatever reason, then it tends to cover a number of subject areas. For example, most of the subjects mentioned above have some elements of law and politics involved, as well as health and medical interests
• information quality. All of the subjects mentioned above are emotive to certain groups of people, who also have their own interests and agendas. This usually results in a lessening of reasonable debate, with a consequential drop in accuracy of data and information put into the public domain

We will examine these two areas more closely, in the context of resource discovery systems.

Cross subjectional interest

As we can see from the examples above, some topics and issues, especially those of high public interest, can cross several subject areas. With regard to subject-based internet resource discovery systems, this raises several issues. The primary issue is whether a resource that is cross-subjectional should be catalogued in just one subject based system e.g. biomedical, or several e.g. biomedical and law.

The experiences of the subject gateways to date indicate that in most cases, the preferred method is for the resource, if it passes all of the evaluation and relevance criteria, to be catalogued in both gateways. This means that the resource description can be targeted at the core audience for that gateway. For example, a document on the legal implications of smoking may be described from a legal perspective for entry into a legal resource discovery system, while it could be described from a medical perspective for entry into a medical resource discovery system.

This leaves us with the problem of cross searching. To date, some limited work on cross searching across multiple gateways has been carried out, by projects such as ROADS [roads]. A demonstrator, allowing people to search across a range of subject-based gateways, with the results subsequently integrated into one uniform results listing, was produced.

However, we are now entering the age of the Resource Discovery Network, and Hubs. The RDN is a venture which will capitalise and build on the experiences and successes of the established subject gateways, such as EEVL, OMNI and SOSIG [rdn]. It will seek to create eight faculty level hubs, covering broad areas such as medical and biological sciences, the social sciences, the arts and industries, and so forth. Each hub will probably consist of a number of collections (gateways) of resource descriptions, distributed or centralised in some manner according to the nuances of the particular hub.

This is where things get interesting. Someone is interested in the legal implications of smoking and lung cancer. Should they search the hub containing law, or the one containing biomedical resources? Or maybe, cross search both hubs and look at all of the results? Or even better, just cross search those bits of the two (or in some cases, more) of the hubs that are of relevance to the query? Alternately, should they just search across all of the gateways/description collections in all of the hubs in one go (high recall, low precision)?

For the user to be offered the best search will mean one or more of the following:

1. the user being able to cope in selecting the correct search mechanism and permutation amongst the options available, and devising a search query appropriate to that mechanism. There was an interesting mini-debate on the issues of user capability on lis-elib in January 1998 [elib]
2. storing some kind of “profile” of the user, so that the permutations of resource databases offered to them are optimised, and reducing the chances of the user making an incorrect selection of database.
3. some combination of both approaches
4. limiting the search option combinations available

It will be interesting to see how the Resource Discovery Network, and other collections of gateways that are starting to band together in consortia arrangements, tackle the issue of cross subjectional interest.

Information quality

Unless you have strictly restricted your media and news coverage to some of the more cerebral outlets over the last few months, you cannot have failed to notice the hysteria over genetically modified foods. The question that arises is: what is newsworthy, and what is information?

One persons viewpoint may not be factually accurate, but does that mean that if they create a resource around that viewpoint, then should not be excluded from, or catalogued in, a “quality resources only” gateway? Two parallel examples:

1) an establised political party may have odd views, that are patently unrealistic and ulterior, on some issues. For example, they may be funded by the nuclear industry, and believe that “if you ignore nuclear pollution, then it goes away over time and is harmless”. However, most resource discovery systems that covered politics would still probably catalogue the party, geographic relevance and content willing, despite its partisan views.

2) a medical company has a web site, where it promotes its drugs as being “the only solution”. Again, a partisan view, and many biomedical resource discovery systems of repute would avoid the resource.

Therefore; do we need different evaluation criteria for different subject areas, possibly even within the biomedical arena itself? Should we treat commercial company products that have been rigorously tested in the same way as we treat some alternative health products, which are less dangerous, but less tested in a scientific manner?

Thankfully, we don’t have to make up these decisions as we go along. Criteria has been developed via which people can assess Internet-based resources, for inclusion into resource discovery systems. We outline some places where you can find such criteria, or discussions about criteria.

AGEC - Advisory Group for Evaluation Criteria

The OMNI advisory group on evaluation criteria helped to develop the core evaluation criteria via which the OMNI service operates. The group has also produced a number of papers and presentations on the issues surrounding resource evaluation and quality. The OMNI service hosts a section on which the evaluation criteria, and several papers, are reproduced [agec].

Comparison of Health Information Megasites

This was a substantial comparison and study of a large range of biomedical resource discovery systems, carried out by a consortium of three US university libraries, in May 1998. The project looked at a large range of criteria pertaining to resource discovery, with issues of quality being prominent within those criteria. One of the products of the project was a listing of “Print and Internet Resources on Web Design and Internet Resource Evaluation” [megasites]; this contains references, and links, to many articles and papers of interest.

DESIRE

The DESIRE project, coordinated by the Institute for Learning and Research Technology in Bristol, UK, has been instrumental in building software tools, reports and guidelines pertaining to resource discovery, especially at a pan-European level. The DESIRE Web site contains reports regarding research into quality and selection criteria issues, as applied to resource discovery [desire].

(Re)developing criteria

Looking back at archives and printouts of resources that were developed a few years ago reveals some interesting characteristics of biomedical resources on the Web:

* resources were generally smaller in size, consisting often of one page. With such resources, people would catalogue or index just the page - simple. Nowadays, with resources often being huge and spread across several servers, we have the issue of whether it is best to catalogue the whole resource, just one part, various sections, and so forth

* resources were simpler; no frames, java, embedded movies and so forth. Also, advanced trickery, such as subtle redirection of the users, or the use of misleading metadata, was much rarer

* companies were slow to use the Internet, and most of the information in the first wave of web sites was produced by keen technically minded academics. Gradually, more formal academic and research material became available over the Web in larger quantities. In the current wave, commercially funded resources make up an increasingly large proportion of new sites; these require closer vetting than academic and research sites, in terms of impartiality and quality.

What these and other characteristics indicate is that there is a strong argument for the constant review and development of evaluation criteria as applied to resource discovery systems. This especially true in the biomedical field, where controversies such as those mentioned at the start of this article lead to people trying more complex ways of getting their viewpoint across over the Internet.

As Internet-based technologies become more complex, and people and organisations use these technologies in more advanced ways (sometime not ethically), so the criteria must be frequently revised, to keep up with these developments, if the gateways are to retain their status as subject based quality filters.

References

1. [agec] OMNI Advisory Group on Evaluation Criteria http://omni.ac.uk/agec/
2. [bio] Biotechnology in our food chain http://www.jic.bbsrc.ac.uk/exhibitions/bio-future/index.htm
3. [bse] BSE enquiry home page http://www.bse.org.uk/
4. [desire] DESIRE http://www.desire.org/results/discovery/
5. elib ebate on internet searching http://www.mailbase.ac.uk/lists/lis-elib/1998-01/0079.html
6. [megasites] Print and Internet Resources on … Internet Resource Evaluation http://henry.ugl.lib.umich.edu/megasite/bibl.html
7. [nih] National Center for Complementary and Alternative Medicine http://altmed.od.him.gov/nccam/
8. [omni] OMNI: Organising Medical Networked Information http://omni.ac.uk/
9. [rdn] Resource Discovery Network - posting to lis-elib http://www.mailbase.ac.uk/lists/lis-elib/1999-01/0039.html
10. [roads] ROADS: Resource Organisation And Discovery System http://www.ilrt.bris.ac.uk/roads/
11. [roslin] Roslin Institute http://www.ri.bbsrc.ac.uk/
12. [Welsh] Arthritis-sufferer jailed for using cannabis
13. http://news.bbc.co.uk/hi/english/world/europe/newsid%5F285000/285148.stm

Author Details