This week in the BMJ
Volume 324, Number 7337, Issue of 9 Mar 2002Unvalidated tools to rate web health information are proliferating
Unvalidated tools to rate web health information are proliferating
Many incompletely developed rating instruments intended to assess the
quality of health information on websites continue to appear and most
stop functioning soon after their release. Gagliardi and Jadad (p 569)
found that in the past four years, 98 tools have been used to rate the
quality of health information on the internet. All were apparently
unvalidated. Strategies to evaluate the quality of health information
on the internet are flourishing but it is not clear they are necessary
or sustainable.
![[To top]](/icons/back.gif){kind=icon}
Internet users say and do different things about quality
Internet users explore only the first few links on general search
engines when seeking health information and do not check out the
"about us" sections of websites or read disclaimer or disclosure
statements. Often they don't remember the site they got their health
information from. In their qualitative study, Eysenbach and Kohler (p
573) found that while the consumers said that they looked for the
source and design to assess the credibility of a site, in practice they
rarely bothered.
Popular breast cancer websites not necessarily of higher quality
The more popular websites providing information about breast cancer are
not necessarily of higher quality. More popular sites are more likely
to contain information on clinical trials and results of trials than
less popular ones. Meric and colleagues (p 577) suggest that type
rather than quality of content determines popularity of websites and it
remains the responsibility of the medical community to direct patients
to sites of known quality.
Apparently credible websites may be inaccurate
Apparently credible websites may not necessarily provide higher levels
of accurate health information. Of 121 websites that provided
information on five common health topics, three credibility features
source, currency, and evidence hierarchyhave only slight or
at best moderate correlation with accuracy of information. Kunst and
colleagues (p 581) suggest that apparently credible websites may not
necessarily provide higher levels of accurate health
information.
Quality of health information on the internet has improved
The quality of health information on the internet has improved over the
past few years despite concerns over poor quality and its possible
consequences. Four years after it was found to be of poor quality, a
re-evaluation of information on managing fever in children at home,
suggests an overall improvement in adherence to guidelines for quality
of content. However, Pandolfini and Bonati (p 582) stress that
monitoring health information on the internet for accuracy,
completeness, and consistency is still fundamental.
Improving public safety in primary care
Safety is a major concern in four main areas of primary care:
diagnosis, prescribing, communication, and organisational change. Wilson and Sheikh (p 584) report that of all adverse incidents reported, 28% are related to problems with diagnosis. Up to 5% of all
prescriptions might cause problems, and one third of these can be
classified as serious. They believe that much can be done now and
propose seven steps towards improving safety: understanding systems,
leadership and culture, research, analysis, establishing best practice,
improving techniques and technology, and monitoring safety.
Consumers should be encouraged to search critically
Consumers should be encouraged to seek out information critically and
to see time invested in critical searching in preference to developing
yet more rating tools. According to Wilson (p 598), although this
will be a challenge, there are already a wide range of tools
to assist website developers to produce good quality sites and to help
users to discriminate between sites. These include codes of conduct,
quality labels, user guides, filters, and third party
certification.
(Credit: SUE SHARPLES)
Microdialysis has wide applicability for in vivo measurements
Microdialysis enables in vivo measurement of tissue chemistry and is
feasible in virtually every human organ. According to Müller (p 588)
it is currently being used to monitor brain ischaemia and metabolic
control. This technique is also about to become a standard tool in drug
monitoring and development. In the future, miniaturisation and online
automisation of chemical assays will allow "bedside" microdialysis
for assessing antibiotic penetration for infected organs and monitoring
tissue metabolism in disease states.
People at risk of cervical cord injury are often not immobilised
Failure to immobilise patients at risk of cervical cord injury is
common and this may have devastating consequences as 75% of injuries
to the spinal cord are incomplete at presentation. In addition to not
initiating immobilisation, problems arise from giving the all clear
after inadequate evaluation. Skellett and colleagues (p 591) report on
60 children at risk of cervical cord injury, 14 (23%) of whom had not
been immobilised before reaching the hospital. The authors recommend
educating healthcare professionals about the importance of
immobilisation and having ready access to immobilising devices that are
appropriate for different ages.
Internet law is muddling along
Technologically mediated health care raises problems of patient
confidentiality, cross border practice, and quality of information. Terry (p 602) reports that US regulators have yet to find the appropriate balance between risk and benefits of cross border practice,
while Stanberry (p 605) says that the emphasis of European policy
in this area ought to switch from resisting online health services to
finding ways to properly supervise and accredit them.
(Credit: MARK OLDROYD)
Student BMJ
Asylum seekers' care
UK medical students have published unreleased government plans to restrict failed asylum seekers' access to medical care
