British Journal of Clinical Pharmacology DOI:10.1046DOI:10. /j.1365DOI:10. -2125DOI:10. .2003
AimsThe generally poor quality of health information on the world wide web ( WWW) has caused preventable adverse outcomes. Quality management of information on the internet is therefore critical given its widespread use. In order to develop strategies for the safe use of drugs, we scored general and content quality of pages about sildenafil and performed an intervention to improve their quality.
MethodsThe internet was searched with Yahoo and AltaVista for pages about sildenafil and 303 pages were included. For assessment of content quality a score based on accuracy and completeness of essential drug information was assigned. For assessment of general quality, four criteria were evaluated and their association with high content quality was determined by multivariate logistic regression analysis. The pages were randomly allocated to either control or intervention g roup. Evaluation took place before, as well as 7 and 22 weeks after an intervention which consisted of two letters with individualized feedback information on the respective page which were sent electronically to the address mentioned on the page.
ResultsProviding references to scientific publications or prescribing information was significantly associated with high content quality (odds ratio: 8.2, 95% CI 3.2, 20.5). The intervention had no influence on general or content quality.
ConclusionsTo prevent adverse outcomes caused by misinformation on the W WW individualized feedback to the address mentioned on the page was ineffective. It is currently probably the most straight-forward approach to inform lay persons about indicators of high information quality, i.e. the provision of references.
IntroductionThe World Wide Web (WWW) has become an important source of health information. Within the European Union 9% of 16067 people aged over 14 years used the WWW to obtain health information at least once a month, another 14% also used it but less frequently [1]. In line with these figures 37/138 patients attending a rheumatology clinic had searched the internet for med- . Several strategies may be followed to prevent such events caused by misinformation present on the internet. Earlier Silberg suggested four criteria to assist users in finding credible information: authorship, attribution (listing of references), disclosure of financial interests, and currency [7]. Meanwhile DISCERN, whereof inter-rater variability has been determined [8,9], and several other rating instruments [10] are available. In a previous study on pages about the herbal drug St John's Wort and exploring eight different markers, we found that two of these markers (citation of scientific sources and absence of possibility to purchase drug) may serve as indicators of reliable information [11]. Alternatively, quality labels, like HON [12], or quality standards, like the e-health code of ethics [13], aim at improving health information quality on the WWW by voluntary participation [14]. In contras...