Context

Currently, many sources, and especially the media, give contradictory information about hormone replacement therapy to women seeking information. The internet may improve concordance and patient satisfaction by facilitating participation in discussion groups, providing support and allowing ideas and experiences to be shared¹. At present the reliability and quality of information presented on an estimated 100 000 medical web sites is variable.²There are several problems concerned with web site evaluation. Firstly, "no omniscient detached observer exists who can simultaneously view an article through the eyes of a specialist researcher, doctor, patient, and member of the public, let alone take into account the different perspectives of orthodox and complementary medicine"³. Secondly, it is possible that organisations offering kite marks to guide the public may face litigation for overvaluation of the site by patients harmed by advice provided or conversely for defamation or disparagement where the site is under-rated. Currently, no published studies exist, evaluating the quality of menopause and hormone replacement therapy web sites.

Objectives

To evaluate quality of web based information on the menopause and hormone replacement therapy, to identify whether site ownership is an indicator of both web site and content quality, and therefore to provide guidance for healthcare professionals over which sites are suitable for recommendation to patients.

Methods

A search of web sites was conducted using eight commonly used search engines and two key terms ("menopause" and "hormone replacement therapy") limited to English language. Thus, a sample of 25 sites (of varied ownership) was generated. Using a pre-tested scoring tool, each site was assessed on factual information provided and quality of site. For each criterion, a score of 1-5 (1 low, 5 high) was assigned, and two mean scores were generated. Each site was also assessed in a semi-structured qualitative manner.

Results

The highest scoring site was http://www.hrtinfo.co.uk (Solvay) with a mean score of 4.6. The weakest site (http://www.plannedparenthood.co.uk) gained a mean score of 1.6, reflecting very low quality information. The mean factual informational value of the sites sampled was 3.129 (out of a maximum of 5) but with a range of 1.6-4.6. The mean value of the sample for the quality of web sites was higher (3.504) but again over a broad range (2.5-4.5). The pharmaceutical industry, community pharmacies, governments, and charities were associated with higher quality sites. Only eight of the 25 web sites analysed displayed any images, but when included, they were generally appropriate and useful for promoting understanding or illustration purposes. Only two web sites displayed any evidence of awards gained. Pharmaceutically associated sites tended to have a slight bias towards uptake of HRT, compared with the alternative of no treatment. Meanwhile, private healthcare providers' and charity-run sites tended to include more information on non-drug treatments and extra information such as contraceptive advice.

Conclusions

The information about menopause and HRT on the web is often of dubious calibre, with incomplete information being commonly provided. Therefore it is recommended that healthcare professionals direct women to sites owned by the pharmaceutical industry, community pharmacies, governments, and charities. In doing so it would be prudent to advise women that commercial sites may be biased towards particular products.

References

1. Jadad AR. Promoting partnerships: challenges for the Internet age. Brit Med J 1999; 319: 761-764.
2. Pallen M. Guide to the Internet: The World Wide Web. Brit Med J 1995; 311: 1552-6.
3. Delamothe T. Quality of websites: kitemarking the west wind: Rating the quality of medical websites may be impossible. Brit Med J 2000; 321: 843-44.