SEARCH

SEARCH BY CITATION

Keywords:

  • Internet;
  • World Wide Web;
  • website;
  • search engine;
  • information score;
  • patient education

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

Objective To determine the use of the Internet by urological patients for obtaining information about their disease, and to conduct an evaluation of urological websites to determine the quality of information available.

Patients and methods Questionnaires about Internet use were completed by 180 patients attending a general urological outpatient clinic and by 143 patients attending a prostate cancer outpatient clinic. The Internet evaluation was conducted by reviewing 50 websites listed by the Hotbot™ search engine for two urological topics, prostate cancer and testicular cancer, and recording details such as authorship, information content, references and information scores.

Results Of the patients actively seeking further information about their health, 19% of the general urological outpatient group and 24% of the prostate cancer group used the Internet to obtain this information. Most websites were either academic or biomedical (62%), provided conventional information (95%), and were not referenced (71%). The information score (range 10–100) was 44.3 for testicular cancer and 50.7 for prostate cancer; the difference in scores was not significant.

Conclusion The use of the Internet by patients is increasing, with > 20% of urology patients using the Internet to obtain further information about their health. Most Internet websites for urological topics provide conventional and good quality information. Urologists should be aware of the need to familiarize themselves with urological websites. Patients can then be directed to high-quality sites to allow them to educate themselves and to help them avoid misleading or unconventional websites.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

There has been a massive expansion in the amount of information available through the Internet over recent years. This expansion has been matched by an increase in the number of individuals using the Internet. The UK now has an estimated 15 million regular users of the Internet, second only to the USA worldwide [1]. Patients who seek to improve their knowledge of health and treatment options can now access in a few seconds data that would previously have required hours of painstaking research. Patients have previously relied heavily upon their doctors for information that was conveyed verbally or through pamphlets, videos or books accessible on an outpatient basis. Those resourceful enough to obtain access to medical libraries and databases could only examine a fraction of the information that is now available on the World Wide Web.

The crucial difference between the information provided through the Internet and more traditional sources is that the latter is usually subject to some form of peer review. The urological patient seeking information only from the Internet may not be aware of the peer-review process and the critical assessment to which published medical material is subjected. The Internet provides no separation of scientifically proven conclusions from anecdotal information or commercially biased reports. Web pages can be set up by anyone and may be altered and/or removed at any time, with no effective way of protecting the general public from misleading medical information. The medical profession is gradually confronting this problem and various instruments have been developed in an attempt to evaluate health information on the Internet [2]. However, these instruments appear incomplete at best, and at worst may threaten the expansion in communication and expression brought about by the Internet.

We assessed the extent to which urology patients are using the Internet and other sources to obtain information about their health. We also evaluated the quality of information available on the Internet for two urological topics, prostate cancer and testicular cancer, based on a previously published but unvalidated scoring system [3].

Patients and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

Questionnaires were given to 180 consecutive new patients attending a general urology outpatient clinic; all questionnaires were completed and returned before the consultation. Questionnaires were also sent to 200 patients who were under follow-up in a separate prostate cancer clinic and were aware of their diagnosis; 143 (73%) completed questionnaires were returned. The questions concerned the nature of information sources consulted by patients, the ease of access to such information and the perceived quality of the information. Patients who had used the Internet to access information were asked a series of more detailed questions about how and where they used the Internet and about specific search engines or sites consulted ( Table 1).

Table 1.  The questionnaire and the results
Characteristic/questionGeneral urology patientsPatients with prostate cancer
Number180143
Mean age (years)  53.470.4
Sex, M/F152/28
% answering ‘Yes’
 Was patient directed to information sources by GP?  1220
 Did the patient actively seek further information  about symptoms and diagnosis?  36 58
 Did the patient use the Internet to obtain this information?  1924
 Was the Internet easy/very easy to use?  6786
 Was a search engine used?  9287
 Was the information obtained useful?  5862
Internet location (%)
 Home  6088
 Work  2812
 Library/other  120

Internet evaluation

The search terms ‘prostate cancer’ and ‘testicular cancer’ were entered into the search engine HotBot™ with the intention of simulating as closely as possible the searches that would be performed by patients with limited Internet or medical knowledge. HotBot has been consistently rated as the best of the Internet search engines [4], covering a third of the indexed Web, which is now estimated at over 1.5 billion pages [5,6]. It uses a program that searches the Web for new addresses, accepting all sites regardless of content. This contrasts with other popular search engines such as Yahoo™ that screen sites before inclusion in their database, a process which may eliminate insidious elements but also subjects the selection of sites to market-driven bias [7].

Search results are displayed as a list of websites, ranked in order of the percentage match to the search term. The searches were not altered or redefined, regardless of the web sites identified. The websites were reviewed sequentially until 50 sites had been assessed for both search terms. The websites were not assessed if they contained the search term but were clearly unrelated to patient education. The websites generated often had hyperlinks to other sites, making it difficult to determine exactly where one site's information moved to another. Therefore, the information was evaluated based on the initial web page, without following links to other sites, a policy advocated by previous reviewers [8]. These restrictions resulted in the need to view 75 sites for prostate and 88 sites for testicular cancer, to review the 50 websites.

Web sites were assessed using a modification of a previously published strategy [3]. Sites were scored according to the following categories.

(i) Authorship; six author classifications were used, i.e. academic (university sites), biomedical (nongovernmental organizations, i.e. charitable organizations), physician (private healthcare groups), commercial, news and personal.

(ii) Contents; the treatment and diagnostic options described by each web site were evaluated and determined to be either conventional or unconventional. Conventional websites advocated therapy that would be present in current textbooks and journals of urology, and unconventional websites described alternative, experimental therapies as the principal mode of treatment.

(iii) References; the web sites were classified according to the reference sources provided, ie as referenced, when peer-reviewed literature was cited, or unreferenced when the authors did not cite their information sources.

(iv) Information score; by modifying the information score (IS) described by Soot et al.[3], a qualitative score was applied to each website selected for review. This score assesses the ability of the website to provide information about the disease, its morbidity and prognosis, treatment options and their complications. The IS ranges from 10 points (poor) to 100 points (excellent), with different weighting given to the different scoring areas ( Table 2). Websites were scored independently by two observers and the mean IS used in the analysis.

Table 2.  Information score categories assessed for information quality
CategoryScoreWeightWeighted score range
Disease summary1–1033–30
Disease complications1–101.51.5–15
Treatment summary1–1033–30
Treatment complications1–101.51.5–15
Disease prognosis1–1011–10

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

Questionnaire replies were received from all 180 new referral patients attending general urology outpatient clinics, and from 143 men under follow-up in a prostate cancer clinic (73%). The results of the questionnaire survey are summarized in Table 1. The categories of the Internet assessment were as follows.

(i) Authorship; the most common author classification overall was biomedical, accounting for 36 of the 100 sites reviewed (36%); more of the prostate cancer sites had biomedical authors (44%) than had the testicular cancer sites (28%). The testicular sites had a higher proportion of academic sites (28%) than the prostate sites (22%). The physician, commercial, news and personal authors were more common in the testicular sites (22, 44%) than in the prostate sites (17, 34%). The authorship results are summarized in Table 3.

Table 3.  Web page author, contents and reference category, and IS for the information quality
CategoryProstate cancerTesticular cancer
Author:
Academic1114
Biomedical2214
Physician56
Commercial84
News18
Personal34
Contents:
Conventional4847
Unconventional23
References:
Referenced1316
Unreferenced3734
Mean IS:
Disease summary17.216.1
Disease complications6.45.9
Treatment summary15.613.6
Treatment complications74.8
Disease prognosis4.23.8
Total IS50.744.3

(ii) Contents; two (4%) of the prostate cancer and three (6%) of the testicular sites recommended unconventional therapy; most of these sites promoted the use of shark cartilage or herbal remedies to prevent disease progression in metastatic disease. The content results are also summarized in Table 3.

(iii) References; most prostate and testicular cancer sites gave no conventional reference sources (overall, 71 of 100). Two of the prostate sites listed unconventional references (anecdotal/personal). The academic sites for prostate and testicular cancer referenced conventional sources of information most of the time (seven of 11 for prostate and nine of 14 for testicular cancer; Table 3).

(iv) Information score; the mean (range) IS for each topic was 50.7 (10–100) for prostate and 44.3 (10–100) for testicular cancer; there was no significant difference in IS between the groups. The IS results are also summarized in Table 3.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

The Internet has been transformed since its original inception in the 1960s as a communications system (ARPANET) designed for American military use in the event of atomic war [9]. The growth of the Internet over the past few years has been phenomenal, with an estimated 1.5 billion web pages currently, compared with 500 million a year ago, and more than 3 million pages being added every day [5]. The UK has the greatest proportion of Internet users over the age of 60 years in Europe, and this ‘silver surfer’ group represents a significant component of the patient population who can now easily obtain medical information [1].

The Internet provides a quick and relatively inexpensive source for the inquisitive patient to seek a second opinion. A 1999 UK survey found that 6 million regular Internet users had obtained further information about their health from the World Wide Web [10]. Concern has arisen because patients are likely to be ignorant of biases inherent in certain web pages. Attempts have therefore been made to develop instruments to evaluate health information on the Internet [11] and previous studies have assessed the quality of ‘on-line’ information for several health topics [3,12]. The present study determined the current use of the Internet by general urological patients attending their first outpatient clinic and by outpatients with prostate cancer who knew their diagnosis.

The patients completing the postal questionnaire (prostate cancer) were older than the outpatients completing the questionnaire (mean age 70 vs 53 years). The prostate cancer group, aware of their diagnoses, had unsurprisingly made a greater attempt to learn more about their symptoms and diagnosis. Overall, few GPs had directed either patient group to other sources of information about their symptoms and diagnoses. This may occur because GPs are not aware of good websites, or it may be an attempt to avoid undue worry and stress in cases of suspected malignancy. However, the greater percentage of patients in the prostate cancer group receiving information about information sources does not support this. Furthermore, there were more benign cases in the outpatient group and yet fewer patients in this group were directed to other sources of information. These findings suggest that a greater provision of patient information leaflets on urological problems for primary care would be justified.

There were more patients using the Internet in the prostate cancer (24%) than in the outpatient group (19%). That more patients with prostate cancer used the Internet at home reflects the higher proportion of retired patients in this group (82% compared with 41% in the outpatient group). Both groups found the Internet ‘easy/very easy’ to use and both used a search engine to obtain information about their symptoms and diagnosis. The use of a search engine is to be expected, given that few of the patients were directed to specific sites for further information. In response to the subjective question as to whether the information obtained was ‘useful’, both groups stated ‘yes’ in 60% of cases overall.

The Internet evaluation indicated that biomedical authors accounted for the largest proportion of prostate cancer websites in the top 50 reviewed. There is a greater public awareness of prostate cancer and this, combined with the higher prevalence of prostate cancer, leads to more biomedical websites. There were reassuringly few ‘unconventional’ websites for either urological topic, consistent with previous findings on other medical topics [3]. Few websites listed reference sources for the information provided; even the academic websites failed to do so in almost 40% of cases.

The IS represented the most objective aspect of the Internet evaluation. There was a markedly higher IS than obtained previously on vascular surgery topics [3]. There was no significant difference between the IS obtained for the two urological topics, and even when the individual components of the IS were compared, there were no significant differences.

The growth of the World Wide Web is inevitable and with this will increase the number of patients using the Internet to gain information about their health. We believe it would be of value to patients if urologists and GPs familiarized themselves with quality medical websites, and would thus be able to recommend one or more websites to their patients. Website recommendations would not only help in educating patients, but also decrease the risk of patients ‘surfing’ onto misleading or unconventional websites.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

Authors

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and methods
  5. Results
  6. Discussion
  7. References
  8. Authors

G.O. Hellawell, MA, FRCS, Research Fellow in Urology.

K.J. Turner, MA, FRCS, Research Fellow in Urology.

K.J. Le Monnier, BA, RN, Research Sister in Urology.

S.F. Brewster, BSc, MD, FRCS (Urol), Consultant Urologist.