There is an interesting article in the November issue of the European Journal of Cancer that addresses the quality of consumer information available through the Internet about the value of PSA-based testing for prostate cancer.
Korfage et al. set out to identify and evaluate “core consumer information” about PSA-based testing for prostate cancer available through an Internet-based search in November 2009 — i.e., several months after the initial results of the PLCO and ERSPC studies had both been published. In evaluating the materials that they found, they made two fundamental assumptions:
- That (mass, population-based) screening for prostate cancer can potentially “reduce prostate cancer mortality at the expense of considerable overdiagnosis and overtreatment”
- That relevant information about PSA-based testing should be “correct, balanced, and supportive of autonomous decision-making”
These assumptions reflect the findings of the ERSPC results rather more than they do the PLCO results, and one can quibble with the wording (if you dislike the term “overdiagnosis”) but on the whole the assumptions seem to be reasonable. We should note that we have been able to review the full text of this paper and not just the abstract.
The authors sought to identify Internet-based materials using a variety of search terms (which they provide in detail in the paper). Their intent was clearly to try to identify materials available on the Internet that could be easily printed out and given to patients (leaflets, brochures, PDF-based handouts, etc.). Only materials from governmental or not-for-profit organizations were (supposedly) accepted for evaluation (i.e., no material from commercially motivated sources). However, the search terms they used may not have given them the best opportunity to find widely used or widely available resources. As just one example, their search criteria did not include the term “brochure” and others that might lead one to relatively high quality US resources. This may reflect intent on the part of the authors (so that they identified a range of materials) or it may reflect a language issue (since the first language of the authors was not English).
In they end they evaluated materials from 23 different sources, including those from:
- The American Cancer Society (USA) [actual document evaluated no longer available]
- The National Cancer Institute (USA) [see online document actually evaluated]
- The Mayo Clinic (USA) [see online document actually evaluated]
- The Centers for Disease Control (USA) [see updated version of document actually evaluated]
- The AUA Foundation (USA) [see online document actually evaluated]
- The Ontario Ministry of Health and Long-Term Care (Canada) [see online document actually evaluated]
- The British Columbia Cancer Agency (Canada) [actual document evaluated no longer available]
- The National Health Services (UK) [see online document actually evaluated]
- CancerHelp UK (UK) [see updated version of document actually evaluated]
- The Prostate Cancer Charity (UK) [see updated version of document actually evaluated]
- The European Association of Urology (Europe) [see potentially updated version of document actually evaluated]
- The Dutch Cancer Society (Europe) [see one of two linked documents actually evaluated]
Interestingly, in the US and Canada, they selected only one document from any form of prostate cancer-specific support organization (PCNG Cincinnati, a chapter of Us TOO [see online document actually evaluated]).
In evaluating the literature that they identified, the authors focused on a number of key points that they considered to be essential in consumer information for men making decisions about PSA-based screening tests:
- The potential for early as opposed to delayed diagnosis of prostate cancer
- Information about reduction in mortality associated with screening
- Adverse events associated with treatment of localized prostate cancer
- How a PSA test actually works (i.e., how the test is carried out)
- The chance that a PSA test will deliver a “false positive” or a “false negative” result
- The risk for “overdiagnosis” (i.e., the diagnosis of tumors that will never have any clinical significance)
- The potential need for other tests (DRE, TRUS, biopsy)
The quality of the materials were evaluated against specific criteria:
- 19/23 of the materials met all key criteria listed above.
- 12/23 of the materials helped men to understand that their preferences were an important component in the decision to have or not have a PSA test.
- 16/23 of the materials helped men to be clear in their own minds about what mattered most to them in making their decisions.
- 16/23 of the materials helped men to discuss their personal preferences with a primary care physician.
- 10/23 of the materials helped men to become involved in the decision process so that their preferences would be thoroughly considered and discussed.
- 11/23 of the materials helped men to improve the likelihood that he would in fact make the choice that best matched the features of importance to him.
In their discussion of the evaluation, the authors note the following:
- Most of the materials “supplied adequate information.”
- Most of the materials did not actually recommend PSA testing, but the degree of suggestion that PSA testing is a good idea tended to reflect variation in opinion about the value of PSA testing and the autonomy of men to participate in the decision process.
- Much of the information did not (at the time of evaluation) contain results from the PLCO and ERSCP trials.
- PSA-based testing for prostate cancer does have side effects.
- “A sufficient level of decision-relevant knowledge is a prerequisite for informed decision-making” (and how such a level of knowledge can be achieved will vary from person to person).
- “Understanding all benefits and disadvantages of PSA screening requires a high level of health literacy.”
The “New” Prostate Cancer Internet believes that this article does have one significant (albeit common) flaw, which is the failure to distinguish clearly between the use of the PSA test in individual assessment of risk in a specific man (i.e., testing) and the use of the PSA test in mass, population-based screening. It would help everyone if we could clearly, carefully, and continuously differentiate between these two concepts.