Because many people look to the media for information on health care, media campaigns can influence patterns of health care service utilization. [1] Given its potential impact, there is general agreement that the information provided by the media should be of good quality; there is less consensus, however, about how the quality should be judged.
In this issue Frank Molnar and colleagues report on 50 articles randomly selected from newspaper advice columns containing medical advice relevant to elderly people (page 393). [2] Five geriatricians assessed the articles on the following 5 criteria: ability to determine to whom the article applied, ability to distinguish opinion from fact, degree to which critical issues were addressed, the safety of the advice, and the appropriateness of the advice given. They concluded that "a significant percentage of the articles contained inappropriate or even potentially dangerous advice."
Do newspaper advice columns really contain serious misinformation? Perhaps these conclusions reflect, at least in part, the different philosophies of the physicians dispensing the advice and those who devised and applied the rating criteria? For each example of advice that the rating physicians considered "potentially dangerous," the more appropriate advice offered was "medical evaluation." This suggests that the columnists and the geriatricians have different views about when and how physicians should attempt to help people understand their own problems, and when and how they should focus on protecting them from harm and encourage consultation with a medical professional.
How worried should we be if the advice given in newspaper columns is sometimes poor? Before we draw conclusions about the harm (or, indeed, the benefit) the advice could cause, we should determine who reads the columns, how they interpret them and how readers act on the advice given, if at all. One need not assume that readers of newspapers are gullible people who will blindly follow any advice given. Most tend to be reasonably well educated; and some seniors are quite skeptical about the accuracy of information provided in newspapers. [3] As an aside, it is interesting that very few columns include disclaimers urging readers to use caution if acting on the information provided to protect against possible litigation.
Most people distinguish between media genres; they have different impressions regarding the accuracy of information provided in advertisements, news reports and fictional serials, for example, and they interpret the information provided therein differently. The advice in newspaper columns is usually explicitly directed toward the one person who asked for advice; we do not know under what circumstances or to what extent readers may apply this advice to themselves.
Several factors might moderate the impact of the advice considered "potentially dangerous" by the raters in the study of Molnar and colleagues. Often, people would not be able to follow the advice given without first consulting a physician. For example, a reader with essential tremor would need a prescription to obtain the propanolol that was recommended; a reader with symptoms of stress incontinence could not do more than "consider" collagen implants (as suggested by the columnist) without obtaining the medical evaluation advocated by the raters. Rather than relying solely on the advice provided in newspapers, readers might be prompted to seek more information elsewhere or to initiate discussions with health care professionals. Consider, for example, a person who had not yet sought medical help for pain and tingling in the legs. Reading that these symptoms could indicate peripheral neuropathy might prompt them to consult a physician even if this course of action was not explicity recommended.
None of this is to deny that advice columns could be improved and that efforts should be made to ensure the advice provided is consistent with the best available research. The quality of the articles might be improved if columnists use a checklist similar to the one used in Molnar's study, but columnists and editors are likely to have other priorities. Peer review might help but will not guarantee quality in return for the professional time invested. Attempts to improve the quality of the articles are more likely to be successful if columnists' aims and the constraints under which they work are considered.
As Molnar and colleagues suggest, it is difficult to be both brief and comprehensive. In many instances, one cannot include all of the information that may, in some cases, be critical without increasing the length of the article. However, a reader can be referred to other sources - books, web sites, support groups and information lines - to find more detailed information.
These issues of information quality are not confined to newspaper advice columns. The quality of information provided in patient leaflets, [4] at Internet sites [5] and by health information services [6] has also been reported to be poor. Thus, attempts to improve the quality of information available to the public must focus on numerous sources. They should be complemented by strategies, via educational programs and the media, to improve peoples' ability to appraise, interpret and use the health care information provided.
Vikki Entwistle is with the Health Services Research Unit, University of Aberdeen, Scotland.
Competing interests: None declared.
Reprint requests to: Dr. Vikki Entwistle, Senior Research Fellow, Health Services Research Unit, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD; fax +44 1224 663087; [email protected]