Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Evidence-Based Dermatology: Some Problems With Research

Evidence-Based Dermatology: Some Problems With Research Sam Gibbs1 is to be congratulated on his research commentary “Breakthrough in the Treatment of Warts” published in the June issue of the ARCHIVES. The flaws in the original article2 are so significant that one wonders how it survived the peer-review process and made it to print. Surely no patient will ever benefit from the findings. It is poignant that the commentary appears in the same volume of the ARCHIVES as an article assessing the frequency of statistical errors in the dermatological literature.3 I applaud the ARCHIVES for being so open and honest about some of its publication shortcomings. Only by assessing our inadequacies can we look forward, improve the quality of dermatological research, and provide better evidence-based patient care. There are several basic principles of research planning, development, and reporting. Without these, research does not help improve patient care: • Ask an appropriate research question that is structured and answerable. It should be patient driven and relate to a specific patient population and condition. Vague questions might get vague answers or, often, no answers at all. • Use an outcome measure that is clinically relevant to the patient. For example, in alopecia areata, clinically meaningful hair regrowth after treatment might (to the patient) mean 100%, or a full head of hair, or the ability to stop wearing a wig. • Use the appropriate research method and design to answer the question. A double-blind, randomized, placebo-controlled trial is not always needed to answer a clinical question. Some randomized controlled trials are performed and reported so poorly that their findings are meaningless.2 • Present and report data in a clear, open, and honest way so that there is no risk of misunderstanding or misinterpretation by the reader. Important information, such as magnitude of treatment benefit, must always be included so that the reader can assess likely patient benefit. • Analyze the data in the correct way. The old adage (attributed by Mark Twain4 to Benjamin Disraeli) “There are three types of lies: lies, damn lies, and statistics” remains alive as long as inappropriate statistical methods are used (either honestly or dishonestly) to interpret data. • Understand the difference between results that are clinically meaningful and those that are purely statistically meaningful. If the results are reported as “significant,” it would nice to think that they were actually significant and meaningful to our patients. Of course readers should critically appraise journal articles3 but many clinicians do not have the time, resources, or expertise to do this. Perhaps journal editors should include a statistician in the peer-review process, but this might be expensive. Researchers should be encouraged to seek advice from epidemiologists and statisticians before undertaking complicated projects. Readers could study the letters section a few months after an article is published to see what the experts really think. Basic epidemiology and statistics should be incorporated into all dermatology training and continuing medical education programs. The ARCHIVES has made an excellent start by publishing these articles,1,3 throwing down the gauntlet, and opening up some informed debate. Correspondence: Dr Sladden, Launceston General Hospital, Charles Street, Launceston, 7250, Australia. Financial Disclosure: None reported. References 1. Gibbs S Breakthrough in the treatment of warts? Arch Dermatol 2006;142767- 768PubMedGoogle ScholarCrossref 2. Gustafsson LLeijonhufvud IAronsson AMossberg AKSvanborg C Treatment of skin papillomas with topical alpha-lactalbumin-oleic acid. N Engl J Med 2004;3502663- 2672PubMedGoogle ScholarCrossref 3. Neville JALang WFleischer AB Jr Errors in the Archives of Dermatology and the Archives of the American Academy of Dermatology from January through December 2003. Arch Dermatol 2006;142737- 740PubMedGoogle ScholarCrossref 4. Twain M The Autobiography of Mark Twain. 1Paine ABed. New York, NY Harper Brothers1924; http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archives of Dermatology American Medical Association

Evidence-Based Dermatology: Some Problems With Research

Archives of Dermatology , Volume 142 (12) – Dec 1, 2006

Evidence-Based Dermatology: Some Problems With Research

Abstract

Sam Gibbs1 is to be congratulated on his research commentary “Breakthrough in the Treatment of Warts” published in the June issue of the ARCHIVES. The flaws in the original article2 are so significant that one wonders how it survived the peer-review process and made it to print. Surely no patient will ever benefit from the findings. It is poignant that the commentary appears in the same volume of the ARCHIVES as an article assessing the frequency of statistical errors in the...
Loading next page...
 
/lp/american-medical-association/evidence-based-dermatology-some-problems-with-research-gDTT1b5yx2
Publisher
American Medical Association
Copyright
Copyright © 2006 American Medical Association. All Rights Reserved.
ISSN
0003-987X
eISSN
1538-3652
DOI
10.1001/archderm.142.12.1650-a
Publisher site
See Article on Publisher Site

Abstract

Sam Gibbs1 is to be congratulated on his research commentary “Breakthrough in the Treatment of Warts” published in the June issue of the ARCHIVES. The flaws in the original article2 are so significant that one wonders how it survived the peer-review process and made it to print. Surely no patient will ever benefit from the findings. It is poignant that the commentary appears in the same volume of the ARCHIVES as an article assessing the frequency of statistical errors in the dermatological literature.3 I applaud the ARCHIVES for being so open and honest about some of its publication shortcomings. Only by assessing our inadequacies can we look forward, improve the quality of dermatological research, and provide better evidence-based patient care. There are several basic principles of research planning, development, and reporting. Without these, research does not help improve patient care: • Ask an appropriate research question that is structured and answerable. It should be patient driven and relate to a specific patient population and condition. Vague questions might get vague answers or, often, no answers at all. • Use an outcome measure that is clinically relevant to the patient. For example, in alopecia areata, clinically meaningful hair regrowth after treatment might (to the patient) mean 100%, or a full head of hair, or the ability to stop wearing a wig. • Use the appropriate research method and design to answer the question. A double-blind, randomized, placebo-controlled trial is not always needed to answer a clinical question. Some randomized controlled trials are performed and reported so poorly that their findings are meaningless.2 • Present and report data in a clear, open, and honest way so that there is no risk of misunderstanding or misinterpretation by the reader. Important information, such as magnitude of treatment benefit, must always be included so that the reader can assess likely patient benefit. • Analyze the data in the correct way. The old adage (attributed by Mark Twain4 to Benjamin Disraeli) “There are three types of lies: lies, damn lies, and statistics” remains alive as long as inappropriate statistical methods are used (either honestly or dishonestly) to interpret data. • Understand the difference between results that are clinically meaningful and those that are purely statistically meaningful. If the results are reported as “significant,” it would nice to think that they were actually significant and meaningful to our patients. Of course readers should critically appraise journal articles3 but many clinicians do not have the time, resources, or expertise to do this. Perhaps journal editors should include a statistician in the peer-review process, but this might be expensive. Researchers should be encouraged to seek advice from epidemiologists and statisticians before undertaking complicated projects. Readers could study the letters section a few months after an article is published to see what the experts really think. Basic epidemiology and statistics should be incorporated into all dermatology training and continuing medical education programs. The ARCHIVES has made an excellent start by publishing these articles,1,3 throwing down the gauntlet, and opening up some informed debate. Correspondence: Dr Sladden, Launceston General Hospital, Charles Street, Launceston, 7250, Australia. Financial Disclosure: None reported. References 1. Gibbs S Breakthrough in the treatment of warts? Arch Dermatol 2006;142767- 768PubMedGoogle ScholarCrossref 2. Gustafsson LLeijonhufvud IAronsson AMossberg AKSvanborg C Treatment of skin papillomas with topical alpha-lactalbumin-oleic acid. N Engl J Med 2004;3502663- 2672PubMedGoogle ScholarCrossref 3. Neville JALang WFleischer AB Jr Errors in the Archives of Dermatology and the Archives of the American Academy of Dermatology from January through December 2003. Arch Dermatol 2006;142737- 740PubMedGoogle ScholarCrossref 4. Twain M The Autobiography of Mark Twain. 1Paine ABed. New York, NY Harper Brothers1924;

Journal

Archives of DermatologyAmerican Medical Association

Published: Dec 1, 2006

References