| Abstract|| |
The need for understanding and reflecting on evidence-based dermatology (EBD) has never been greater given the exponential growth of new external evidence to inform clinical practice. Like any other branch of medicine, dermatologists need to acquire new skills in constructing answerable questions, efficiently searching electronic bibliographic databases, and critically appraising different types of studies. Secondary summaries of evidence in the form of systematic reviews (SR), that is, reviews that are conducted in a systematic, unbiased and explicit manner, reside at the top of the evidence hierarchy, because they are less prone to bias than traditional expert reviews. In addition to providing summaries of the best external evidence, systematic reviews and randomized controlled trials (RCTs) are also powerful ways of identifying research gaps and ultimately setting the agenda of future clinical research in dermatology. But like any paradigm, EBD can have its limitations. Wrong application, misuse and overuse of EBD can have serious consequences. For example, mindless pooling together of data from dissimilar studies in a meta-analysis may render it a form of reductionism that does not make any sense. Similarly, even highly protocolised study designs such as SRs and RCTs are still susceptible to some degree of dishonesty and bias. Over-reliance on randomized controlled trials (RCT) may be inappropriate, as RCTs are not a good source for picking up rare but important adverse effects such as lupus syndrome with minocycline. A common criticism leveled against SRs is that these frequently conclude that there is lack of sufficient evidence to inform current clinical practice, but arguably, such a perception is grounded more on the interpretation of the SRs than anything else. The apparent absence of evidence should not paralyze the dermatologist to adopt a state of therapeutic nihilism. Poor primary data and an SR based on evidence that is not up-to-date are also limitations that can only improve with better primary studies and updated reviews such as those done by the Cochrane Collaboration. Most dermatologists are interested in integrating the best external evidence with the care of individual patients and have been practicing good EBD without realizing it.
Keywords: Bias, dermatology, evidence-based, systematic review, trials, uncertainty
|How to cite this article:|
Williams HC. Strengths and limitations of evidence-based dermatology. Indian J Dermatol 2014;59:127-33
| Is There Any Point in Discussing Evidence-Based Dermatology?|| |
The idea that doctors should base their treatment decisions on good evidence seems such an obvious notion that it might be taken for granted. Who can really be opposed to the central tenet of evidence-based medicine (EBM) - that of using the best evidence wisely?  It is not as if there is one group of dermatologists "over there" practicing EBD and another group "over here" who choose not to [Figure 1]. If the practice of EBD goes without saying, what is point of discussing and promoting EBD?
|Figure 1: It's not as if there are two distinct groups of dermatologists separated by a brick wall: One evidence - based and the other not. All dermatologists practice evidence - based dermatology to some degree. But like learning how to perform skin surgery, learning the basics of formulating questions, searching, critical appraisal, and interpretations are skills that have to be acquired and practiced|
Click here to view
Medicine is advancing very rapidly, creating major changes in the way dermatologists treat patients. There is a clinical need to keep up to date with the exponential growth of such new external evidence, yet dermatologists frequently fail to keep up if they only rely on an occasional flick through the main dermatology journals or a visit from a pharmaceutical representative. Such a policy leads to a deterioration of knowledge with time. Attempts to overcome this deficiency by attending clinical education programs have been shown to fail to improve our performance, whereas the practice of EBM has been shown to keep its practitioners up to date.  New skills need to be learnt on how to search electronic bibliographic databases efficiently and how to critically appraise the different types of study, as outlined in the "toolbox" section of the book on which this article is based. The problem is that time for such activities is very limited indeed in a busy clinical schedule. Dermatologists cannot be expected to read and appraise every new randomized controlled trial published in dermatology, especially as reliance on a single early positive randomized controlled trial can be hazardous, even when published in a prestigious journal. , Instead, high-quality systematic reviews that have searched for all relevant data are needed to update dermatologists on current best treatments.
It is important to point out that the acquisition of new information technology skills does not mean that they replace the attributes of being a good doctor, such as history taking and examination skills, often described as the "art" of medicine.  Instead, the practice of EBM must be seen as an integration of knowledge management skills with clinical skills. The aim is to create an emergent, knowledgeable, up-to-date, skillful, compassionate, and efficient doctor.
Given that no doctor in their right mind sets out to deliberately practice "non-evidence-based dermatology", it is important to explore the advantages and limitations of promoting the approaches that have been used to date to develop the practice of EBM, which is the focus of this article.
| Advantages of EBD|| |
Being systematic and explicit
Systematic reviews, that is, reviews that have been conducted in a systematic and unbiased way, reside at the top of the evidence hierarchy. Systematic reviews of randomized clinical trials (RCTs) developed after it was realized that traditional reviews about treatments were done in quite arbitrary ways.  Traditional expert reviews may be appropriate for raising issues for discussion and debate, but they are much less suitable for summarizing treatment effects. The unsystematic approach used in most traditional reviews means that they are highly prone to bias and influence from hidden agendas.  Many have written "traditional" review articles in the past containing a biased selection of citations to support predetermined views. Indeed, the author confesses to having used the "file drawer" method to search for papers for a review of atopic eczema in the past [Figure 2]. 
|Figure 2: The good old "file drawer" method for locating studies is still the method used and preferred by some authors of traditional "expert" reviews|
Click here to view
Systematic reviews, such as those produced by the Cochrane Collaboration, summarize accurate, up-to-date, high-quality external evidence of the benefits and harms of interventions for treating and preventing human disease.  Cochrane reviewers use and refer to a prepublished protocol to describe precisely how they will search, appraise, and synthesize data concerning a specific clinical question. Such an explicit structure and methodology means that another researcher could replicate the review if necessary, even down to the level of planned subgroup analysis. Like any good randomized controlled trial, the protocol of a good systematic review will be published before the main review is done in order to prevent the analysis from being driven by data when putting the studies together. For example, it is easy within the context of a systematic review to end up colluding with seven RCTs that have measured a clinically dubious outcome measure by providing a meta-analysis of such data, instead of pointing out that none of the studies had measured what is really important to patients according to a prespecified primary outcome published in a protocol. A recent clinical trial of levamisole measured serum tumor necrosis factor levels as the sole outcome in patients with oral lichen planus rather than symptom control - when did you last see a serum tumor necrosis factor walk into your clinic?  An international database of systematic reviews protocols has been set up called PROSPERO with the aim of a providing a "one stop shop" for prospectively registering all systematic reviews in health and social care http://www.crd.york.ac.uk/Prospero/(accessed September 13 th , 2013).
Keeping up to date and increasing precision
Evidence from cardiovascular medicine has shown that doctors have failed to use effective treatments, such as intravenous streptokinase, for acute myocardial infarction even when there was overwhelming evidence for their effectiveness.  Conversely, they continued to recommend medicines such as intravenous lidocaine for post-infarction arrhythmias long after the evidence suggested that they were ineffective or even harmful.  With over 280 specialist dermatology journals (http://skin.cochrane.org/accessed 13 th September 2013), it has become increasingly difficult for the dermatologist to keep up with the literature.  Systematic reviews such as those supported by the Cochrane Skin Group that track down all possible published and unpublished studies are needed to keep dermatologists and other healthcare professionals up to date.  Cochrane Skin Group reviews have been shown to be of higher quality than non-Cochrane reviews dealing with dermatology,  and they are also updated as new evidence and criticisms become available. Over 200 systematic reviews relevant to skin disease now exist on the Cochrane Library (http://www.thecochranelibrary.com/accessed 13 th September 2013). Given that systematic reviews have become a rapidly increasing publication type in dermatology, some groups such as the Center of Evidence-Based Dermatology at Nottingham have produced freely available maps of systematic reviews (Cochrane and non-Cochrane) relevant to topics such as eczema and acne (http://www.nottingham.ac.uk/research/groups/cebd/resources/index.aspx, accessed 13 th September 2013), so that users do not have to waste time searching for such reviews themselves.
Systematic reviews can also reduce uncertainty and confusion created by the apparently conflicting results of several small inconclusive studies by combining their results-provided they are sufficiently similar. Such a facility may overcome the common tendency to divide all clinical trials into those that are significant at the arbitrary 5% level and those that are not [Figure 3], instead of pooling studies that are sufficiently alike in terms of patients, interventions, and outcomes into summary estimates that indicate a range of plausible treatment effects by means of confidence intervals. Studies that reach the "magic" P < 0.05 significance level are commonly claimed as being "positive" and those that fail to reach that level are often considered "negative," whereas in reality many of the latter trials are far too small to detect even quite large changes. , Instead of concluding that such studies are "conflicting," a meta-analysis performed within the context of a carefully conducted systematic review may show that they are all compatible with a clear overall treatment benefit. The Cochrane Collaboration logo shows a good example of this [Figure 4].
|Figure 3: Research users must overcome the slavish obsession for dividing the results of all clinical trials into those that are statistically significant at the 5% level and those that are not, and instead use confidence intervals to estimate a range of likely effects|
Click here to view
|Figure 4: The Cochrane Collaboration logo depicts a systematic review of seven placebo - controlled trials evaluating the efficacy of a short course of oral corticosteroids for women in premature labor to prevent fetal death. Each horizontal line represents a single RCT; the shorter the line, the more certain are the results. If an RCT touches the vertical line, it means that particular trial found no clear evidence of treatment benefit. The diamond at the bottom represents the combined results, and its position to the left of the vertical line of no treatment difference indicates that the treatment was clearly beneficial in reducing premature infant mortality by 30-50% The first of these RCTs was done in 1972. The figure depicts what would have been revealed had a systematic review of the available evidence been done a decade later. By 1991, another seven RCTs had been done. Because no systematic review of these studies had been produced until 1989, most obstetricians had not realized that the treatment was so effective, but instead interpreted each individual study as "conflicting." As a result, tens of thousands of premature babies around the world have probably died or suffered unnecessarily. This is an example of the human costs of failing to perform an up - to - date systematic review of apparently "conflicting" studies. (Source: the Cochrane Collaboration.)|
Click here to view
Minimizing bias and identifying research gaps
EBD is very much concerned with reducing bias. Systematic reviews are powerful tools for minimizing bias, because they use explicit methods and because they seek to include all relevant studies rather than ones selected on the basis of their results or how easy they are to find. When assessing an intervention for skin disease, a prospective protocol for a systematic review provides an opportunity to state which participants should ideally be studied, which comparators are appropriate, and which outcomes would make a difference clinically and at what time point. This "bottom-up" and non-reductionist approach also provides the opportunity of consulting users (people with a condition or their carers) to ensure that the perspective of the review and outcome measures are appropriate. Specifying analysis plans in protocols also helps minimize the problem of data-driven reviews that amplify features of studies that may interest the pharmaceutical industry more than patients or healthcare professionals. Even for rare skin diseases, producing a systematic review that finds no reliable evidence to inform practice may still be useful, in that it reassures practitioners that they are not missing some important new development. , Systematic reviews also have an important role to play in highlighting possible research gaps for future study, ,, and many funding bodies now will not commit to funding an RCT without a systematic review being done first. It should be noted that, like any other study design, there are good and bad systematic reviews, and even the better ones have room for improvement. 
The basic unit of analysis in most systematic reviews is the RCT. Like any study design, RCTs can also be done badly and used in the wrong situations.  Nevertheless, the RCT remains one of the strongest designs in modern medicine for assessing treatment efficacy, because of its potential to minimize bias.  As Bigby points out in his essay on "Snake oil for the 21 st century," studies of inferior design, such as case series, have many times led to overly optimistic claims of treatment efficacy in dermatology that were not borne out by subsequent RCTs. 
Influencing the agenda of future dermatology trials
Perhaps the most important and subtle advantage of EBD is its propensity to help change the agenda for future clinical trials in dermatology. It is quite clear from a systematic review of 272 RCTs in atopic eczema that most trials have reflected the agenda of the drug industry in order to license new product, and that many questions that are important to clinicians and patients remain unanswered.  Similar conclusions have been voiced in a review of all RCTs conducted in the field of psoriasis by the European Dermato-Epidemiology Network (EDEN).  Many of the outcome measures used in these trials are clinical scoring systems that may show up minor differences in disease scores in the short-term, yet their clinical significance in practice is often obscure, especially in chronic diseases where disease remission and quality of life may be more important. EBD is thus a key instrument in informing the need for comparative effectiveness research using pragmatic clinical trial designs and active comparators when trying to answer questions about which treatment is best.
| Potential Limitations of EBD|| |
The threat of reductionism
Meta-analysis, which refers to the statistical pooling of results drawn from several studies, is prone to dangerous reductionism if used to add studies together which it is not sensible to combine in the first place.  Thus, before contemplating playing with any statistical techniques, it may be wise not to consider combining studies of childhood atopic dermatitis with those dealing exclusively with adult atopic dermatitis, as they may belong to a different disease group in terms of the etiology or treatment responsiveness. It may also not be sensible to combine studies of atopic eczema that evaluate one sort of dietary exclusion with another. It may not make any sense to pool a clinically obscure outcome measure such as a "doctor-assessed itch," simply because it was the only outcome that was common to all trials.  Meta-analyses are only as good as the data from the individual studies that make up such analyses, and great care needs to be taken to avoid adding together things that should not be combined, especially when the statistical output may give rise to a spurious air of precision when read by those less experienced in assessing the quality of such analyses. It is for this reason that meta-analyses should always be performed within the context of a systematic review. Involvement of users is another important aspect that may protect against reductionism-for example, in the choice of outcomes. Another criticism of meta-analyses is that, as more and more data are pooled and aggregated, it becomes more difficult to apply such "average" summary measures back to individual patients-in other words, EBM applies to populations and not to individuals.  This is a somewhat specious argument, given that extrapolation from groups to individuals is a paradox that is fundamental to all scientific studies on groups of people. The paradox of groups and individuals is also applicable to the notion of "clinical experience," which by definition relies on recall of aggregated experiences with groups of patients which is then applied to new individual patients. It can also be argued that combining several studies containing a broad range of people with different ethnic backgrounds, sex, and comorbidities actually increases external validity to a wider and more typical group of patients, rather than relying on a single trial that describes a narrow range of trial participants.
Reductionism can also occur if EBD practitioners become too preoccupied with the critical appraisal of research without considering the many human factors that are central to the successful practice of EBD. 
As with any research methodology, it is possible for those with a vested interest to twist the methods and conclusions of a systematic review to their own advantage.  Thus, one could conveniently fail to include one crucial study that contradicted the results one wanted to show, or if the study is declared within the review, find a weak excuse to exclude it post hoc. Because many trials never see the light of day in terms of publication, or are held as "data on file," it is possible for a review to include or exclude such additional unpublished data in a way that favors a particular product. , As with any written document assessing drug treatments, there is plenty of scope for undue emphasis on positive effects and lack of discussion of relevant adverse events. Readers need to develop a "good nose" for what constitutes a good systematic review and clinical trial such as starting with a peep at the acknowledgments to see who sponsored the review or study in order to assess possible conflicts of interest  -if, in fact, sponsorship has been declared. ,
Overemphasis on RCTs
Whilst RCTs may be the most robust study design for minimizing bias in conventional evaluation of the effectiveness of interventions for skin diseases, they have their limitations. In some circumstances, it may be impossible or unethical to perform an RCT. For example, it is unlikely that mothers will agree to be randomized to breast feeding or bottle-feeding to see whether either prevents atopic eczema. Similarly, it would be impractical to randomize medical students to one form of education and others to another within the same class, because they would not be blinded to the interventions, and there might be considerable "contamination" of the intervention from one group as students talk together. Just because a study is an RCT does not mean that it is a good RCT. Attention to quality and relevance is important, rather than just blindly following the concept of the hierarchy of evidence. Rare but serious events, which are extremely important when evaluating the pros and cons of a new treatment, are not well-characterized in RCTs, but instead require other approaches such as case reports, case-control studies, and wide-scale pharmaceutical surveillance methods. Frequently, there is asymmetry in the way that systematic reviews devote a lot of space to treatment efficacy and less or none to issues such as potentially serious side effects. 
The concept of only using RCTs as the basic building blocks of evidence for systematic reviews has been criticized because it implies that all other evidence that contributes to our understanding of treatment efficacy, such as case series, case reports, and "clinical experience" are not valid.  This is clearly inappropriate. Ideally, the totality of evidence should be considered when conducting a systematic review, so that evidence from observational studies can contribute to the conclusions from RCTs. Approaches such as hierarchical modeling, likelihood estimations, and Bayesian statistics that inform indirect comparisons have been used in attempts to address these gaps. It is likely that the concept of a good informative study design is more of a continuum representing the risk of bias, rather than a dichotomy of "good" (i.e. RCTs) and "bad" (for example, a large case series). Such a continuum needs to be tempered by the added but crucial dimension of study quality.
Another potential drawback of only incorporating RCTs into systematic reviews is that older traditional therapies such as curettage and electrocautery for actinic keratoses, which were rarely assessed to the same RCT standards as modern therapies when first introduced but which have stood the test of time, will often lose out to newer industry-funded and more expensive treatment simply because they fail to pass the required level of evidence.  Lack of evidence of a well-established treatment does not necessarily imply that such treatments are inferior, although it is always desirable to test such treatments as active comparators when new treatments are introduced.
There are challenges, therefore, for future systematic reviewers to find ways of incorporating informative data from nonrandomized studies, and a methodology groups addressing such issues exist within the Cochrane Collaboration. In the meantime, it is best that to walk before running, by adhering to the RCT as the basic study design for assessing treatment efficacy in dermatology, at least until better methodological approaches have evolved.
Reviews always end with the phrase "insufficient evidence"
A common criticism of Cochrane skin reviews by dermatology trainees is that they always end up with the same conclusion-that there is "insufficient evidence" to inform current practice. Whilst this may be true for some reviews, a glance at the reviews in the Cochrane Database of Systematic Reviews shows that at least 40% of those relevant to a practicing dermatologist make specific and clear recommendations for therapy.  Even "null reviews" that do not find any good evidence to make specific treatment recommendations have their uses. Thus, a systematic review evaluating the evidence for antistreptococcal treatments for guttate psoriasis found no reliable evidence, despite confident textbook recommendations in favor of such a treatment approach.  Not only does such a review identify a major gap needing research, but it also reassures doctors and their patients that they are not missing some important study. It empowers doctors to feel more confident in relying on other levels of evidence, such as case series and empirical reasoning based on mechanism, until better studies are done. It also empowers patients with guttate psoriasis, by allowing them to challenge doctors who insist that they must take prolonged courses of antibiotics or who threaten to take out their tonsils [Figure 5].
|Figure 5: One Cochrane systematic review found no good evidence to support the use of antistreptococcal interventions (prolonged antibiotics or tonsillectomy) for treating guttate psoriasis. Sometimes such a "negative" systematic review can be useful, in that it can empower patients to question doctors on the evidential basis for their treatment decisions|
Click here to view
Absence of RCT evidence does not mean that dermatologists should become paralyzed into doing nothing-so-called therapeutic nihilism.  Failure to find any RCTs for the treatment of acne agminata, for instance, does not mean that patients should be told to go away because there is no treatment. It is possible that a well-conducted case series exists, or a convincing case report, or simply the sharing of anecdotal evidence from senior colleagues will result in valuable information. All of forms of evidence are entirely appropriate to use in the absence of better sources. It is just that for many years, the evidence hierarchy has been inverted in everyday dermatology practice-starting with muddling along and experimenting on individual patients on empirical grounds, or asking a colleague, or referring to an out-of-date textbook.
Evidence not up-to-date and poor primary data
Whilst it is the aspiration of the Cochrane Collaboration to update all of its systematic reviews periodically as new primary data emerges, the reality is that such updating activity is incomplete, which is not surprising given than it is largely a voluntary activity. Keeping right up to date remains a challenge and those using systematic reviews often need to undertake additional electronic searches to make sure that no new key studies have emerged since the reviews as a result.
Although the methods of undertaking systematic reviews have become increasingly sophisticated, the raw materials of reliable data is often lacking, and it will take some time before the field of dermatology recognizes the importance of good trial design, prospective trial and systematic review registration, and complete reporting.  Some promising signs have emerged, but it is too little and too late for many research users.  Reducing future research wastage by encouraging clinicians and patients to prioritize research questions, ensuring good study designs that minimize bias, and making sure that all results are published and that they are published fully and honestly and interpreted in the context of other relevant research are all needed urgently if medical and dermatological research is to hold any credibility and trust with the patients who volunteer to participate in such research. ,
| Conclusion|| |
This article has described some of the advantages and potential limitations of EBD and its constituent building blocks of systematic reviews and randomized controlled trials. It is clear that most dermatologists are interested in integrating the best external evidence with the care of individual patients and are thus practicing good EBD without calling it that. EBM is about increasing rather than dictating choice.  The key is to learn more about secondary evidence sources such as systematic reviews and how to distinguish a good one from a bad one, and to change the legacy of small poorly designed and incompletely reported dermatology clinical trials by collaborating on a national and international scale to address the big questions that our patients deserve to know the answer to.
| Acknowledgment|| |
Chapter extract from Williams HC. Where do we go from here? In: Williams HC, Bigby M, Furue M, Herxheimer A, Naldi L, Ran Y, zany B (eds). Evidence-Based Dermatology 3rd edition. Oxford, Wiley Blackwell Publishing and BMJ Publishing Group 2014 (In Press, ISBN 9781118357675).
As per a mutual agreement between Wiley-Blackwell and Wolters Kluwer Medknow, Wiley-Blackwell will retain copyright on Prof Williams' contribution and it will be referenced with the book title and ISBN.
| References|| |
|1.||Sehon SR, Stanley DE. A philosophical analysis of the evidence-based medicine debate. BMC Health Serv Res 2003;3:14. |
|2.||Sackett DL, Richardson WS, Rosenberg Q, Haynes RB. Evidence-based Medicine. How to Practise and Teach EBM. London: Churchill Livingstone; 1997. |
|3.||Ioannides JP. Contradicted and initially stronger effects in highly cited clinical research. JAMA 1995;294:218-28. |
|4.||Williams HC. Two "positive" studies of probiotics for atopic dermatitis: Or are they? Arch Dermatol 2006;142:1201-3. |
|5.||Cook HJ. What stays constant at the heart of medicine. BMJ 2006;333:1281-2. |
|6.||Ladhani S, Williams HC. Management of established postherpetic neuralgia: A comparison of the quality and content of traditional vs. systematic reviews. Br J Dermatol 1998;139:66-72. |
|7.||Williams HC. Atopic eczema - we should look to the environment. BMJ 1995;311:1241-2. |
|8.||Williams HC, Adetugbo K, Po AL, Naldi L, Diepgen T, Murrell D. The cochrane skin group. Preparing, maintaining, and disseminating systematic reviews of clinical interventions in dermatology. Arch Dermatol 1998;134:1620-6. |
|9.||Sun A, Chia JS, Wang JT, Chiang CP. Levamisole can reduce the high serum tumour necrosis factor-alpha level to a normal level in patients with erosive oral lichen planus. Clin Exp Dermatol 2007;32:308-10. |
|10.||Lau J, Antman EM, Jimenez-Silva J, Kupelnick B, Mosteller F, Chalmers TC. Cumulative meta-analysis of therapeutic trials for myocardial infarction. N Engl J Med 1992;327:248-54. |
|11.||Echt DS, Liebson PR, Mitchell LB, Peters RW, Obias-Manno D, Barker AH. et al. Mortality and morbidity in patients receiving encainide, flecainide, or placebo. The Cardiac Arrhythmia Suppression Trial. N Engl J Med 1991;324:781-8. |
|12.||Delamere FM, Williams HC. How can hand searching the dermatological literature benefit people with skin problems? Arch Dermatol 2001;137:332-5. |
|13.||Freeman SR, Williams HC, Dellavalle RP. The increasing importance of systematic reviews in dermatology clinical trial research and publication. J Invest Dermatol 2006;126:2357-60. |
|14.||Collier A, Heilig L, Schilling L, Williams H, Dellavalle RP. Cochrane Skin Group systematic reviews are more methodologically rigorous than other systematic reviews in dermatology. Br J Dermatol 2006;155:1230-5. |
|15.||Williams HC, Seed P. Inadequate size of "negative" clinical trials in dermatology. Br J Dermatol 1993;128:317-26. |
|16.||Williams HC, Dellavalle RP. The growth of clinical trials and systematic reviews in informing dermatological patient Care. J Invest Dermatol 2012;132:1008-17. |
|17.||Xue S, Gu R, Wu T, Zhang M, Wang X. Oral potassium iodide for the treatment of sporotrichosis. Cochrane Database Syst Rev 2009;4:CD006136. |
|18.||Owen CM, Chalmers RJ, O'Sullivan T, Griffiths CE. Antistreptococcal interventions for guttate and chronic plaque psoriasis. Cochrane Database Syst Rev 2000;(2):CD001976. |
|19.||Futamura M, Thomas KS, Grindlay DJ, Doney EJ, Torley D, Williams HC. Mapping systematic reviews on atopic eczema: An essential resource for dermatology professionals and researchers. PLoS One 2013;8:e58484. |
|20.||Griffiths CE, Clark CM, Chalmers RJ, Li Wan Po A, Williams HC. A systematic review of treatments for severe psoriasis. Health Technol Assess 2000;4:1-125. |
|21.||Hoare C, Li Wan Po A, Williams H. Systematic review of treatments for atopic eczema. Health Technol Assess 2000;4:1-191. |
|22.||Jadad AR, Cook DJ, Jones A, Klassen TP, Tugwell P, Moher M, et al. Methodology and reports of systematic reviews and meta-analyses: A comparison of Cochrane reviews with articles published in paper-based journals. JAMA 1998;280:278-80. |
|23.||Rees J. The nature of clinical evidence: Floating currencies rather than gold standards. J Invest Dermatol 2007;127:499-500. |
|24.||Barton S. Which clinical studies provide the best evidence? The best RCT still trumps the best observational study. BMJ 2000;321:255-6. |
|25.||Bigby M. Snake oil for the 21 st century. Arch Dermatol 1998;134:1512-4. |
|26.||Naldi N, Svensson A, Diepgen T, Elsner P, Grob JJ, Coenraads PJ, et al. European Dermato-Epidemiology Network. Randomized clinical trials for psoriasis 1977-2000: The EDEN Survey. J Invest Dermatol 2003;120:738-41. |
|27.||Egger M, Smith GD, Sterne JA. Uses and abuses of meta-analysis. Clin Med 2001;1:478-84. |
|28.||Sweeney KG, MacAuley D, Gray DP. Personal significance: The third dimension. Lancet 1998;351:134-6. |
|29.||Hajjaj FM, Salek MS, Basra MK, Finlay AY. Nonclinical influences, beyond diagnosis and severity, on clinical decision making in dermatology: Understanding the gap between guidelines and practice. Br J Dermatol 2010;163:789-99. |
|30.||Jorgensen AW, Hilden J, Gotzsche PC. Cochrane reviews compared with industry supported meta-analyses and other meta-analyses of the same drugs: Systematic review. BMJ 2006;333:782-5. |
|31.||Gøtzsche PC. European governments should sue Roche and prescribers should boycott its drugs. BMJ 2012;345:e7689. |
|32.||Melander H, Ahlqvist-Rastad J, Meijer G, Beermann B. Evidence b (i) ased medicin - selective reporting from studies sponsored by pharmaceutical industry: Review of studies in new drug applications. BMJ 2003;326:1171-3. |
|33.||Williams HC, Naldi L, Paul C, Vahlquist A, Schroter S, Jobling R. Conflicts of interest in dermatology. Acta Derm Venereol 2006;86:485-97. |
|34.||Lundh A, Krogsbøll LT, Gøtzsche PC. Sponsors' participation in conduct and reporting of industry trials: A descriptive study. Trials 2012;13:146. |
|35.||Perlis CS, Harwood M, Perlis RH. Extent and impact of industry sponsorship conflicts of interest in dermatology research. J Am Acad Dermatol 2005;52:967-71. |
|36.||Ernst E, Pittler MH. Assessment of therapeutic safety in systematic reviews: Literature review. BMJ 2001;323:546. |
|37.||Straus SE, McAlister FA. Evidence-based medicine: A commentary on common criticisms. CMAJ 2000;163:837-41. |
|38.||Gilchrest BA, Martin G. Pitfalls of evidence-based medicine: The example of actinic keratosis therapy. Arch Dermatol 2012;148:528-30. |
|39.||Parker ER, Schilling LM, Diba V, Williams HC, Dellavalle RP. What's the point of databases of reviews for dermatology if all they compile is "insufficient evidence"? J Am Acad Dermatol 2004;50:635-9. |
|40.||Hill D. Efficacy of sunscreens in protection against skin cancer. Lancet 1999;354:699-700. |
|41.||Panda S, Williams HC. Consorting with CONSORT 2010. Indian J Dermatol 2013;58:2-5. |
|42.||Heneghan C, Godlee F. Where next for evidence based healthcare? BMJ 2013;346:f766. |
|43.||Chalmers I, Glasziou P. Avoidable waste in the production and reporting of research evidence. Lancet 2009;374:86-9. |
|44.||Eleftheriadou V, Whitton ME, Gawkrodger DJ, Batchelor J, Corne J, Lamb B, et al. Vitiligo priority setting partnership. Future research into the treatment of vitiligo: Where should our priorities lie? Results of the vitiligo priority setting partnership. Br J Dermatol 2011;164:530-6. |
|45.||Dickersin K, Straus SE, Bero LA. Evidence based medicine: Increasing, not dictating, choice. BMJ 2007;334:s10. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]