Question
Asked 27th Mar, 2014

What proportion of medicine is evidence-based?

There is a high bar for introducing a new intervention (social/behavioral, drug, device, etc). However, I suspect that many interventions, treatments, and decisions in medicine do not have a high level of evidence base (i.e. no formal trials, decisions are based on experience, practice, etc). Have there been studies that estimate what proportion of medicine is evidence-based?

Most recent answer

17th Jul, 2014
Karl West
Academy of Lymphology llc
As good as evidence is -- to use it as a paradigm is wrong. Evidence is merely the foundation upon which the process and principles may be relied upon.
1 Recommendation

Popular answers (1)

30th Mar, 2014
Constantine Kaniklidis
No Surrender Breast Cancer Foundation (NSBCF)
My own review suggests there are large areas of clinical practice that are in poor compliance with evidence-based guidance, and that even in cases of compliance, the underlying clinical practice guidelines often exhibit low methodological quality, thus robbing even wider swatches of medical practice with a much needed strong evidence-based foundation.
COMPROMISED COMPLIANCE WITH EVIDENCE-BASED GUIDELINES
In cardiology, a critical review of evidence-based guidelines in hypertension [1] found that despite the widespread availability of evidence-based guidelines for treating hypertension, recent evidence suggests that physicians may not be prescribing first-line drugs for their patients with high blood pressure, and for patients with high blood pressure alone, only 38 percent being on a diuretic, and less than a third prescribed a beta-blockert (he JNC VI recommended first-line antihypertensives for essential hypertension), while approximately half of individuals with high blood pressure and certain comorbidities received non-first-line interventions.
It's been independently shown that GPs are in poor compliance with evidence-based hypertension guidelines and are undertreating hypertension [2] and, despite being aware of the risks of hypertension in the elderly and the benefits of its treatment, with fewer than half complying with the broad recommendations of even the most conservative evidence-based guidelines [3,4], a problem widespread among GPs in the UK and elsewhere [5].
In OBGYN, another UK review [6] examined the practice of induction of labour (IOL) to determine whether induction was performed as per the Royal College of Obstetricians and Gynaecologists/National Institute for Clinical Excellence (RCOG/NICE) guidelines, finding only 60 – 70% compliance with guidelines.
If we turn to the domain of emergency medicine, in particular Emergency Oxygen Guidelines in emergency departments, another British study [7] found that as many as 46% of patients were inappropriately receiving excess oxygen, and as many as 40% were inappropriately not receiving oxygen, so it is clear that uptake of authoritative evidence-based guidelines has been poor as in the inpatient setting.
Another recent (2012) retrospective analysis of compliance with evidence based protocols in cases admitted to the ICU [8] found that in 45% of the severe pre‐eclampsia patients and in 46% of sepsis cases, the guidelines were not followed and there was exceedingly poor adherence - a mere 10.8% - to guidelines of massive hemorrhage cases.
Within trauma medicine, another recent (2014) review of compliance with evidence-based guidelines in patients with traumatic brain injuries [9], it was found that the overall compliance rate was 73%, with only 3 out of 11 Level I trauma centers achieving a compliance rate exceeding 80%, despite the fact that multivariate analysis showed that increased adoption of EBM was associated with a reduced mortality rate. This cross-confirms the previous observation [10] of patients admitted to Level I trauma center (2006–2008) with moderate to severe injuries, where little over half of evidence-based recommended care was delivered to trauma patients with moderate to severe injuries, just 17% for neurosurgical interventions, and alarmingly, with those patients with increasing severity of traumatic brain injuries being the least likely to receive optimal evidence-based care.
In oncology, another recent (2014) review of melanoma treatment in Australia and New Zealand [11] evaluated both the extent of evidence-based support for clinical practice guideline recommendations concerning cutaneous melanoma follow up and the methodological quality of these guidelines; the review found that melanoma follow-up recommendations concerning frequency of physical examinations, duration of follow-up appointments and use of imaging or diagnostic tests are based mostly on low-level evidence or consensus expert opinion; in addition, recommendations were often inconsistent between different guidelines, and to this day, there is no international evidence-based consensus regarding what constitutes best practice for follow up of melanoma survivors.
CONCLUSIONS - AND ANOTER PROBLEM: COMPROMISED GUIDELINES
These and over a hundred other studies I reviewed across the spectrum of medical specialties collectively and strongly suggest that large proportions of medical clinical practice demonstrate low to at most modest compliance with evidence-based recommendations and practice guidelines. Aggravating this problem is the related - and underlying - problem, of clinical practice being informed by putative evidence-based clinical practice guidelines that themselves are of significantly compromised methodological quality. Thus, a Canadian study [12,13] evaluated the strength of the evidence underlying therapy recommendations (n=338) in evidence-based clinical practice guidelines in three domains (diabetes, dyslipidemia, and hypertension), finding that overall, less than one-third of treatment recommendations (and less than half of those citing RCTs in support of the advocated treatment) were based on high-quality evidence.
And although evidence-based medicine regards RCTs as the strongest form of evidence for clinical decision making, a recent intensive review [14] found that overall, at least 20.2% of all published medical research has significant methodological flaws, and perhaps alarmingly, prospective studies appear to have as many methodological limitations as nonprospective studies, and RCTs have as many limitations as non-RCTs, with as much as 38.7% of published RCTs receiving detailed review found to have methodological issues.
Similarly, among 100 orthopedic articles analyzed for compliance with CONSORT and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines [15], 96% failed to report trial registration, 63% failed to discuss study limitations, 59% failed to report the source of funding, 48% failed to report details of the research setting, 38% failed to report adverse events and unintended effects, 37% inadequately reported the number of patients enrolled, and 23% reported no measurement of error for the primary outcome.
And again: up to 30% of articles in the thoracic surgery literature have been reported to have limitations, with the most frequently cited problem being inappropriate use of statistical tests or violation of principles of research design, and the experimental design of many surgical investigations often do not adequately account for bias or type II error [16]. Statistical tests were used or reported incorrectly in 27% of articles evaluated in the surgical literature and were not used at all in an additional 10% of articles [17].
The data I have presented above, which can be extended by dozens and dozens of additional studies, collectively suggests that substantial portions of clinical practice have decidedly questionable foundations in robust evidentiary medicine, with low compliance and low study methodology aggravating the evidence-based foundations of modern medicine. We need to do better as a profession.
REFERENCES
1. Holmes JS, Shevrin M, Goldman B, Share D. Translating research into practice: are physicians following evidence-based guidelines in the treatment of hypertension? Med Care Res Rev 2004; 61(4):453-73.
2. Cranney M, Barton S, Walley T. The management of hypertension in the elderly by general practitioners in Merseyside: the rule of halves revisited. Br J Gen Pract 1998; 48: 1146–1150.
3. Cranney M, Warren E, Walley T. Hypertension in the elderly: attitudes of British patients and general practitioners. J Hum Hypertens 1998; 12: 539–545.
4. Sever P, Beevers G, Bulpitt CJ. Management guidelines in essential hypertension: report of the second working party of the British Hypertension Society. Br Med J 1993; 306: 983–987.
5. Cranney M, Warren E, Barton W, Gardner K, Walley T. Why do GPs not implement evidence-based guidelines? A descriptive study. Family Practice (2001) 18 (4): 359-363.
6. Nooh A, Baghdadi S, Raouf S. Induction of labour: how close to the evidence-based guidelines are we? J Obstet Gynaecol 2005; 25(5):451-4.
7. Wallace SM, Doy LE, Kedgley EN, Ricketts WM. Evidence-based emergency oxygen guidelines are not being followed in the emergency department. Thorax 2010;65:A114-A115.
8. Kondov K, Sharpe P, Mousa, H. Are we following the guidelines ‐ a retrospective analysis of compliance with evidence based protocols in cases admitted to Intensive Care Unit (ICU) from Labour Ward (LW): 1AP2‐2. Eur J Anaesthesiol 2012; 29:9.
9. Shafi S, Barnes SA, Millar D, et al. Suboptimal compliance with evidence-based guidelines in patients with traumatic brain injuries. J Neurosurg 2014; 120(3):773-7.
10. Shafi S, Barnes SA, Millar D, et al. Suboptimal compliance with evidence-based guidelines in patients with traumatic brain injuries. J Neurosurg 2014; 120(3):773-7.
11. Marciano NJ, Merlin TL, Bessen T, Street JM. To what extent are current guidelines for cutaneous melanoma follow up based on scientific evidence? Int J Clin Pract 2014 Feb 18.
12. van Diepen SF, McAlister FA, Padwal R, Johnson JA, Majumdar SR. How Evidence-Based Are Evidence-Based Guidelines. 60th Annual Meeting of the Canadian Cardiovascular Society and CCCN Annual Meeting and Scientific Sessions. October 2007, Volume 23, Supplement SC. Can J Cardiol. Abstract 0053.
13. McAlister FA, van Diepen S, Padwal RS, Johnson JA, Majumdar SR. How evidence-based are the recommendations in evidence-based guidelines? PLoS Med 2007; 4(8):e250.
14. Steen RG, Dager SR. Evaluating the evidence for evidence-based medicine: are randomized clinical trials less flawed than other forms of peer-reviewed medical research? FASEB J 2013; 27(9):3430-6.
15. Parsons, N. R., Hiskens, R., Price, C. L., Achten, J., and Costa, M. L. (2011) A systematic survey of the quality of research reporting in general orthopaedic journals. J. Bone Joint Surg. Br. 93, 1154–1159.
16. Ferraris, V. A., and Ferraris, S. P. (2003) Assessing the medical literature: let the buyer beware. Ann. Thorac. Surg. 76, 4–11.
17. Kurichi, J. E., and Sonnad, S. S. (2006) Statistical methods in the surgical literature. J. Am. Coll. Surg. 202, 476–484.
23 Recommendations

All Answers (135)

27th Mar, 2014
Nelson Elias
Vila Velha Hospital
Dear Rachel
This paper may be useful for you
Arthroscopy. 2014 Mar;30(3):362-371.
Current Status of Evidence-Based Sports Medicine.
Harris JD, Cvetanovich G, Erickson BJ, Abrams GD, Chahal J, Gupta AK, McCormick FM2 Bach BR
1 Recommendation
The papers below show that for paediatric surgery, general practice and dermatology studies shows considerable shortcomings in interventions being based on solid evidence.
Abeni D, Girardelli CR, Masini C, Aprea R, Melchi CF, Puddu P, Pasquini P. What proportion of dermatological patients receive evidence-based treatment? Archives of Dermatology 2001; 137(6): 771-776
Baraldini V, Spitz L, Pierro A. Evidence-based operations in paediatric surgery. Pediatric Surgery International 1998; 13: 331-335
Ellis J, Mulligan I, Rowe J, Sackett DL. Inpatient general medicine is
evidence based. Lancet 1995;346: 407-410.
2 Recommendations
28th Mar, 2014
Muhammad Imran Ahmad
University of Adelaide
Thanks for asking this question and for the answer by Pieter.
28th Mar, 2014
Eivind Aakhus
Innlandet Hospital Trust
It depends on what you mean by 'evidence based'. Do you mean practice that adhere strictly to scientific evidence or do you mean practice that adhere to clinical practice guidelines? As you may be aware of, clinical practice guidelines, although aiming to provide evidence based recommendations to health professionals, give recommendations that are based on clinical expertise or good clinical practice when evidence is lacking. In psychiatry there are substantial evidence that practitioners follow clinical guidelines to a limited extent only. The following references might get you started within the field of psychiatry
Rhodes L, Genders R, Owen R, O'Hanlon K, Brown JSL. Investigating barriers to implementation of the NICE Guidelines for Depression: a staff survey with Community Mental Health Teams. Journal of Psychiatric and Mental Health Nursing 2010 Mar 1;17(2):147-51.
Smolders M, Laurant M, Verhaak P, Prins M, van Marwijk H, Penninx B, et al. Adherence to evidence-based guidelines for depression and anxiety disorders is associated with recording of the diagnosis. Gen Hosp Psychiatry. 2009 Sep-Oct;31(5):460-9.
1 Recommendation
28th Mar, 2014
Muhammad Imran Ahmad
University of Adelaide
I don't think that clinical practice guidelines can be considered as evidence at all.
1 Recommendation
28th Mar, 2014
Nicolas Gaston Moreno
Asociación de Anestesia, Analgesia y Reanimación de Buenos Aires
Dear Rachel,
Your inquire got me thinking. I would like to give a little twist at your question and ask you and everyone else a few new questions: Should we aim for a 100% evidence-based medicine? Is evidence-based medicine the cornerstone of medical practice? As you said many of the things we do everyday in clinical practice are not fully supported by the evidence. Does that mean that we shouldn't give antibiotics to treat a bacterial pneumonia? There are no randomized controlled trials nor meta-analysis for many of the interventions we put in practice everyday (and that we know for sure that they work just fine, based on the history of medicine or our colleagues' experience or even our own). Should we stop using them?
On the other hand, many of the practices that used to be done with the support of the evidence have turned to be harmful or not that effective after the years have passed (because of flaws in trials design, dishonest authors or just because there were missed considerations at the moment that the studies have taken place).
I am not saying that we shouldn't aim for a more 'science-based' practice, with targeted interventions and measurable outcomes, with solid evidence of their results. I am just saying that, sometimes, it is a little tricky trying to base your whole practice on the newest, most solid and sometimes 'cool' evidence and discarding some interventions just because they are not (and will never be) evidence-based.
2 Recommendations
29th Mar, 2014
Arthur Leibovitz
Shmuel Harofe Hospital , Geriatric Medical Centre 1984 - 2011
Medicine evolved as Experience Based and is now expected to develop as
Evidence Based but it will always remain PROVIDENCE BASED ....
2 Recommendations
29th Mar, 2014
Susanne Bernhardsson
Region Västra Götaland, Göteborg, Sweden
Muhammad - I agree with you that CPGs are not evidence in themselves, but if they are developed to high standards they should be based on evidence (EBCPGs). Historically, guidelines were developed by experts and based on their experience (GOBSAT = Good Old Boys Sat Around the Table :)). These days, when the evidence base is growing rapidly in most fields, guidelines are indeed primarily evidence-based and only influenced minimally by the guideline development group. So while EBCPGS aren't evidence in themselves they summarize and provide evidence for clinicians, facilitating that their practice can be evidence-based. Some versions of the evidence hierarchy even put guidelines above RCTs, systematic reviews and meta-analyses, because they, ideally, also incorporate clinical expertise and patient preferences, important parts of EBM. And, importantly, guidelines constitute pre-appraised evidence, further facilitating for clinicians to use/apply the evidence in their practice.
Rachel - I don't think the study you're looking for exists. It seems it would be very difficult to measure what proportion of practice is evidence-based, in such a wide perspective such as what you ask for. There are a number of challenges to measure this. One would hope though, that after 20+ years of EBM, the answer to your question would be 100%. Since the definition of EBM includes also clinical expertise and patients’ values and preferences, it would be difficult for anyone not to base their practice on evidence! I think “evidence-based” these days is more of a cultural approach to medicine and health care and integrated into most clinicians’ thinking and practice, and therefore not easily measurable. But that’s of course my opinion, and not evidence :).
4 Recommendations
29th Mar, 2014
Thomas Karl Hillecke
SRH Hochschule Heidelberg
Interesting question.
I think that the main problem in EBM is a wrong unterstanding of the power of science. Science is powerful if it is used to falsificate hypotheses. But in EBM the positivistic perspective of inductive proof dominates. Therefor a lot of studies in therapy research tend to deliver positive results (see Fanelli, positive results increase ...). Looking deeper into concrete research studies and especially into meta analyses the results of studies and the strength of so called evidences are often overinterpreted. In consequence I think that EBM changed from a nice and motivated approach to a hypocritical control system of health care systems. One symptom may be that the expertise seem to play not anymore the same role as in the beginning of the EBM discussion (see the work of Sackett), and the role of external evidences is meanwhile exaggerated. In my opinion we should enhance efforts to falsificate interventions to come closer to the goal of better practice? But perhaps my proposal here is too much influenced by critical rationalism (Popper).
Regards Thomas
3 Recommendations
29th Mar, 2014
Ru-Jeng Teng
Medical College of Wisconsin
To me there is just a small portion is evidence-based. Still recall one of my teacher told me that there is no evidence that my patient management is right so his way was the right one. I could not fight back even though I know at that time there was no evidence that his way is the right one either.
1 Recommendation
30th Mar, 2014
Constantine Kaniklidis
No Surrender Breast Cancer Foundation (NSBCF)
My own review suggests there are large areas of clinical practice that are in poor compliance with evidence-based guidance, and that even in cases of compliance, the underlying clinical practice guidelines often exhibit low methodological quality, thus robbing even wider swatches of medical practice with a much needed strong evidence-based foundation.
COMPROMISED COMPLIANCE WITH EVIDENCE-BASED GUIDELINES
In cardiology, a critical review of evidence-based guidelines in hypertension [1] found that despite the widespread availability of evidence-based guidelines for treating hypertension, recent evidence suggests that physicians may not be prescribing first-line drugs for their patients with high blood pressure, and for patients with high blood pressure alone, only 38 percent being on a diuretic, and less than a third prescribed a beta-blockert (he JNC VI recommended first-line antihypertensives for essential hypertension), while approximately half of individuals with high blood pressure and certain comorbidities received non-first-line interventions.
It's been independently shown that GPs are in poor compliance with evidence-based hypertension guidelines and are undertreating hypertension [2] and, despite being aware of the risks of hypertension in the elderly and the benefits of its treatment, with fewer than half complying with the broad recommendations of even the most conservative evidence-based guidelines [3,4], a problem widespread among GPs in the UK and elsewhere [5].
In OBGYN, another UK review [6] examined the practice of induction of labour (IOL) to determine whether induction was performed as per the Royal College of Obstetricians and Gynaecologists/National Institute for Clinical Excellence (RCOG/NICE) guidelines, finding only 60 – 70% compliance with guidelines.
If we turn to the domain of emergency medicine, in particular Emergency Oxygen Guidelines in emergency departments, another British study [7] found that as many as 46% of patients were inappropriately receiving excess oxygen, and as many as 40% were inappropriately not receiving oxygen, so it is clear that uptake of authoritative evidence-based guidelines has been poor as in the inpatient setting.
Another recent (2012) retrospective analysis of compliance with evidence based protocols in cases admitted to the ICU [8] found that in 45% of the severe pre‐eclampsia patients and in 46% of sepsis cases, the guidelines were not followed and there was exceedingly poor adherence - a mere 10.8% - to guidelines of massive hemorrhage cases.
Within trauma medicine, another recent (2014) review of compliance with evidence-based guidelines in patients with traumatic brain injuries [9], it was found that the overall compliance rate was 73%, with only 3 out of 11 Level I trauma centers achieving a compliance rate exceeding 80%, despite the fact that multivariate analysis showed that increased adoption of EBM was associated with a reduced mortality rate. This cross-confirms the previous observation [10] of patients admitted to Level I trauma center (2006–2008) with moderate to severe injuries, where little over half of evidence-based recommended care was delivered to trauma patients with moderate to severe injuries, just 17% for neurosurgical interventions, and alarmingly, with those patients with increasing severity of traumatic brain injuries being the least likely to receive optimal evidence-based care.
In oncology, another recent (2014) review of melanoma treatment in Australia and New Zealand [11] evaluated both the extent of evidence-based support for clinical practice guideline recommendations concerning cutaneous melanoma follow up and the methodological quality of these guidelines; the review found that melanoma follow-up recommendations concerning frequency of physical examinations, duration of follow-up appointments and use of imaging or diagnostic tests are based mostly on low-level evidence or consensus expert opinion; in addition, recommendations were often inconsistent between different guidelines, and to this day, there is no international evidence-based consensus regarding what constitutes best practice for follow up of melanoma survivors.
CONCLUSIONS - AND ANOTER PROBLEM: COMPROMISED GUIDELINES
These and over a hundred other studies I reviewed across the spectrum of medical specialties collectively and strongly suggest that large proportions of medical clinical practice demonstrate low to at most modest compliance with evidence-based recommendations and practice guidelines. Aggravating this problem is the related - and underlying - problem, of clinical practice being informed by putative evidence-based clinical practice guidelines that themselves are of significantly compromised methodological quality. Thus, a Canadian study [12,13] evaluated the strength of the evidence underlying therapy recommendations (n=338) in evidence-based clinical practice guidelines in three domains (diabetes, dyslipidemia, and hypertension), finding that overall, less than one-third of treatment recommendations (and less than half of those citing RCTs in support of the advocated treatment) were based on high-quality evidence.
And although evidence-based medicine regards RCTs as the strongest form of evidence for clinical decision making, a recent intensive review [14] found that overall, at least 20.2% of all published medical research has significant methodological flaws, and perhaps alarmingly, prospective studies appear to have as many methodological limitations as nonprospective studies, and RCTs have as many limitations as non-RCTs, with as much as 38.7% of published RCTs receiving detailed review found to have methodological issues.
Similarly, among 100 orthopedic articles analyzed for compliance with CONSORT and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines [15], 96% failed to report trial registration, 63% failed to discuss study limitations, 59% failed to report the source of funding, 48% failed to report details of the research setting, 38% failed to report adverse events and unintended effects, 37% inadequately reported the number of patients enrolled, and 23% reported no measurement of error for the primary outcome.
And again: up to 30% of articles in the thoracic surgery literature have been reported to have limitations, with the most frequently cited problem being inappropriate use of statistical tests or violation of principles of research design, and the experimental design of many surgical investigations often do not adequately account for bias or type II error [16]. Statistical tests were used or reported incorrectly in 27% of articles evaluated in the surgical literature and were not used at all in an additional 10% of articles [17].
The data I have presented above, which can be extended by dozens and dozens of additional studies, collectively suggests that substantial portions of clinical practice have decidedly questionable foundations in robust evidentiary medicine, with low compliance and low study methodology aggravating the evidence-based foundations of modern medicine. We need to do better as a profession.
REFERENCES
1. Holmes JS, Shevrin M, Goldman B, Share D. Translating research into practice: are physicians following evidence-based guidelines in the treatment of hypertension? Med Care Res Rev 2004; 61(4):453-73.
2. Cranney M, Barton S, Walley T. The management of hypertension in the elderly by general practitioners in Merseyside: the rule of halves revisited. Br J Gen Pract 1998; 48: 1146–1150.
3. Cranney M, Warren E, Walley T. Hypertension in the elderly: attitudes of British patients and general practitioners. J Hum Hypertens 1998; 12: 539–545.
4. Sever P, Beevers G, Bulpitt CJ. Management guidelines in essential hypertension: report of the second working party of the British Hypertension Society. Br Med J 1993; 306: 983–987.
5. Cranney M, Warren E, Barton W, Gardner K, Walley T. Why do GPs not implement evidence-based guidelines? A descriptive study. Family Practice (2001) 18 (4): 359-363.
6. Nooh A, Baghdadi S, Raouf S. Induction of labour: how close to the evidence-based guidelines are we? J Obstet Gynaecol 2005; 25(5):451-4.
7. Wallace SM, Doy LE, Kedgley EN, Ricketts WM. Evidence-based emergency oxygen guidelines are not being followed in the emergency department. Thorax 2010;65:A114-A115.
8. Kondov K, Sharpe P, Mousa, H. Are we following the guidelines ‐ a retrospective analysis of compliance with evidence based protocols in cases admitted to Intensive Care Unit (ICU) from Labour Ward (LW): 1AP2‐2. Eur J Anaesthesiol 2012; 29:9.
9. Shafi S, Barnes SA, Millar D, et al. Suboptimal compliance with evidence-based guidelines in patients with traumatic brain injuries. J Neurosurg 2014; 120(3):773-7.
10. Shafi S, Barnes SA, Millar D, et al. Suboptimal compliance with evidence-based guidelines in patients with traumatic brain injuries. J Neurosurg 2014; 120(3):773-7.
11. Marciano NJ, Merlin TL, Bessen T, Street JM. To what extent are current guidelines for cutaneous melanoma follow up based on scientific evidence? Int J Clin Pract 2014 Feb 18.
12. van Diepen SF, McAlister FA, Padwal R, Johnson JA, Majumdar SR. How Evidence-Based Are Evidence-Based Guidelines. 60th Annual Meeting of the Canadian Cardiovascular Society and CCCN Annual Meeting and Scientific Sessions. October 2007, Volume 23, Supplement SC. Can J Cardiol. Abstract 0053.
13. McAlister FA, van Diepen S, Padwal RS, Johnson JA, Majumdar SR. How evidence-based are the recommendations in evidence-based guidelines? PLoS Med 2007; 4(8):e250.
14. Steen RG, Dager SR. Evaluating the evidence for evidence-based medicine: are randomized clinical trials less flawed than other forms of peer-reviewed medical research? FASEB J 2013; 27(9):3430-6.
15. Parsons, N. R., Hiskens, R., Price, C. L., Achten, J., and Costa, M. L. (2011) A systematic survey of the quality of research reporting in general orthopaedic journals. J. Bone Joint Surg. Br. 93, 1154–1159.
16. Ferraris, V. A., and Ferraris, S. P. (2003) Assessing the medical literature: let the buyer beware. Ann. Thorac. Surg. 76, 4–11.
17. Kurichi, J. E., and Sonnad, S. S. (2006) Statistical methods in the surgical literature. J. Am. Coll. Surg. 202, 476–484.
23 Recommendations
31st Mar, 2014
Eivind Aakhus
Innlandet Hospital Trust
Thanks to all for contributing to this thread. I still think that the discussion needs to clarify what EBM is about, in order to answer Rachel's initial question. To my knowledge EBM initially was concerned about how to translate scientific evidence into clinical practice, acknowledging the huge performance gap between these fields and to elaborate tools to address this gap. The understanding of EBM has evolved during the last 10-15 years though and add other factors to the concept, such as clinical expertise and the patient's values and preferences, which is not necessarily evidence-based in the traditional understanding of the concept. Thus, merely measuring whether clinicians follow the conclusions from the latest systematic reviews or meta-analysis is not sufficient to qualify for EBM practice according to current understanding.
High quality clinical practice guidelines, with all their limitations and weaknesses, still is probably the best source for clinicians to update their practice in accordance with EBM. Thus, measuiring adherence to CPGs is a reasonable way to address Rachel's question: What proportion of medicine is evidence-based? For developers of CPGs, the primary challenge is to present the recommendations with sufficient clarity and unambiguity, in order to make clinicians understand the recommendations in accordance with the intention. Gordon Guyatt and colleagues have published a neat handbook, "Users' guides to the medical literature" which I recommend.
2 Recommendations
31st Mar, 2014
Susanne Bernhardsson
Region Västra Götaland, Göteborg, Sweden
Thanks, Eivind, for clarifying the evolution of the EBM concept. It seems that we agree that high quality EBCPGs are at a higher level in the evidence hiearchy, because they incorporate clinical expertise and patient values and preferences, and also because they are pre-appraised and presented in a digested format for the clinician, which most likely increases adherence. I agree that measuring guideline adherence may be an appropriate way of measuring EBP. However, many challenges remain in the measuring of guideline adherence also. Many studies only measure self-reported adherence, or use, of guidelines, which makes for a high risk of social desirability bias. There is also the issue of definitions of various kinds of adherence, or use, such as implicit vs explicit, instrumental vs conceptual. That is: Is it considered use of a guideline if you have read it and incorporated it into your clinical decision making? Probably implicit but maybe not explicit use?
I can add to the excellent summary above, that in my field, physiotherapy, guideline adherence in various countries and contexts has been measured to between 40 and 90%. Too much variation, and yes, we can do better! But the measuring difficulties I mentioned make comparisons difficult ...
2 Recommendations
31st Mar, 2014
Mark Bell
Leeds Beckett University
And what proportion is poor biased research?
1 Recommendation
31st Mar, 2014
Alan V. Schmukler
Hpathy.com
The idea of Evidenced Based Medicine suggests that we can achieve some scientific perfection and that we can control outcomes in medicine...and in life! My four years working in a hospital including considerable time in ICU, PCU and EU helped me understand that human nature will trump science every time. In the end it's imperfect, ego laden, emotional, humans who carry out in clinical practice, the final stage of all that research. What would you think of doctors in the 21st century handling a patient's wounds with unwashed hands? Or wrong medications being prescribed because of failure to read the chart? Or of nurses changing dials on ventilators without undestanding what they're doing? Each year 100,000 die from medical errors, another 100,000 from hospital acquired infections and another 100,000 from side effects of prescription drugs. (That last statistic is due to the fact that drug research is carried out entirey by the same companies that will profit from sale of the drugs. That research is totally compromised.)
Healing, like life, is complex and involves an interaction of many variables. We nobly try to make it better and control outcomes, but that's a goal which is rarely achieved. One of the best things we can do as scientists and healers, is come to the playing field with good intentions.
1 Recommendation
31st Mar, 2014
Vinicius Pedrazzi
University of São Paulo
Dear Rachel, I think that unfortunately due the run for production there is a lot of papers sprayed all around the world without evidence-based health. In all fields of health. So you can try Cochrane Collaboration, Central, Cochrane Library and you will find the best systematic reviews with RCT and evidence-based research. I wish u good luck! And the same advise I give to all my students: KEEP YOUR SOUL, IN A TIME, SOMEONE WILL TRY HARD TO STOLE IT FROM YOU.
31st Mar, 2014
Rachel E Patzer
Emory University
Thanks all for this excellent discussion.
Per Nicolas's comment above, that "There are no randomized controlled trials nor meta-analysis for many of the interventions we put in practice everyday (and that we know for sure that they work just fine, based on the history of medicine or our colleagues' experience or even our own). Should we stop using them?" This is exactly the point I am trying to make. Absolutely we should strive for evidence-based medicine in practice as much as we possibly can, but I agree that not all aspects of clinical work can be done according to guidelines or protocols. And, subjecting current protocols that are NOT supported by RCTs, etc may be be inefficient, expensive, and a disincentive for innovation.
2 Recommendations
1st Apr, 2014
Lynne Staff
University of Tasmania
I agree with you mark. We really have to examine what 'evidence' the evidence is based on before we can make clinically sound decisions, that have to be made with patient choice and context as a primary consideration in informed decision making. Some of the evidence makes for a very shaky base indeed. I also agree with the comment that it's intent has changed from best practice to control and malpractice management.
1st Apr, 2014
Lynne Staff
University of Tasmania
PS we should be using the term 'evidence informed practice' to truly reflect that practice involves more than the practitioner.
2 Recommendations
1st Apr, 2014
Anne Mullin
National Health Service
There are usually sound clinical reasons for patients in general practice who are not being treated according to hypertension guidelines (for example-there are others e/g heart failure/AF )-this is built into the GP QOF contract and we have to give reasons as part of the audit trail for chronic disease management as to why certain patients are not prescribed certain meds. This is compounded by often having frail elderly patients prescribed multiple medications (polypharmacy) -there are real issues of compliance/safety etc.
Evidence base itself is a slightly problematic term in that often the evidence base that is driving policy is itself incomplete and excludes in the trial data many patients that I see in my patient population.
1 Recommendation
1st Apr, 2014
Livia Puljak
Catholic University of Croatia
When you start looking into gold standard for systematic reviews, i.e. those produced by The Cochrane Collaboration, you get true picture about the lack of high-quality evidence for many interventions. Primary studies are absent or there are few of them, or too short (studies on chronic diseases that last few weeks or few months), or with poor methodology and high risk of bias. By improving quality of primary studies, and studying what really matters to patients, we can improve our evidence base.
5 Recommendations
1st Apr, 2014
Lynne Staff
University of Tasmania
Yes Anne and Livia. My point exactly about the amount (or lack of) and quality of evidence on which practice is based. Well constructed studies with power take time and money, and many RCTs will never be able to be conducted because of ethics, human rights and consent issues. I see a place for the preference trial in situations like this - however I am a budding researcher, only and very interested in this thread.
2nd Apr, 2014
Rejina Kamrul
University of Saskatchewan
I agree with Lynne regarding using the term of evidence informed practice. I also agree with the Rachel that we should strive for that practice as much we possibly could specially in the situation where there is high grade evidence, recommendation is present but not all aspect of clinical practice could be done according to guideline.
2nd Apr, 2014
Paul Silverston
Anglia Ruskin University
Good question, Rachel! However, before one can answer that question one has to ask how much of evidence-based medicine is truly evidence-based and whether you can truly trust that it is? The concept of there being an evidence-base to medicine is relatively new, historically-speaking. Empiricism and experience led to clinicians adopting practices that they would probably have called evidence-based, too: Cupping & leeching and Typhoid Mary would probably have been written-up from a good-looking clinical trial , particularly if they had been sponsored by a therapeutics company and I guess we shouldn't forget that a generation (or two) of children lost their tonsils, or had grommets inserted in more recent times, based upon what was considered to be good evidence. So, we need to take a healthy scepticism of what the quality of the evidence base is. We should not be so arrogant as to believe that we are immune from the same mistakes, as history is full of such examples.
Even when we think that we have good evidence, we still need to be aware of what we are not being told in the studies. (I am not a conspiracy-theorist but someone who has been in involved in assessing the quality of evidence-based material.) We have had scandals related to the withholding of adverse results by drug companies and we know that meta-analysis studies can depend upon which studies are included within the analysis and which are not. The political and financial influences on the health care systems in which the studies are performed also have some influence on the inclusivity and interpretation of the results. I am currently evaluating the evidence-base for the treatment of a common condition for a well-respected journal and we have found that the treatments used on each continent are totally different. It is hard not to conclude that the biases involved in the presentation of evidence are considerable.
Even in didactic teaching, such as teaching Basic & Advanced Life Support, the material that one presented as evidence-based fact 30 years ago has changed radically over the years and continues to change over time. We have a saying in England that "today's news is tomorrow's fish and chips wrapping paper" (from the days when fish & chips came wrapped in newspaper)! The same could be said of some evidence-based material, I'm afraid! So, to answer your question, who knows how much medicine is truly evidence-based, given that what appears to be evidence-based today will become un-evidence-based by a subsequent study tomorrow. History has much to teach us about evidence-based practice, I'm afraid…...
2 Recommendations
2nd Apr, 2014
Daryle Gardner-Bonneau
Bonneau and Associates
My answer to the question posed would be "The proportion that is evidence-based is not nearly high enough." It seems to take forever to change accepted practice, even when evidence exists that shows it to be wrong. In addition, too many medical decisions/recommendations are made based on reasons that have nothing to do with the well-being of the patient. Obviously, I may be a bit of a cynic, but personal experience has taught me to do my own homework (when that is possible) and not to unquestionably accept the advice of a physician.
2nd Apr, 2014
June H Romeo
Ybarra Research Group
Many of the subspecialty medical societies, such as the American College of Cardiology, publish guidelines for management of specific diseases. In these guidelines (specifically the ACC/AHA guidelines, ex: for management of heart failure) you will find recommendations. Each recommendations is "graded" according to the evidence supporting it as well as the quality of that evidence. Ex, if the evidence is from several large randomized clinical trials the "grade" will be higher than if it is from a few small studies. Take a look at these guidelines for heart failure management and I think you'll be surprised at the work and effort that went in to their development. Whether a physician chooses to follow the guidelines is another story altogether. However, CMS and others are now grading hospitals on whether or not patients are discharged on treatment based on the guidelines, and if not, a medical justification must be made.
3rd Apr, 2014
Robert H. Eibl
Universität Heidelberg
Medicine at the university level in Germany is evidence-based. This is standard since several years. Students learn this very early and know how to ask the right questions. Clinical studies, for example on anti-cancer treatments, need to follow strict regulations. Unfortunately, there is not much action against the harming natural healers; just recently in the media, parents basically killed their child with not allowing an evidence-based approved real therapy with 80% chance of survival, but no chance to cure cancer by a questionable healer.
1 Recommendation
3rd Apr, 2014
Eivind Aakhus
Innlandet Hospital Trust
Then Germany may serve as an example and inspiration to us all, Robert! But still some questions arise: How do the Germans ensure that the strict regulations are truly EB (that is, how are the regulations developed)? How do you measure any gap between what healtcare professionals report that they do (in order to prove adherence to the regulations) and what they actually do? And how do you ensure that healthcare professionals are able to find the regulations when they need them? Do the regulations apply to the whole country or for regions? I agree that medical students today provide skills in EBM, literature searching and critical appraisal, but in Norway there is a saying that "It takes six years to learn medical students the skills of EBM, and six months in clinical practice to unlearn them!"
3rd Apr, 2014
Robert H. Eibl
Universität Heidelberg
Treating patients with anything else than evicence-based medicine easily could get lawers file some lawsuits.
3rd Apr, 2014
John L Wallace
The University of Calgary
One has to be careful. There is a fine line between evidence-based and evidence-biased. Meta-analyses that include only published studies, for example: there is a clear 'publication bias' towards papers showing positive results. Pharma-sponsored clinical trials that turn out to be negative are often not published.
2 Recommendations
3rd Apr, 2014
Robert H. Eibl
Universität Heidelberg
Eivind, I have never been in Norway, unfortunately, but the Norwegians I know (MDs and non-scientists) appear to have a very similar background and similar standards. I can only assume that medicine in Norway is also evidence based and not 1000 years behind. I understand that sometimes, pharma industry makes slight modifications to old substances with come to their end of patent protections and then try to introduce a basically identical substance and mechanism with a new patent, but at the end medicine is based on physicians and science not on side-effects of merchandise. So, basically there may be room for improvement, but the knowledge in the medical field is evidence based and not based on just beliefs, estimates, wrong assumptions (OK, maybe except homeopathy, which should be outside of medicine, and some medical psychology, which is based on "schools"...).
1 Recommendation
4th Apr, 2014
Thomas Karl Hillecke
SRH Hochschule Heidelberg
It is still an interesting discussion here.
I agree with John. 'One has to be careful.'
There is no doubt that current medicine compared to the medicine of former times can be described as more evidence based. But the question was, ‘What proportion … ?’, and in the discussion some contributors also pointed out some critical aspects and limitations. I also think that no one would believe that EBM as it is proceeded and institutionalized today is the end of medical history.
We perhaps can go on asking where are the limitations of current EBM? Or, is it possible to rely only on current definitions of EBM? Or, does EBM fit better with some medical (sub-)disciplines? Or, how to optimize EBM?
In any case we have to admit, that there are epistemological, methodological and practical limits.
• It is impossible to manualize / operationalize every procedure / intervention exhaustively. Descriptions are always limited, because they are base on words and language. And language is less complex compared to concrete (crude) reality (e.g. diagnostics, clinical decisions, surgeons action, psychotherapists behavior).
• Research methods that work in one specific field like e.g. pharmaceutical therapy are not necessarily useful in other disciplines like surgery, psychotherapy or physiotherapy.
• Relying on complex methodological procedures like meta-analysis brings up interpretation problems (validity) in many medical disciplines (e.g. control group bias, garbage-in-garbage-out bias, apples-and-oranges-bias, publication-bias, times-changing-realities-changing bias, false-positive-bias, stake-holder-bias).
• There are some additional possible errors we have to respect: 1. Naturalist fallacy, merelogical fallacy.
• In many medical fields a gap between science / research and clinical practice can be observed.
• Institutionalization of EBM has advantages but also disadvantages. For example it is increasingly hard to apply new promising interventions. So EBM perhaps lead to a conservative attitude.
• Alternative concepts like evidence-informed treatment, empirical supported treatment are discussed because a lot of clinicians are dissatisfied with current EBM.
• There is evidence that physicians (also in Germany) have severe problems with the interpretation of research results (see the work of Gerd Gigerenzer). Therefor despite there is evidence it is often not used in an adequate way. In consequence competences in dealing with research methods should be better integrated in medical training programs (also in Germany).
In consequence, I believe that by staying critical we are perhaps able to optimize treatments.
Regards Thomas
4th Apr, 2014
Paul Silverston
Anglia Ruskin University
We would all accept that illness is a dynamic process and it is equally important that we appreciate that the same applies to evidence, which is often based upon our current medical knowledge of a disease process. Evidence is comforting and reassuring and we all like to feel that we are doing the very best for our patients, based upon a sound rationale but our desire to create evidence that can then be followed is not without pitfalls. Most of these pitfalls arise when we take today's evidence as being written in tablets of stone, rather than simply saying that we are basing what we are doing today on the best evidence that is currently available. In the desire to be scientific, we can sometimes lose the perspective that is afforded us by history. For example:
In the Middle Ages, tinea corporis became known as ringworm because it was believed that the circular lesions were the site of entry of a worm into the body. They had no concept of micro-organisms at that time so they sought an explanation for the phenomenon, based upon their current scientific knowledge. How many illnesses today are thought to be caused by "viruses" because we have no other plausible explanation? In the future, when the true aetiology is discovered, folks will look back at our explanation of these "virally-associated" conditions with the same degree of incredulity as we do of our predecessors. So, evidence-based treatment that is based upon our current knowledge of a disease process will change as our understanding of that disease process changes.
If one looks at the field of emergency medicine, where so much that is taught has to be didactic, given that there is no time to procrastinate in such situations, the guidelines have still changed radically over the past 30 years and at each point that new guidelines have been issued has come the evidence-base for that change. An evidence base does not stay still but changes over time, so we also need to incorporate an understanding of this into both our teaching and our clinical practice. Some of the first proponents of CPR were Scandinavian societies and most of the patients whom they tended to had suffered a cardiac arrest as a result of drowning in cold water. Mouth-to-mouth was not performed but drawing the arms above the chest & then pushing them down onto the chest was, as this mimicked the movement of the chest in inspiration and expiration. The next step was to light a fire and draw warm air into a bellows before blowing it into the patient's rectum. (Try getting that one past the local Ethics Committee now, let alone gaining the co-operation of your local paramedic units for a RCT!) This may seem utterly bizarre but, of course, some patients survived because they were hypothermic and this represented "core warming", so that's why it worked. There was an evidence-base for its effectiveness, even if the method & rationale for doing so were unusual.
I have a book on the management of emergencies in general practice, written in the 1920's, in England, in which it describes the actions to be taken by a GP when presented with a patient with internal haemorrhage. It counsels that the doctor should do nothing to either increase the volume or speed of the heart rate, as this will cause the patient to exsanguinate faster but simply to accompany the patient to hospital and provide the receiving team with a good history of what has happened. (This sounds remarkably like controlled/permissive hypotension, in modern parlance and was advice that was based on sound principles, up to a point.) Unfortunately, the book goes on to say that the best way of managing a patient with respiratory compromise is to invert a soda syphon so that the CO2 is available to be drawn out, before placing the nozzle in the patient's mouth, on the basis that CO2 stimulates the respiratory centre. (That's another one that probably won't get past the Ethics Committee right now.)
All I am saying is that basing our treatment on sound principles and the best available evidence is a worthy principle but we should always temper this with an appreciation that scientific principles change with new discoveries and that evidence-gathering is a dynamic process in which the evidence is often much weaker than one might think, regardless of how well it is presented and what level of quality it is said to possess. Look at the claims that were made more recently about the safety of a certain drug, only to find that the "evidence" that we had all taken to be the truth was biased by a failure to declare all the evidence that was available. Look at the guidelines for patient management that are constantly changing as new "evidence" appears, only to come full circle as the latest evidence is subsequently refuted by further studies. All we can say is that we are simply doing our very best to "First, do no harm" by trying to practice medicine that is based upon the evidence that is available to us today.
There are "lies, damned lies and statistics" as they say, so using these statistics to form a scientific evidence base can have significant drawbacks. Evidence-based medicine is an excellent principle but often lets us down in practice, due to the inadequacies of our scientific knowledge, or the true quality of the studies that are conducted and the interpretation that we place upon the results.
4th Apr, 2014
Robert H. Eibl
Universität Heidelberg
The major difference between medicine and non-medicine, like so-called alternative medicine, is that medicine is evidence-based. Our knowledge in science today is quite good for understanding many processes in aetiology, pathology of diseases, including viruses, bacteria, genetics of cancer. The choice of antibiotics is based on resistance mechanisms. So it would not be a good advice to neglect the current knowledge of science in medical treatment. That may include additional knowledge in the future and a more differentiated view for example in the treatment of new sub-types of tumors not known before the molecular and genetic mapping a few years ago.
5th Apr, 2014
Hans Peter Aleff
Independent Researcher
In the medical approach to baby-blinding retinopathy of prematurity (ROP), the proportion of evidence is zero, due to blatant research frauds and the medical system's inherent inability to recognize or correct them. The epidemic of ROP started in the U.S. in 1940 , the year after the introduction of fluorescent lamps, and broke out just as suddenly after World War 2 in many other industrial countries as these lamps became available there. However, a group of eugenicist American ophthalmologists believed the blinding was due to "defective germ plasm" and decided to eliminate its carriers. To do so, they blamed the life-saving oxygen breathing help that kept many of the babies most at risk alive, and in the 1950s they rigged a multi-hospital trial to withhold oxygen from all babies for the first two days. Then they enrolled only the survivors after having so killed off the babies who would otherwise have survived with eye damage.
The trial designers falsely claimed to have achieved this reduction in blind survivors without increasing the death rate and touted their result as a great success of then then relatively new concept of controlled clinical trial. This led to routine oxygen withholding from premature babies around the world. According to later medical estimates, these oxygen withholding policies killed in the first two decades about 150,000 infants in the U.S. alone. Other clinical trials were unable to reproduce the claimed effect but many still increased the death rate.
When critics pointed out the overexposure of the still developing baby retinae to the excess light from the fluorescent nursery lamps, two pediatric retinal surgeons rigged another bogus experiment in the 1990s. Knowing that the "blue-light-hazard" spike in the spectrum of these lamps over-exposed the babies unprotected retinae in just a few minutes to more damage-weighted retinal irradiance than what the Industrial Safety Guidelines have set as the danger limit for adult workers during an eight-hour shift, these "researchers" patched the eyes of the babies only up to 24 hours after birth, thus knowingly over-exposing both their trial and control groups equally. Then they claimed that the lack of difference in blinding among the two groups proved the innocence of the nursery lights. This bogus evidence is enshrined at the National Eye Institute and so continues the still ongoing epidemic which assures a steady stream of customers for pediatric retinal surgery.
The U.S. Institutes of Health did not react to alerts about this fraud, nor did the Office for Research Integrity or the Office for Human Research Protections. Meanwhile, the recent "SUPPORT" baby suffocation experiment tried again to find the sweet spot of just the right amount of oxygen to give and predictably killed 23 "extra" babies by doing so. When challenged about this premeditated killing in their attempt to prevent a non-fatal condition, so-called medical "ethicists" defended the experiment by saying the risk had not been foreseeable, although they knew the researchers at described that risk in their study protocol and predicted even the correct percentage of "extra" deaths. This experiment was conducted without parental consent to the known risks
For a documentation of this medical disdain for actual evidence and patient health, see http://retinopathyofprematurity.org/01summary.htm and http://retinopathyofprematurity.org/KnowinglyHarmfulMedicalResearch.htm.
As long as such gross pre-Nuremberg-Code abuses are tolerated by the medical profession and the government agencies supposed to regulate it, their alleged commitment to "evidence" and "ethics" is not worth the paper on which its solemn declarations are written.
1 Recommendation
5th Apr, 2014
Ulyana Lushchyk
National Academy of Sciences of Ukraine
Evidence-based medicine. The question is quite interesting and quite simple if you understand the profound essence of medicine: "Who have a good knowledge in the diagnostic, that have success in treatments!" It is an evidence of instrumental diagnostics and availability of analytical component in medical technology, which allows the doctor to get reliable and useful information to form a treatment plan. Simultaneously, evidence based medical technologies help to understand process of recovering due to monitoring of patient health during all process of treatment. http://angio-veritas.com/technologies/individual-angiotherapy-with-instrumental-monitoring/?lang=en
1 Recommendation
6th Apr, 2014
David A Kault
James Cook University
I would hope that medicine would be scientifically based, not evidence based. The latter term refers only to statistical evidence and ignores evidence based on commonsense, expert opinion and mechanism. For example see "Parachute use to prevent death and major trauma related to gravitational challenge: systematic review of randomised controlled trials by Smith GCS & PellJP BMJ 2003;227:1459-1461. EBM dangerously undervalues low cost treatments with small but real benefits.
3 Recommendations
6th Apr, 2014
Maduram Kamaraj
Shri Sathy Sai Medical College and Research Institute
Evidence based medicine is right but the question stand what type of evidences is it statical or scientifically evidences truly helping the researcher or the society
6th Apr, 2014
Susanne Bernhardsson
Region Västra Götaland, Göteborg, Sweden
David, you are of course right that not every healthcare decision can be based on statistical significance or RCTs. That's why EBM has evolved, quite dramatically in the past decade, to now increasingly emphasize the other two circles in the commony used venn diagram which integrates evidence from external research with clinical expertise and patient values and preferences (sometimes there's a fourth circle also considering available resources and/or context, whose roles are increasingly identified as important). The concept of EBM has also widened to now often be referred to as EBP (evidence-based practice) and EBHC (evidence-based health care). One could therefore argue that it's the other way around; the term scientifically based would emphasize science and statistics more than EBM/EBP! Only one of the circles is directly based on science, although one would of course hope that also clincial expertise is at least particially based on scientific evidence, where available. Patient preferences COULD be based on science but are most often based on personal experiences and values.
The parachute article, while very entertaining, misses the point. Very few EBM proponents, I would think none, argues that all medical interventions need to be verified in RCTs. Instead, different evidence hiearchies are acknowledged, where you go "down the pyramid" when no RCTs or controlled studies are available.
6 Recommendations
6th Apr, 2014
Ulyana Lushchyk
National Academy of Sciences of Ukraine
Evidence based observation is necessary for doctors for understanding patient's problem as an indicative points of patogenesis. Results of randominised investigations are necessary for researchers as total opinion. In case with concrete patient we need to use individual approach to him problems and control of illness dynamics process.
2 Recommendations
6th Apr, 2014
Lynne Staff
University of Tasmania
Hello Susanne
Yes patient preference trials can be based on science, but importantly human agency is a fundamental premise where consent is obtained after deliberation and thoughtful consideration of the proposed treatment/care path. It could be construed as more ethical to be part of something preferred rather than consenting to be randomised into a group and perhaps be part of something that may have repercussions down the line. I am not saying that preference trials do not have repercussions, but agency and the capability of making one's own decisions should be fundamental in research on humans and for humans. Sometimes the patient is rendered invisible in science - at least a preference trial gives the patient a voice.
7th Apr, 2014
Vladimir Baksheev
"The time will come and life will prove that many postulates of evidence based medicine was the usual myth of the 20th century." (Vladimir Vizob)
1 Recommendation
7th Apr, 2014
Mojtaba Heydari
Shiraz University of Medical Sciences
Our computerized survey on "UpToDate" recommendations (as a provider of EBM recommendation for clinical practice) has following results:
1- Grade A recommendations: 19%
2- Grade B: 27%
3- Grade C or without grading: 54%
This can show near 20% of medical recommendations are supporting by high quality evidences. This fraction seems to be lesser in the clinical practice.
12 Recommendations
10th Apr, 2014
Stephen P. Gough
At present it is difficult to say with certainty that any medical practice/clinical decision is truly evidence based when there is no certainty that all data from trials has been included in the systematic process of review. It can only be said that based on available data....... EBM is still in it's infancy.
2 Recommendations
10th Apr, 2014
Roger Jahnke
INSTITUTE OF INTEGRAL QIGONG AND TAI CHI
Will be brief and defer to those that know specifics.
Research on extent of EB in medicine:
At the conferences over as many as 10-15 years, there has been constant reference to the fact that conventional procedures have not actually been put through the rigor of EB investigation. I am sure some of the respondents will point to specifics.
Another Key Point:
In my field -- wellness, health promotion, disease prevention, non-pharmacological interventions, stress mastery. There is a real problem with the EB process. Because there are no negative side effects to activities like meditation and relaxation practice or ultra gentle movement, the level of rigor in the research is logically lower. Big trails have proven this or that technique as safe and effective. That is good, really good.
The problem is with policy.
For many very safe and effect forms of neuro-self-regulatory practice (Tai Chi, Yoga, Qigong, Mindfulness), a certain form has been declared as EB'd, yet many forms that are at least as effective are neglected because the purveyors are not associated with a university or entity that has the funds to to the RCTs and get published.
The point -
There are many forms of wellness or health maximization that are just like the modalities that have been found EB'd - BUT, they are not being utilized because they don't have the $$ to get into the EB registries.
This is a huge waste.
Sorry if this seems off topic. Consider this - medicine must be specific to a correct diagnosis. Wellness is a general field in which the interventions are often benign and no diagnosis is necessary - so any gentle Tai Chi is as safe and effective as any other gentle Tai Chi.
2 Recommendations
10th Apr, 2014
Robert H. Eibl
Universität Heidelberg
We may all agree that medicine -in a theoretical sense- always should be evidence-based. Each medical discipline (intensive care, surgical procedures, antibiotic treatments, cancer therapies, even the diagnostics of cancer etc.) in each country may develop their own gold standards and recommendations and they may change during time, as well as some medical centers may go ahead and may differ to develop future standards, but again: evidence-based. Sometimes, a gold standard becomes obsolete which was an evidence-based recommendation for years. This is part of any developing science and knowledge, it changes. But the original question here may refer more to practical medicine. In the theoretical base of medicine, 100% of any diagnostics and therapy should be evidence-based, although the standards may develop and change to better and improved standards due to better diagnostics and therapies.
3 Recommendations
11th Apr, 2014
Thomas Karl Hillecke
SRH Hochschule Heidelberg
Dear Robert,
I am sure that we agree on the argument that the practical discipline medicine should primary rely on high scientific standards.
The technical term ‘evidence-based’ seem to me has at least to perspectives or perhaps meanings a broader one that means ‘medicine should be guided by sound empirical research’ and a narrow sense that means ‘the concrete approach as it is proceeded currently’.
And I want to add that it is not science alone that guides medicine, ethics is also relevant, and rationality, and economics, and cultural decisions, and laws and experience, and competence.
I additionally think that in different countries the proportion of EBM is different.
Regards Thomas
2 Recommendations
14th Apr, 2014
Sharon Mickan
Gold Coast University Hospital and Griffith University
In a systematic review I led, we found that there is progressive 'leakage' from guideline recommendations; because clinicians were not aware, did not agree and they did not adopt in their own practice. Our summary was that only 34% adhered to the guideline for all patients. However, we saw different patterns for different types of guidelines. You can find the paper at http://pmj.bmj.com/content/87/1032/670.short.
In short, research evidence is necessary but not sufficient for practice or policy change. Over the last 20 years the EBM movement has dramatically improved knowledge and practice about designing and critically appraising research, so that it can be used appropriately. The next challenge is understanding how to translate this research knowledge and actually improve practice and inform policy.
6 Recommendations
14th Apr, 2014
Lynne Staff
University of Tasmania
Hello Thomas
I agree. Add to this melting pot sociological influences and what fields the evidence sits in and who has invested interests that determine whether to utilize the evidence or not...and to question its soundness or not.
1 Recommendation
16th Apr, 2014
Carlos Alberto Marín Correa
Instituto Nacional de Salud
Other problems related to this theme are generated because in many cases the evidence generated by the researchers is not shared in the best way with people taking decisions (knowledge management), or that the research is not a priority of public health for their country, or that decision-makers ignore this evidence or have different interests that are not related to public health.
2 Recommendations
18th Apr, 2014
Shakil Siddiqui
Rafaheaam Herbal Laboratories, Karachi, Pakistan
Dear Patzer, in my opinion, there a lake of evidence based medicinal practice, through out the world. if you are asking about the scientific based evidence, it is impossible to develop evidence of each medicine, even it is is from natural source and if you are asking about the evidence clinically or on the basis of experience than all medicine have a specific evidence. Many of the invention in world are based on natural experiences in background.
1 Recommendation
18th Apr, 2014
Jorge Joven
Universitat Rovira i Virgili
Rachel,
This is certainly a question that may promote strong debate. In the real world, Medicine is always based on evidences, but what evidence? The quality of the evidence is not always clear and even impossible to test. For example, calorie restriction may prevent cancer? Physicians we do as better as we can and this is ambiguous. A brief reference to a relatively simple situation (i.e. treatment of hyperlipidemia) may illustrate the fact that even easy questions require protracted discussion by Committees. Then if you feel well represented in such a Committee (say for example a cardiologist) then you will cite and follow the unavoidable guidelines. Why guidelines are not simple statements in a single paragraph? If not (say for example a vascular surgeon or a general practitioner) then commitment is presumably low. It is easy to predict the value (again in the real world) of supposed evidences. In this equation, however, I may foresee a second and more important factor, the patient, which is usually neither educated nor questioned on the effectiveness of a prescribed treatment. Low adherence is therefore easily predictable. In a recent survey (n=312; not evidence-based) of aged patients, unknown for the investigator, with intermittent claudication and at least 3 “treatable” medical conditions, the adherence to the respective treatment ranged between 30% and 52% and the difference was apparently due to the perceived importance of the condition… by the patient. I am pessimistic on the value of “evidences” in health care.
1 Recommendation
18th Apr, 2014
Robert H. Eibl
Universität Heidelberg
There are some good points and views in this discussion, thanks Thomas, Sharon and others. Since EBM is part of real medicine and not part of all of the suspicious "alternative" approaches, we should regard real medicine as the one with a real base of scientific knowledge. It is up to permanent controls by everybody to improve real medicine and to fight against quacks.
1 Recommendation
21st Apr, 2014
Ron Gazze
NEE/FPL (NextEra Energy and Florida Power and Light corporations)
As a clinician, I'd love it if the answers that I give to all of my patient's issues were the result of rigorous application of others work to apply the Scientific Method well to my patient's exact problem. Unfortunately, that scenario almost never happens. For such a decision to be truly directly informed by the Scientific Method, the question must be EXACTLY the same one asks in the experiments, the relevant experiments must have been successfully reproduced with multiplicity in similar environments to mine and with great statistical power, the inclusion/exclusion criteria must apply to my patient, the process must have been performed without bias (IE: greed, ego, politics, shoddy worksmanship, time/money pressures, publication bias, anchoring heuristics, etc. must not have significantly influenced the results), and comorbidities related to my patient must have been actively considered during the research process. Even in the unlikely event where that has occurred, the answer must still be specifically relevant to my patient based upon their own available resources, personal philosophies, previous experiences, allergies, genetics, et alii.
Nonetheless (as a thorough primary-care physician interested in preventive care, wellness, health-maintenance, community health, and diagnosing/treating disease), I need to address daily a barrage of individualized questions from unique patients, their families/friends, and other community stakeholders. The best way for me to do so is to ever be a voracious student myself by accessing and critically reviewing as much research as is possible, learning/reviewing the current state of knowledge regarding the underlying processes that affect the answers to the questions above (Anatomy, Physiology, Microbiology, Biochemistry, Pharmacology, Epidemiology, Epistemology, Theology, Philosophy, Macroeconomics, Microeconomics, etc.), and then doing my best to critically apply all of the above to the very specific personal situation of my patient in a way that does the most good an the least harm to the patient and the system in a pragmatic sense.
To me, that is the epitome of evidence-BASED medicine--To be aware of the base of evidence, to understand it's strengths and weaknesses, and to apply those into complex individual cases. I practice it almost every day. Ironically, however, to do so means that I frequently choose NOT to follow guidelines. Interestingly, that would make me part of the problem in many of the analyses given of the studies described above. Oh well, until there's a MUCH better answer, I guess that I'll keep doing it my way because I'm pretty sure that I'm doing a good job an most of my patients and employers have really seemed to like it.
3 Recommendations
21st Apr, 2014
Gordon Banks
Studies that provide the evidence have only been done usually in common diseases affecting large populations. They are expensive to do and hard to do right. So we are left to less well established evidence for most diseases. It doesn't mean NO evidence, even if the evidence is anecdotal. In addition, large studies usually don't take into account generic variation. For example, a study showing metformin to be the treatment of choice for mild type II diabetics wouldn't apply to someone who has the genes that make metformin ineffective. Yet guidelines typically ignore this.
5 Recommendations
21st Apr, 2014
Lynne Staff
University of Tasmania
Another dimension to this discussion is that evidence does is not only relevant disease...it is to wellness and interventions used in healthy individuals as well. Very insightful post BTW Ron...
1 Recommendation
22nd Apr, 2014
Stephen Gordon
European Central Council of Homeopaths
BMJ Clinical Evidence established that out of 3000 treatments reported in RCTs just 11% were clearly shown to be beneficial and a further 24% likely to be beneficial + 50% are of unknown effectiveness Put this together with factors mentioned above such as leakage and compliance and the answer to the question posed becomes quite challenging.
2 Recommendations
23rd Apr, 2014
Christine M Kurtz Landy
York University
Very interesting discussion. Here are a few of my throughts on the matter.
From my perspective the issue is not necessarily about evidence based practice but rather about evidence INFORMED practice/ decision making. I would hope that health care providers strive to inform their patient care with the best evidence available. Evidence based practice is not practice based on the best evidence alone, but rather should always be Clinician Decision Making informed by the best research evidence available, the patients wishes and preferences, the clinician's experience and the resources available.
The emergence of Clinical Practice Guidelines, Best Practice Guidelines, etc has resulted in one knowledge mobilization strategy to bring research evidence that has been preprocessed by experts to the practice setting. It has helped make some research evidence more accessible and useful to providers who do not have the time, access or sometimes skill to appropriately process primary research, systematic reviews etc to guide practice. However many CPGs and BPGs are not based on high quality evidence. Yet some users of BPGs, etc erroneously assume they are based on high quality evidence.
Much of medicine is not based on high quality research evidence...the high quality research is lacking. In addition, the external validity of many 'high quality' studies (RCTs) is questionable. More research is required in many areas of health care and better methodologies need to be developed to produce knowledge that is generalizable and useful to patient care.
4 Recommendations
30th Apr, 2014
Joachim Sturmberg
The University of Newcastle, Australia
Little, I would say. What we belief to be evidence is socially constructed, and the history of medicine is full of certainties that turned out to be erroneous.
2 Recommendations
Joachim
You say "certainties that turned out to be erroneous". Well, that is science for you. Imperfect, but it is the best system we have for getting to the truth. Do you think that the social constructs science (and EBM) is a worthless pursuit in your opinion? What do you think the 'truth" is in scienceand EBM?
1st May, 2014
Joachim Sturmberg
The University of Newcastle, Australia
Pieter,
let's start with what is science. Science, we "socially constructed", to be characterized by its method, namely the scientific method, which is based on very precisely defined parameters, a very precisely defined (thus simple) intervention (or experiment), and the observation (note: this is subjective/perceptional, not objective) of the outcome, AND this has to be repeatable by others AND achieve the same outcome. The method goes back to Newtonian times, when "science" was largely concerned with the physical world where "these crude methods" worked well, however, the event of quantum mechanics has put all of this into the "false" basket, even though, for pragmatic purposes (like in engineering) it works fine without being the truth.
Truth is the illusion and really needs to stay out of the discussion, I called a paper once The illusion of certainty, a deluded perception (DOI:10.1111/j.1365-2753.2011.01667.x) which offers some more explanations.
Living things, better called living systems, don't follow the mechanistic rules, living things behave in complex adaptive ways and are non-deterministic. Living things can respond in a bounded way, not every possible response is possible, but within the bounds a variety of different outcomes do occur repeatedly, i.e. outcomes are patterned and thus "mutually agreeable", however, which outcome one will observe is not predictable from the factors defined at the beginning of an "experiment". In addition every interaction within a complex adaptive system irreversibly changes the members (technically called agents) of the system, and repeating the same intervention will result in different and untested behaviors and outcomes (see the Mandelbrot set as an example how the iterations of "the same" intervention results in divergent but self-similar, i.e. patterned, outcomes).
I would pose the question you ask slightly different, how should we best interpret the observations (some would call it results) we make in the living world, and how closely do these "experimental observations" match with experienced patterns, and if not, is what we have observed truly a new pattern. It is always worth the effort to observe, analyse and critically reflect - in all domains of inquiry, however, the dogma of having found the holy grail based on a "scientific experiment" and thus being able to infer future actions is uncritical and unscientific.
BTW - we probably shouldn't talk about re-search, rather search or inquiry. Re-search entails the presupposition of re-confirming what is already known, thus stands in the way of truly gaining new knowledge.
2 Recommendations
I agree with much of what you say, Joachim thanks for your answer. You say "the dogma of having found the Holy Grail based on a "scientific experiment" and thus being able to infer future actions is uncritical and unscientific". That does not sound right to me. What do you mean here by "uncritical" and "unscientific”? Are you suggesting prediction is impossible? I am probably misunderstanding you. The causal relationships that experiments reveal can be used to predict the future to a certain extent. For example, if in a clinical trial that tests morphine for pain we find that morphine is effective. Based in this, we can with some certainty predict that morphine will be effective in the future too. Do you believe that we can predict the future, based on patterns in the past? Now I know that predictions are not perfect, there are too many unknown factors that we don’t account for. It must be! Otherwise I would not want to fly in an aircraft, just because the aeronautical engineer predicts the plane will stay in the air based on his aeronautical experiments!
Speaking of social constructs, here is a very light-hearted and fun article on the social construct of gender. Yes, indeed, some think gender is a social construct, I kid you not! http://wmbriggs.com/blog/?p=10796
1 Recommendation
2nd May, 2014
Joachim Sturmberg
The University of Newcastle, Australia
Yes, predictions in the sense of "this cause will result in this precise outcome" is impossible. The causal relationships you imply arise from linear relationships in physical systems where output is proportional to input. In living systems inputs invariably result in disproportional outputs which are not predictable. That said I agree with your observation that one can "predict the future to a CERTAIN extent", which is to say that most outcomes in living systems fall within a limited range of possibilities but in some cases outcomes WILL be complete unforeseeable.
Your comparison to an aircraft highlights a common misconception between complexity and complicatedness, the former being unpredictable, the latter highly/precisely predictable, otherwise neither you nor I would put a foot onto a plane. Whilst experts can precisely understand a plane, experts of living systems always only will have a good educated hunch that gets better with increasing experience.
2 Recommendations
Thanks Joachim, I agree with all you stated. It is tempting to think the universe is like a big billiard table, where everyhing can be predicted almost perfectly depending on initial conditions, as Laplace thought.
1 Recommendation
2nd May, 2014
Nelson Albuquerque de Souza e Silva
Federal University of Rio de Janeiro
I think it is clear from the several comments to the question posed, and the existent data, that a small proportion of decisions we take in medicine is supported by well conducted research. I add a conclusion from the Congressional Office of Tecnology Assessment - USA that only 15% of the interventions in the medical practice have support from the informations comming from controlled randomized trials. To aggravate the problem, only 1% of medical publications represent "good science" contrasting to the distorted or misinforfation comming from propaganda, specific interests and "bad science" to support almost everything.
To compllicate the problem we now have guidelines full of conflicts and opinions withouth scientific evidences to support.
Evidence based medicine brought inquestionable advances, but has tremendous limitations to apply the conclusions from studies to medical practice. The selection of patients that participate in trials are rarely representative of the patients encountered in the medical practice, problems frequently occur in large randomized trials with multiple centers involved, the cointerventions are frequent and may distort the results, the efficacy of the drugs or other interventions tested are very small and the great majority of patients will not benefit from the interventions. Interventions that affect us as a whole, as physical activity, diet and environment (including social organization and working environment) are usually more effective than drugs that try to act on a single physiologic system. For instance, in a recent study our group found that in the city of Rio de Janeiro, the area with the lowest Human Development Index had a cerebrovascular mortality 10 times greater and 10 years earlier than the mortality found in the region with the greatest HDI. This finding is more important than any of the "famous risk factors" (diabetes, hypertension, cholesterol and so on). The better intervention is not with drugs but in the elimination of social inequalities.
Thus, I feel that every hospital and medical practice must maintain a data bank of the patients under care with constant surveyance and analysis of their practices and results obtained discussed with the entire staff in order to take decisions to change the practices according to the local results, compared to other places, in order to reflect and comprehend the differences or similarities encountered and decide the direction to go, including social interventions.
2 Recommendations
2nd May, 2014
Arthur Leibovitz
Shmuel Harofe Hospital , Geriatric Medical Centre 1984 - 2011
Medicine is practiced differently in different health systems and catchment areas. Not all evidenced based knowledge is carried out . So the question shoud specify the
meaning of " medicine " : does it refer to medical knowledge or to medical practice
and to what extent medical knowledge is " translated " into medical practice.
1 Recommendation
2nd May, 2014
Susanne Bernhardsson
Region Västra Götaland, Göteborg, Sweden
What an interesting discussion to follow, with so many different perspectives! I think there is no right or wrong - evidence means different things to different people, and so does knowledge. In addition to the concept "evidence" and all its definitions, "knowledge" can also be broken down, for instance in research-based knowledge and practice-based knowledge; often intertwined in practice. I also find it fascinating that the whole EBM movement sometimes is perceived as something rather new, that emerged in the 1990s... while on the contrary the basic principles of EBM (basing a decision on results from systematic trials) had been practiced for many hundred, if not thousands, of years! One of the first trials I have read about described two runners that were sent off to run the same distance - one had eaten ginseng before the run and the other not. This was in ancient China around 1000 AD!
2 Recommendations
2nd May, 2014
Hans Peter Aleff
Independent Researcher
Joachim said above that "the dogma of having found the holy grail based on a "scientific experiment" and thus being able to infer future actions is uncritical and unscientific." Indeed, even some of the so-called "gold-standard" double-blind randomized controlled clinical trials are so hopelessly biased that they fill that holy grail with poison. I posted earlier the example of ascribing the baby-blinding retinopathy of prematurity to excess oxygen breathing help, based on a rigged multi-hospital experiment with all the appearance of modern science but so obviously flawed that any impartial observer can spot its inherent falsehood.
Yet, there are no mechanisms in the judging of "evidence" that would flag such distortions, and accordingly, even the "gold-standard" of evidence is nothing but fools' gold. Until the proponents of "evidence-based" medicine can recognize and correct such flaws in their system, I have to agree with Joachim's above comment that "What we belief to be evidence is socially constructed, and the history of medicine is full of certainties that turned out to be erroneous". Believing in those alleged certainties is worse than simply acknowledging our ignorance.
2 Recommendations
4th May, 2014
Thomas Kersting
Technische Universität Berlin
Have a look at http://www.testingtreatments.org and/or the very good booklet "Testing Treatments. Better research for better healthcare" by Imogen Evans et al.
It might answer some portion of your questions. A very critical article with tons of literature (unfortunately in German - but the literature is mostly English): Müller JM. Minimal-invasive Chirurgie unter dem Blickwinkel der Evidenz basierten Medizin. Passion Chirurgie. 2011 November; 1(11): Artikel 02_01.
Worldwide there are 234 million operations per year, 7 million complications and 1 million surgery-associated deaths - but only 4% of all published randomized controlled trials (RCTs) comparing surgery and only 24% of all surgical therapies are based on the results of RCTs. Estimates give us a number of >70% of all surgical procedures not beeing evaluated due to the standards of evidence based medicine.
2 Recommendations
7th May, 2014
Andre Catrice
Department of Health Victoria
Hi Rachel. You could look at BMJ's Clinical Evidence aims to provide a systematic review of the evidence for treatment of a variety of clinical conditions: http://clinicalevidence.bmj.com/x/set/static/cms/about-us.html.
The evidence base for effectiive treatments that are beneficial for a number of conditions is 11% with a further 24% likely to be benficial. 50% of treatments have unknown effectiveness. They do have a caveat on this much like the issues you raise, including where the evidence base is evolving.
Cheers Andre
1 Recommendation
7th May, 2014
Shakil Siddiqui
Rafaheaam Herbal Laboratories, Karachi, Pakistan
Very first we have to classify the term of evidence based. 1. Scientific. 2. Experience 3. Observational. Based, then we shall be clear the ratio between.
1 Recommendation
7th May, 2014
Sharon Mickan
Gold Coast University Hospital and Griffith University
Facts or information are the starting point of evidence. Yet, information can mean different things to different people, and in different circumstances. It may be that information constitutes evidence in one setting, but not in another.
I have debated the difference between the terms evidence and evidence-based in a recent blog at http://knowledgetranslationoxfordblog.wordpress.com/2014/03/11/what-is-evidence/
1 Recommendation
7th May, 2014
Joachim Sturmberg
The University of Newcastle, Australia
Sharon, you explain this well. I think there is another way of looking at knowledge, namely through a complexity lens. We all know things about an item in different ways, influenced by our way of thinking and our prior expertise. Knowledge is complex and emergent (also known as learning). If you are interested here is a paper that explores this in more detail: Knowing - in medicine.
3 Recommendations
8th May, 2014
Andre Catrice
Department of Health Victoria
Seems to me that EBM guidelines have significant value in critical conditions demand the best responses or in procedural work where predictable outcomes are required to ensure predictable outcomes. In defense of the GPs they are often required to manage multimorbidity and chronic diseases in frail and ageing populations. If they were to apply EBM guidelines to all conditions then it is more than likely that they would cause more harm than good. I refer people to an old article but one that still resonances: http://www.ncbi.nlm.nih.gov/pubmed/9797743
Does anyone know of any articles using an EBM approach to managing complexity in multimorbidity? I am particularly interested where a holistic approach to care involves decisions ar made that are person centred and incorporate patient preferences, optimise benefits and miniimse harm and aim to enhance quality of life in the context of the individual. This seems to go to the heart of decision making that Joachim's post. I suspect that there is a limited base.
2 Recommendations
8th May, 2014
Helen Georgiou
Victoria University Melbourne
I totally agree, evidence based medicine is another 'tick box', and in my experience it in my experience evidence based medicine is lip service. For eg in the treatment of haemoglobinopathies, individual haematologists follow their own protocols and not that of 'evidence based medicine, ' what ever that means. In the treatment of for eg sickle cell and thalassaemia major, it is well know that iron chelation and hypertransfusion are key to the well being and longevity of such patients, yet treatment is diverse. So where is the practice of evidence based medicine?
1 Recommendation
8th May, 2014
Sharon Mickan
Gold Coast University Hospital and Griffith University
Andre and Helen, Guidelines are just one strategy of EBM that have been corrupted by human inertia and power plays. They were designed to synthesise high quality evidence to guide practical use. It should also be recognised that, using critical appraisal skills, there are valid occasions when the research is not applicable to the patient or situation and it is appropriate for clinicians not to adhere to a particular recommendation.
HOWEVER with the pressure to develop guidelines in (many) areas of practice where there is limited or contested evidence, low quality and experiential research evidence is often included to make the guideline more complete. Therefore it is difficult and almost impossible to critically appraise the guideline for use - and many clinicians choose not to bother!
ALSO organisations are using guidelines to denote minimum standards of care - and this is totally at odds with their original purpose
3 Recommendations
8th May, 2014
Helen Georgiou
Victoria University Melbourne
I agree with you Sharon, however I maintain there are competing interests. Experimental evidence is rarely considered despite more than 90% of western medicine being based on it eg transfusion, hip replacements stents in cardiac surgery etc.
2 Recommendations
8th May, 2014
Anne Mullin
National Health Service
This is in reply to Andre's post-
Norbury, M., Mercer, S. W., Gillies, J., Furler, J., and Watt, G. C. Time to care: tackling health inequalities through primary care. Fam Pract. 28(1), 1-3. 2011
This study addresses some of your questions & if you google 'GPs at the Deep End' you can see some of our positional papers which cover issues related to multimorbidity.
1 Recommendation
9th May, 2014
Andre Catrice
Department of Health Victoria
Thanks for the comments. Like with breaking the rules you need to know what they are first. Thus, it takes skilled critical assessment to know when the evidence base is likely to cause more harm than good, especiaqlly where there is competing care issues. This may go some way to explain the relatively low adherence to EBM.
Thanks for the reference Anne. Sadly, it seems our current conservative government is pushing on with healthcare reforms (ways to reduce government spending and avoid having a vision really) that are likely to increase inequality and widen the gap with the inevitable outcomes.
It does not appear that the pollies are looking at the evidence base.
Sorrry, a long way from the original discussion - or is it? Evidence base medicine can provide a basis upon which govenment can make the best decisions. It does happen at levels of government that work closely with the care deliverers. It gives us working in government the evidence to make decisions and support policy/programs/ processes that show the best outcomes - can help to take the politics out of decision making.
Cheers Andre
1 Recommendation
29th May, 2014
Atif Sitwat Hayat
Suleman Roshan medical college,Tando-Adam Sindh, Pakistan
In fact, EBM is very difficult to be applied in critically ill pts. I suspect that many interventions, treatments, and decisions in medicine do not have a high level of evidence base (i.e. no formal trials, decisions are based on experience, practice, etc).
Atif Hayat
Isra university Hyderabad.
1 Recommendation
4th Jun, 2014
Boris Katsnelson
Yekaterinburg Medical Research Center Prevention and Occupational Health of Industrial Workers
We should not forget that medical "decisions based on experience, practice, etc" were helping patients and saving lives during many tens of centuries and are still a firm foundation of diagnostics and treatment. The experience accumulated by generations of doctors and new (scientifically justified) drugs, methods etc constitute a reciprocally enriching tandem, and I do not believe that calculation of respective percentages is very wise. What is really sad: in many civilized countries blatant quackery is flourishing outside the medicine however evidence-based is the latter.
2 Recommendations
12th Jun, 2014
Andre Catrice
Department of Health Victoria
Perhaps consider that complex clinical decision making needs integration of EBM, person centred care and clnician's expertise.
Andre
4 Recommendations
12th Jun, 2014
Alan V. Schmukler
Hpathy.com
The EBM standard, while often not followed in actual practice, is used to critique holistic therapies, thus denying patients legitimate and time tested healing methods. It's an interesting double standard which helps maintain the pharmaceutical drug paradigm.
2 Recommendations
12th Jun, 2014
Boris Katsnelson
Yekaterinburg Medical Research Center Prevention and Occupational Health of Industrial Workers
Dear Alan, I seem to have said something like it 8 days ago.
18th Jun, 2014
David A Kault
James Cook University
I have just had a paper published in the Journal of Medical Statistics and Informatics (http://www.hoajonline.com/journals/pdf/2053-7662-2-5.pdf) which shows statistically that EBM is wrong about two thirds of the time when it declares a treatment ineffective on the basis of the observed effect being only weakly positive (ie. p>0.05).
1 Recommendation
18th Jun, 2014
Boris Katsnelson
Yekaterinburg Medical Research Center Prevention and Occupational Health of Industrial Workers
Dear David: If nobody takes offence, I' ll say that it is a typical example off what I call the statistical fetishism deep-rooted in common mentality of medical and some other researchers (clinical, epidemiological, experimental) : if some result is not significant statistically it is no result at all and, a any rate, is not worth discussion.
18th Jun, 2014
Alan V. Schmukler
Hpathy.com
What's happening in homeopathy nowadays is a good example. We have 200 years of clinical successes by some of the best medical doctors who ever practiced. Many thousands of people were saved from some of the worst epidemics in history. Yet, all this is ignored while more RCT's are demanded. Reality takes a back seat to laboratory results.
18th Jun, 2014
Daryle Gardner-Bonneau
Bonneau and Associates
I will be very interested to read David's paper, as it goes against the grain of what I'd been taught with respect to EBM, and also with respect to my experience doing a big meta-analysis project on the use of bracing in the treatment of scoliosis some years back. True to what I'd been taught, lower quality (i.e., less controlled) studies tended to show small positive effects of bracing (along the lines of "any treatment is better than no treatment"), but when you looked at higher quality, more controlled studies, you would see a somewhat different picture. A properly conducted meta-analysis would seem to tease out the issues and biases to tell you what is truly going on and whether the effect that seems to be there across studies is "real" or not.
19th Jun, 2014
David A Kault
James Cook University
Dear Boris
there has been much concern about publication bias and unduly optimistic evaluations of new and costly treatments with later "medical reversals". My paper is the first to quantify the converse phenomenon of undue rejection by EBM of cheap established and at least mildly effective treatments. I believe that this is an aspect of the careless application of EBM which may be of comparable cost in terms of unnecessary suffering and deserves more attention. It is to be expected that some effective treatments yield negative trial results - this is simply Type II error. For any particular trial Type II error might be set at say 20%, but this does not tell us what proportion of negative trial results are due to Type II error - this is estimated in my new paper (http://www.hoajonline.com/medicalstat/2053-7662/2/5).
26th Jun, 2014
Kuldeep Dhama
Indian Veterinary Research Institute
May be less than 25% since nature medicines like Herbs, ayurveda, homeopathy and other complementary therapeutic regimens and new generation therapies like stem cell therapy, nanomedicines and others still need much scientific validations and trust - evidenced based principles
2 Recommendations
28th Jun, 2014
Sandip Chakraborty
Feed Mixing Plant, ARDD & Deptt. of Vety. Microbiology, C.V.Sc & A.H., R.K. Nagar, West Tripura
Herbs, Ayurveda along with new generation therapies like stem cell therapy as well as nanomedicines still need much scientific validations and trust. So, in my opinion only 20 per cent of medicine is evidence-based.
2 Recommendations
1st Jul, 2014
MANUEL-MARIA ORTEGA-MARLASCA
Servicio Andaluz de Salud, Jerez de la Frontera, Cádiz, Spain
A patient in 4 minutes, only....I think is more survival medicine, isn't I?.
1 Recommendation
3rd Jul, 2014
Yohannes Woubishet Woldeamanuel
Stanford Medicine
The difference between statistical significance and statistical insignificance is not statistically significant. Plus, many of behavioral and biological parameters (which happen to be the foundations of evidence) follow chaos and nonlinear phenomena which is difficult to model or assume.  
2 Recommendations
3rd Jul, 2014
Mel Hopper Koppelman
University of Western States
While I think this discussion is interesting, I wonder if the term "evidence-based medicine" is being used correctly. In 1997, David Sackett described EBM as: "the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients. The practice of evidence-based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research." In other words, using clinical experience to make decisions about individual patients is an important part of EBM. Systematic reviews, which gained in popularity on the back of the EBM movement, are not the only aspect. 
What this discussion seems to be about is the proportion of clinical practice that is based on high quality evidence such as that found from large scale RCTs and high quality systematic reviews, such as from Cochrane, which is another important question. But I feel it's important to agree on a definition of EBM (the Sackett quote above is usually given as the standard definition) when addressing this important question. 
4 Recommendations
4th Jul, 2014
Mostafa Ghanei
Baqiyatallah University of Medical Sciences
While many drug companies try to push physician for prescription of their drugs and provide lot of papers and books in these regards, it is very hard to talk about evidence based medicine. They have money and attract many scientists to help them for their objects. you find drugs that after they make rich comany then expertists say there is little or no effect for that. I think this infrastructure born bad medicine.
8th Jul, 2014
Arnold Eiser
University of Pennsylvania
There is no direct linear correlation between clinical evidence and practice guidelines, some subjectivity necessarily enters guideline formation. Otherwise there would not be so many differing guidelines on the same subject. That being said, the above point regarding pharma influence is significant as well.
Another point is that humans are a heterogenous group so guidelines for a large group may not be advantageous for individuals who are in clinical outlier subsets that were not present in significant numbers in the studies.
Finally cost, side effects, personal experience and other factors influence adherence  to guidelines.
14th Jul, 2014
Khalid F Tabbara
Saudi Commission For Health Specialties
Comparative effectiveness protocols are common in the medical literature and represent a module for Evidence Based Medicene (EBM). Unfortunately these protocols are deficient and do not provide true EBM. Most of the protocols look at the response of groups and not individuals. Human investigations should be personalized and must take into consideration the individual differences in metabolism , genetics, and pharmacogenomics, etc.. 
1 Recommendation
17th Jul, 2014
Karl West
Academy of Lymphology llc
As good as evidence is -- to use it as a paradigm is wrong. Evidence is merely the foundation upon which the process and principles may be relied upon.
1 Recommendation

Similar questions and discussions

Related Publications

Got a technical question?
Get high-quality answers from experts.