Evidence-based practice
Confidence are needed in managing medical information and making decisions. Traditional approaches relied on pathophysiologic reasoning, local practice patterns and opinion leaders, their education in medical school, reviews written by experts, and continuing education courses taught by experts. Pathophysiologic reasoning means that by understanding the biochemistry or physiology or anatomy, one can choose the most appropriate treatment.
For examples:
Local practice patterns are also very important. Physicians are generally more comfortable practicing in a pattern or style that is similar to those around them. Often, physicians train at a hospital or university and then remain in that community, further reinforcing these local practice patterns. What clinicians learn in school is obviously important in shaping clinical decision-making. However, as medical knowledge has advanced, they often have trouble knowing when to discard discredited practices and adopt new, effective ones.
Whoever in discussion adduces authority uses not intellect but rather memory. - Leonardo DaVinci
Bias
The statistical definition of the bias: a systematic distortion of statistical results due to a factor not allowed for in its derivation (simply means a tendency to vary from the truth).
Unintentional bias
As human beings we are wired a certain way, and therefore have certain inherent cognitive biases. These are unintentional, but are biases nonetheless that have a powerful effect on clinical decision-making.
- Cognitive biases: “The fool doth think he is wise, but the wise man knows himself to be a fool” - Wm. Shakespeare. 1). Conjunction rule bias (“Occam’s Razor”): Several different signs or symptoms are more likely to be caused by a single unifying diagnosis then by several separate diagnoses. 2). Anchoring or “Diagnostic momentum”: Given an initial estimate or diagnosis, we tend to “anchor” subsequent evaluation to that initial diagnosis or number (most people are optimizing prior work rather to judge them). 3). Availability bias: More recent and readily available answers are preferentially favored because of ease of recall. 4). Confirmation bias: Diagnosticians tend to interpret the information gained during a consultation to fit their preconceived diagnosis, rather than the converse. 5). Failure to systematically measure outcomes and limits of pathophysiologic reasoning.
Intentional bias
Intentional bias is an increasingly important factor in how medical information is communicated to physicians. Corporations with a fiduciary responsibility to their shareholders to increase market share for a medical product have an incentive that may not always be aligned with that of patients and may not be consistent with the best available evidence. Continuing medical education is largely financed by corporate partners, as are journal supplements, monographs, and videos.
Increasingly, clinical trials are a collaboration between a corporation and an academic institution, resulting in studies designed to market a product rather than find scientific truth.
The modern clinical trial has several key elements: 1). comparison of two or more treatments, one of which may be a placebo. 2). prospective data collection. 3). randomization. 4). masking of patients, physicians, and/or researchers to treatment assignment (if possible).
Strategies employed to intentionally bias clinical trials include: 1). inadequate comparison (i.e. low dose of the generic or comparison drug). 2). comparison to placebo rather than existing standard practice. 3). changing key outcomes in mid-study (occurred in 51 of 82 clinical trials in one study) (Chan, 2004). 4). stopping a study early or extending it (if the desired effect is not seen). 5). active run-in periods to filter out patients who do not tolerate or respond to the drug. 6). limiting the study to patients who have failed standard therapy
Out of that effort grew the Cochrane Collaboration, named after Archie Cochrane and consisting of thousands of researchers working in dozens of groups all over the world to perform systematic reviews.
{width=10%}
PHARMACEUTICAL ADVERTISING: non-rational appeals. 1). Appeal to Authority. 2). Bandwagon Effect：“This is the most widely prescribed antibiotic in the U.S.”. 3). Disease oriented evidence. 4). Red Herring. 5). Appeal to Fear and Minimizing Harms
Evidence-Based Medicine
To be evidence-based, a recommendation for clinical practice should have all of these characteristics: 1). All relevant studies on a topic have been identified. 2). Studies have been evaluated for how well the research is designed to avoid bias. 3). The recommendation is based on the best designed, most relevant, and valid studies. 4). Limitations of the evidence are clearly stated, and the strength of evidence for the recommendation is noted.
An evidence-based approach means that the clinician has made the effort to identify the strongest, most valid studies, is able to change his or her mind about a test or treatment when the evidence supports a change in practice, and acknowledges when the evidence available for making a decision is less than ideal.
An evidence-based approach to patient care acknowledges that medicine is a probabilistic enterprise. Our goal should be to choose tests and treatments that have the greatest probability of helping our patients. For example, we know that lowering blood pressure in hypertensive patients will make them, on average, have fewer cardiovascular complications than if their blood pressure elevation was left unchecked. We don’t know, though, for a particular patient, that treatment of his or her blood pressure will actually make a difference. For patients with mild hypertension, 700 have to be treated each year to prevent one bad thing from happening to any one of them. For severe hypertension, this number decreases to 15. Armed with this information, clinicians and their patients can make better decisions.
Evidence-based practice is not “cookbook medicine”. In fact, an evidence-based approach allows one to discuss a range of effective options with patients and then help them choose one that is best for them. Evidence-based practice is a somewhat “art of medicine” compared to current too much art in medicine. Evidence-based practice need more effectiveness trials and more community-based clinical trials. The evidence of Evidence-based practice often from many RCTs done in carefully selected populations that do not reflect the messy, undifferentiated world of real clinical practice. something that works well in the ideal setting of an efficacy trial may not work as well in a real world “effectiveness study”.
Patient Oriented Evidence: Things patients would care about; Improved symptoms, reduced duration of illness, reduced morbidity and mortality, lower cost
Disease Oriented Evidence: Things a scientist should care about; Improved BP, blood sugar, flow rate; But: physiologic measures may not translate into helping a patient live a longer or better life.