Why Reading the Title Isn't Good Enough: An Evaluation of the 4S Approach to Evidence-Based Medicine

Why Reading the Title Isn't Good Enough: An Evaluation of the 4S Approach to Evidence-Based Medicine
of 24
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  THE INTERNATIONAL JOURNAL OF  FEMINIST APPROACHES TO  BIOETHICS   Vol. 2, No. 2 (FALL 2009). © 2009   WHY READING THE TITLE ISN’T GOOD ENOUGH: AN EVALUATION OF THE 4S APPROACH TO EVIDENCE󰀭BASED MEDICINEK󰁩󰁲󰁳󰁴󰁩󰁮 B󰁯󰁲󰁧󰁥󰁲󰁳󰁯󰁮 Abstract Proponents o evidence-based medicine (EBM) have recently suggested a “4S” approach to clinical decision making in which physicians are advised to rely on increasingly abstract summaries o the available research evidence. Tis retreat rom the srcinal data o medical research is ill-advised: it extends an unjustified evidence hierarchy, overestimates the role o computer systems, divides communi-ties, discards evidence, ignores contexts, and devalues broad critical evaluation. I draw upon eminist social epistemology to evaluate the 4S approach to EBM and to suggest means or improving the evidence base o medical research and practice. I n 2004 the deputy editor o the  Journal of the American Medical Association , Drummond Rennie, remarked, “Medicine is becoming a sort o Cloud Cuckoo Land, where doctors don’t know what papers they can trust in journals, and the public doesn’t know what to believe” (Brownlee 2004). Proponents o   KIRSTIN BORGERSON 153 evidence-based medicine (EBM) suggest that they have the solution to this problem—they have a method or determining what physicians should believe. Unortunately, they are mistaken. And, unortunately, these mistakes have ethical implications. What physicians know and how they know it governs the inormation provided to patients and subjects during inormed consent, sets the standards or uncertainty underlying assessments o equipoise, and shapes the appraisal o the scientiic validity o proposed research trials. As such, epistemic claims matter to the ethics o medical research and practice.In what ollows, I explain the core commitments o EBM and then outline the recently proposed “4S” approach o EBM. Following this, I raise six concerns with the 4S approach. I then draw on eminist social epistemology—specifically critical contextual empiricism—to explain why the 4S approach is problematic and to make suggestions about what might be done to improve the situation. Feminist social epistemology, drawing on longstanding eminist concerns about the organization o social institutions and the potential or biased and oppressive social practices, has much to offer in proposing corrections to the ethically loaded, epistemic errors o the latest orm o EBM. Evidence-based medicine and the 4S approach Evidence-based medicine was introduced seventeen years ago at McMaster University in Canada and has now been adopted by health-care proessionals around the world. EBM requires that physicians integrate the best available clinical-research evidence into decisions made in the clinical care o individual patients (Sackett et al. 1996).At the core o the EBM movement is the evidence hierarchy, designed to reflect the strength o the evidence produced by various research methods. It is assumed that higher-ranked evidence on this scale provides greater justification or treatment decisions made in the care o individual patients. Te hierarchy places systematic reviews o randomized controlled trials (RCs) above individual RCs, which are then ranked above cohort studies. Below cohort studies are case-control studies and case-series, ollowed by expert opinion and bench research. Practitioners o EBM, then, are not merely committed to using research evidence in their clinical practice; they are also committed to assessing research evidence according to the evidence hierarchy. Tere is more to being evidence-based than simply basing decisions on evidence.According to the earliest ormulation o EBM, a physician searching or a treatment or a particular patient should ormulate a question, do a literature  154    I󰁮󰁴󰁥rnai󰁯na󰁬 Journa󰁬   OF FEMINIST APPROACHES TO    B󰁩󰁯etics 2:2 search or relevant studies, and consult the evidence hierarchy to determine the quality and strength o the results obtained. Afer this critical appraisal has been perormed, he or she can then weigh the particularities o the individual case against the determined objective strength o the evidence. Te evidence hier-archy was created in large part to assist physicians in weeding out low-quality evidence. EBM is thought to promise greater certainty o decisions and a scien-tific and objective approach to medicine (EBM Working Group 1992).In recent years, evidence-based medicine has been modified and no longer recommends that physicians ollow the exact steps outlined above. Practitioners o EBM today have contracted out the critical appraisal step and are now much more reliant on evidence syntheses produced by others. It is not uncommon to find physicians claiming to practice EBM who rely entirely on reviews o the litera-ture produced by others (Brody, Miller, and Bogdan-Lovis 2005; Upshur 2002).Te tendency to preer analyses that bring together a body o literature, exemplified in the placement o systematic reviews at the top o the srcinal evidence hierarchy, culminates in the position taken by one o the ounders o EBM, Brian Haynes, in 2001 and adopted by Straus and colleagues in the latest edition o an authoritative EBM handbook (2005). Haynes, and now Straus and colleagues, suggest a “4S” approach to medical decision making. It is this pro-posed approach that is the ocus o my critical analysis. In light o “increasing numbers o clinically important studies, increasingly robust evidence synthesis and synopsis services, and better inormation technology and systems,” propo-nents o the 4S approach have suggested a way to guide physicians toward the “most evolved” evidence more quickly (Haynes 2001, A11). Naturally, the 4S approach is hierarchically organized:SystemsSynopsesSynthesesStudiesAt the lowest level o this hierarchy are individual studies : RCs, cohort studies, and case-control studies. Note that there is no mention o the lowest-ranked evidence rom the srcinal hierarchy o evidence (or instance, case studies and bench re-search), which appears to have dropped off the bottom o the new 4S hierarchy.Above studies, and more highly valued, are syntheses  such as meta-analyses o RCs, o the sort produced by the Cochrane Collaboration. A meta-analysis—a   KIRSTIN BORGERSON 155 orm o systematic review—is a statistical synthesis o the quantified results o several trials, each o which addressed the same question (Greenhalgh 2006).Moreover, EBM advocates argue that even syntheses (systematic reviews) are too inaccessible to the busy practitioner: “What busy practitioner has time to use evidence-based resources i the evidence is presented in its srcinal orm or even as detailed systematic reviews?  ” (Straus et al. 2005, 37, emphasis added). A solution is proposed: when syntheses are still too detailed, turn to synopses.  Synopses are the abstracts o systematic reviews ound in evidence-based jour-nals such as the  ACP Journal Club  and Evidence-Based Medicine . According to Straus and colleagues, “the perect synopsis o a review . . . would provide only, and exactly, enough inormation to support a clinical action” (38).Further, i the synopsis, or abstract, is still too much inormation to pro-cess, Straus and colleagues helpully point out that “in some circumstances, the title [of a review] provides enough information” (38, emphasis added). Teir inter-est is in conveying the results o research—not the methods or messy details—as efficiently as possible.Finally, at the top o the 4S hierarchy are computerized, decision-support systems , which would ideally integrate the lower levels o evidence in one loca-tion. Computer systems are not yet able to do this but according to Straus and colleagues, this would be an ideal, or “perect,” approach (34).Straus and colleagues stress that, “You should begin your search or best evidence by looking at the highest-level resource available or the problem that prompts your search” (34). Higher-level evidence “trumps” lower-level evidence (Haynes 2006, 163). So, as with the srcinal evidence hierarchy according to which you hope or RCs but might end up settling or less (a cohort study, or even just a case study, or instance), using the 4S approach you hope to find a synopsis (or title o a synopsis) and will only look urther down the hierarchy i this initial search ails. Tere is no need to look urther down the hierarchy once you have ound something at a high level.Te 4S approach is motivated primarily by practical concerns and new methodological and technological developments. Tough its motivations may be practical, there are epistemic costs to the conveniences it offers. In what ol-lows I argue that the 4S approach is ar rom ideal.  156    I󰁮󰁴󰁥rnai󰁯na󰁬 Journa󰁬   OF FEMINIST APPROACHES TO    B󰁩󰁯etics 2:2 A critical evaluation of the 4S approach Continuity Because the 4S hierarchy is an extension o the srcinal EBM hierarchy o evidence, the arguments made against the srcinal hierarchy provide reasons or questioning the epistemic value o the extended hierarchy. Tere is an ex-tensive (and growing) body o critical literature on EBM, which has argued persuasively that there is no epistemic justification or the strict hierarchical ranking o research methods ound in the srcinal evidence hierarchy (Bluhm 2005; Borgerson 2008, 2009; Feinstein and Horwitz 1997; Goldenberg 2006; onelli 1998; Upshur 2002; Worrall 2002, 2007).Te main conclusions o these critical arguments are as ollows. Te clearest distinction in the hierarchy is that between randomized and non-randomized trials, with randomized trials consistently ranked higher. Proponents tried to argue that this was justified because non-randomized trials tend to overestimate effects, but the circularity o this argument was quickly noted: it would be consistent with empirical evidence indicating differences between the results o the two methods to say that RCs underestimate effects (i the empirical evidence were to be ound, which is also contested). Proponents also attempted to argue that non-randomized trials are “misleading” (implying that RCs, by contrast, were not misleading), but this seems to radically misrepresent the nature o scientific research, which must always be open to the possibility that any method may be misleading or produce alse results (Grossman and Mackenzie 2005; Worrall 2002).Although claims about the special powers o randomized trials in establishing causation have been more persistent, philosopher o science John Worrall has pro- vided a thorough argument against this line o reasoning in his paper, “Why there’s no cause to randomize” (2007). He is particularly critical o the tendency to attribute the benefits brought about by ideal RCs to real-world RCs (such as the ability o randomization to guarantee balanced treatment and control groups, which o course they cannot do in any actual study). Finally, I have argued that claims about the rela-tive ability o various research methods to control or bias have been unounded (Borgerson 2009). Note that none o these are arguments against the RC, which may very well be an excellent method or answering certain questions in medicine. Also, these are not arguments against the use o research evidence in medical deci-sion making or against distinguishing between better and worse studies o the same type. Instead, these are arguments against the strict lexical ordering o RCs above other research methods as is the case in the srcinal hierarchy o evidence.
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!