answered by different types of research designs. For example, systematic reviews of crosssectional studies would best respond to diagnostic issues. Questions about the likelihood of causing harm would be best answered by either a systematic review of randomized n-of-1 trials for common harms and rare harms, by a systematic review of case-control studies, or by studies revealing dramatic effects. Prognosis questions would be best answered by systematic review of inception cohort studies. There often is no best evidence, but a clinical guideline is still needed by health-care practitioners. Therefore, in addition to attending to the kind of study needed, the reviewer needs to explicitly describe the strength of the evidence supporting a specific guideline. Such description provides the health-care practitioner an indication of the level of certainty of a guideline recommendation (ADA, 2010; Howick et al., 2011). Guidelines with weak supporting evidence need to be updated as new evidence becomes available.
The Bradford-Hill criteria provide a framework for epidemiological research demonstrating causality between environment and disease states (Hill and Bradford, 1965). These criteria can be applied to the development of EBGs because they help determine whether an association between an intervention and an outcome is causal. For the task of this committee, establishing a causal relationship between a nutritional intervention and a TBI outcome would rely on the following: strength of evidence, consistency of evidence, specificity, temporal relationship (temporality), biological gradient (dose-response relationship), plausibility (biological plausibility), coherence with existing knowledge, experimental testing, and analogy (consideration of alternative explanations).
The second principle of evidence-based medicine is that the clinician uses judgment when weighing the trade-offs associated with alternative management strategies, including consideration of patient values and preferences as well as societal values (Guyatt, 2002; Guyatt et al., 2000).
Evidence-based guidelines are developed by professional organizations, health-care organizations, or other nonprofit, disease- or condition-specific organizations (DoD, 2008; Grilli et al., 2000; Knuth et al., 2005; Thomas et al., 1999). There are well-established procedures to systematically review and synthesize the research best suited to answer the clinical questions faced by health-care practitioners. More than 40 clinical guidelines for TBI were identified at the National Guideline Clearinghouse online database.1 However, many of these guidelines focus on emergency department treatment or evaluating for the presence or absence of TBI in primary care or sports settings, and only a few address nutritional concerns.
In addition to guidelines developed specifically for TBI, generic evidence-based clinical practice guidelines for critical care of adults in intensive care units may also be appropriate in acute TBI. For mild TBI, other EBGs might also be appropriate based on additional conditions, such as obesity.
For this report, EBGs from the following organizations were selected for more comprehensive evaluation because of their relevance to TBI: the American Society of Parenteral and Enteral Nutrition (ASPEN), the Society of Critical Care Medicine (SCCM), the American Dietetic Association (ADA), the Brain Trauma Foundation, the National Neurotrauma Soci-
Available online: http://www.guideline.gov/ (accessed October 26, 2010).