Clinician certainty with the evidence: In practice, it's complicated

Abstract

Clinical decisions about nutrition care are made every day. Unlike the highly controlled selective environments of research settings, nutrition decisions are typically performed within the complex, comorbid, messy world of hospitals, communities, families and homes. Evidence-based medicine has been defined as “the conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients”.1 While dietitians have reasonable confidence in appraising individual studies, there is an acknowledged gap in workforce confidence in knowing how and when to implement change in practice when new or updated evidence emerges.2 As health professionals, we want to make good evidence-based decisions about nutrition care, but not all evidence is created equal, and what does one do when working in an evidence-free zone? Creating an acceptable hierarchy of evidence can make it easier for clinicians to appreciate the quality of research at a glance.3, 4 Definitions for levels of evidence for clinical decision making are based largely on the quality of experimental design (eg, a randomised controlled trial [RCT] is deemed a higher level of evidence than case-control or cohort studies). Although double-blind RCTs are considered the highest level of experimental design, they have their limitations in human nutrition science and, indeed in some circumstances, may be implausible, unethical or unrealistically expensive to even conduct. Even the best-designed studies are never perfect. Pragmatic compromises on study design, inclusion criteria, randomisation and completion rates can introduce bias and a level of uncertainty in the results. The risk of bias is based on the likelihood that features of the study design or conduct of the study may give misleading results. Assessing risk of bias can indicate the degree of certainty one can have in the results presented, and there is a variety of tools to assist.6-10 A robust synthesis of the strengths and weaknesses of results from all levels of evidence is important in nutrition science, especially when we may need to rely on studies with less-than-perfect designs. Understanding levels of, and certainty in, evidence is an important component of evidence-based practice. Systematic reviews of the effects of nutrition care or service delivery, which synthesise the available evidence and critique the bias and certainty across studies, provide essential information for dietitians making well-informed decisions about nutrition care.11, 12 This issue of Nutrition and Dietetics includes a series of recently accepted systematic reviews that all offer clinicians a critique of currently available evidence (or lack thereof) that can assist decision-making during the course of clinical practice and dietetic service delivery. A well-conducted systematic review can enhance clinical practice in many ways. It can provide clear support for or against steps in the nutrition care process; it can highlight gaps in our current knowledge, therefore driving research agendas; or it can provide a leverage to lobby for systems change to align with best evidence. Systematic reviews often start with curiosity over an unanswered clinical question, and through the use of robust methodologies, a synthesis of the available evidence can influence clinical practice. The systematic review by Lukomskyj et al13 investigates the widely held belief about cardiometabolic benefits of breakfast consumption. This is an example of a popular clinical question (‘tis the most important meal after all) that cannot be answered by assessing any single time point but requires long-term, longitudinal cohort studies that follow participants for many decades of life. Perhaps unsurprisingly, only four cohort studies from two countries were identified. The synthesis of results support the benefits of breakfast consumption across the lifespan, particularly in regard to body composition outcomes, but conceded the critique that quality of evidence identified low certainty of results with a high risk of bias. The intense promotion of low-carbohydrate diets across multiple platforms stimulates the need for clinicians to be aware of the quality of evidence assessing the effectiveness of low-carbohydrate, high-fat diets in patient populations at risk of chronic disease. The systematic review by Ross et al14 restricted examination to only those studies of overweight and obese participants within RCTs (the highest level of experimental design) and explored diabetes- and cardiovascular disease-related risk markers from eight available studies. While short-term benefits to weight, blood glucose and blood pressure were universal, adherence is likely to be poor beyond 3 months, as is the case with most restrictive nutrient-focused diets. This review brings to the forefront the dietitians' dilemma of balancing data of efficacy with effectiveness in real-world practice and how best to translate efficacy outcomes into sustainable practice. Dietitians are an integral part of multidisciplinary teams providing care to manage side effects of chemotherapy and radiotherapy for cancer. Oral mucositis is a painful inflammation or ulceration of the oral mucosal membrane and is experienced by almost all patients undergoing radiotherapy for head and neck cancer. Edwards et al's 15 systematic review aims to evaluate the quality of evidence for nutritional interventions for oral mucositis. Their review captures a broad spectrum of studies, including five systematic reviews with meta-analyses, six systematic reviews, two RCTs, one non-randomised controlled trial and four prospective cohort studies. They found varying certainties in evidence for a total of nine nutritional intervention areas for the management of oral mucositis, which may have important clinical implications. Arming oneself with the most up-to-date evidence for commonly requested nutrition interventions allows the clinician to have open and transparent conversations about how best to achieve patient-important outcomes. The relationship between antioxidant supplements and all-cause mortality has typically been plagued by heterogeneity across studies of adults. This highlights complexities in comparing assessment methodologies across studies and the confounding effect of comorbidities in diverse patient populations. The systematic review by Das et al16 reviews for the first time the results of antioxidant supplementation in older adults (>65 years) and the relationship with all-cause and cause-specific mortality. In identifying 22 studies (of which 6 were RCTs), they identified that evidence was low level and heterogeneous, with 9 studies showing significant decreases, 4 finding significant increases and 9 reporting no association between antioxidant intake and risk of mortality. Two of the systematic reviews in this issue of Nutrition and Dietetics highlight the underinvestment in research designed specifically to target key patient populations to inform specialist dietetic services.17, 18 Medically compromised patients with eating disorders experience some of the highest mortality rates of any psychiatric disorder,19 and yet high-quality research to inform dietetic practice is sparse. The systematic review by Jaffe et al17 is a prime example of dietitians asking important practice-based questions about management decisions within an environment of conflicting clinical practice guidelines. Their systematic review aimed to assess evidence for the use of high-energy goal intakes for medically compromised eating disorders and identified only four studies comparing to lower-energy goal intakes, all of which were non-randomised cohort studies, three of which were retrospective in design. By using a robust critique of the levels and certainty of evidence and risk of bias, they unveiled a stark gap in high-quality evidence informing current management of hospitalised people with eating disorders. A call to action for addressing the need for high-quality research studies in vulnerable populations is needed. Alston and Partridge18 systematically reviewed dietary interventions conducted specifically for rural Australian populations over the last two decades. Over this time period, there were 12 unique studies identified with an RCT or pre-post intervention design, predominantly focused on obesity and chronic disease management strategies or improving food environments. While they identified that the scope of literature is limited, their critique highlights the potential for considerable benefit when interventions are specifically targeted to regional communities. It was also clear that inequities and inadequacies in the evidence base remain, which are likely barriers to developing good rural health policy for the prevention and treatment of preventable diet-related chronic diseases. And finally, Rushton et al20 demonstrate how a systematic review can assist decision-making related to workforce demands and dietetic service design. Rushton et al20 approached their review as an investigation to whether delegation of malnutrition care activities to dietitian assistants impacted patient, healthcare and/or workforce outcomes in adult hospital inpatients with or at risk of malnutrition. Of the 11 studies identified (2 of which were randomised in design), there was support for broad benefits of delegation models to improve nutrition care delivery, with likely positive impact on patients, healthcare systems and the multidisciplinary workforce. While good evidence alone may be inadequate to stimulate systems change,21 it is a critical component to progressing the translation of research into practice in a complex and resource-limited clinical setting. This review should speak to the leaders designing how the broader clinical dietetic profession will function into the future. When faced with limited or low-quality evidence, dietitians need to be flexible in their decision-making regarding appropriate practice: drawing on professional experience, being aware of and critiquing data as it becomes available and ensure patient preferences have a substantial influencing role in decisions about nutrition care. As health professionals, our relationship with evidence is complicated. We need to integrate the best available scientific information with clinical judgement and patient-centred care to guide decision-making about nutrition management. Be reflective on what preconceptions, assumptions and dogma you bring to your practice. When faced with new or updated evidence, consider what it might take to influence and evolve your practice to align with best evidence. At a time when trust in science and scientists is so critical, we cannot simply trust that all research is equal when it comes to influencing practice. Systematic reviews are a valuable way for health professionals to integrate the most up-to-date data into their daily decision-making about care and how it is delivered.