Author: Dr Veronique Chachay
As a key component of evidence-based healthcare, nutrition research provides insight into a number of significant issues: the metabolic consequences of under- and over-nutrition; the food constituents with health-promoting properties; and the complex diet-disease relationships that underlie conditions such as cardiovascular disease, diabetes and cancer. However, nutrition research is often considered a “soft” field of research. This is due to the limitations of instruments used to investigate the dietary intake of individuals, and the difficulty in demonstrating causality when many nutrition-associated biomarkers are slow to progress, or may be influenced by other factors. For example, while an antioxidant-promoting diet is correlated with cardiovascular health, the benefits of this in a controlled research trial may not be visible for several years. Surrogate biomarkers are often used to assess the effect of the diet, such as metabolites of oxidative stress (F2-isoprostanes), as indicators of the imbalance between free radicals and antioxidant compounds.
Scholarly investigation of a patient’s diet-disease relationship requires a thorough record of their usual dietary intake, which is subsequently analysed with food composition software. A number of methods are used, including:
24-hour recall, in which individuals verbally report their dietary intake over the previous 24 hours.
A diet diary, in which an individual records their food and beverage consumption over a specified period, often 3-7 days.
A diet history interview, in which a nutrition professional aims to capture patients’ long-term eating habits and patterns, including cooking methods.
A food frequency questionnaire, through which patients report their long-term frequency and quantity of intake of specific foods.
However, any quantitative analysis derived from these assessments must be interpreted in the context of several limitations. Firstly, all of these methods rely on self-reporting, raising a number of possible biases. For example, is a given individual’s report accurate, or could recall bias or fear of prejudice be at play? Is the individual accurately recalling all consumed items, including all ingredients? Is the individual estimating portion sizes reliably, and in standardized units?
Secondly, these methods often fail to adequately capture participants’ actual dietary intake. For example, diet diaries are well recognised as behaviour-modifying because of the associated self-monitoring that takes place when recording one’s food intake. Similarly, 24-hour recalls are often poorly representative of a patient’s usual daily food intake, as circumstances may change from one day to another. Ideally, a combination of methods should be used in order to “triangulate” the data obtained; however, this raises the complexity of the procedure, increasing the amount of resources required, as well as potentially increasing the likelihood of participant drop-out.
Can N-of-1 designs improve the quality of data collection in nutrition research?
N-of-1 trials and Single-Case Experimental Designs (SCEDs) use repeated measurement of an outcome of interest, often collect via digital methods (e.g. wearables, mobile apps, online questionnaires). N-of-1 studies can be combined with sampling frameworks that focus on real-time measurement in an individual’s natural environment (e.g. Ecological Momentary Assessment). Investigators can prompt an individual for their nutrition data at frequent intervals, which could lead to a reduction in recall bias and estimation errors (e.g. portion size and food composition). Furthermore, N-of-1 methods can facilitate ‘personalized interviewing’, where the design of the study is tailored to the unique preferences and needs of the individual. This would not be feasible in Randomised Controlled Trials (RCTs), which typically require large sample sizes.
Furthermore, as clinical nutrition counselling already applies a personalised approach to patient prescriptions to enhance adherence, formally applying this through N-of-1 trials and SCEDs makes perfect sense. Altering dietary habits can represent a significant burden to patients, because food and eating are so prevalent in our daily lives. Therefore, dietitians tailor a dietary plan to address health goals, based on information they gather initially from the patient: dietary pattern and habits, food preferences, readiness for change etc. in order to enhance adherence for optimal outcomes.
Opportunities in weight loss research
N-of-1 trials and personalised research designs appear particularly promising in the field of weight loss. A number of head-to-head RCTs have demonstrated that there is no one “best approach” to achieve weight loss, but that different methods suit different people. Results are often diluted by large dropout rates (with a 10 to 80% attrition rate being reported in weight loss studies), but results suggest that participants who complete their trials often enjoyed their treatment they received. However, a holistic approach to weight loss must consider maintenance of weight loss achieved. Rebounding to previous excess weight is common after a successful weight loss phase and has been shown to occur for around 50% of individuals. Therefore, intervention personalisation in the maintenance phase deserves thorough investigation because it may hold the key to long term success in keeping weight off.
In interventions looking at increasing adherence through personalisation, accurate participant reports are what guides the personalisation of intervention. While nutrition research may continue to rely on self-reporting, N-of-1 trials have the potential to increase data quality, intervention precision, and thus participation in future trials.
About the Author
Dr Veronique Chachay is a lecturer in Nutrition Science at the University of Queensland and is the ICN Nutrition Theme Co-ordinator.
Comentários