Abstract
The use of clinical trials to demonstrate effect of foods consumption on human health has increased significantly in recent years at the global level. As in other areas of human health, some authors choose to use parallel trial designs, while others prefer to use crossover designs for these trials. Because crossover trials have the advantage of reducing the number of subjects needed and the economic cost to be performed, they have many advocates within the scientific community. However, these types of tests also have numerous drawbacks, due to the difficulty of carrying out adequate statistical analyses, the lack of reliable standards adapted to them or confounding factors. In this chapter, the advantages and disadvantages of crossover designs and whether they are a recommended option for human nutrition research are shown. The usefulness of design of experiments coupled to crossover trials, especially when comparing various levels of the dependent variable, are also discussed.
TopIntroduction
Human clinical trials are often designed to assess the effectiveness of an intervention or treatment on human health. Typically, these trials are used to test the efficacy of different therapeutic options in medicine, pharmacology or nutrition. However, in the last decades the use of such trials is increasingly common for the investigation of the effect of other types of habits and interventions on human health, as may be the case with nutrition baseline research. In addition to basic research, one of the most common purposes for conducting nutrition trials is to demonstrate the beneficial effects of a specific food component in order to use this information in the in food labeling. Food products that are labeled with messages related to the promotion of human health are marketed at a price 30–50% higher than their counterparts that cannot use these health claims, and in some cases can reach even more (Miranda et al., 2018). In order to obtain authorization for the use of one of these health claims, it is usually necessary to present a meta-analysis demonstrating the beneficial effects of the food on human health. As an example, in the European Union it is compulsory to include human trials to obtain authorization for the use of health claims in marketing food products (European Food Safety Authority, EFSA, 2011). Following this obligation imposed by the European Union, the number of nutrition-related clinical trials published worldwide increased significantly. As it can be seen in Figure 1, clinical trials related to human nutrition showed gradual growth between 2000 and 2010, which accelerated substantially from that date onwards for both cross-clinical trials and for other types of clinical trials.
Figure 1. Total clinical trials and crossover trials related to human nutrition published from 2000 to 2018 (source: ISI web of knowledge®). Search criteria: total clinical trials were search by “clinical trial” in the field “title” and “nutrition” in the field “topic”. Crossover trials were search by “crossover trial” in the field “title” and “nutrition” in the field “topic”.
Randomized clinical trials are considered the “gold standard” to evaluate therapeutic effectiveness due to its ability to avoid or minimize bias associated with imbalance in potentially confounding variables (Leonard, Lafrenaye & Goffaux, 2012). The most employed randomized clinical trials are the so-called parallel-group trial or design (Harris & Raynor, 2017), in which subjects are randomized to a unique intervention or control group during the entire trial, that occur simultaneously in time. Identical dependent outcome variables are measured in all groups included in the trial, with outcomes compared between groups, or between subjects, with the aim to determine the intervention effectiveness. In contrast, in a so-called crossover design, all subjects receive all levels of the independent variable at some point in the study, but subjects do not receive all levels at the same time (Figure 2). To determine intervention effectiveness, dependent outcome variables are measured for the two levels of the intervention and afterwards, then they are compared within the same subject (Harris & Raynor, 2017).
Figure 2. Basic (2x2) crossover trial design
Key Terms in this Chapter
Randomization: Method based on chance alone by which study participants are assigned to a treatment group.
Carry-Over: Effect that persist from one experimental condition to another. Whenever subjects perform in more than one condition (as they do in within-subject designs) there is a possibility of carryover effects.
Sequence Effects: Potential confounding influences in experiments where subjects are exposed to multiple conditions.
Clinical Assay: Medical research conducted on people who voluntarily participate in these studies and who help discover better ways to treat, prevent, diagnose and understand diseases that affect humans.
Design on Experiments: Set of active techniques that manipulate a process to induce it to provide the information required to improve it through changes in its variables and their interaction or sequence of execution.
Nutrition: Food intake in relation to the dietary needs of the human body.
Cross-Over: Observational epidemiological design to assess whether a given intermittent or unusual exposure may have triggered an immediate short-term, acute event.
Confusing Factor: Variable that influences both the dependent variable and independent variable, causing a mistaken association.