Systematic Reviews: Clearing the confusion in nutrition messages

Wine, chocolate, soy products, and more have starred on both the "eat this" list and the "not that" list. How can scientific studies done on the same food yield such contradictory results!? Well, there are factors that affect overall quality, applicability, and strength of research results. If researchers don't account for these in their study design, they could skew results.

1. Baseline exposure – Unlike a drug trial, people already have a level of nutrients in their diets when they enter a study. This level can be affected by food choices, supplements, and at times endogenous synthesis (e.g. vitamin D from sun).  
2. Nutrient status – Refers to whether a person is nutrient replete at the start of the study.  Current diets are not always an indicator of nutrient status.  For example, someone may have a micro-hemorrhage that could affect nutrient status.
3. Bioequivalence of different chemical forms of nutrients – Nutrients occur in multiple forms that differ in biological activity. It's important to calculate nutrient equivalents (like acknowledging folate to be essentially the same as folic acid).
4. Bioavailability of nutrients – There are factors that affect the body’s absorption of the nutrient. For example, tannins blocks the absorption of iron. If you drink tea during your steak dinner, you're getting far less iron than you might think from that slab of meat.
5. Multiple and interrelated biological functions of a nutrient – Since most nutrients have multiple biological functions, it’s important to clearly define the nutrient-specific scope of the study. 
6. Undefined nature of nutrient intervention – Food-based nutrient interventions make it hard to quantify change in intake. For ex, if you’re measuring EPA and DHA by asking subjects to eat fish, there are variable levels of these nutrients within species of fish, or depending on the time of year the fish were caught, or the animal husbandry practices for farm-raised fish.
7. Uncertainties in assessing dose-response relationships – Measurement procedures across studies can be subject to error, bias, or inconsistencies. 

What's the solution? Well, systematic reviews exist to clear up some confusion. A systematic review is a paper that tries to sum up the best available research on a specific science question. Like, "Are soy products good or bad for your health? To answer that question, a research team scours the scientific literature for reliable studies. They apply critical appraisal skills to understand the methods and results of studies and assess the research quality. These reviews can be done on nutrition topics too, and in that case the research team weighs factors 1-7 above to decide just how influential results from a particular study should be on their final consensus. The procedure for researching and writing the systematic review is thoroughly documented, completely transparent and fully reproducible at every stage, proving the research team that writes it has no vested interest in the outcome, and that the report is free of bias. This is important because systematic reviews are often referred to when policy makers need to develop health practice guidelines.

References:
1. LichtensteinAH, et al. Application of systematic review methodology to the field of nutrition. J Nutr 2008; 138 (12):2297-2306.

2. Hill A, et al. What is critical appraisal? Skills Programme, Institute of Health Sciences, Hayward Medical Communications 2001; 3 (2):1-8.