Understanding Nutrition Research
Have you ever noticed that the media are constantly reporting findings from yet another nutrition research study? Knowing which types of studies are the most reliable is helpful, according to Rachel Johnson, PhD, MPH, RD, who presented on this topic at an American Dietetic Association (ADA) meeting. IDEA author Cathy Leman, RD/LD, draws on the ADA session to explain the different types of research, from the most to least reliable.
Multicenter, randomized control clinical trials, typically conducted on human subjects to evaluate the benefits and risks of a given treatment, provide the most reliable research. Study groups are randomized; there is a control group; investigators are blinded/masked; and the sample size must be large enough to allow differences to be detected. One example is the ongoing Women’s Health Initiative, which recently reported its findings on the risks of taking hormone replacement therapy that combines estrogen and progestin.
This type of study has its advantages: Conditions are at least partially controlled, and a cause-and-effect relationship can be observed. However, large-scale studies are expensive, and maintaining enough participants can be a challenge.
Longitudinal prospective cohort studies are observational examinations that compare the outcome of one group that is exposed to a potential hazard or disease with the outcome of another group that is not exposed. Unlike clinical trials, which focus on the risks and benefits of experimental treatments, cohort studies focus on hazard exposure versus nonexposure. One example is the Nurses’ Health Study, which, among other things, has provided information on body mass index and its relationship to stroke and heart disease.
One advantage of these studies is that they focus on exposure before a disease develops. On the other hand they require a large number of subjects; a lengthy follow-up period to allow the disease time to develop; and a lot of money.
In case/control studies the focus is on subjects who have already developed a given disease. Researchers compare the afflicted participants’ past exposure to suspected risk factors with the exposure of controls who do not have the disease. Case studies are analyses of persons or groups and usually focus on social, psychological or medical phenomena. They serve as catalysts for critical thinking, but they are not as scientifically rigorous as other forms of research. Anecdotal studies provide information from personal experiences. They may spur further research, but in themselves are not based on fact or proper study design.
To help make sense of media reports, stay current with reliable resources such as the ADA (www.eatright.org), Tufts University’s Nutrition Navigator (www.navigator.tufts.edu) and life science journals. (To access these journals click on “PubMed Central” at the National Center for Biotechnology Information, www.ncbi.nlm.nih.gov.) Always keep in mind which type of study is being
discussed.
There’s a unique component to nutrition research: It involves recording food intake to establish links between nutrition and disease. Researchers use food records, food-frequency questionnaires, 24-hour recall and dietary histories to collect data. Unfortunately research subjects commonly underreport their dietary intakes.Food underreporting happens for several reasons. First, tracking intake requires a high degree of cooperation from the subjects. Second, subjects often want to appear to be eating less than they really are. They may underreport food intake because of psychological factors, such as body dissatisfaction or an unwillingness to mention “bad” foods. So what’s the bottom line? Conventional wisdom among nutrition experts is that underreporting makes it difficult to obtain a realistic picture of what clients are eating, and any conclusions based on underreported data may not be reliable.
