“What you mean, he don’t eat meat? Oh, it’s OK, I make him lamb!” (quote from “My Big Fat Greek Wedding”, 2002 movie, released by IFC films).
In our house, my family frequently works favorite movie quotes into the daily dialogue…Maybe more than other houses (a topic for yet another post)…
What does this have to do with Epidemiology? And why I care…
In the recent news, there have been several articles posted (see links & titles below) regarding a newly released meta-analysis in the March issues of the Journal of the American College of Cardiology, entitled “The Effect of Mediterranean Diet on Metabolic Syndrome and its Components”.
Mediterranean diet linked to lower heart disease risk (Kirk McKoy / Los Angeles Times)
March 8, 2011 |By Mary Forgione, Tribune Health
Why to Eat Like a Greek By RON WINSLOW
Several terms here require definition. Metabolic Syndrome (MS) is a fancy name for a combination of risk factors for cardiovascular (heart and blood vessels) disease. This includes high blood pressure (hypertension), high blood fats (hypercholesterolemia), being fatter than you should be (obese, especially around the belly-area, shaped like an apple, some definitions say women with a waist measuring 40 inches, or over, for men, and 35 inches, or over, for women), and having increasing blood sugar (insulin-resistance) (Kastorini, et al., 2011). The World Health Organization (WHO) has said that MS is increasing in prevalence (how much disease/condition exists in a specific population during a specific time) (Gordis, 2009, p. 43) throughout the world, along with diabetes and obesity. This is an increasingly serious, global public health problem (Kastorini, et al., 2011), with prevalence estimates of 13.4% to 70% (Kastorini, et al., 2011).
The Mediterranean Diet has been talked about since the 1960’s. It is loosely defined as a eating a higher amount of olives/olive oil (monounsaturated fats, avocados have this too), along with larger daily portions of vegetables, whole grains, legumes (beans), fruits, and low-fat dairy products. Every week, fish & poultry are included, along with nuts from trees (almonds, walnuts, pecans, pistachios). Moderate daily alcohol intake is included (some studies say primarily red wine with meals). Red meat consumption (such as beef, lamb, & pork) is limited to once, or twice, a month (Kastorini, et al., 2011). Higher garlic consumption has also been included in some definitions. Numerous studies have linked this type of diet as being “cardioprotective” (protecting the heart & blood vessels from damage and disease).
(FOOD ALERT….By the way, this is an easy way to eat! Our menu last night consisted of olive-oil & garlic sautéed mahi-mahi, spinach salad with almonds, strawberries, low-fat feta cheese (drizzled with balsamic vinaigrette/olive oil dressing…should have added avocado, or olives!), brown rice, and steamed broccoli, with a glass of red wine, was very, very tasty! The focus is on keeping foods simple, using as few “processed” foods as possible…This was a very quick meal to fix as well).
So, just to be clear, the “exposure” (epidemiological word for the thing being studied to see if it is related to a good, or bad, health situation/outcome) in this study is the Mediterranean Diet. The “disease” (word for health situation/outcome being studied) here is Metabolic Syndrome.
Only this isn’t just one study. It is a combination of 50 different studies, with over 500,000 participants, called a meta-analysis. A meta-analysis looks for common findings. If it is done well, it searches scientific literature for the exposure and disease being examined, weeds out the studies that were poorly done, and then measures the strength of the associations noted in the studies (Gordis, 2009, p. 342). At first glance, it is tempting to believe that a meta-analysis enables one to draw more powerful conclusions (after all, this study includes 50 studies, with over 500,000…IT IS BIG!). However, its reliability and generalizability (can you trust its conclusions? can you apply the results to many populations/situations?) are only as good as the studies you put into it…as the old saying goes, “Garbage in, garbage out”. Also, where a more rigid definition of a meta-analysis might only use a combination of similar studies of one type (such as randomized control trials), more and more meta-analyses are combining several study types (clinical trials, case studies, cross-sectional studies, etc.) (Gordis, p. 342). So, if you are comparing “apples to oranges”, then, how reliable are the results of the meta-analysis?
One particular noted limitation of meta-analysis is that the results can be affected by publication bias, as in only published studies are included. The reasons some studies get published in the first place may not be un-biased…such as readers are more interested in a specific topic with specific results (Gordis, p. 343)…(a current non-scientific media example would be all the attention garnered by the antics of actor Charlie Sheen! Mental illness and substance abuse are more worthy topics, however, nothing satisfies the American public better than a big, juicy public meltdown…). So the amount of published studies on a specific topic may not truly reflect risk, or results).
(More on this particular meta-analysis later today…).
No comments:
Post a Comment