Bellevue Nutritionist, Angela Pifer writes: A new study, based on the Nurses’ Health Study (NHS), was recently published with the finding that “women who drink moderately are less likely to gain weight over time than those who don’t.” The study did not find a causal link and only offers loose theories to explain their findings. The authors cautioned that they are not suggesting people go out and drink more alcohol as a weight-control strategy. I am wondering why this is even news. This is just one more example of a lead ‘health news’ headline that will further confuse people who are trying to make healthier decisions. My issue with this headline doesn’t stop at its ‘non-news’ status. It lies deeper into the very fabric of the study design.
Many recent headlines have been based on studies, based off the data collected from the Nurses’ Health Study I, II and III. The Harvard Fertility Diet among them. The NHS has had extremely large study populations, following well over 100,000 registered nurses over the past few decades. This study population is almost unheard of in the scientific community where a good majority of the studies are based on populations well below 50 participants. This makes the NHS studies particularly enticing to the scientific community.
Registered nurses enrolled in the studies received questionnaires pertaining to a wide range of health topics. One of these was a food frequency questionnaire. Basically participants were asked to recollect how many servings of this food or that food they ate on average over a period of time. This method of recall is simply flawed bringing many of the studies based off the NHS data linking health conditions, labs and disease to eating patterns under question.
If at the end of your day, you sit down and try to recall what and how much you ate and drank, you will probably get some things wrong. You may have forgotten that you ate something or miscalculated the amount you did eat. Would you be able to accurately identify if you had two servings of fruit and three servings of vegetables? Imagine then, going back and trying to accurately recollect how many servings of fruit or vegetables you had over a week, a month or a year. Now consider how accurately you could identify your patterns over four years, as was done with the Nurses’ Health Study II. Food frequency questionnaires were mailed to participants once every four years.
In my private practice I experience daily a person’s inability to accurately recall what they eat and drink. During our first sit down consultation, my new patient is all too eager to share with me their ‘typical’ day. I obligingly write this down knowing that after they journal for me for four days eating as normally as possible and recording their intake in real time, their actual intake will differ greatly from their ‘typical’ day. My point is that no week is really ‘typical.’ People seem to believe that they eat a specific way when in reality it is often another. They may like to think they eat on average three servings of fruit each day, but in reality it may only be one.
The creators of this study believed that because of their training, nurses would be able to respond with a higher degree of accuracy to brief, technically worded questions. I disagree with this point specifically in regards to the food frequency questionnaires. I took the undergraduate course Nutrition and Nursing at the University of Washington. This is the only nutrition class nurses receive during their training. The class loosely covers the food pyramid, the exchange system and offers a very basic education on nutrition fundamentals. With all the amazing things that nurses’ do in their job, offering nutritional advice, specific to portions, servings and meal plans are not among them. Hospitals use dieticians and diet techs to perform this task, so the information that nurses learned in the class is not practiced, making them no more likely to recollect how many servings of fruits or vegetables they had in their diet over the last four years, than any other person.
These same points carry over to how this study was performed; the study looking at alcohol intake and weight involved 19,220 registered nurses from the NHS over the age of 38 who were of normal weight. Researchers asked the women (questionnaire) about their alcohol consumption over the past year and recorded how much of four different types of alcoholic beverages they consumed — beer, red wine, white wine and liquor. This group of nurses was followed for thirteen years.
See if you can answer these two questions: If you drink wine in the evening how many glasses do you have? In that same evening, how many servings of wine did you have? You may or may not be surprised to learn that the answer to these two questions is most likely not the same answer. Most wine glasses hold two to three servings. So you may think that you drank one glass of wine and had one serving, but it is more likely that you had one glass of wine and two servings. Most nurses may know that 3.5 ounces of wine is a serving, but do they also know how many ounces of wine their wine glass holds and how many ounces they actually pour? I have my patients pour themselves their usual serving and then pour that into a Pyrex container to measure their pour. This exercise is very enlightening and is really the only way to get an accurate measure of how much someone drinks. I would be more impressed with a study design that had participants keep a daily journal of their drink intake (after being taught how to properly measure a pour or better estimate the alcohol content in a mixed drink). This can easily be done online with a daily email reminder.
I do acknowledge that it is very difficult to have controlled studies with human participants. Let’s face it; it would be hard to find 100 people who want to stay in a science lab for one week let alone a month. We do need to rely on other means of collecting data. I respect the work that my fellow colleagues are doing. I simply feel we should place less emphasis on these types of studies and not create a media frenzy around unfounded, uncorrelated findings that lead to further confusing people trying to make healthier choices.
Angela Pifer, MSN, CN Bellevue Nutritionist
Seattle Weight Loss Programs – I work Nationally through Skype
Tags: Drink, Study, Thinner, Weigh Less, Women
Categories: Current Affairs, Uncategorized
