So many questions, so few answers!

By Lori Bricker, MS, RD

I hate giving people answers they don't want to hear.  As a tea shop owner, I get a lot of questions about the health effects of tea.  Since I have a background in nutrition, I have some insight into how complicated the question really is, so my answers are never what people expect.  People ask pretty specific questions, but sadly, the specific answers they want are not available yet.

Here are some really good questions that have no known answers:

    •    Which tea is the most healthful?
    •    Which tea has the least caffeine?
    •    Which tea will cure my [fill in the blank]?
    •    How much tea should I drink in a day?

The science of tea (Camellia sinensis)  is too young to offer definitive answers to these and other popular questions. But there are a lot of studies being done, and many get written up in the popular press.  So many, in fact, that it's easy to get the impression that tea is a real cure-all.  I'd like to talk about why you should be cautious when you read these articles.  For this post, I'll just talk about cohort studies, and another post will discuss research studies.

Here's what you need to know about cohort studies:

  • Cohort studies involve large numbers of people (not rats!) who have something in common and following them for several years, then studying data collected during that time and to look for trends in the various outcomes.  For example, you might look at everyone born before 1950 in a particular town, and follow them for 15 years to determine the rates of a certain disease.  You compare the disease trends with the information the participants give you about their diet, health history, medications, etc.
  • Their major strength is that because cohort studies last so long, you can see real outcomes, like death and disease, that you just don't get with shorter experiments.
  • The downside is that lots and lots of things happen to the people in your cohort, and you have less control over it than you would in an experiment.
  • Cohort studies usually use questionnaires to gather data from the participants.  For food and beverage intake data, researchers use a tool called a food frequency questionnaire, which will ask you potentially hundreds of questions such as, "How often do you eat broccoli in a month and what is the serving size?"  By the time you get to the end, your mind is mush.  Even in a well-designed food frequency tool, self-reported data is only as good as the participants' memories, patience, and honesty.
  • Cohort studies usually do not show cause and effect.  They show correlations, but they don't tell if one thing caused the other. To illustrate the point, you might notice a correlation between the occurrence of rain the the appearance of umbrellas.  Of course, you would be wrong to conclude that umbrellas caused the rain.

Another confounding factor for ANY tea research is that tea preparation methods can affect how concentrated a brew you are consuming, which in theory should affect the results.  I might brew an 8 oz. cup of tea for 8 minutes in boiling water using 3 tsp of leaves.  You might drag a tea bag through 8 oz. of tepid water and be done.  There could be a huge difference in the amount of tea catechins and other active compounds in these two beverages.

Now, of course, the scientists doing this research know all these things, but sometimes these concerns don't get addressed in media stories about the research. 

When reading an article in the popular press about a research article involving a cohort:

  •  Read carefully to see if they use the word, "correlation," and if so, don't misread causation into that.
  • Notice details about the cohort itself– is it all women, all men, all people from one country or another, do they have any diseases, what kind of tea do they drink?
  • What do they say about the kind of questionnaire used?  Was it filled out just once at the beginning or multiple times?  Do they tell you what they asked regarding tea consumption?
  • Read more than one article about the same research to see if you get a consistent message.

Be aware that it takes mounds of research before you can really say something with confidence. In this online article reporting on a heart disease study that involved tea and coffee, Dr. Frank Hu, professor of nutrition and epidemiology at the Harvard School of Public Health, is quoted as saying,

"Based on current evidence, it is very difficult to come up with an optimum amount of coffee or tea for the general population." 

After years of research on tea (and coffee), this is where we are?  Yes, and we are likely to be here for a while.