Saturday, September 26, 2015

Week 1, Journal Entry

Once when I did this 8-week challenge we were supposed to write a blog post each week too. I don't know why this part of it isn't in my list of points this time around.

Anyway I'll still try to write something every week about how I'm doing with this 8-week challenge.

First, why did I choose to do this in a time period when Halloween will be happening (second week I'm supposed to cut out sweets and junk food---how will I do that?!?)? Answer: I had to get started, and now was the time, and I'll just have to deal with all that as I go along.

Second, why is keeping a food journal the very first challenge? Because it actually helps a person who is trying to get control of his/her eating habits.

The Web MD people  cite a study which found that "...people keeping a food diary six days a week lost about twice as much weight as those who kept food records one day a week or less."

What's weird about the Web MD link when citing this study, which I expected to link to the actual study, does not actually link to the original study but to another Web MD page with helpful hints for keeping a food diary. Here they are:

  • Write as you go. Don't wait until the end of the day to record what you ate and drank. "We recommend they write it down as soon as they can after they eat," says Stevens.
  • Focus on portion size. Practice at home with measuring cups, measuring spoons, or food scales. And be aware that people tend to underestimate how much food they're served.
  • Use whatever type of food diary works for you. It doesn't matter whether you use scrap paper, a personal digital assistant (PDA), or a notebook. What matters is that you use it, says Stevens.
  • Don't skip your indulgent days. "We encourage people to keep records especially on days when they're tempted to eat," says Stevens. "What gets measured tends to get changed."
  • Cook at home. You'll have more control over what you consume, and you know what that food contains, and how much of it you're eating. That makes for a more detailed entry in your food diary.
 Okay, great hints from Web MD, but not great for people like me who want to read an actual study instead of someone else's summary of it. So I found the study myself, and here it is:

"Weight Loss During the Intensive Intervention Phase of the Weight-Loss Maintenance Trial"

And, oops, there are some problems with the data and the statistical analysis thereof: Over 130 "missing weights" at the end of the study had to be "replaced." Which is not okay. They used the famous (?) "Markov Chain Monte Carlo sampling" method  to guess---yes, I said it---GUESS---what the missing values were and "replace" them. Then, "Parameter estimates were then averaged, standard errors were adjusted with a function of the between-imputation variation, and the degrees of freedom were adjusted to obtain unbiased p-values."

So, what I think is that when you average a bunch of estimates and adjust the standard errors AND degrees of freedom, you're just showing how desperate you are to get your study published. (If I did that for my own scientific work, I wouldn't have said scientific work any more.)

But does keeping a daily food journal help people lose weight? I still think it does, though admittedly the basis for my conclusion is even less solid than the statistical methods used in that study. I think it does because it has helped me in the past.

And I think it does because, as the Web MD people note, "What gets measured tends to get changed."

No comments:

Post a Comment