Science of Eating Disorders Tumblr
Tetyana here. I run the Science of Eating Disorders blog. This is the SEDs-associated Tumblr. I post about ED research, (mental) health, psychiatry, and medicine. I reblog pretty art and photography, promote critical thinking, and rant about stuff. Previously answered questions are here. Content is not always on topic and may be triggering.

Ask Archive Random FAQ + READ ME Theme
❝ We aimed to compare subgroups of BN with and without a history of AN in terms of trait- and state-level eating-related and general psychopathology, as well as to examine patterns of negative affect surrounding eating and activity behaviors in these two groups. We found that [the groups] did not differ on trait-like measures of dietary restraint, shape and weight concerns, psychiatric comorbidity, or anxiety. The groups also did not differ on momentary measures of anxiety, dietary restriction, eating disorder behavior frequency, or negative affect preceding or following eating disorder behaviors.


Trait-level and momentary correlates of bulimia nervosa with a history of anorexia nervosa

This paper is also open access.

One of these days we’ll progress from talking about neurotransmitters and start talking about neurotransmitter receptors. Ah, one can dream :-).

Also, I’m back. I’ll get on answering Q’s and filling up the queue soon :-)

From the Archives: Walking a Mile in Your Shoes: Treating Eating Disorders with a Personal History of Eating Problems

Chronic starvation secondary to anorexia nervosa is associated with an adaptive suppression of resting energy expenditure.

Anonymous:  can you explain what statistical significance is since a lot of people get it wrong?


I can! First we have to understand two concepts:

  1. Null hypothesis - suggests there is no relationship or effect between the independent and dependent variable. So if we are testing an antidepressant, the null hypothesis suggests the medication has no effect on depression levels.
  2. Alternative hypothesis - suggests there is a relationship or effect between the independent and dependent variable. So for that antidepressant, this might suggest the medication has some effect on depression levels.

Statistical significance in its proper label is called null hypothesis significance testing. When testing for statistical significance between our independent and dependent variables, we follow a specific process:

  1. Assume the null hypothesis is true and that there is no effect (fun fact: this is never the case, but we assume it anyway).
  2. We apply a statistical model to our data in a way that represents the alternative hypothesis (i.e., that there is some effect) and see how strong of a fit it is while still assuming the null is true.
  3. We then calculate the probability of this new model “fitting” when we assume that the null hypothesis is true (i.e., “if there truly was no effect, then what is the probability of getting the results we see here?”)
  4. If that probability is sufficiently small (often when p < .05), then we assume that the alternative hypothesis’ model fits the data well and we can reject the null.

Ultimately, statistical significance suggests that the observed results are very likely to be inconsistent with the null hypothesis.

A lot of people think that our p value represents the probability that our results are due to chance, or that statistical significance means we can be “95% confident our results are accurate and not due to chance.” These are not exactly correct. Instead, it’s just the probability that our attained results would be very extreme or unlikely should the null hypothesis be true, and because of the low likelihood, we can probably conclude that there is some effect.

What’s important to remember is statistical significance in no way hints at the importance or size of the effect being observed. Some recent article quoted a researcher as calling it “statistically discernible,” and I think that’s a fantastic term. Something can be immensely significant but be completely meaningless, while something can be very insignificant but be incredibly meaningful. Statistical significance plays virtually no role in making that call.

❝ These results suggest that habitual binge/purge behavior may have some influence on circulating plasma ghrelin levels in both BN-P and AN-BP. Habitual binge/purge cycles with vomiting as opposed to binge-eating episodes without vomiting may have a greater influence on fasting plasma ghrelin concentration in eating disorders.

— Habitual binge/purge behavior influences circulating ghrelin levels in eating disorders

I’ll be away for the next few days without internets, so if I haven’t responded to your ask yet, it may take a while :/.