org355: Unexpected Ice Cream

From the New York Times. This article is notable for being about an epidemiologist with a perceived conflict of interest.

1. Why do you think a trial was justified?

2. Should scientists be prohibited from managing studies when they already favour one outcome?

And an episode of the Milk Street podcast. The bit in question comes very early in the episode. (And here's a longer article from The Atlantic, where the journalist interviewed on Milk Street gets deeper into the epidemiological weeds.)

1. What is the RR figure to get 22%? What is the PICO (population, intervention group, comparison group, outcome).

2. What is the hypothesised mechanism?

3. What would you do if you did a study and had an unexpected finding like this?

Re: Unexpected Ice Cream

by | GEORGE LAWSON - Tuesday, 5 December 2023, 3:53 PM

NYT:
1. Systematic reviews have in some cases suggested some protective effect of low dose alcohol, eg <3 units per day, in cardiovascular disease.
If it is thought that even at low doses alcohol is actually harmful then maybe a study focusing especially on low dose alcohol and especially attempting to control confounders - social elements, affluence etc could be warranted. But would have to look at literature more deeply to see if that question had been effectively answered already.

2. I don't think that favouring a particular outcome should prohibit a scientist from managing a study, I imagine most studies are run by scientists who think there is going to be a particular outcome and are setting out to prove it. However industry funding complicates the matter, especially when there have been concerns raised previously that have been sufficient to stop a trial due to conflict of interest and "violation of federal policy."

MS:
1. men eating 2 servings of ice cream vs men not having skimmed/semi-skimmed milk, risk of diabetes mellitus
P: men with no comorbidities
I/E: dairy fat intake as proportion of total energy intake
C: difficult to find - is it low dairy intake i.e. the lowest quintile of dairy intake
O: T2DM

PECO for ice cream specifically
men with no comorbidities, E two or more servings of ice cream every week, C: lowest quintile of dairy intake, O: T2DM

2. only happy healthy people eat ice cream!
milk fat globule membrane and relative abnudance of whole food in ice cream vs other apparently more healthy food
3. check the numbers, consider confounders, look at other literature, check the numbers again!

Re: Unexpected Ice Cream

by | Sujit Rathod - Friday, 8 December 2023, 10:19 AM

Thanks George, for tackling this tricky issue!

More on the topic, from the New York Times.

Re #3: What will you do after you're satisfied with the numbers, confounders, and literature? (Also, I'll encourage everyone to do these checks even if you get the findings you expected/hypothesised, and moreover to pre-specify the checks in an analysis plan...)