What Test Should I Use?

Hi there,
I am analysing a large data set involving a food-training study, and I am struggling with the last question.
Here are the key facts:

The participants were 83 people aged 18–65 with a body mass index (BMI) of at least 18.5.
All participants were weighed at baseline (Week 1, prior to training).
The training (Week 2) involved four sessions of an online go/no-go response inhibition task, in which participants were required to press a key in response to images paired with ‘go’ signals but withhold their response to images paired with ‘no-go’ signals.

The participants were randomly allocated to one of two treatment groups:
A ‘food inhibition’ treatment in which high-energy-density foods were always paired with no-signals, whereas healthy foods were always paired with go signals
A ‘control’ treatment in which the food images were replaced by images of household objects(randomly paired with go/no-go signals).

After training (end of Week 2), the participants were weighed again and given a final (fifth) training session.
The proportion of errors (GoErrors, NoGoErrors) and reaction times for correct responses (GoRT)on the first and last sessions were recorded.

The participants were then given a free choice test in which their consumption (in kcal) of chocolate (choc_consumption) and crisps (crisps_consumption) was measured. They were contacted again 1 month and 6 months later and asked to report their current weight.

The question is:
People with a BMI of 30 or more are classified as ‘obese’. Did consumption of crisps, chocolate or both predict obesity at the baseline measurement?

I initially performed three linear regressions, one for each snack and then one for interactions. But is there another way I could figure this out?
Any help is highly appreciated!

Kind regards,