There are some other issues with this study that might strain credibility, but yes, I think you are mistaken about the size of the effect being reported.
At low prevalence probabilities, odd ratios are more or less just multipliers on the prevalence probability.
So if there is a 0.1% probability of developing Crohn's without childhood antibiotics, an odds ratio of 3.5 would mean that there is approximately an 0.35% probability of developing Crohn's with childhood antibiotics. That's 100 in 100k vs. 350 in 100k, and in my opinion that's not a small difference.
More generally, an odds ratio is the ratio of the odds of something under two different sets of circumstances.
At low prevalence probabilities, odd ratios are more or less just multipliers on the prevalence probability.
So if there is a 0.1% probability of developing Crohn's without childhood antibiotics, an odds ratio of 3.5 would mean that there is approximately an 0.35% probability of developing Crohn's with childhood antibiotics. That's 100 in 100k vs. 350 in 100k, and in my opinion that's not a small difference.
More generally, an odds ratio is the ratio of the odds of something under two different sets of circumstances.