How do we know that longer use causes worse withdrawal?

Could it be the case that those that find it harder to stop just stay on the drugs for longer, creating a selection bias?

Would also love to see more studies on stopping and starting multiple times as that could be worse than staying on.