The article introduces a new way to measure the effects of changing the chances of receiving a treatment, rather than just looking at the average effect of getting the treatment or not. This method can be helpful when the usual assumptions about treatments don't hold. The researchers show how this method can be used with different types of treatments and time points, and provide ways to estimate and interpret the results. This approach can be useful in real-world situations and may lead to new ways of understanding causal effects.