In this article of van Klaveren, David, et al. "Models with interactions overestimated heterogeneity of treatment effects and were prone to treatment mistargeting." Journal of clinical epidemiology (2019), the authors mentioned:
" ‘‘risk model’’ (with a constant effect of treatment assignment)"
" ‘‘effect models’’ (including interactions of risk predictors with treatment assignment)"
without giving the mathematical formula.
I am a little confused as to the mathematical definition and the difference between these models and I would be grateful if someone could clarify this point with mathematical formulas or give me clear references and which explains these two models well.
Thank you for your help.