Sum-product networks (SPNs) have recently emerged as a novel deep learning architecture enabling highly efficient probabilistic inference. Since their introduction, SPNs have been applied to a wide range of data modalities and extended to time-sequence data. In this paper, we propose a general framework for modelling sequential treatment decision-making behaviour and treatment response using recurrent sum-product networks (RSPNs). Models developed using our framework benefit from the full range of RSPN capabilities, including the abilities to model the full distribution of the data, to seamlessly handle latent variables, missing values and categorical data, and to efficiently perform marginal and conditional inference. Our methodology is complemented by a novel variant of the expectation-maximization algorithm for RSPNs, enabling efficient training of our models. We evaluate our approach on a synthetic dataset as well as real-world data from the MIMIC-IV intensive care unit medical database. Our evaluation demonstrates that our approach can closely match the ground-truth data generation process on synthetic data and achieve results close to neural and probabilistic baselines while using a tractable and interpretable model.