Both statistical and neural methods have been proposed in the literature to predict healthcare expenditures. However, less attention has been given to comparing predictions from both these methods as well as ensemble approaches in the healthcare domain. The primary objective of this paper was to evaluate different statistical, neural, and ensemble techniques in their ability to predict patients' weekly average expenditures on certain pain medications. Two statistical models, persistence (baseline) and autoregressive integrated moving average (ARIMA), a multilayer perceptron (MLP) model, a long short-term memory (LSTM) model, and an ensemble model combining predictions of the ARIMA, MLP, and LSTM models were calibrated to predict the expenditures on two different pain medications. In the MLP and LSTM models, we compared the influence of shuffling of training data and dropout of certain nodes in MLPs and nodes and recurrent connections in LSTMs in layers during training. Results revealed that the ensemble model outperformed the persistence, ARIMA, MLP, and LSTM models across both pain medications. In general, not shuffling the training data and adding the dropout helped the MLP models and shuffling the training data and not adding the dropout helped the LSTM models across both medications. We highlight the implications of using statistical, neural, and ensemble methods for time-series forecasting of outcomes in the healthcare domain.

No abstract

No abstract

Understanding how relationships are structured in physician networks provides insights into how these networks influence physicians' beliefs and behaviors. This understanding would help improve strategies for disseminating medical information and guidelines. But most physician social networks mainly focus on a binary relationship where either one physician is connected or not connected to another physician without any description about the strength of the relationship. This binary relationship can lead to misinformation in the network (as acquaintances and close friends may be treated equally). In this paper, we overcome the limitation of the binary relationship by proposing a weighted influence approach among a network of physicians. A physician network is a social graph comprising of nodes (physicians) and edges between nodes (social relationships). Specifically, we attach weights to the edges to quantify the strength of the relationship between two connected physicians. In one network, we assume an un-weighted (binary) link between two connected physicians; whereas, in a second network, we assume a weighted link between two physicians. In both networks, edges are created between physicians who are affiliated with the same organization-group and affiliated or working in the same hospital within the same specialty or specialty-group. We compare both the weighted and un-weighted approaches in the network by considering the diffusion of four highly prescribed pain medications in the US. Results reveal that the weighted approach is superior compared to the un-weighted approach network in explaining the diffusion of pain medications inside the social network. Additionally, our results help us identify that affiliation to the same organization-group and affiliation to the same hospital are important attributes to the diffusion process. Additionally, weights with high values do not necessarily lead to large diffusions inside the social network. We highlight the implication of our results for the diffusion of innovations in physician networks.

Machine learning (ML) offers a wide range of techniques to predict medicine expenditures using historical expenditures data as well as other healthcare variables. For example, researchers have developed multilayer perceptron (MLP), long short-term memory (LSTM), and convolutional neural network (CNN) models for predicting healthcare outcomes. However, recently proposed generative approaches (e.g., generative adversarial networks; GANs) are yet to be explored for time-series prediction of medicine-related expenditures. The primary objective of this research was to develop and test a generative adversarial network model (called "variance-based GAN or V-GAN") that specifically minimizes the difference in variance between model and actual data during model training. For our model development, we used patient expenditure data of a popular pain medication in the US. In the V-GAN model, we used an LSTM model as a generator network and a CNN model or an MLP model as a discriminator network. The V-GAN model's performance was compared with other GAN variants and ML models proposed in prior research such as linear regression (LR), gradient boosting regression (GBR), MLP, and LSTM. Results revealed that the V-GAN model using an LSTM generator and a CNN discriminator outperformed other GAN-based prediction models, as well as the LR, GBR, MLP, and LSTM models in correctly predicting medicine expenditures of patients. Through this research, we highlight the utility of developing GAN-based architectures involving variance minimization for predicting patientrelated expenditures in the healthcare domain.