Introduction: Digital therapeutics (DT) emerged and has been expanding rapidly for pain management. However, the efficacy of such approaches demonstrates substantial heterogeneity. Machine learning (ML) approaches provide a great opportunity for personalizing the efficacy of DT. However, the ML model accuracy is mainly associated with reduced clinical interpretability. Moreover, classical ML models are not adapted for the longitudinal nature of the DT follow-up data, which may also include nonlinear fluctuations. Objectives: This study presents an analytical framework for personalized pain management using piecewise mixed-effects model trees, considering the data dependencies, nonlinear trajectories, and boosting model interpretability. Methods: We demonstrated the implementation of the model with posture biofeedback training data of 3610 users collected during 8 weeks. The users reported their pain levels and posture quality. We developed personalized models for nonlinear time-related fluctuations of pain levels, posture quality, and weekly training duration using age, gender, and body mass index as potential moderating factors. Results: Pain levels and posture quality demonstrated strong improvement during the first 3 weeks of the training, followed by a sustained pattern. The age of the users moderated the time fluctuations in pain levels, whereas age and gender interactively moderated the trajectories in the posture quality. Train duration increased during the first 3 weeks only for older users, whereas all the users decreased the training duration during the next 5 weeks. Conclusions: This analytical framework offers an opportunity for investigating the personalized efficacy of digital therapeutics for pain management, taking into account users' characteristics and boosting interpretability and can benefit from including more users' characteristics.

This study describes four statistical models—Poisson; Negative Binomial; Zero-Inflated Poisson; and Zero-Inflated Negative Binomial—which were devised in order to examine traffic accidents and estimate the best probability estimating model in terms of future risk assessment at interurban road sections. The study was conducted on four sets of fixed-length sections of the road network: 500, 750, 1000, and 1500 m. The contribution of transportation and spatial parameters as predictors of road accident rates was evaluated for all four data sets separately. In addition, the Empirical Bayes method was applied. This method uses historical accidents information, allowing regression to the mean phenomenon so as to improve model results. The study was performed using Geographic Information System (GIS) software. Other analyses, such as statistical analyses combined with spatial parameters, interactions, and examination of other geographical areas, were also performed. The results showed that the short road sections data sets of 500 and 750 m yielded the most stable models. This allows focused treatment on short sections of the road network as a way to save resources (enforcement; education and information; finance) and potentially gain maximum benefit at minimum investment. It was found that the significant parameters affecting accident rates are: curvature of the road section; the region and traffic volume. An interaction between the region and traffic volume was also found.