Sequential Model for Predicting Patient Adherence in Subcutaneous Immunotherapy for Allergic Rhinitis

TL;DR

This work tackles long-term adherence to subcutaneous immunotherapy for allergic rhinitis by applying two sequential models—a latent-variable SLVM/SLAC framework and a classic LSTM—to predict nonadherence and local symptom scores over a $3$-year horizon using data from $205$ patients collected at six time points. The results show that LSTM achieves higher adherence-prediction accuracy ($66$–$84\%$) than SLVM ($60$–$72\%$), while SLVM provides stronger score prediction; both yield RMSE substantially below random baselines. Additionally, SLAC can function as a simulator to explore how different treatment actions affect outcomes, and interpretability analyses identify distance to the clinic and Der $f$ allergen metrics as key predictors of adherence. Overall, the study demonstrates the value of sequential modeling for dynamic, personalized SCIT management in AR, enabling earlier interventions and potential integration into clinical workflows.

Abstract

Objective: Subcutaneous Immunotherapy (SCIT) is the long-lasting causal treatment of allergic rhinitis (AR). How to enhance the adherence of patients to maximize the benefit of allergen immunotherapy (AIT) plays a crucial role in the management of AIT. This study aims to leverage novel machine learning models to precisely predict the risk of non-adherence of AR patients and related local symptom scores in three years SCIT. Methods: The research develops and analyzes two models, sequential latent-variable model (SLVM) of Stochastic Latent Actor-Critic (SLAC) and Long Short-Term Memory (LSTM) evaluating them based on scoring and adherence prediction capabilities. Results: Excluding the biased samples at the first time step, the predictive adherence accuracy of the SLAC models is from 60\% to 72\%, and for LSTM models, it is 66\% to 84\%, varying according to the time steps. The range of Root Mean Square Error (RMSE) for SLAC models is between 0.93 and 2.22, while for LSTM models it is between 1.09 and 1.77. Notably, these RMSEs are significantly lower than the random prediction error of 4.55. Conclusion: We creatively apply sequential models in the long-term management of SCIT with promising accuracy in the prediction of SCIT nonadherence in AR patients. While LSTM outperforms SLAC in adherence prediction, SLAC excels in score prediction for patients undergoing SCIT for AR. The state-action-based SLAC adds flexibility, presenting a novel and effective approach for managing long-term AIT.

Figures (10)

• Figure 1: Flowchart on 205 patients treated with SCIT for Der p/Der f allergy during a 36-month treatment period. The adherence was assessed with sequential latent variable models focusing on patient's demographic characteristics and clinical follow-up data.
• Figure 2: Histogram of scores across six time steps. Score value (horizontal axis) vs. count (vertical axis).
• Figure 3: RMSE of the prediction step by step. The red dashed line is the RMSE of random prediction with Uniform distribution. See Fig. \ref{['fig:all_feature_onestep_slac']} and \ref{['fig:all_feature_onestep_lstm']} for more details.
• Figure 4: Accuray of the prediction step by step. The red dashed line is the accuracy of random prediction with Uniform distribution. See Table \ref{['table:classification']} for more details.
• Figure 5: RMSE of SLAC one-step prediction across various scores and time steps.