Several countries established health technology assessment (HTA) processes to support decision-making. Considering the high volume of submissions processed by HTA agencies, approaches to determine factors associated with the approval would be beneficial. This study aimed to predict the final recommendation of the National Committee for Health Technology Incorporation (Conitec) using a natural language processing (NLP) algorithm for text extraction.

Conitec’s 2012 to 2022 reports (n=389) were split into 75 percent training and 25 percent testing data. Tokenization enabled NLP models: Least Absolute Shrinkage and Selection Operator (LASSO), logistic regression, support vector machine (SVM), random forest, neural network, and Extreme Gradient Boosting (XGBOOST). Evaluation criteria included accuracy, area under the receiver operating characteristic curve (ROC AUC) score, precision, and recall. Cluster analysis with k-modes identified two clusters (group 0 = approved, group 1 = rejected).

The neural network model demonstrated the best accuracy metrics with a precision of 0.815, accuracy of 0.769, ROC AUC of 0.871, and a recall of 0.746. Some tokenization identified that linguistic markers could contribute to the prediction of incorporation decision by the Brazilian HTA Committee, such as international HTA agencies’ experience and the government as the main requester. Cluster and XGBOOST analysis identified similar results with approved technologies with a predominance of drugs assessment, mainly requested by the government, and not approved mostly assessing drugs, the industry as the main requester.

The NLP model could identify predictors for the final decision process on the incorporation of health technologies in Brazil’s Unified Health System, opening paths for future work using HTA reports coming from other agencies. This model could potentially improve the throughput of HTA systems by supporting experts with prediction/factors/criteria for approval or nonapproval as an earlier step.

Health technology assessment (HTA) plays a vital role in healthcare decision-making globally, necessitating the identification of key factors impacting evaluation outcomes due to the significant workload faced by HTA agencies.

The aim of this study was to predict the approval status of evaluations conducted by the Brazilian Committee for Health Technology Incorporation (CONITEC) using natural language processing (NLP).

Data encompassing CONITEC’s official report summaries from 2012 to 2022. Textual data was tokenized for NLP analysis. Least Absolute Shrinkage and Selection Operator, logistic regression, support vector machine, random forest, neural network, and extreme gradient boosting (XGBoost), were evaluated for accuracy, area under the receiver operating characteristic curve (ROC AUC) score, precision, and recall. Cluster analysis using the k-modes algorithm categorized entries into two clusters (approved, rejected).

The neural network model exhibited the highest accuracy metrics (precision at 0.815, accuracy at 0.769, ROC AUC at 0.871, and recall at 0.746), followed by XGBoost model. The lexical analysis uncovered linguistic markers, like references to international HTA agencies’ experiences and government as demandant, potentially influencing CONITEC’s decisions. Cluster and XGBoost analyses emphasized that approved evaluations mainly concerned drug assessments, often government-initiated, while non-approved ones frequently evaluated drugs, with the industry as the requester.

NLP model can predict health technology incorporation outcomes, opening avenues for future research using HTA reports from other agencies. This model has the potential to enhance HTA system efficiency by offering initial insights and decision-making criteria, thereby benefiting healthcare experts.