Last dental visit and severity of tooth loss: a machine learning approach

Abstract

Abstract Objective: To investigate the time of last dental visit, as a mediator of socioeconomic status, associated with severe tooth loss and presence of functional dentition (FD) and use a machine learning approach to predict those at higher risk of tooth loss in adults and older adults. Methods: We analyzed data from a nationally representative sample of 88,531 Brazilian individuals aged 18 and over. Tooth loss was the outcome by; 1) functional dentition (lost 1-12 teeth) and 2) severe tooth loss (lost 23-32 teeth). Structural Equation models were used to find the time of last dental visit, as a mediator of socioeconomic statuses, associated with the outcomes. Moreover, machine learning was used to train and test predictions to target individuals at higher risk for tooth loss. Results: For 65,803 adults (aged 18 to 59), more than two years of last dental visit was associated with lack of functional dentition. Age was the main contributor in the machine learning approach, with an AUC of 90%, accuracy of 90%, specificity of 97% and sensitivity of 38%. For older adults, more than two years of last dental visit was associated with higher severe loss. The main contributor to the machine learning approach was the last dental visit with an AUC of 77%. Conclusions: More than two years of last dental visit appears to be associated with a severe loss in older adults and lack of functional dentition in adults. The machine learning approach had a good performance to predict those individuals.