Machine-learning algorithms may help in predicting tooth loss
BOSTON, U.S.: Little is known about socioeconomic factors in relation to tooth loss. In a new study, researchers at the Harvard School of Dental Medicine (HSDM) have developed machine-learning algorithms for predicting tooth loss in adults that—in addition to the obvious parameters such as age and dental care—included patients’ socioeconomic factors. The findings suggest that these tools may help identify teeth at risk in order to ensure early intervention.
Generally, tooth loss can be prevented if dental disease is caught and treated at an early stage. This is confirmed by studies that found that patients who attend regular checkups are less likely to lose teeth. However, barriers such as access to dental care and high costs can discourage patients from seeing a dentist. In the U.S., a decisive factor may be that adult dental coverage is not an essential health benefit in most public health insurance programs. Owing to this lack of routine care, by the time these patients see a dentist, it is already too late to save the tooth, and extraction becomes the most affordable option. This is where the screening tool could help to identify high-risk patients in time.
According to the researchers, machine-learning methods have been applied in medicine to provide information for clinical decisions; however, they have not been developed to predict oral health outcomes yet. Therefore, the researchers developed and tested five algorithms using different combinations of parameters—such as medical conditions and socioeconomic background—to predict tooth loss in adults and to compare the performances of the different tools. To develop the screening tools, the research team used data from nearly 12,000 adults from the National Health and Nutrition Examination Survey.
Socioeconomic characteristics decisive
Comparing the performances of the different algorithms, the researchers found that those models which incorporated socioeconomic characteristics, such as race and education, were better at predicting tooth loss than those models relying on traditional dental clinical indicators alone.
“Our analysis showed that while all machine-learning models can be useful predictors of risk, those that incorporate socioeconomic variables can be especially powerful screening tools to identify those at heightened risk for tooth loss,” said lead author Dr. Hawazin Elani, assistant professor of oral health policy and epidemiology at HSDM, in a university press release.
“This work highlights the importance of social determinants of health. Knowing the patient’s education level, employment status, and income is just as relevant for predicting tooth loss as assessing their clinical dental status,” she added.
In addition to the socioeconomic background of patients, the research team also determined preexisting medical conditions as predictors of tooth loss. “We found that medical conditions—such as arthritis, diabetes, high cholesterol, hypertension and cardiovascular diseases—are among the predictors of tooth loss. Clinicians could use this information to screen patients at high risk for tooth loss and coordinate their referral and dental care,” they stated.
Developed tool may be used by different health care providers
The screening tool was designed to be applied worldwide and in a variety of health care settings, even by nondental professionals, as it assesses the risk for tooth loss without the need for a dental examination. However, any patient deemed at high risk for losing a tooth would still have to undergo an actual examination.
The study, titled “Predictors of tooth loss: A machine learning approach,” was published online on June 18, 2021, in PLOS ONE.