New research led by researchers at Harvard School of Dental Medicine suggests that machine learning tools can help identify people at highest risk for tooth loss and refer them for further dental evaluation in an effort to ensure early interventions for avoid or delay the condition .
The study compared five algorithms that use a different combination of variables to detect risk. The results showed that those that took into account medical characteristics and socioeconomic variables, such as ethnicity, education, arthritis, and diabetes, outperformed algorithms that relied solely on dental clinical indicators .
The approach could be used to assess people globally and in a variety of healthcare settings, including by non-dental professionals.
In the study, the researchers used data including nearly 12,000 adults from the National Health and Nutrition Examination Survey to design and test five machine learning algorithms to assess how well they predicted complete and incremental tooth loss among adults in function of socioeconomic factors, health and medical characteristics. In particular, the algorithms were designed to assess risk without a dental exam. However, anyone who is deemed to be at high risk for tooth loss would still need to undergo an actual exam.
The results of the analysis point to the importance of socioeconomic factors that shape risk beyond traditional clinical indicators.
Our findings suggest that machine learning algorithm models that incorporate socioeconomic characteristics were better at predicting tooth loss than those based solely on routine clinical dental indicators. This work highlights the importance of the social determinants of health. Knowing the patient’s education level, employment status, and income is as relevant to predicting tooth loss as assessing their clinical dental status.
In fact, it has long been known that low-income and underserved populations experience a disproportionate share of the burden of tooth loss, likely due to lack of regular access to dental care, among other reasons .