Tooth loss is a major problem for many people. However, it does not affect everyone equally. Scientists have been looking for a way to better identify individuals who are more vulnerable to losing their teeth, without having them undergo a dental examination.

The Harvard School of Dental Medicine has been conducting research using machine-learning methods for improved identification of those most susceptible to tooth loss, so that they can be referred for additional assessments that allow them to receive preventative dental care to halt, or delay, tooth loss.

The June 18 study involved the comparison of five different algorithms using numerous variables for the screening of risk assessment. The results of the study revealed that algorithms that factored in medical traits and socioeconomic variables like education, race, and the presence of diabetes and arthritis, offered better results than algorithms relying strictly upon dental indicators.

Anticipated Study Benefits

It is hoped that this approach can be used on a global basis to screen vulnerable individuals. It is believed that it could be utilized in a wide range of healthcare settings, even by those who are not professionally trained as dentists or dental assistants.

Effects of Tooth Loss

The loss of teeth can be both physically and psychologically damaging. It has a strong negative effect on the patient’s quality of life, well-being and social interactions. Fortunately, the process can be slowed, or even halted, when early signs are detected so that the patient can receive preventative dental care.

Unfortunately, many of those with tooth decay or periodontal disease may not schedule an appointment with their dentist until their condition has progressed to the point that the tooth cannot be saved. This type of screening tool should help in identifying those who are most vulnerable, allowing them to be referred for additional assessment.

Study Methodology

The researchers who conducted the study tracked nearly 12,000 adults to create five machine-learning algorithms. They were then tested using these algorithms to determine their ability to predict both total and incremental tooth loss in categories like socioeconomic, medical and health traits.

Early indications are that the machine-learning algorithms are superior in predicting tooth loss than simply relying upon normal clinical dental indicators. The researchers highlighted the fact that understanding the education level of a given patient, as well as their income and employment status, is equally relevant in the prediction of tooth loss as a physical examination by a dental professional.

Low-Income and Susceptibility to Tooth Loss

The idea that low-income, marginalized individuals share an unequal burden of tooth loss has been known for some time now. It is hoped that the results of the study will help provide dental professionals, and non-dental professionals alike, with valuable tools to assess a given patient’s risk factors in developing conditions that can lead to tooth loss. In turn, this should help in providing early.

DISCLAIMER: The advice offered is intended to be informational only and generic in nature. It is in no way offering a definitive diagnosis or specific treatment recommendations for your particular situation. Any advice offered is no substitute for proper evaluation and care by a qualified dentist.