Population-based early childhood caries (ECC) models are useful when trying to understand large-scale causes of caries; however, they are not as relevant when looking at individual cases. In a new critical review, author Kimon Divaris, DDS, PhD, evaluates where ECC models fall short and outlines how they can be improved.

In his review, Dr. Divaris explained four reasons why dental practitioners cannot apply current ECC risk assessment tools on an individual basis. He also detailed seven steps the dental community can take to create new tools as we learn more about caries (Journal of Dental Research, December 8, 2015).

"This article focuses on early childhood caries and critically reviews available approaches and tools to inform prediction of ECC development and guide clinical decision making," wrote Dr. Divaris, who is an associate professor of pediatric dentistry at the University of North Carolina at Chapel Hill.

In the review, Dr. Divaris evaluated the quality of current ECC risk assessment tools, such as the cariogram. He noted that there is weak evidence on the validity of such tools for personalized care or "precision dentistry," which he defined as "providing optimal customized oral healthcare according to individuals' specific oral health needs."

Dr. Divaris specifically pointed out four problems with many ECC risk assessment tools:

• Privatization of risk
• Outcome definition
• Proximal causes
• Missing dimensions

Privatization of risk

First, population-based risk assessment tools cannot be used on an individual basis because of privatization of risk.

"Although frequently interpreted as such, risk does not communicate the likelihood of individual case occurrence," Dr. Divaris explained. "The fallacious attempt to transfer and apply risk estimates or population risk factors to individuals is termed the privatization of risk."

While risk factors may be important at a broad level, that does not necessarily mean they are risk factors for an individual.

"This explains why ECC risk factors consistently and strongly associated with caries prevalence (or incidence) in large population studies are poor predictors of individual case occurrence," he wrote.

Outcome definition

One problem with caries is that a clinician cannot identify the disease unless it has visibly manifested itself, for example, with a lesion. In the review, Dr. Divaris pointed out that by the time a lesion appears, the person already has the disease.

"If the disease is correctly defined at the person level, risk is not applicable when an ECC diagnosis is made -- the disease is already present," he wrote.

Therefore, until dentistry has a better definition of caries -- one that can be applied before caries becomes visible -- different risk assessments will have different (and potentially not sufficient) caries definitions.

Proximal causes

Population-based assessments also overemphasize "upstream" caries factors, such as poverty or access to dental care. However, for individuals, "proximal" causes, such as salivary flow or tooth structure, are perhaps equally important.

"The genome, along with the oral microbiome (metagenome), and their interactions (transcriptome, proteome, and metabolome) at the tooth surface level remain largely unappreciated in caries risk assessment and personalized clinical decision-making," Dr. Divaris wrote.

Missing dimensions

Finally, Dr. Divaris explained that current risk assessments often miss quantification and time, but they are important for asses the risk of a disease. For example, he explained how a 19% likelihood of getting a disease within a two-year time period may be seen as low risk for caries but high risk for oral cancer.

"For this reason, derivation and communication of absolute estimates are essential," he wrote. "Otherwise, interpretations of low, moderate, and high risk are guaranteed to differ between and among clinicians and families, according to their context and subjective criteria."

Dr. Divaris pointed out that the cariogram is the only risk assessment tool that "communicates time-at-risk and actual disease propensities explicitly."

What we can do about it

It is important to note Dr. Divaris pointed out several times that population-based risk assessments are still valuable to society as a whole, even if they cannot be used for individuals.

"Although the existing caries risk assessment and ECC prediction tools have limited practical clinical utility, they serve as a valuable resource in dental education (training of clinicians) and facilitate communication with patients and their families," he wrote. "In addition, they serve as guides for the development of public health programs and the allocation of resources in vulnerable segments of the population."

However, he still outlined seven steps dentistry can take to development of valid and efficient ECC predictive and clinical decision-making aid tools (see sidebar).

Fortunately, Dr. Divaris concluded his article emphasizing his belief that accurate caries risk assessment for individuals and populations are possible.

"In these exciting times of rapid knowledge generation in the oral health domain, accurate caries risk assessment at the population level and 'precision dentistry' at the person level are both desirable and achievable," he wrote, "but they must be based on high-quality data and rigorous methodology."