Who Needs More than 1,000 ppm? The Epidemiology of High-Risk Populations Open Access

Oral health is necessary to overall health and well-being. Most oral diseases are preventable. Global improvement in oral health is one of the major public health success stories of the past century. Public health measures such as fluoridation of water, individual preventive approaches available for self-care such as fluoride toothpaste and rinses, and professional dental services including fluoride varnishes and dental sealants have resulted in dramatic reductions in dental caries across most populations. These successes, however, have not been experienced uniformly by all populations. Marcenes et al. [2013a] in their systematic analysis reported that in 2010, globally oral conditions still affected 3.9 billion people and accounted for 15 million disability-adjusted life-years. The most prevalent condition for the entire Global Burden of Disease study was untreated caries in permanent teeth, with a global prevalence of 35.3%. Untreated caries in primary teeth was another common condition, ranking the 10th most prevalent condition and affecting 9% of the global population [Marcenes et al., 2013a]. There were 3 oral conditions in the ‘top ten' Global Burden of Disease study (severe periodontitis ranked 6th). Being on the ‘top ten' list is usually something to aspire to and be proud of; however, in this particular instance, as oral health professionals we have nothing to be proud of! It is well documented that minorities, individuals with low socioeconomic status and other vulnerable populations continue to battle with oral health disparities by experiencing higher oral disease levels and health problems due to difficulties in accessing oral health services including preventive care.

Dental caries is well recognized as an ancient disease and was first documented and described through examination of ancient skulls [Moore and Corbett, 1973]. The pattern of caries described was mostly cervical rather than coronal caries and corresponded to the time when agriculture replaced hunting and gathering as a food source. Over the centuries, with civilization, the pattern of caries changed dramatically and with modern research techniques and abilities, our understanding of caries has changed. The more contemporary description of dental caries is a common infectious disease that starts in childhood and is entirely preventable - infectious because the transmission of the caries causing bacteria is from the mother or caregiver to the child, and preventable as there are good scientific evidence, knowledge and approaches on preventing caries, which when used have led to the prevention of caries. Because dental caries continues across the life span and its effects are irreversible and incurable, it has also been considered by many to be a chronic disease, as demonstrated by the data shown in table 1, demonstrating that the prevalence of caries increases across the life span. Although not generally common knowledge, globally dental caries still remains one of the most common chronic diseases of children, as seen in figure 1 [Petersen et al., 2005]. In the USA, even though there has been a substantial decline in caries prevalence, starting with the introduction of community water fluoridation in 1945 and steadily continuing over the last 3 decades, the most recent US data on the national prevalence of dental caries still shows a 36.7% prevalence in 2- to 8-year-olds in primary teeth [Dye et al., 2015], while asthma, considered the most common childhood disease, has a prevalence of only 8.3% in the same age group [CDC, 2013]. These data, first reported in ‘Oral health in America: a report of the surgeon general', continue to suggest that dental caries is still almost 5 times more common than asthma, although it gets little attention [US Department of Health and Human Services, 2000]. Similar to this is the finding with hay fever, also considered to be a common childhood disease.

To effectively address the epidemiology of high-risk populations for any disease or condition, we first need to define and understand ‘risk'. The word ‘risk' in the English dictionary can be used as a noun, verb or idiom. In the context of any of these, it generally defines ‘risk' as an ‘exposure to the chance of injury or loss; a hazard'. In medicine ‘risk' is defined as ‘the possibility of suffering a harmful event' [American Heritage® Dictionary, 2011]. In any of these definitions ‘risk' denotes vulnerability [Beck, 1998]. The overarching principles of being considered ‘at risk' are as follows: not all people who are considered ‘at risk' are at risk; just by visual presentation or appearance we cannot identify who is ‘at risk'; and not all ‘at risk' are at the same level of risk. Several medical dictionaries then define ‘high-risk group/populations' as ‘a group of people in the community with a higher-than-expected risk for developing a particular disease' [http://medical-dictionary.thefreedictionary.com]. The group or groups of people are defined on measurable characteristics including sociodemographic characteristics such as age, gender, income, education and other characteristic such as physical attributes, lifestyle, habits and environment.

Another definition of ‘at risk' at the population level is those groups that have a higher than average prevalence of a particular disease compared to the general population. In the context of our topic of interest for this conference and paper: ‘Who needs higher than 1,000 ppm fluoride', the disease then is dental caries. This report also focused on the population level, although the factors that influence caries development at the population level and the individual level are the same [Fejerskov, 2004]. The frequently used epidemiological measures to describe and quantify caries on a population level are caries prevalence, caries incidence and caries severity. A literature search of the global caries scientific literature using PubMed (MEDLINE) was conducted to identify high-risk populations that would benefit from 1,000 ppm fluoride. The search terms used were ‘caries', ‘caries prevalence', ‘caries incidence', ‘caries severity', ‘epidemiology' and ‘high-risk'. The global scientific literature on dental caries prevalence, incidence and severity is vast, therefore this report focused on the scientific literature published from 2000 onwards. This report also used literature on caries prevalence rather than incidence, as that was the predominant literature. Those reporting caries incidence were more likely to be intervention studies of caries preventive agents.

The global scientific literature on dental caries has consistently described several groups within a population that have a higher than average prevalence, incidence and severity of dental caries than the general population. These groups are mainly based on age, gender, race and ethnicity, socioeconomic or poverty status, geography and disability status. It needs to be remembered that the comparison of caries prevalence, incidence and severity across studies is complicated as the caries diagnostic criteria and the reporting of results vary between studies.

Data from large national studies, such as the National Health and Nutrition Examination Survey (NHANES) in the USA, show that caries prevalence differs by group, as seen by the data presented in table 1 that reports caries prevalence data from the 1999-2004 NHANES. As seen in this table, caries prevalence for coronal caries in primary and permanent teeth, and root caries in permanent teeth across all age groups differed by gender, race and ethnicity, poverty status and education. This scenario is generally no different in other countries, except that the risk is not consistent nor has it the same directionality across categories for all countries. For example, when considering poverty status, in industrialized and developed countries caries prevalence in children from low-income families is higher than in children from higher-income families (table 1). However, in less industrialized or developed countries, the caries prevalence in children from high-income families is higher than in children from low-income families. Caries data from multivariate analyses of 12-year-old Sudanese children showed that those who were from the middle-income group had higher caries prevalence than the low-income group [Nurelhuda et al., 2009]. Caries data from the USA, Australia and Denmark for this same time period show that caries prevalence in children from low-income families is higher than in children from higher-income families [Hallet and O'Rourke, 2003; Psoter et al., 2006; Dye et al., 2007; Christensen et al., 2010]. This pattern is quickly changing, with children from low-income families beginning to show higher risk even in developing countries. Therefore, care needs to be taken when generalizing trends.

Although earlier thought to be mainly a disease of children, caries is now considered a disease that occurs across the life span, as demonstrated by the data in table 1. However, several population studies have shown that the risk of caries across the lifespan is not equal [Thomson, 2004; Dye et al., 2007]. Young children and the elderly (those aged 65 years and older) are more at risk of caries than adults between the ages of 20 and 64 years.

Caries was mainly considered a disease of childhood. Figure 1 shows the global map of dental caries (decayed, missing and filled permanent teeth, DMFT) in 12-year-olds [Petersen et al., 2005]. In some countries, including the USA, after decades of a decreased caries prevalence trend, a rise in caries prevalence in primary teeth is being observed. Between NHANES 1988-1994 and 1999-2004 prevalence increased from 24.2 to 27.9%, a 3.7% difference, for children aged 2-5 years [Dye et al., 2007], and the most recent data for 2011-2012 report a further increase to 55.7% for children aged 6-8 years from 51.2% for children aged 6-11 years 1999-2004 [Dye et al., 2007, 2015]. Bagramian et al. [2009], in a review of global caries trends using data from fairly large studies between the years 2001 and 2008, concluded ‘a marked increase in prevalence of dental caries'. This increase was across developed and developing countries, with a high caries prevalence ranging from 90 to 100% reported in the Philippines, China, Mexico and Taiwan.

Several national prevalence studies of caries in adult populations have shown that the elderly within these populations are at a higher risk for coronal and root caries, as more individuals are retaining their teeth through their life span. Thomson [2004], in a review of the literature on individuals aged 50 and older, concluded that the disease rate in this age group was ‘at least as great as that of adolescents'. Many reasons for this increased risk have been identified. Mostly, these are due to consequences of aging such as physical limitations and cognitive decline, deterioration of general health due to the many chronic diseases, and oral health decline such as attachment loss and mouth dryness or xerostomia (either physiological or due to medications). In the dentate elderly, prevalence studies across the globe show a caries prevalence of 65% and higher, with some studies reporting over 90% [Dye et al., 2007; Rihs et al., 2009; Liu et al., 2013]. The elegant meta-analysis by Griffin et al. [2005] using six studies on coronal caries in the elderly, after adjusting for study length and examiner reversals, reported that the annual incidence of caries in the elderly was 39.8% (95% CI 33.1-46.5), with an annual increment of 0.86 surfaces (95% CI 0.66-1.07). The review by Thomson [2004] came to a similar conclusion that the annual increment for coronal caries was between 0.5 and 0.8 surfaces per year, and that for root caries was between 0.2 and 0.4 surfaces per year.

In countries and populations that are made of single race or ethnicity, this factor does not determine risk. However, in populations made up of more than one race or ethnicity such as the USA and UK, studies have shown that differences in disease levels exist by racial or ethnic groups. Differences have also been reported in indigenous and native populations compared to nonnative populations. These differences are not generally attributed to inherent racial and ethnic difference per se, but due to socioeconomic or lifestyle differences.

Several studies have documented racial and ethnic differences in caries prevalence and severity. For example, in the UK, ethnic differences exist. In a study of 3- to 4-year-old children living in Inner North East London, White European (mean = 1.91), Pakistani (mean = 1.11) and Bangladeshi (mean = 1.05) children had a significantly higher number of untreated carious teeth than White British children (mean = 0.56) [Marcenes et al., 2013b]. In the USA, in the younger age groups and for root caries, racial and ethnic minority populations, particularly African-Americans and Mexican-Americans, are at higher risk for caries, as seen from the data presented in table 1. In adults, the caries prevalence is higher in White non-Hispanics, largely because of how we measure caries to include past disease and treatment. In Brazil, among the elderly, the Caucasian group also had higher DMFT [Rihs et al., 2009].

Socioeconomic status is measured using three individual characteristics - income, education or occupation. Other measures of socioeconomic status are composite indices such as the Social Vulnerability Index (SVI) [Bonanato et al., 2010], the Scottish Index of Multiple Deprivation (SIMD) [Levin et al., 2010] and the Neighborhood Deprivation Score [Willems et al., 2005]. When the subject studied is a child, these characteristics of the parents are used as determinants. When the subject is an adult, the characteristic of the individual is used.

In the early and mid-20th century, dental caries was considered a disease of modernization, with a higher prevalence in higher socioeconomic countries and individuals with higher socioeconomic status. In the late 20th and the 21st century, this pattern of caries prevalence and severity changed in developing and developed countries, with the contemporary caries literature showing that low income [Psoter et al., 2006; Armfield, 2007; Dye et al., 2007; Christensen et al., 2010], low education levels [Psoter et al., 2006; Armfield, 2007; Dye et al., 2007; Traebert et al., 2009; Christensen et al., 2010] and lower classes of occupations [Ferro et al., 2010] are consistently related to high caries risk. Data from table 1 for the USA showed that children from low-income families had higher caries prevalence; however, low-income adults had lower caries prevalence. The higher prevalence of caries in the higher-income adults with income greater than 200% of the federal poverty level (FPL) is again because of how we measure caries to include past disease and treatment, reflecting better access to care in those with higher incomes. In those with income greater than 200% of the FPL, the ‘filled' component of the DMFT is a large part of it, accounting for the overall increase in DMFT compared to those with income less than 200% of the FPL.

When considering geography to determine risk, several measures are used. These can be as broad as those reported by the WHO by regions of the world (such as AFRO, AMRO/PAHO, EMRO, EURO, SEARO and WPRO) to continents to individual countries. Further, within a country measures such as urban-rural or regions are often used to describe geographical differences and risk.

Figures 1 and 2 show the global distribution of caries severity in 2004 for 12-year-old children and adults aged 35-44, respectively [Petersen et al., 2005]. Although these maps need updating as they are now over 10 years old, again these data show that caries is a universal disease. For 12-year-olds, in the majority of countries much of the severity is designated as very low or low caries (DMFT <2.6), large parts of Eastern Europe, Russia and South America as moderate, and few pockets in Central and South America and Europe as high (DMFT ≥4.4). For adults aged 35-44, this is reversed, with the majority of countries designated as moderate or high (DMFT ≥9.0) and parts of Africa, Asia and China designated as very low or low caries (DMFT <8.9).

Within a country or geographic region, at a local level, caries risk can also differ, as seen in the recent study of a large sample of children aged 3-4 years in 20 nurseries living in three of the most deprived boroughs in Inner North East London, UK: Hackney (mean = 0.63), Tower Hamlets (mean = 1.06) and Newham (mean = 1.06) [Marcenes et al., 2013b]. Studies of the elderly in China have also reported urban-rural differences, with a higher prevalence in urban areas [Liu et al., 2013].

Special needs populations and medically compromised subjects are high-risk populations, as seen from several studies. Medically compromised populations are those with systemic medical diseases and conditions that affect oral health and dental treatment. Some of these medical conditions include diabetes, multiple drug use, cardiovascular diseases, neurological disorders, cancer, radiation and infectious disease such as HIV.

Special needs populations are those with special health care needs. In the USA, children with special health care needs are defined by the Maternal and Child Health Bureau as those ‘who have or who are at risk for a chronic physical, developmental, behavioral, or emotional condition and who also require health and related services of a type or amount beyond that required by children generally'. Many of these conditions, including mental retardation, developmental disabilities, cerebral palsy, craniofacial abnormalities and seizure disorders, can impact a child's oral health. The causes of increased caries risk in these populations are the following: diet, xerostomia, difficulties performing oral hygiene because of limitations, gingival hyperplasia and crowding of the teeth, and medications containing sugar. In a study of disabled children and young adults in Kuwait, the prevalence of caries was higher in those with special needs compared to children in the general population [Shyama et al., 2001]. Similarly, several studies in Saudi Arabia reported caries prevalence from 65 to 97% [Al-Ansari, 2014].

At the population level, the global literature on the epidemiology of high risk is varied, with no real guidance on at what caries prevalence, incidence or severity do we consider someone at high risk, beyond the broad generic definition of ‘having a higher than average dental caries prevalence, incidence or severity than the general population'. A curious issue, then, is quantifying ‘average prevalence, incidence or severity' and using it to determine ‘high risk'. In table 1, in the USA, the average prevalence of caries in primary and permanent teeth of 6- to 8-year-old children was 53.2%. By the above definition, any population group that has caries prevalence higher than 53.2% should be considered high risk. Based on the data in table 1, that would include males, African-American non-Hispanic children, Mexican-American children and those whose parents' income was less than 200% of the FPL. On the other hand, the average prevalence of caries in the permanent teeth of dentate seniors 65 years and older was 93%. Again, based on the data in table 1, males, white non-Hispanics adults, those with incomes greater than 200% of the FPL and those with higher education levels (high school or more) would be considered high risk. However, given the very high prevalence of caries in this population, should all dentate seniors aged 65 years and older not be considered high risk? The US data in figure 3 may make the argument that at least for adult populations in developed countries we should not do this, as what accounts for this very high prevalence is the amount of past caries, as seen by the filled and missing components of the DMFT, rather than current caries in the decayed component. This would argue that incidence is a better measure. However, epidemiological studies reporting incidence are very few or practically nonexistent, because they are expensive to conduct as individuals need to be followed over time. Similar to this is the case of untreated caries, which might be a better measure; however, again it reflects access to care. In children, prevalence is a useful measure as it is more practical and data are generally available. If prevalence was high, it would be appropriate to consider the whole population as high risk.

Another way to consider this may be the cost of identifying the high risk. If the costs to identify those populations at high risk are substantial, then targeting all may be the cheaper option. With a lack of any hard evidence to make broad generalizations and based on the current global prevalence and severity of dental caries in populations where the majority of the population has caries, then defining a high-risk population is moot, as the entire population should be considered high risk and needs to be targeted. In populations with lower risk, those groups with a higher than average risk should be identified using the available data, so that they can be targeted easily. Much of this assumes that current caries data at the population level are available via sociodemographic and other characteristics such as lifestyle, habits and environment to identify those at risk. This is always the concern, as substantial resources are needed to collect this level of data.

In conclusion, at the population level, high-risk caries populations should be designated based on the prevalence or incidence of caries in that population. In a population with very high prevalence or incidence, then the entire population should be considered high risk. However, in populations with lower prevalence or incidence of caries, those at high risk for caries should be considered based on age, gender, race and ethnicity, socioeconomic status, geography and disability status.

Prof. Mascarenhas has received a stipend from Colgate Palmolive for presenting this paper at the conference and for the preparation of the manuscript. The paper was not reviewed by Colgate prior to submission nor have they had any role in the editorial process.