| Sign In to gain access to subscriptions and/or personal tools. |
Overview of the Impact of Changing Global Patterns of Dental Caries Experience on Caries Clinical TrialsOral Health Services Research Centre, University Dental School and Hospital, Wilton, Cork, Ireland; h.whelton{at}ucc.ie
Key Words: caries • clinical trials • epidemiology • adults • children INTRODUCTION Measurement of change in disease levels is an important component of epidemiology. Since the first publication of Oral Health Surveys Basic Methods in 1971, the World Health Organisation (WHO) has facilitated the collection of standardized oral health data internationally. Dental caries is a disease that is amenable to measurement at different stages. The most common stage of measurement, and that promoted by the WHO, would appear to have been at the cavitation level, with visual or tactile confirmation of progression of the lesion to dentin. However, researchers have also measured enamel or white spot pre-cavitation caries by visual assessment. Although the WHO confines its methods to visual-tactile criteria, other methods used have included radiographs, electrical resistance, lasers, and transillumination to aid the diagnostic process. This paper will focus initially on the global changes in caries, at the cavitation level, which have taken place for children since the 1970s (Beltrán-Aguilar et al., 1999). Following this overview of the changes, a more detailed review of the relative attack rates on pits and fissures and smooth surfaces in children and young adults in Established Market Economies (EMEs) will be undertaken. The implications of the current pattern of caries attack on the conduct of clinical trials of caries-preventive agents will be considered in the final section. This global overview of changes in dental caries levels from the 1970s to 2001 focuses on 12-year-olds, since they are the most commonly studied age group for the purposes of comparative dental epidemiology. Recent caries data for 12-year-olds have been sought for eight different demographic regions: Established Market Economies (EMEs) (OECD countries), Formerly Social Economies of Europe, Sub-Saharan Africa, Middle Eastern Crescent, India, China, Other Asia and islands, and Latin America and the Caribbean. The literature has been reviewed to track the changes in caries levels in these regions. In the case of Europe, a recent review (Marthaler et al., 1996) is incorporated and updated. In the case of Sub-Saharan Africa, Middle Eastern Crescent, and the Americas, recent reviews of caries trends were available (Beltrán-Aguilar et al., 1999; Cleaton-Jones, 2001). The results of these reviews are summarized, and data from some countries in the regions are presented as examples of the caries experience in the region. For the other regions, the most recent data retrieved from the search are presented. Local variation is a feature of dental caries that cannot be captured in this type of overview. The aim of this paper is to indicate trends on a "macro" scale, where possible national data have been used and, where unavailable, data from smaller regions within countries have been used. Hence, the data presented should be interpreted with caution, since this review represents a broad overview of trends and not an in-depth analysis. CHANGES IN CARIES IN 12-YEAR-OLDS
Table 1
The decline in dental caries in children in established market economies (EMEs) is well-documented. The EMEs include most of North America, Canada, Western Europe, Australia, New Zealand, and Japan. By the end of the 20th century, caries prevalence and incidence among 12-year-olds in these regions had declined dramatically. Some of these changes are shown in Table 1  . For example, in England and Wales, the mean DMFT among 12-year-olds declined from 4.8 in 1973 to 1.2 in 1993 (O’Brien, 1993). The difference in DMFT levels between 12- and 13-year-old children decreased from 1.3 in 1973 to 0.4 in 1993. The smaller difference in DMFT levels between these two age cohorts in recent years may be indicative of a slower rate of caries development, although longitudinal studies would be required to substantiate this hypothesis. A review of caries levels in Europe between 1990 and 1995 showed "a general trend towards a further decline" (Marthaler et al., 1996). The countries with the lowest caries levels among 12-year-olds in this review were the Netherlands in 1992–93, with a DMFT of 0.9, and Finland in 1991, with a DMFT of 1.2. Earlier studies report a mean DMFT for this age group of 1.1 in the Netherlands in 1989 and 2.0 in Finland in 1988 (Marthaler et al., 1996). In Ireland, caries has continued to decline in this age group also (Whelton et al., 2001). In 1984, the mean DMFT among 12-year-olds in the North Western region of Ireland was 2.4; this had declined to 1.2 by 1997–98. Similar changes have been reported for the USA (Beltrán-Aguilar et al., 1999), Canada (Speechley and Johnston, 1996), and Australia (Davies et al., 1997). Data collected in the Hague (Truin et al., 1998) suggest a variation in the changes in caries levels over time, according to socio-economic status (SES). In 1993, the mean DMFT for children in low-, medium-, and high-SES groups was 0.9, 0.9, and 0.6 respectively. In 1996, the estimated mean DMFT was 1.1 for low- and 0.3 for medium- and high-SES groups (Truin et al., 1998). In some Central and Eastern European countries, caries prevalence in children and adolescents was still high in the early 1990s. For 12-year-olds in 30 European countries, the range of mean DMFT reported by Marthaler et al.(1996) was 1.1 to 7.7. Six countries had DMFT levels < 2.0, 13 had DMFT levels 2.0 to 2.9, 7 had 3.0 to 3.9, and 4 had DMFT greater than 4.0. The two countries with high DMFT were Latvia, with a mean DMFT of 7.7 in 1992, and Poland, with a mean DMFT of 5.1 in 1991. More recent data indicate a reduction in caries levels in these two countries, to 4.0 (Annual Report Book, 1999) and 3.8 DMFT (Petersen et al., 2001), respectively. Other Eastern European countries have shown no change in caries levels in recent years. In Bulgaria, for example, a DMFT of 3.1 was recorded in 1991 (Kunzel, 1996), and another study reported a DMFT of 3.0 in 1997 (Kukleva and Kondeva, 1999). The mean DMFT for 12-year-olds in Hungary was 4.1 in 1991 (Marthaler et al., 1996) and 3.8 in 1996 (Szoke and Petersen, 1998). In Croatia, caries levels increased from 2.6 to 3.5 between 1991 and 1999 (Rajic et al., 2000). The authors suggested that the war was a likely factor in this change, since earlier studies had shown a distinct decline in caries levels in Croatia up to 1991.
Studies of caries levels in Sub-Saharan Africa were systematically reviewed by Cleaton-Jones (2001). He reported no overall change in caries levels in 10 countries in the region between 1988 and 1998. Of a total of 300 studies of dental caries levels reviewed, 45 fit the strict inclusion criteria. These studies reported caries levels in 10 of the 42 countries in the region. Non-linear regression curves were fit to the data, which were plotted against time (year of study) according to age group for both 5- and 6-year-olds and 11- to 13-year-olds. A linear regression analysis was then conducted to look for significant trends over time. The mean DMFT score trend was relatively constant over time in the 11- to 13-year-old group. Results of studies conducted in Tanzania (Fejerskov et al., 1994) and Senegal (Sembene et al., 1999) are presented in Table 1 Cleaton-Jones (2001) conducted a similar review and statistical summation for the Middle Eastern Crescent (including North Africa). This review covered 21 countries in the region. The author identified 193 studies of caries levels in the selected age groups. Of these, 25 studies conducted mostly between 1988 and 1998 in 10 countries met the inclusion criteria. There was no change in the mean DMFT scores over the period of the study. A study in Kuwait included in this review showed a change in dental caries levels from 2.0 in 1982 to 2.6 in 1993 (Vigild et al., 1996). This increase was also evident in other age groups. A recent review of the oral health status in Syria (Beiruti et al., 2001) demonstrates the variability of survey findings. Mean DMFT scores for 12-year-olds are reported for 1980, 1981, 1983, 1985, 1988, 1994, and 1998. The DMFT figures were 2.0, 2.3, 1.4, 2.0, 1.9, 2.5, and 2.3, respectively. Confidence intervals for the data were not presented. The authors pointed out that the variance in the data due to period and sampling effects could be reduced by removing the extremes of data. Without these extreme point estimates (1.4 for 1983 and 2.5 for 1994), the estimated DMFT score for 12-year-olds for the period 1980 to 1998 ranges from 1.9 to 2.3. From these data, it was concluded that secular change in caries levels among 12-year-olds in Syria was not apparent. Studies in Israel (Zusman and Crawford, 1995; Kelman, 1996) also show some variation but do not indicate major changes in caries levels in the region overall. Studies in China (Fejerskov et al., 1994; Wong et al., 2001), India (Chawla et al., 2000), Pakistan (Khan, 1992), Central America (Beltrán-Aguilar et al., 1999), South America (Beltrá-Aguilar et al., 1999), and Mexico (Beltrán-Aguilar et al., 1999; Irigoyen and Sánchez-Hinojosa, 2000) all indicate declining caries levels. RELIABILITY OF DATA
The data presented in Table 1
Overall, it would appear from Table 1 In many instances, the changes in caries levels have been attributed to the use of fluorides in either toothpaste, water, or salt. Only a few reports tried to assess factors other than fluorides which are known to affect caries activity. THE IMPACT OF THE CHANGES IN CARIES LEVELS ON CARIES CLINICAL TRIAL DESIGN The impact of changes in the pattern and level of caries on caries clinical trials is considered under four headings: caries incidence in children, caries in adults and rate of progression, differential surface effect of fluoride, and impact of fissure sealant programs. CARIES INCIDENCE When one considers the conduct of randomized controlled trials on caries-preventive products, caries incidence is of great importance. Demonstration of product effectiveness will be determined by several factors, including incidents of caries in the study population, size of the true effect, sample size, compliance of study population, diagnostic threshold, validity of measurement, and reproducibility of measurement. In addition, other study design characteristics, such as randomization and blinding, will have an impact. Major changes in two of these factors which will affect the design of caries clinical trials in the future are the changes in caries incidents and developments in caries diagnosis.
As caries levels in 12-year-olds have fallen, the numbers required to power a randomized controlled trial adequately have risen. For example, based on data from published RCTs (Stookey et al., 1993), Table 2
It can be seen that, for these data, as the actual caries increment decreases, the sample size required to demonstrate a statistically significant 10% difference increases. Investigators have been able to compensate for these population changes by using inclusion criteria which allow for the selective enrollment of children at greater risk of developing caries—for example, children with existing caries at the baseline examination. Such pre-selection is useful to keep the size of the clinical trial at a manageable level and to test the relative efficacy of the product in those with above-average caries risk. It is reasonable to assume that if the caries process is the same throughout the population, then results of such studies are relevant for all age groups and risk categories.
A graph compiled by Backer Dirks (1974) (Fig.
However, in populations with lower levels of caries activity, as in Aurora, the slope of the line illustrating caries development is much more gradual and does not flatten off as quickly as in the high-caries populations. This feature is again illustrated by Chesters et al.(2004) in a low-caries population in England in 1993.
Thus, while the rationale for selecting 11- to 12-year-old children for inclusion in clinical trials in high-caries populations is clear, the data presented by Backer Dirks (Fig. RISK OF PRIMARY CARIES IN ADULTS One outcome of the reduction of caries in children is the presence of more surfaces at risk of developing caries in adulthood. A paper by Luan et al.(2000) describes the 10-year incidence of dental caries in adult and elderly Chinese and demonstrates a "distinct increase in the number of new DFT with age being 2.5 among the 20–29 year olds and 5.3 among the subjects over 60". The paper concludes that, in the adult population study, dental caries was a major problem, and that caries incidents increased with old age. Although these data pertain to a rural population with low levels of access to dental care and poor oral hygiene, they illustrate the lifelong nature of caries risk in a population with relatively low levels of caries in childhood (mean DMFT age 20 to 29 was 4.3). In the Culemborg Tiel study (Backer Dirks, 1974) a delay in the incidence of pit and fissure cavities among children in fluoridated Tiel was reported. In children at 9 years of age, 70% of the fissures were filled in Culemborg compared with 30% in Tiel. Also, whereas in Culemborg the incidence of pit and fissure caries tended to decrease at 9–13 and level off thereafter, in fluoridated Tiel the incidence tended to increase steadily for as long as the children were followed, up to age 17. This report may have been one of the earliest to describe a delay in caries due to fluorides. More recent studies of adolescents and young adults who have grown up in the fluoride toothpaste era indicate a prolonged risk of caries development on both pit and fissure and smooth surfaces. For example, Mejàre et al.(1998) followed caries development in subjects from 11 to 22 years of age in a radiographic study of a cohort of Swedish children born in 1972 or 1973. Both occlusal and approximal caries levels showed a constant rate of increase in the percentage of erupted surfaces which developed caries up to 22 years of age. The time lag between the initiation of caries and its detection radiographically is a consideration in interpretation of the results of this study. It would be useful to replicate this investigation with a prospective study design incorporating clinical examination including modern diagnostic aids.
Data from the UK (Kelly et al., 2000) and Ireland (O’Mullane and Whelton, 1992) (Table 3
The continued susceptibility of pit and fissure surfaces has been studied by numerous investigators. It has been of interest to policy-makers when considering the appropriateness of fissure sealant placement. Ripa et al.(1998) followed 1920 10- to 13-year-old children for 3 years and found that, in contrast to earlier studies in the 1950s and ’60s, the occlusal surfaces of molars suffered a relatively constant attack rate between age 10 and age 16, the rate being 10.5% of surfaces per annum. These findings were in agreement with those of Chestnutt et al.(1996), who reported three-year prospective data on a cohort of 4294 adolescents (mean age, 12.5 years). They found that 41% of occlusal surfaces at risk on maxillary molars and 39% of occlusal surfaces at risk on mandibular molars decayed over the three-year study period, thus indicating the continued susceptibility of pits and fissures in teeth up to 8 or 9 years after eruption. Arthur and Swango (1987) retrospectively examined dental records over 4 years for a US Navy population aged 17 to 26 years. They found that young adults were experiencing an appreciable amount of pit and fissure caries and would benefit from fissure sealing. These findings were in contrast to those of Eden (1976), who concluded that caries rates in molars of a young Navy population were very low at 4%, and that they had passed the period of high caries susceptibility. Thus, it would appear that from the 1970s to the 1980s, the attack rate for first permanent molar pits and fissures changed, and that these sites are now decaying at a later age. In 1991, Foreman conducted a cross-sectional study on 327 military personnel aged 17 to 26 years (Foreman, 1994). The examiners recorded whether occlusal caries was diagnosed while the patient was on active duty. This was interpreted as "new" primary caries during the subject’s time in the military. Eleven percent of 17- to 19-year-olds and 51% of 24- to 25-year-olds were found to have new caries during their time in the military. This study has limitations in its cross-sectional design and lack of training and calibaration of examiners; however, the order of the difference between the two age cohorts suggests substantial caries activity in this 17- to 26-year-old group. In a prospective study design, Richardson and McIntyre (1996) studied the development of caries in Royal Airforce (RAF) recruits from 1988 to 1992. It was found that 24% of the recruits developed occlusal caries over 5 years, and 36% developed caries on approximal surfaces. These studies indicate a continuing risk of primary caries in adults who have enjoyed the benefits of caries prevention in childhood. RATE OF PROGRESSION Hintze (2001) studied caries progression on radiographs taken 1 to 7 years apart in 640 Danish recruits. She concluded that, "Generally, the development of new approximal lesions and the progression of enamel caries was a slow process during the late teens." In a review in 1983, Pitts estimated the mean time for a lesion to progress through the enamel to be 3 to 4 years. A study of caries progression in adolescents during the five-year period between the ages of 14 and 19 years (Gustafsson et al., 2000) showed the median rate of progression of caries through enamel to be greater than five years in Sweden in the 1990s. Similarly, Mejàre et al.(1998), in a radiographic study, reported that the majority of lesions confined to the enamel do not reach the outer dentin within 4.8 years. Slower progression of caries is likely to mean more presentation of new dentin lesions in adults, although further research on the behavior of caries in adults is required to validate this suggestion. Hence, occlusal and approximal caries progression has slowed in countries where fluoride is used regularly. The nature of primary caries appears to have changed from a rapidly progressing disease of childhood to a slowly progressing disease which commences in childhood but progresses steadily in adulthood. As a result, the goalposts for prevention have widened. The formulation of caries-preventive products and clinical trial design must respond accordingly. DIFFERENTIAL SURFACE EFFECT OF FLUORIDE Accompanying the change in caries levels has been a change in the relative contributions of pit and fissure and smooth-surface caries to the overall caries experience. Backer Dirks (1974) reported the relatively greater effect of fluoride on approximal surfaces compared with pit and fissure surfaces in the Culemborg and Tiel study of water fluoridation in the Netherlands. There was a 75% difference in proximal cavities between study (no water fluoridation) and control groups (water fluoridated to 1.0 ppm) as compared with a 36% difference in pit and fissure cavities in 15-year-old children in 1969. Backer Dirks (1974) cited the results of the Hastings Fluoridation Project (Ludwig, 1971), which reported similar findings. From Backer Dirks’ data, 38% of cavities were on approximal surfaces in non-fluoridated Culemborg. In fluoridated Tiel, where the incidence of caries was lower, the relative effect on approximal surfaces was greater, since only 22% of cavities were found on approximal surfaces. In the USA, Ripa and co-workers (1998) reported that 83% of all caries in US schoolchildren in 1979–80 occurred in pits and fissures. Similarly, Mejàre et al.(1998) reported that 83% of caries was on occlusal surfaces in 12-year-old children in Stockholm. However, by the time these children were aged 21, the relatively greater increase in approximal caries during this period had decreased this proportion to 52%. McDonald and Sheiham (1992) reported that the increase in the relative contribution of pit and fissure caries to overall caries levels was not solely as a result of fluoride. They studied the distribution of caries on different tooth surfaces at various levels of caries and in both the presence and the absence of fluoride. The results showed that, even in the absence of fluoride, approximal caries levels decreased more than occlusal caries as overall caries levels declined. USE OF FISSURE SEALANTS An analysis of Irish data collected in 1997 (Whelton et al., 1998) shows that, for 15-year-olds in fluoridated areas, 81% of the DMFS of 4.0 occurs on pit and fissure surfaces. In non-fluoridated areas, 72% of the DMFS of 6.1 occurs on these surfaces. The relative contribution of pit and fissure caries to overall DMFS may be less in this population because of pit and fissure sealant programs delivered by the dental public health service. Sixty percent of 15-year-olds had some pit and fissure sealants on their teeth. Hence, when designing clinical trials of caries-preventive agents, investigators should consider the estimated effect on projected caries increments for approximal and pit and fissure surfaces separately and take this caries pattern into account when planning the study. Potential confounding due to the placement of fissure sealants should also be anticipated. Development of new oral hygiene products formulated to improve caries prevention on pit and fissure surfaces would be worthwhile. Currently, prevention of caries on these surfaces depends heavily on visits to the dental surgery for the professional placement of pit and fissure sealants. DEVELOPMENTS IN CARIES DIAGNOSTIC POWER As well as the changes in the pattern and level of caries, the second important change in recent years has been in the area of caries diagnosis. The research community has responded to the changing caries pattern by the development of new methods of measuring caries objectively and reliably at earlier stages of development. The majority of studies reviewed for this overview were conducted according to WHO criteria or visual-tactile methods. There is also a body of literature which deals with the impact on reported caries levels of lowering the diagnostic threshold for caries and recording it at pre-cavitation levels (Rimmer and Pitts, 1992; Machiulskiene et al., 1999). There are clear advantages to this approach when one is studying the efficacy of products on a disease which may take years to reach cavitation level and where early intervention and prevention may extend the success of caries prevention into adulthood. Estimates of changes in caries progression are problematic due to the paucity of longitudinal data in the literature for children, adolescents, and young and older adults. The cohort effect, combined with sampling effects and diagnostic differences, confounds the investigation of the true changes in caries progression with time. The elucidation of the age-related pattern and rate of caries development in successive age cohorts will be important in informing future clinical trial design. In summary, the changes in caries patterns which have an impact on the design of caries clinical trials are:
These changes indicate that caries continues to be a challenge throughout life. The conduct of clinical trials of caries-preventive agents must now incorporate more sensitive diagnostic methods capable of valid and reliable measurement of caries initiation and progression in its early stages. The application of sophisticated statistical analysis which takes account of the pattern of caries attack will also help to overcome the difficulties posed by these changes in caries patterns. The application of such techniques to dental datasets which have large numbers of tooth-surface variables and multiple observations has been made possible by the increasing capacity of and accessibility to high-speed computers.
FOOTNOTES Presented at the International Consensus Workshop on Caries Clinical Trials, Glasgow, Scotland, January 7–10, 2002 REFERENCES
Journal of Dental Research, Vol. 83, No. suppl 1,
C29-C34 (2004) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
