.
Journal of Dental Research
Volume 69 (Special Issue); Page 723-727.
February 1990
Recent Trends in Dental Caries in U.S. Children and the Effect of Water Fluoridation
J.A. Brunelle and J.P. Carlos
Epidemiology Branch, National Institute of Dental Research, National Institutes
of Health, Westwood Building, Room 538, Bethesda, Maryland 20892
Presented at a Joint IADR/ORCA International Symposium on Fluorides: Mechanisms of Action and Recommendations for Use, held March 21-24, 1989, Callaway Gardens Conference Center, Pine Mountain, Georgia.
The decline in dental caries in US schoolchildren, first observed nationwide in 1979-1980, was confirmed further by a second national epidemiological survey completed in 1987. Mean DMFS scores in persons aged 5-17 years had decreased about 36% during the interval, and, in 1987, approximately 50% of children were caries-free in the permanent dentition. Children who had always been exposed to community water fluoridation had mean DMFS scores about 18% lower than those who had never lived in fluoridated communities. When some of the "background" effect of topical fluoride was controlled, this difference increased to 25%. The results suggest that water fluoridation has played a dominant role in the decline in caries and must continue to be a major prevention methodology.
Introduction.
At the beginning of this decade, results from epidemiological studies conducted in several Western countries demonstrated that dental caries was rapidly declining among school-age children (Glass, 1982). This was also true in the United States; examination of a national sample of approximately 38,000 children aged 5-17 years, completed in 1980, showed a reduction of about 32% in the point estimate of mean decayed, missing, and filled permanent tooth surfaces (DMFS) nationwide, as compared with data collected in an earlier survey during 1971-1973 (Brunelle and Carlos, 1982). The reduction had occurred at every age and in all regions of the U.S. Although the reasons for this decrease could not be determined with certainty, most observers concluded that widespread exposure to fluoride in various forms had played a major role. For example, it was demonstrated that children who had always lived in fluoridated communities had a mean DMFS in 1980 approximately 33% lower than those who had never been exposed to fluoridated drinking water (Brunelle et al., 1983).
In 1987, a second national survey of caries in US children was completed by the NIDR. An attempt was made to use a sampling protocol and clinical diagnostic methods identical to those employed in the 1980 study in order that direct, quantitative comparisons of changes in caries prevalence could be made with confidence. Additionally, data were collected to monitor further the effect of exposure to fluoride in community water supplies.
Materials and methods.
The sample for the 1986-87 US survey was drawn from the population of all US children aged 5 through 17 years who were enrolled in public or private schools. A multi-stage probability sample of children was drawn to represent the 43 million children in these age groups.
The country was stratified into the same seven geographic regions used in the previous survey. Each of the regions was stratified further into Standard Metropolitan Statistical Area (SMSA) (urban) and non-SMSA (rural) areas and then into additional strata based on median county income. Five Primary Sampling Units (PSUs) were drawn from each of these 14 areas. Each PSU consisted of a public school district and nearby non-public schools. Additional PSUs (for a total of 83) were drawn from the more populous regions to increase the precision of the estimates for those areas. Measures were used in 1986 to maximize the retention of the PSUs selected in 1979 in order to reduce the number of children needed to detect a certain level of difference in DMFS between the two surveys. Within PSUs, a random selection of schools and classes was made.
Children in the selected classes whose parents had given signed consent were examined for dental caries and other oral health conditions by 14 dental teams trained and calibrated by NIDR staff according to standardized written diagnostic criteria (National Caries Program: NIDR, 1981). Inter-examiner reliability was assessed by replicate exams performed after training and was repeated several times during the course of the survey. Reproducibility of diagnoses compared with those of a reference examiner was generally high, with most values of the intraclass (reliability) correlation coefficient greater than 0.6.
In addition, the parents were asked to provide a complete residential history for their child, and to answer questions about the child's use of fluorides in the form of drops, tablets, treatments in the dental office or school-based fluoride programs. The residential histories were coded "complete" (no gap greater than three months) or "incomplete", and each residence was classified as fluoridated (natural or adjusted), nonfluoridated, or not ascertained. The Worldwide Geographic Codebook (1984), the Rand-McNally Road Atlas (1987), and Fluoridation Census (Centers for Disease Control, 1984) were used to make these determinations. Areas which could not readily be classified as to fluoridation status were referred to the Centers for Disease Control for resolution. The coded histories, along with birthdates and information on public water supplies, were used to determine whether a child had always, sometimes, or never lived in an area with community water fluoridation.
Results
Trends in dental caries. - The number of children examined, and the number to which the results can be extrapolated, are shown in Table 1 for each year of age. Examinations were completed for more than 39,000 schoolchildren who represented over 43 million children aged 5-17 years.
|
TABLE 1
NUMBER EXAMINED AND ESTIMATED POPULATION REPRESENTED, BY AGE (U.S. 1986-87) |
||
|
Age
|
Number Examined
|
Population Represented
|
|
5
|
1851
|
2,552,751
|
|
6
|
3098
|
3,980,732
|
|
7
|
3149
|
3,578,063
|
|
8
|
3256
|
3,211,415
|
|
9
|
3361
|
3,332,326
|
|
10
|
3463
|
3,357,708
|
|
11
|
3443
|
3,179,166
|
|
12
|
3186
|
3,206,386
|
|
13
|
2983
|
3,229,289
|
|
14
|
2983
|
3,473,894
|
|
15
|
2771
|
3,552,049
|
|
16
|
2912
|
3,581,737
|
|
17
|
2750
|
3,045,456
|
|
All Ages
|
39,206
|
43,280,974
|
Fig. 1 shows the mean number of decayed, missing, and filled permanent tooth surfaces (DMFS) by age and gender. As has often been reported, females had a higher caries experience than males at every age. Contrary to some earlier observations, however, white children had lower mean DMFS scores than non-whites (blacks and all others) at most ages (Fig. 2).
Regional differences in caries in the US have been noted in epidemiological studies since 1941 (Britten and Perrott, 1941). Similar differences were also observed in this investigation, as demonstrated in Fig. 3. Children living in the New England states (Region 1), the Northeast (Region II), and the Pacific Coast (Region VII) had the highest mean DMFS scores, and those in the Southwest (Region V) had the lowest. Although these differences were less marked than those observed in the study completed in 1980, the pattern was similar.
Caries had continued to decline in US children, as shown in Table 2, which gives the age-specific mean DMFS estimates from both the 1979-80 and 1986-87 surveys. For all ages, for the entire US, the relevant statistics are 4.77 DMFS per person in 1979-80 and 3.07 in the recent study, or an overall decrease of 36%. Nearly 50% of children in 1987 were completely caries-free in their permanent dentition, which was also a major change from the earlier survey, in which 37% were caries-free (Table 3).
|
TABLE 2
U.S. AGE-SPECIFIC CARIES EXPERIENCE (DMFS), 1979-80 AND 1986-87 |
||
|
Mean DMFS
|
||
|
Age
|
1979-80
|
1986-87
|
|
5
|
0.11
|
0.07
|
|
6
|
0.20
|
0.13
|
|
7
|
0.58
|
0.41
|
|
8
|
1.25
|
0.71
|
|
9
|
1.90
|
1.14
|
|
10
|
2.60
|
1.69
|
|
11
|
3.00
|
2.33
|
|
12
|
4.18
|
2.66
|
|
13
|
5.41
|
3.76
|
|
14
|
6.53
|
4.68
|
|
15
|
8.07
|
5.71
|
|
16
|
9.58
|
6.68
|
|
17
|
11.04
|
8.04
|
|
All Ages
|
4.77
|
3.07
|
|
TABLE 3
PERCENT OF U.S. CHILDREN CARIES-FREE IN TWO NATIONAL SURVEYS |
||
|
Percent of DMFS
|
||
|
Age
|
1979-80
|
1986-87
|
|
5
|
95.4
|
97.3
|
|
6
|
89.7
|
94.4
|
|
7
|
76.5
|
84.2
|
|
8
|
58.6
|
75.0
|
|
9
|
50.6
|
65.5
|
|
10
|
37.9
|
55.7
|
|
11
|
33.7
|
45.0
|
|
12
|
26.9
|
41.7
|
|
13
|
21.1
|
34.0
|
|
14
|
19.6
|
27.7
|
|
15
|
14.9
|
21.8
|
|
16
|
11.8
|
20.0
|
|
17
|
10.7
|
15.6
|
|
All Ages
|
36.6
|
49.9
|
The percent decrease in mean DMFS during this time interval is shown for each year of age in Fig. 4, and for each geographic region in Fig. 5. Clearly the decline in caries was a nationwide phenomenon and, despite some irregularities, had continued to occur in children at all ages studied.
Shown in Table 4 are the percentages of total DMFS scores attributable to decayed, missing, and filled surfaces, with the same statistics from the earlier survey included for comparison. Of once-carious surfaces, 82% had been restored, and only 4% were missing, the latter translating to about 0.1 missing surface per person, or about one missing permanent tooth for every 50 children. The percentages of both filled and missing surfaces also represent improvements compared with 1979-80.
|
TABLE 4
PERCENT OF DMFS DUE TO DECAYED, MISSING, AND FILLED SURFACES, U.S. CHILDREN, 1979-80 AND 1986-87 |
||
|
Percent of DMFS
|
||
|
1979-80
|
1986-87
|
|
|
Decayed
|
16.8
|
13.4
|
|
Filled
|
76.1
|
82.3
|
|
Missing
|
7.1
|
4.3
|
The decrease in caries experience was not uniform for all surface types. As shown in Fig. 6, the decline in caries, compared with 1979-80, was about 32% on both occlusal and buccal-lingual surfaces, but about 54% on mesial and distal surfaces. Considering that buccal-lingual DMFS scores included caries in buccal pits and lingual grooves in molars, it seems likely that caries on smooth surfaces is approaching eradication.
To summarize the overall trends in dental caries in US schoolchildren during the past 15-16 years, Fig. 7 shows the age-specific DMFS means from both studies conducted by the NIDR, together with those from an earlier survey conducted by the National Center for Health Statistics (National Caries Program: NIDR, 1981). During the time-interval covered, approximately 14 years, the overall fall in caries experience was about 53%, or an annual average decreased rate of approximately 4%.
The effect of water fluoridation. - Complete residence histories were available for about 92% of subjects, and 8165 were classified as having lifelong continuous exposure to water fluoridation. A similar number (8233) had never lived in an area which was fluoridated. The distribution of these two groups, by age, is given in Table 5.
|
TABLE 5
NUMBER OF U.S. CHILDREN EXAMINED WITH OR WITHOUT LIFE-LONG WATER FLUORIDATION BY AGE (1986-87) |
||
|
Age
|
Life-long Water Fluoridation
Exposure
|
No Water Fluoridation
Exposure
|
|
5
|
227
|
229
|
|
6
|
705
|
645
|
|
7
|
764
|
780
|
|
8
|
782
|
757
|
|
9
|
766
|
811
|
|
10
|
802
|
710
|
|
11
|
716
|
756
|
|
12
|
649
|
687
|
|
13
|
616
|
613
|
|
14
|
590
|
600
|
|
15
|
504
|
559
|
|
16
|
529
|
551
|
|
17
|
515
|
535
|
|
All Ages
|
8165
|
8233
|
As shown in Table 6, the mean DMFS of children with continuous residence in fluoridated areas was about 18% lower than in those with no exposure to fluoridation. This pattern was fairly consistent over age, except at 5 and 6 years old, where very little caries had occurred in either group.
|
TABLE 6
MEAN DMFS OF U.S. CHILDREN WITH PERMANENT TEETH BY AGE AND WATER FLUORIDATION EXPOSURE |
|||
|
Life-long Water Fluoridation
Exposure
|
No Water Fluoridation
Exposure
|
||
|
Age
|
Mean DMFS*
|
Mean DMFS*
|
Percent Difference
|
|
5
|
0.03
|
0.10
|
70
|
|
6
|
0.14
|
0.14
|
0
|
|
7
|
0.36
|
0.53
|
32
|
|
8
|
0.64
|
0.79
|
19
|
|
9
|
1.05
|
1.33
|
21
|
|
10
|
1.64
|
1.85
|
11
|
|
11
|
2.12
|
2.63
|
19
|
|
12
|
2.46
|
2.97
|
17
|
|
13
|
3.43
|
4.41
|
22
|
|
14
|
4.05
|
5.18
|
22
|
|
15
|
5.53
|
6.03
|
8
|
|
16
|
6.02
|
7.41
|
19
|
|
17
|
7.01
|
8.59
|
18
|
|
All Ages
|
2.79
|
3.39
|
18
|
| * All means adjusted to age and gender distribution of toal U.S. population ages 5-17. | |||
No noteworthy differences were observed between groups in relation to restorative care received (%F/DMFS) or required (%D/ DMFS), as shown in Table 7. However, with respect to the types of tooth surfaces affected, children with exposure to fluoridation had proportionally less caries on mesial and distal surfaces (Table 8). Although it might be expected that a similar difference would be evident on buccal and lingual surfaces, this was not the case. It is probable that the inclusion of pit and fissure lesions in the categories of "buccal and lingual" caries obscured the expected difference on these free smooth surfaces.
|
TABLE 7
DISTRIBUTION OF DMFS IN U.S. CHILDREN BY COMPONENTS AND WATER FLUORIDATION EXPOSURE |
|||
|
%D/DMFS
|
%F/DMFS
|
%M/DMFS
|
|
| Life-long Water Fluoridation Exposure |
13.0
|
83.7
|
3.3
|
| No Water Fluoridation Exposure |
14.0
|
82.3
|
3.7
|
|
TABLE 8
DISTRIBUTION OF DMFS IN U.S. CHILDREN BY SURFACE TYPE AND WATER FLUORIDATION EXPOSURE |
|||
|
Percent of Children
by Type of Surface
|
|||
|
Buccal-Lingual
% |
Mesial-Distal
% |
Occlusal
% |
|
| Life-long Water Fluoridation Exposure |
31.0
|
9.0
|
60.0
|
| No Water Fluoridation Exposure |
30.0
|
14.0
|
56.0
|
As shown in Table 9, mean DMFS scores in children with life-long exposure to water fluoridation were also lower than in their unexposed counterparts in most geographic regions of the country. An exception was in the Midwest, Region III, where exposed children had a slightly higher mean DMFS. There is no obvious explanation for this departure from the overall pattern. However, Region III had, by far, the largest percentage of the population on fluoridated water supplies (Table 9), and the mean DMFS estimate for unexposed children is therefore based on relatively small numbers of subjects (n = 304).
|
TABLE 9
MEAN DMFS* FOR EACH U.S. REGION BY COMMUNITY WATER FLUORIDATION STATUS (1986-1987) |
|||
| Region |
Life-long Water Fluoridation
Exposure
|
No Water Fluoridation
Exposure
|
Percent of U.S. Population
Served Fluoridated Water
|
| I |
3.11
|
3.45
|
55
|
| II |
3.08
|
3.42
|
49
|
| III |
2.86
|
2.69
|
74
|
| IV |
2.75
|
3.60
|
54
|
| V |
2.49
|
2.71
|
59
|
| VI |
2.36
|
3.07
|
34
|
| VII |
1.42
|
3.61
|
19
|
| All U.S. |
2.79
|
3.39
|
53
|
Observed differences in caries experience between children with life-long exposure to fluoridated water and those with no exposure are, of course, partly influenced by exposure to topical fluorides, which was common in both groups (Table 10). In an attempt to assess more clearly the effect of water fluoridation, children with a reported history of exposure to supplemental or topical fluorides were removed from the analysis. The mean DMFS in those continuously exposed to water fluoridation (n = 3574) was then 25% lower than in those with no fluoridation exposure (n = 2380).
|
TABLE 10
PERCENT OF U.S. CHILDREN AGED 5-17 YEARS REPORTING USE OF SUPPLEMENTAL OR TOPICAL FLUORIDES BY RESIDENCE STATUS |
||
|
Supplemental or Topical
Fluoride
|
Life-long Water Fluoridation
Exposure
|
No Water Fluoridation
Exposure
|
|
Drops/Tablets
|
14.8
|
54.2
|
|
Dentist's Office
|
48.6
|
54.3
|
|
School Program
|
9.8
|
22.0
|
|
Children with At Least
One Supplemental/Topical
|
55.6
|
70.7
|
Discussion.
The sharp decline in dental caries among US schoolchildren, which was first documented in 1980, evidently continued at approximately the same rate during the subsequent 6-7 years. Neither of these NIDR-conducted national epidemiological studies was designed to enquire into the cause of this decline. In fact, because of the multifactorial etiology of caries, it is doubtful that an unequivocal explanation for this trend can be found, though there can be little argument that the ubiquity of fluoride in the environment has been a dominant factor.
By 1986, an estimated 130 million persons, or over 53% of the US population, were receiving optimally fluoridated domestic water (Centers for Disease Control, 1988). However, other methods of fluoride delivery had also become common. Fluoride-containing dentrifices accounted for nearly 90% of all dentrifices sold during 1981-1985 in the US, compared with less than 75% a decade earlier. A large, though unknown, number of children were participating in school-based fluoride mouthrinse programs, and a fluoride-containing mouthrinse was available over the counter for home use. Exposure to fluoride was also occurring in the form of tablets, drops, topical applications in dental offices, and in the food chain. When some of this "background" effect of supplemental or topical fluoride exposure was removed from the data, the protection against caries attributable to water fluoridation increased. However, this effect is still underestimated, since there was no way to control for either exposure to fluoride in dentrifices or incidental exposure to fluoride in the diet. It seems safe to assume that practically all US children are exposed to fluoride to some degree, and on a continuing basis. It is therefore not surprising that the caries-preventive effect of water fluoridation has appeared to decrease over time, and it can be expected that this will continue.
No convincing evidence has been offered to refute the conclusion that the dramatic decrease in dental caries observed in the US children during recent decades is primarily the result of this exposure to fluoride. Furthermore, as far as we know, the caries-preventive effects of fluoride are both subject- and time-specific, in the sense that they are not experienced by persons not directly exposed to the fluoride ion and, unlike disease-prevention methods which result in the eradication of some etiological agent, the benefits of fluoride are not transferable to future generations. Nevertheless, it is likely that if caries in children remains at low levels or declines further, the necessity of continuing the current variety and extent of fluoride-based prevention programs will be questioned. Unfortunately, such a debate will be lacking in scientific substance, since the relative contribution to overall caries prevention of the different forms of fluoride delivery cannot be measured, except under experimental conditions. Therefore, the effect of terminating some of these programs cannot be predicted, though opportunities occasionally arise for relevant data to be gathered. Thus, Kunzel (1980) demonstrated that a steady downward trend in the age-specific mean DMFS was halted and in some age groups reversed, when technical problems resulted in several years of sub-optimal fluoridation in Karl-Marx-Stadt, DDR. When optimal levels of fluoride were restored to the water supply, the decline in caries resumed. More recently, Stephen et al. (1987) reported that caries in deciduous teeth (mean dfs) had increased by 40% five years after the termination of fluoridation in Wick, Scotland, even though fluoride dentrifices were widely available. Thus, the only available recent data, although limited, suggest that caries levels in children can be expected to rise where water fluoridation is interrupted or terminated. Unfortunately, little information is available on which to base estimates of the results of curtailing other forms of fluoride delivery.
Finally, it must be emphasized that, with respect to the overall benefits of water fluoridation, the question of cost-effectiveness is frequently overlooked. Furthermore, whether the protective mechanism of fluoride is exerted systemically or topically remains unanswered, although the weight of accumulated evidence suggests that the topical effect is predominant. However, delivery of fluoridated water is by far the most cost-effective method to ensure the topical exposure of teeth continuously from eruption, and to maximize patient compliance (or eliminate this critical variable).
At present there are many unanswered questions regarding the reasons for the continued downward trends in dental caries in US children, but there can be little doubt that for this decline to be maintained, fluoride-based prevention must be continued. Further, to ensure optimum delivery of fluoride and to minimize costs, it may be concluded that water fluoridation must continue to be the major component of this effort, at least in the United States.
REFERENCES
BRITTEN, R.H. and PERROTT, G. (1941): Summary of Physical Findings on Men Drafted in the World War, Pub Hlth Rep 56:41-62.
BRUNELLE, JA and CARLOS, JP (1982): Changes in the Prevalence of Dental Caries in US Schoolchildren, 1961-1980, J Dent Res 61 (Sp Iss): 1346-1351.
BRUNELLE, JA; MILLER. A.J.; and SMITH, Ll. (1983): DMFS in US Children With and Without Life-Long Exposure to Water Fluoridation, J Dent Res 62:203, Abst. No. 302.
Center for Disease Control (1984): Fluoridation Census 1980, Atlanta, GA: US Dept. of HHS. U.S.P.H.S. (June 1984).
Center for Disease Control (1988): Fluoridation Census 1985, Atlanta, GA: US Dept. of HHS, U.S.P.H.S. (July 1988).
GLASS, R.L., Ed. (1982): First International Conference on the Declining Prevalence of Dental Caries, J Dent Res 61 (Sp Iss):1304-1383.
KUNZEL, W. (1980): Effect of an Interruption in Water Fluoridation on the Caries Prevalencc of the Primary and Secondary Dentition, Caries Res 14: 304-310.
NATIONAL CARIES PROGRAM: NIDR (1981): The Prevalence of Dental Caries in United States Children. NIH Publ. No. 82-2245.
Rand-McNally Road Atlas - United States, Canada, Mexico (1987). Rand-McNally and Company, Chicago, IL.
STEPHEN, K.W.; McCALL, D.R.; and TULLIS, J.I. (1987): Caries Prevalence in Northern Scotland Before, and 5 Years After, Water Defluoridation, Br Dent J 163:324-326.
Worldwide Geographic Location Codes (April
1984). US General Services Administration Office of Finance, Washington.
DC.