- Email: [email protected]
Fluorine in Drinking Water: its Effect on Dental Caries
F L U O R IN E IN D R IN K IN G W A T E R : IT S E F FE C T O N D E N T A L C A R IE S
Francis'A. Arnold, Jr., D.D.S., Bethesda, Md.
D u r i n g the last decade dentistry has
been presented with evidence suggest ing a practical method for partially controlling dental caries in large popula tion groups ; that is, by means of fluorination of domestic water supplies. Such a suggestion originated principally from studies of the effects of fluoride-bearing waters on the prevalence of caries. In view of the fact that hitherto no effective means of mass control of this disease has been known, both the dental profession and the laity are extremely interested in the evidence presented. Often the hopes and expectations of both groups go be yond logical interpretation of the scien tific evidence. It is the purpose of this paper to re view briefly some of the major points of evidence reported to date, with special reference to the epidemiologic studies of water-borne fluorides. A more complete review of the dental caries-fluorine rela tionship can be found in the recent pub lication of the American Association for the Advancement of Science.1 Prior to 1938, most dental research in regard to water-borne fluorides centered on the problem of dental fluorosis. T o day the cause of this disease and the methods of its prevention are known. Of special importance in the dental cariesfluorine phenomenon is the fact that the concentrations of fluorides that produce
fluorosis also are known.'Aside from the establishment of these basic facts, addi tional knowledge has been obtained. As stated by Dean,2 “ Studies in endemic dental fluorosis provided the proving ground whence evolved many of the ideas and methodologies later utilized in the studies designed to outline the epi demiological characteristics of the fluo rine-dental caries relationships.” Before the etiologic agent of dental fluorosis, fluorine, was known and prior to the first report of low caries rates in endemic areas of dental fluorosis, men tion was made of the fact that caries was no more prevalent, and probably less so, in persons who exhibited evidence of fluorosis.3' 5 Special interest of dental re search workers was focused on the cariesfluorine relationship a decade ago after fluorine was recognized to be the cause of mottled enamel and the earlier ob servations were supported by the results of both laboratory and epidemiologic studies. Epidemiologic Studies After Dean6 had reported concerning the lowered dental caries rates in areas of endemic dental fluorosis in South D a 2. Dean, H . T ., Some General Epidemiological Con siderations, in Dental Caries and Fluorine ,1 pp. 1-4. 3. M cKay, F . S., Mottled Enamel: A Fundamental Problem in Dentistry. D. Cosmos 67:847 (September)
19*5-
4. Bunting, R . W., and others,( Further Studies of the Relationship of Bacillus Acidophilus to Dental From Dental Research Section, Division of Physi Caries ( III) . D. Cosmos 70:1002 (October) 1928. ology, National Institute of Health, Bethesda, Md. 5. M cKay, F . S., The Establishment of a Definite Read at the eighty-eighth annual session of the Relation Between Enamel That Is Defective in Its American Dental Association, Boston, August 6, 1947. Structure, as Mottled Enamel, and the Liability to Dental Surgeon, U . S. Public Health Service. Decay ( II) . D . Cosmos 71:747 (August) 1929. . 1. Dented Caries and Fluorine. Washington, D . C .: 6. Dean, H . T ., Endemic Fluorosis and Its Relation American Association for the Advancement of Science, to Dental Caries. Pub. Health R ep. 5 3 :14 4 3 - (August 1946. 19 ) »938.
Arnold
kota, specific epidemiologic studies were inaugurated -to gain further knowledge of this phenomenon. These studies were de signed to check by detailed local inves tigations the relation between fluorosis and caries. The observations were limited to spe cific groups according to age and sex. White school children, aged 12 to 14 years, were selected. Since it was logical to assume that this phenomenon might be related to differences in domestic wa ter supplies, only children who used con tinuously one common water supply were included. The cities selected for study had domestic water supplies which met the necessary prerequisite as to fluoride content and had made no changes in their water supplies for a period greater than the age of the children studied. Fac tors such as dietary habits (especially carbohydrate intake), latitude and sun light were taken into account in so far as possible. Complete details of the epidemi ologic technics employed in the selection of children and areas of study have been presented by Dean.7 The first of these investigations was made in four cities in north central Illi nois.8 Two of these cities, Galesburg and Monmouth, had communal water sup plies which contained 1.8 and 1.7 parts of fluoride per million, respectively. Quincy and Macomb, the control cities, used domestic water supplies which were fluoride-free; that is, they contained less than 0.2 p.p.m. of fluoride. The prevalence of dental caries in the children of these four communities dif fered greatly. For example, the average 13 year old child in Galesburg had ex-' perienced caries in only about two per manent teeth, but in Quincy the average 13 year old child had about six teeth showing caries experience. It also was 7. Dean, H . T ., Epidemiological Studies in the United States, in Dental Caries and Fluorine ,1 pp. 5-3i. 8. Dean, H . T ., and others^ Domestic Water and Dental Caries, Including Certain Epidemiological As pects of Oral Lactobacillus Acidophilus. Pub. Health Hep. 5 4 : 86q (May 26) 1939.
J.A.D.A., Vol. 36, January 1948 . . .
29
noted that in Galesburg and Monmouth approximately 36 per cent of the chil dren were caries-free as compared with 14 per cent in Macomb and 4 per cent in Quincy. Other interesting findings of this study were the differences in the distribution of high and low Lactobacillus acidoph ilus counts, the striking differences in proximal and smooth surface caries and the relation between macroscopic flu orosis and dental caries. In the case of the last, it was observed that the low ered caries rates in Galesburg children were not associated with the presence of. dental fluorosis by itself. This find ing confirmed the earlier observations by Dean.6 The evidence from this study together with other information available at that time pointed to water supplies and pre sumptively to the fluoride content of those water supplies as the primary asso ciated factor. This raised the question of whether a decreased prevalence of dental caries is associated with the use of water supplies whose fluoride content is less than the concentration necessary to pro duce any significant amount of dental fluorosis. In order to answer this question a de tailed study was made of school children in eight suburban Chicago communities. The communities and the children were selected in the manner previously de scribed.9 The results of this study con firmed the findings at Galesburg and Quincy in regard to the marked differ ences in the prevalence of dental caries between children using fluoride waters as compared with children using fluoride-free water. As. shown in Table 1, the caries experi ence rates in Elmhurst, Maywood, Au rora and Joliet, whose public water sup plies contain 1.8, 1.2, 1.2 and 1.3 p.p.m. 9. Dean, H. T ., and others, Domestic Water and Dental Caries. II. A Study of 2,832 White Children, Aged 12-14 Years, of 8 Suburban Chicago Communi ties. Including Lactobacillus Acidophilus Studies of 1,761 Children. Pub. Health Rep. 56:761 (April 11) 1941*
30
of fluoride, respectively, were 252, 258, 281 and 323, respectively. The children of Evanston, Oak Park and Waukegan, using fluoride-free water, showed caries experience rates of 673, 722 and 810, re spectively. The results of the examina tions at Elgin, where the public water supply contains about 0.5 p.p.m. of fluo ride, indicate a caries experience rate (444) that is about midway between that of the fluoride and fluoride-free areas. The observations in this study were suggestive of an inverse variation of den tal caries and the fluoride concentration of the domestic water. It also was noted that the inhibitory factor, presumably fluorine, was operative at concentrations below what is now the accepted per missible standard for water, 1.5 p.p.m.10 At these low concentrations dental fluo rosis is no problem. Evidence of this fact was obtained in Aurora where al though 15 per cent of the children were classed as having fluorosis, only 5 per cent of all the teeth and only about 0.4 per cent of the anterior teeth showed even so much as the mildest forms of fluorosis. This minor amount of affectioij, which is noticeable only to a trained observer, is of no esthetic or public health signifi cance. As may be noted in T able 1, there were differences other than the fluoride con tent in the water supplies of the com munities studied. For example, the cities in which the caries rates were low used well waters while those with high caries prevalence rates used surface (Lake Michigan) water. Thus, comparisons were made on water supplies of varying mineral content. In order to check this variable and to confirm the observa tion regarding the effect of the lower ranges of fluoride concentrations (1 p.p.m. and less), epidemiologic studies in other communities were conducted. By the use of similar technics of exam ination and the same criteria of selec-
The Journal of the American Dental Association
tion, children 12 to 14 years of age in nineteen additional communities were studied. The communities selected had water supplies which varied from a high fluo ride content with a low mineral content to fluoride-free waters with a high min eral content. The sources of these water supplies were deep wells, lakes and rivers, and cisterns. Including the children in the eight Chicago suburbs, a total o f 8,576 children in twenty-seven cities o f eight states have been studied. The results of these examinations are shown in the il lustration and in Table 2. M ore com plete details concerning these study areas were presented by Dean, Arnold and Elvove.11 The data suggest the following conclu sions: First, children who are born and reared in communities whose domestic water supplies contain increased amounts of fluorine (more than 0.5 p.p.m.) ex perience less dental caries than do chil dren living in areas where the domestic water supply is fluoride-free (Tables 1 and 2). Second, this phenomenon is op erative whether or not the teeth show macroscopic evidence of fluorosis. Third, the decreased prevalence of dental caries is evident in areas where the fluoride content of the water is below those levels which produce dental fluorosis of any esthetic significance. A careful analysis of the data strongly suggests fluorine as the active inhibitory factor present in these water supplies. As stated previously, factors other than those present in the water supplies which probably could have influenced the den tal caries experience were taken into ac count and in so far as possible were equalized by careful selection of the study areas. As for the variability in water sup plies, one characteristic other than fluo ride which has been suggested as low
11. Dean, H . T .; Arnold, F- A ., Jr., and Elvove, Elias, Domestic Water and Dental Caries. V . Addi tional Studies of the Relation of Fluoride Domestic Waters to Dental Caries Experience in 4 ^ 2 5 White 10. Public Health Service Drinking Water Stand Children, Aged 12 to 14 years, of 13 Cities in 4 States. ards, 1946. Pub. Health Rep. 6 1:3 7 1 (March 15) 1946. Pub. Health R ep. 57:115 5 (August 7) 1942.
Arnold
J.A.D.A., Vol. 36, January 1948 . . .
ering dental caries experience is the total hardness of the water. In the study of the twenty-seven cities, however, there is lit tle evidence of correlation between the total hardness of the water and caries prevalence (Table 2). For example, M id dletown and Zanesville, Ohio, and Elk hart, Ind., have waters of relatively high-' total hardness (329, 291 and 220 p.p.m,
31
respectively) and show caries experience rates of 703, 733 and 823, respectively. On the other hand, Colorado Springs, Colo, and Maywood, 111., using waters of much lower total hardness (27 and 75 p.p.m., respectively) had lower caries rates (246 and 258, respectively). The one variable, however, which does corre late with caries prevalence is the fluoride
Table I.— Summary of dental caries findings in 2,832 selected white children, aged 12 to 14 years, in eight suburban Chicago communities in relation to the fluoride content of the public water supply9
Elm May hurst wood Aurora Joliet
873
1,625
1,412
1,030 2,125
1,662 1,354
Number of 12-14 year old white chil dren whose histories on repeated questioning indicated continuity of exposure and who were examined 170
171
633
447
403
329
26.9
19.6
Watei supply
Source Permanent hardness in parts per million 323.4
1.8
Dental caries experience, permanent teeth, per 100 children examined 252
Clinical aminati
39.0
31.7
39.1
Deep wells
Meanfluoride(F) content, 1939- 1940, in parts per million
X V
Evans Oak Wau ton Park kegan
Total number of 12-14 year old chil dren present at time of sampling 633
Percentage of the total present who were examined
G O
Elgin
Dental caries experience, proximal surfaces, superior incisors, per 100 surfaces
0.60
75.0 328.5 1.2*
258
0.59
1.2
281
0.78
256 12.0
19.8
CO CN
Sampling
City or village
31.2
Lake Michigan
349.3
102.6
1.3
0.5
323
444
131.0 132.2 134.4 0.0
673
0.0
722
0.0
810
1.3
4.1
10.7
9 .0.
17.7 79.9
First permanent molar mortality, per 100 children examined
11.8
11.7
14.5
19.5
20.3
42.6
31.0
Percentage of children with no den tal caries experience
25.3
29.8
23.5
18.3
11.4
3.9
4.3
3. 1 '
Percentage incidence of endemic dental fluorosis (mottled enamel)
40.0
33.3
15.0
25.3
4.2
1.6
0.6
0.2
Percentage of those ex- Negative amined bacteriologically and < 10 0
36.7
39.6
38.2
35.6
21.2
15.4
18.2
14.9
24.7
20.9
25.6
26.6
33.2
41.8
43.8
54.6
ctf £O dophilus counts were: M
30,000 ánd>
*There is both presumptive and direct evidence that prior to a few years ago the Maywood water contained probably 1.4-1.6 p.p.m. of fluoride (F ).
32
The Journal of the American Dental Association
content of these water supplies. Attention is called to the possibility that other factors or combinations of fac tors associated with water may play a role in the carious process or in the cariesfluorine relationship. At present compara tively little information is available re garding the physiologic role of such a major part of our daily diet as our drink ing water. It is known that there is a great variation at least in the chemical composition of the different waters; fur ther study may disclose physiologic effects ,
resulting from some of these variables. In addition to indicating that fluorine is the active inhibitory factor, the obser vations in the twenty-seven cities also suggest that almost full benefit is ob tained if the fluoride content is only about i p.p.m. At this low range of fluoride concentration the dental fluorosis index is less than 0.5 (see illustration). A de scription of the method12 of establishing 12. Dean, H . T ., The Investigation of Physiological Effects of the Epidemiological Method, in Fluorine and Dental Health. Washington, D . C .: American Association for the Advancement of Science, 1942.
Table ¿.— Summary of dental caries and fluorosis findings in 8,576 selected white school children, 12 to 14 years of age, in twenty-seven cities of eight states in relation to fluoride content, total hardness and source of domestic water supply
Water supply
Clinical
City
Hereford, Texas. . . . Colorado Sprgs, Colo. Galesburg, 111.......... Elmhurst, 111........... Maywood, 111.......... East Moline, 111. . . . Joliet, 111................ Aurora, 111............... Kewanee, 111............ Vicksburg, Miss.. . . Nashville, Tenn.. . . Clarksville, Tenn.. . Pueblo, Colo........... Marion, Ohio.......... Elgin, 111.................. Lima, Ohio............. Evanston, 111........... Zanesville, Ohio. . . . Escanaba, Mich... . Portsmouth, Ohio. . Middletown, Ohio.. Oak Park, 111.......... Elkhart, Ind........... Quincy, 111............... Waukegan, 111......... Michigan City, Ind.
Caries experience Chil Total dren in permanent teeth Per cent children Fluoride, hard ex p.p.m. ness, with amined Per cent Number with per 100 fluorosis p.p.m.* children none
60 404 273 170 171 152 447 633 123 1-72 662 60 614 263 403 454 256 459 270 469 370 329 278 330 423 236 95
38.3 28.5 27.8 25.3 29.8 20.4 18.3 23.5 17.9 S.7 9.4 16.7 10.6
5.7 11.4 2.2
3.9 2.6 1.1
1.3 1.9 4.3 1.4 2.4 3.1 0.0 0.0
147 246 236 252 258 303 323 281 343 587 461 458 412 556 444 652 673 733 877 772 703 722 823 706 810 1,037 1,070
100.0
3.1
73.8 47.6 40.0 33.3 31.6 25.3 15.0
2.6
12.2 11.0 8.6
0.9
8.3 6.5 6.1
4 .2 . 2.2 1.6
1.5 1.5 1.3 1.1 0.6
0.4 0.3 0.2 0.0 0.0
1.9 1.8 1.2f 1.2t
1.3 1.2 0.2 0.0 0.2 0.6
0.4 0.5 0.3 0.0 0.2 ' 0.2 0.1 0.2 0.0 0.1 0.1 0.0 0.1 0.1Î
383 27 247 323 75 276 349 329 445 94 79 89 302 209 103 223 131 291 123 80 329 132 220 88
134 141 36Î
Source
Deep wells Surface (Pike’s Peak) Deep wells Deep wells Deep wells Deep wells Deep wells Deep wells Deep wells Yazoo and Miss. River Cumberland River Cumberland River Arkansas River Deep wells Deep wells Impounded surface Lake Michigan Deep wells Lake Michigan Ohio River Deep and shallow wells Lake Michigan Deep wells Mississippi River Lake Michigan Lake Michigan Cisterns
*T otal hardness calculated'in accordance with Standard Methods for the Examination of Water and Sewage. New York: American Public Health Association. 1936. fT here is both presumptive and direct evidence that these water supplies contained about 1.5 p.p.m. of fluoride prior to a few years ago. 4 ^ < ^ ^Average of five water samples from cisterns considered typical of those used by most K ey West inhabitants.
Arnold
J.A.D.A., Vol. 36, January 1948 . . . 33
Relation between the amount of dental caries in the perm anent teeth of 8,sy6 white children 1 2 to 14 years old in twenty-seven cities of eight states and the index of dental fluorosis
a fluorosis in d e x has been g ive n b y D e a n . taine d 0.25 p .p .m . O ck erse15 re p o rte d T h e p ro b le m of d e n ta l fluorosis in these s im ila r findings in S o u th A fric a . areas is n e gligib le as in d ica te d b y the ob T h is evidence, like m ost e p id em io lo gic servation in A u ro ra , 111., as p revio usly evid en ce, is p re s u m p tive in ch a ra cte r,16 described. In cre a s in g the flu o rid e content b u t in a d d itio n to these studies on h u m a n m u c h above i p .p .m . produces little a d beings there are reports fro m la b o ra to ry d itio n a l in h ib itio n o f caries b u t does m a research w h ic h in d ica te th a t flu o rin e in te ria lly increase the a m o u n t a n d degree fluences the carious process in e x p e ri o f fluorosis.13 m e n ta l anim a ls. A n u m b e r of authors I n studies in this a n d o th e r countries h a ve s h o w n th a t in d u c e d caries in the observations of a sim ila r ch a ra cte r h a ve m olars o f rats m a y be in h ib ite d b y the been m a d e . W e a v e r14 m a d e a stud y of a d d itio n o f fluorides to the d rin k in g w a d ental co n d itio ns in the B ritis h Isles an d ter o r the foo d a n d th a t this ca rie s-in fo u n d th a t the c h ild re n o f S o u th Shields h ib itin g actio n is related to the a m o u n t of h a d o n ly a b o u t h a lf as m u c h d ental caries flu o rid e ingested. A ls o , d ental caries in experience as d id the c h ild re n o f N o rth S y ria n ham sters has been in flu en ced in a Shields, w h ic h is ju s t across the R iv e r like m a n n e r. A re v ie w of e xp e rim e n ta l T y n e . H e a ttrib u te d the v a ria b ility in caries a n d fluorides w as m a d e b y H o d g e caries pre va le n ce to differences in the a n d Sognnaes.17 dom estic w a te r su p p ly, specifically to the A lth o u g h it is n o t k n o w n th a t d ental flu o rid e co n ten t. T h e w a te r sup ply of caries in h u m a n beings is co m p a ra b le to S o u th Shields co n taine d 1.4 p .p .m . of 15. Ockerse, T ., Fluorine and Dental Caries in South flu o rid e a n d th a t of N o rth Shields c o n Africa, in D e n t a l C a r ie s a n d F lu o r in e ,1pp. 36-42. 13. In Hereford, Texas (Table 2 and illustration) the number of continuous residents is small, and the low caries rate observed may be due to sampling error. 14. Weaver, Robert, Epidemiological Studies in the British Isles and India, in D e n t a l C a rie s a n d F l u o r in e / PP.
32- 35*
16. Although the epidemiologic evidence is pre sumptive in character, this fact does not decrease the validity of the evidence. Chlorination of water sup plies for the control of typhoid fever is based on sim ilar presumptive evidence. 17. Hodge, H. C ., and Sognnaes, R . F ., Experi mental Caries and a Discussion of the Mechanism of Caries Inhibition by Fluorine, in D e n ta l C a r ie s a n d F lu o r in e ,1pp. 53*73*
34
The Journal of the American Dental Association
induced caries in animals, such labora tory evidence lends weight to the pre sumption that fluorine is the active in gredient in the water supplies in areas of decreased caries rates.
differences were noted in the o-ioo range (inactive caries) and in those counts over the 20,000-30,000 range (active caries). The differences in the counts paralleled the differences in caries experience rates observed in these children. Mechanism Studies Cities using waters with a fluoride con • At the present time the mechanism of tent of more than i p.p.m. averaged 37.5 the action of fluoride waters on the cari per cent in the o-iop range compared ous process is not known; however, the with 17.8 per cent in the fluoride-free limited information available sheds some areas (Table 3). In the intermediate fluo light on this phenomenon. One of the ride level (0.5 p.p.m.) 21.2 per cent of more important observations made in the counts were in the low range. Con connection with the epidemiologic studies versely, the high counts (20,000 and is in regard to changes in the oral bac more) were 28.7 per cent in the fluoride tériologie flora. It is a generally accepted areas and 52.4 per cent in the fluoridebelief that dental caries activity can be free areas. These differences in the per diagnosed by determining the number centage distribution of high and low Lac of Lactobacilli in the saliva. The oral tobacillus counts suggest that the cariesflora of persons in whom caries is active fluorine relationship may be based on shows a large number of Lactobacilli per action of the fluoride waters on the oral cubic centimeter, but in persons in whom flora. In other words, they point to a de caries is inactive or absent the oral flora crease in the activating forces of the is negative for Lactobacilli, or the organ carious process. This result would seem to ism is present only sporadically and in indicate a theory of action opposed to that suggested by laboratory evidence on low numbers. The results of such determinations in chemical composition of the teeth.13 Additional information on the mecha the fluoride and nonfluoride areas, shown in Table 3, indicate that there is a definite nism of action of fluoride waters has been difference in the percentage distribution gained by studying persons who gave a of Lactobacillus counts in the saliva of history of change from fluoride to fluo children included in some of the epidemi ride-free water supplies or vice versa. In ologic studies.8’ 9' 18 The most striking Bauxite, Ark., it was observed that chil18. Jay, P h ilip, and A rn o ld , F . A ., J r., E p idem io logical A spects o f O ral Lactobacillus Counts in F lu o rid e an d N on -F lu oride Areas, in Dental Caries and Fluorine,1 p p . 43-46.
19. Arm strong, W . D ., C h em ical D ifferences o f C a ries Susceptible and Im m u ne T e e th and a Consideration o f . F o o d Sources o l F luorine, in Dental Caries and Fluorine,1 p p . 47-52.
Table 3.— Percentage distribution of oral Lactobacillus counts on 12 to 14 year old children living in fluoride and nonfiuoride areas*
Fluoride ' , ■ content of , domestic water, p.p.m. 1.0 0.5 0.0
-
L. acidophilus counts Children examined
1,052 250 1,087
0-100, per cent
20,000 and more, per cent
37.5 21.2 17.8
28.7 40.0 52.4
Dental caries experience (rate per 100) f 280 444 760
♦Includes children from the follow in g cities: Q u in cy , Evanston, O ak Park, W aukegan, E lgin, Elm hurst, M a y w ood , A u rora , Joliet and G alesburg, 111., and Escanaba, M ich . fB a sed on all children given dental exam ination in respective com m unities.
Arnold
dren whose teeth were formed on a fluo ride water and erupted into a fluoridefree environment experienced less dental caries than children of a neighboring community who had used fluoride-free water throughout life.20 This finding is in accord with the results reported by Deatherage21 on selectees in Illinois. Observations somewhat opposite in character were made in Maywood, 111., Garrettsville, Ohio and Escanaba, M ich.22’ 23 In Maywood a study was made of children who had moved to that city at varying ages from areas using fluoridefree Lake Michigan water. With the ex ception of those children who had moved to Maywood prior to 4 years of age, there was no significant evidence of any inhibi tion of caries incidence following the change from a fluoride-free to a fluoride environment. At Garrettsville, Ohio, and Escanaba, Mich., the respective water supplies were changed from fluoride-free water to that containing increased amounts of fluorine. Dental and bac tériologie examinations in Garrettsville and bacteriologic examinations in Es canaba gave no indication of a change in caries activity in children whose teeth had formed on fluoride-free water. Analysis of these results suggests the necessity of an increased ingestion of fluo ride from birth or at least during the period of tooth formation in order to ob tain beneficial effects. These results, how ever, are not entirely in accord with the findings of Deatherage.24 20. D ean, H . T . , and others, D om estic W ater and D ental Caries. I . A D ental Caries Study, In clu din g L. Acidophilus Estim ations, o f a Population Severely A ffected b y M ottled Enam el an d W h ich fo r the Past 12 Y ears H as U sed a F lu oride-F ree W ater. Pub. Health Rep. 5 6 :3 6 5 (February 28) 1941« q i. D eatherage, C . F ., F luoride D om estic W aters and D ental Caries E xperience in 2026 W h ite Illinois Selective Service M en . / . D. Res. 2 2 :1 2 9 (A p r il) 1943. 22. A rn old , F . A ., J r.; D ean, H . T . , and Elvove, E ., D om estic W ater and D ental Caries. I V . E ffect o f Increasing th e F lu oride C on ten t o f a C om m on W ater Supply o n th e Lactobacillus Acidophilus Counts o f the Saliva. Pub. Health Rep. 5 7 :7 7 3 (M a y 22) 1942. 23. A rn o ld , F . A ., Jr., T h e Possibility o f R ed u cin g D ental C aries b y Increasing F lu oride Ingestion, in Dental Caries ana Fluorine,1 p p . 99-107. sa. D eatherage, C . F ., A Study o f F lu oride D om es tic W aters an d D ental Caries E xperience in 263 W hite Illinois Selective Servicem en Living^ in F luoride Areas F ollow in g th e P eriod o f C alcification o f the Perm a nent T eeth . / . D. Res. 2 2 :1 7 3 (Ju n e) 1943-
J.A.D.A., Vol. 36, January 1948 . . . 35 Discussion
A logical assumption, in view of the evidence so far presented, is that dental caries may be controlled by fluorination of water supplies. The unusually high prevalence of this disease makes it par ticularly suited to a type of control measure which is on a communal basis and is not dependent on individual voli tion. Fluorination of water is economical. According to Harris,25 the per capita cost per year is about 7J/2 cents. This figure varies, depending on the price of the fluo ride used and the amount already present in the water supply. The results that might be expected from such a control measure were dis cussed in a previous article.26 In comparing the dental caries picture of the children of Aurora where the do mestic water contains 1.2 p.p.m. of fluo ride with that of children using fluoridefree water, it was noted that in Aurora ( 1) about six times as many children had had no dental caries experience; (2) there was about 60 per cent less dental caries experience; (3) there was a de crease of almost 75 per cent in mortality of the permanent first molar, and (4) ap proximately 95 per cent less caries oc curred on the proximal surfaces of the four upper incisors. Attention is called to the fact that the evidence to date in regard to the effect of water-borne fluoride on human dental caries is of an epidemiologic nature. At present one can only assume that the in gestion of a water which has fluoride added artificially'will produce the same effects as the ingestion of a natural fluo ride water. All evidence suggests that it will, but the final answer to this question awaits the results of research studies now 25. H arris, W . L ., E xperience in the A p plication o f F lu oride t o a P u blic W ater Supply, in Inservice Train ing Course for Waterworks Personnel Lectures. Ann A r b o r : U niversity o f M ich igan S ch ool o f Public H ealth , M a y 22-24, *94526. A rn old , F . A ., Jr., T h e R o le o f Fluorides in P reventive D entistry. J.A.D.A. 3 0 :4 9 9 (A pril 1) 1943.
36
in progress. In eight communities in the United States and Canada (Grand R ap ids, M ich; Newburgh, N. Y .; Midland, M ich.; Brantford, O nt.; Sheboygan, Wis.; Ottawa, K an .; Evanston, 111., and Marshall, Texas) fluorine is being added to the domestic water supply for the pur pose of studying the effect on dental caiies. The first of these communities to start adding the fluorine was Grand R ap ids, Mich. This study is sponsored by the U. S. Public Health Service with the co operation of the Michigan Department of Health and the University of Michi gan. Fluorination was begun in-January 1945, and to date no major mechanical difficulties have been encountered. More complete technical details of this prob lem were given by Harris.25 The influence of fluorination on the teeth of present inhabitants of these communities poses another question. Both Deatherage24 and Klein27 have reported beneficial effects in persons who had moved from supposedly fluoride-free areas to fluoride areas. However, the lim ited epidemiologic evidence in Maywood suggests that fluorination of water sup plies may haye little or no effect on the carious process in teeth formed on fluoride-free waters. In view of the latter evidence, an ab rupt beneficial change in dental caries in cidence would not be expected to follow the fluorination of a domestic water supply. Preliminary tabulation of results of annual examinations in Grand Rapids supports this contention. It appears dif ficult to explain, however, why numerous 27. K lein , H enry, D ental Caries E x perience in R e located C h ildren E xposed to W ater C on ta in in g F lu o rin e. I . In cid en ce o f N ew Caries after 2 Years o f Exposure A m on g Previously C aries-Free Perm anent T eeth . Pub. Health Rep. 6 0 :1 4 6 2 (D e ce m b e r 7) 1945.
The Journal of the American Dental Association
topical applications of fluoride, as would occur from the ingestion of fluorinated waters, would not have over a period of years a beneficial effect somewhat similar to topical application of stronger solu tions of fluoride. It may be that the diagnostic methods used were not critical enough to demonstrate the minor changes in the incidence of caries and that more time will be required for clarification of the results. Attention is called to the fact that other factors such as climatic conditions, culinary habits and unusual mineral con stituents of water supplies may play an important role. Further research studies together with the results of projects now in operation should shed light on some of these aspects of the problem. Summary
Epidemiologic evidence indicates that the ingestion during the formative period of the teeth of natural waters containing as little as i part per million of fluorine is accompanied by a decrease in the preva lence of dental caries. Presumptive evi dence points to fluorine as the active inhibitory agent and suggests the possi bility of adding controlled amounts of fluorine to water supplies as a practical method of preventing or inhibiting caries. Studies of such a procedure are now in progress, but final evaluation must be deferred during^ the time required for teeth to develop and erupt under this changed environment. At present it is deemed advisable to await the results of these research projects before an un qualified recommendation for mass fluorination of water supplies is made.
Prepaid Medical Care.-—The Council on Medical Service of the American Medical Association reports that 5,000,000 persons are covered by medical society approved prepayment medical care plans.—Illinois M. J. 92: 4, October 1947.
Francis'A. Arnold, Jr., D.D.S., Bethesda, Md.
D u r i n g the last decade dentistry has
been presented with evidence suggest ing a practical method for partially controlling dental caries in large popula tion groups ; that is, by means of fluorination of domestic water supplies. Such a suggestion originated principally from studies of the effects of fluoride-bearing waters on the prevalence of caries. In view of the fact that hitherto no effective means of mass control of this disease has been known, both the dental profession and the laity are extremely interested in the evidence presented. Often the hopes and expectations of both groups go be yond logical interpretation of the scien tific evidence. It is the purpose of this paper to re view briefly some of the major points of evidence reported to date, with special reference to the epidemiologic studies of water-borne fluorides. A more complete review of the dental caries-fluorine rela tionship can be found in the recent pub lication of the American Association for the Advancement of Science.1 Prior to 1938, most dental research in regard to water-borne fluorides centered on the problem of dental fluorosis. T o day the cause of this disease and the methods of its prevention are known. Of special importance in the dental cariesfluorine phenomenon is the fact that the concentrations of fluorides that produce
fluorosis also are known.'Aside from the establishment of these basic facts, addi tional knowledge has been obtained. As stated by Dean,2 “ Studies in endemic dental fluorosis provided the proving ground whence evolved many of the ideas and methodologies later utilized in the studies designed to outline the epi demiological characteristics of the fluo rine-dental caries relationships.” Before the etiologic agent of dental fluorosis, fluorine, was known and prior to the first report of low caries rates in endemic areas of dental fluorosis, men tion was made of the fact that caries was no more prevalent, and probably less so, in persons who exhibited evidence of fluorosis.3' 5 Special interest of dental re search workers was focused on the cariesfluorine relationship a decade ago after fluorine was recognized to be the cause of mottled enamel and the earlier ob servations were supported by the results of both laboratory and epidemiologic studies. Epidemiologic Studies After Dean6 had reported concerning the lowered dental caries rates in areas of endemic dental fluorosis in South D a 2. Dean, H . T ., Some General Epidemiological Con siderations, in Dental Caries and Fluorine ,1 pp. 1-4. 3. M cKay, F . S., Mottled Enamel: A Fundamental Problem in Dentistry. D. Cosmos 67:847 (September)
19*5-
4. Bunting, R . W., and others,( Further Studies of the Relationship of Bacillus Acidophilus to Dental From Dental Research Section, Division of Physi Caries ( III) . D. Cosmos 70:1002 (October) 1928. ology, National Institute of Health, Bethesda, Md. 5. M cKay, F . S., The Establishment of a Definite Read at the eighty-eighth annual session of the Relation Between Enamel That Is Defective in Its American Dental Association, Boston, August 6, 1947. Structure, as Mottled Enamel, and the Liability to Dental Surgeon, U . S. Public Health Service. Decay ( II) . D . Cosmos 71:747 (August) 1929. . 1. Dented Caries and Fluorine. Washington, D . C .: 6. Dean, H . T ., Endemic Fluorosis and Its Relation American Association for the Advancement of Science, to Dental Caries. Pub. Health R ep. 5 3 :14 4 3 - (August 1946. 19 ) »938.
Arnold
kota, specific epidemiologic studies were inaugurated -to gain further knowledge of this phenomenon. These studies were de signed to check by detailed local inves tigations the relation between fluorosis and caries. The observations were limited to spe cific groups according to age and sex. White school children, aged 12 to 14 years, were selected. Since it was logical to assume that this phenomenon might be related to differences in domestic wa ter supplies, only children who used con tinuously one common water supply were included. The cities selected for study had domestic water supplies which met the necessary prerequisite as to fluoride content and had made no changes in their water supplies for a period greater than the age of the children studied. Fac tors such as dietary habits (especially carbohydrate intake), latitude and sun light were taken into account in so far as possible. Complete details of the epidemi ologic technics employed in the selection of children and areas of study have been presented by Dean.7 The first of these investigations was made in four cities in north central Illi nois.8 Two of these cities, Galesburg and Monmouth, had communal water sup plies which contained 1.8 and 1.7 parts of fluoride per million, respectively. Quincy and Macomb, the control cities, used domestic water supplies which were fluoride-free; that is, they contained less than 0.2 p.p.m. of fluoride. The prevalence of dental caries in the children of these four communities dif fered greatly. For example, the average 13 year old child in Galesburg had ex-' perienced caries in only about two per manent teeth, but in Quincy the average 13 year old child had about six teeth showing caries experience. It also was 7. Dean, H . T ., Epidemiological Studies in the United States, in Dental Caries and Fluorine ,1 pp. 5-3i. 8. Dean, H . T ., and others^ Domestic Water and Dental Caries, Including Certain Epidemiological As pects of Oral Lactobacillus Acidophilus. Pub. Health Hep. 5 4 : 86q (May 26) 1939.
J.A.D.A., Vol. 36, January 1948 . . .
29
noted that in Galesburg and Monmouth approximately 36 per cent of the chil dren were caries-free as compared with 14 per cent in Macomb and 4 per cent in Quincy. Other interesting findings of this study were the differences in the distribution of high and low Lactobacillus acidoph ilus counts, the striking differences in proximal and smooth surface caries and the relation between macroscopic flu orosis and dental caries. In the case of the last, it was observed that the low ered caries rates in Galesburg children were not associated with the presence of. dental fluorosis by itself. This find ing confirmed the earlier observations by Dean.6 The evidence from this study together with other information available at that time pointed to water supplies and pre sumptively to the fluoride content of those water supplies as the primary asso ciated factor. This raised the question of whether a decreased prevalence of dental caries is associated with the use of water supplies whose fluoride content is less than the concentration necessary to pro duce any significant amount of dental fluorosis. In order to answer this question a de tailed study was made of school children in eight suburban Chicago communities. The communities and the children were selected in the manner previously de scribed.9 The results of this study con firmed the findings at Galesburg and Quincy in regard to the marked differ ences in the prevalence of dental caries between children using fluoride waters as compared with children using fluoride-free water. As. shown in Table 1, the caries experi ence rates in Elmhurst, Maywood, Au rora and Joliet, whose public water sup plies contain 1.8, 1.2, 1.2 and 1.3 p.p.m. 9. Dean, H. T ., and others, Domestic Water and Dental Caries. II. A Study of 2,832 White Children, Aged 12-14 Years, of 8 Suburban Chicago Communi ties. Including Lactobacillus Acidophilus Studies of 1,761 Children. Pub. Health Rep. 56:761 (April 11) 1941*
30
of fluoride, respectively, were 252, 258, 281 and 323, respectively. The children of Evanston, Oak Park and Waukegan, using fluoride-free water, showed caries experience rates of 673, 722 and 810, re spectively. The results of the examina tions at Elgin, where the public water supply contains about 0.5 p.p.m. of fluo ride, indicate a caries experience rate (444) that is about midway between that of the fluoride and fluoride-free areas. The observations in this study were suggestive of an inverse variation of den tal caries and the fluoride concentration of the domestic water. It also was noted that the inhibitory factor, presumably fluorine, was operative at concentrations below what is now the accepted per missible standard for water, 1.5 p.p.m.10 At these low concentrations dental fluo rosis is no problem. Evidence of this fact was obtained in Aurora where al though 15 per cent of the children were classed as having fluorosis, only 5 per cent of all the teeth and only about 0.4 per cent of the anterior teeth showed even so much as the mildest forms of fluorosis. This minor amount of affectioij, which is noticeable only to a trained observer, is of no esthetic or public health signifi cance. As may be noted in T able 1, there were differences other than the fluoride con tent in the water supplies of the com munities studied. For example, the cities in which the caries rates were low used well waters while those with high caries prevalence rates used surface (Lake Michigan) water. Thus, comparisons were made on water supplies of varying mineral content. In order to check this variable and to confirm the observa tion regarding the effect of the lower ranges of fluoride concentrations (1 p.p.m. and less), epidemiologic studies in other communities were conducted. By the use of similar technics of exam ination and the same criteria of selec-
The Journal of the American Dental Association
tion, children 12 to 14 years of age in nineteen additional communities were studied. The communities selected had water supplies which varied from a high fluo ride content with a low mineral content to fluoride-free waters with a high min eral content. The sources of these water supplies were deep wells, lakes and rivers, and cisterns. Including the children in the eight Chicago suburbs, a total o f 8,576 children in twenty-seven cities o f eight states have been studied. The results of these examinations are shown in the il lustration and in Table 2. M ore com plete details concerning these study areas were presented by Dean, Arnold and Elvove.11 The data suggest the following conclu sions: First, children who are born and reared in communities whose domestic water supplies contain increased amounts of fluorine (more than 0.5 p.p.m.) ex perience less dental caries than do chil dren living in areas where the domestic water supply is fluoride-free (Tables 1 and 2). Second, this phenomenon is op erative whether or not the teeth show macroscopic evidence of fluorosis. Third, the decreased prevalence of dental caries is evident in areas where the fluoride content of the water is below those levels which produce dental fluorosis of any esthetic significance. A careful analysis of the data strongly suggests fluorine as the active inhibitory factor present in these water supplies. As stated previously, factors other than those present in the water supplies which probably could have influenced the den tal caries experience were taken into ac count and in so far as possible were equalized by careful selection of the study areas. As for the variability in water sup plies, one characteristic other than fluo ride which has been suggested as low
11. Dean, H . T .; Arnold, F- A ., Jr., and Elvove, Elias, Domestic Water and Dental Caries. V . Addi tional Studies of the Relation of Fluoride Domestic Waters to Dental Caries Experience in 4 ^ 2 5 White 10. Public Health Service Drinking Water Stand Children, Aged 12 to 14 years, of 13 Cities in 4 States. ards, 1946. Pub. Health Rep. 6 1:3 7 1 (March 15) 1946. Pub. Health R ep. 57:115 5 (August 7) 1942.
Arnold
J.A.D.A., Vol. 36, January 1948 . . .
ering dental caries experience is the total hardness of the water. In the study of the twenty-seven cities, however, there is lit tle evidence of correlation between the total hardness of the water and caries prevalence (Table 2). For example, M id dletown and Zanesville, Ohio, and Elk hart, Ind., have waters of relatively high-' total hardness (329, 291 and 220 p.p.m,
31
respectively) and show caries experience rates of 703, 733 and 823, respectively. On the other hand, Colorado Springs, Colo, and Maywood, 111., using waters of much lower total hardness (27 and 75 p.p.m., respectively) had lower caries rates (246 and 258, respectively). The one variable, however, which does corre late with caries prevalence is the fluoride
Table I.— Summary of dental caries findings in 2,832 selected white children, aged 12 to 14 years, in eight suburban Chicago communities in relation to the fluoride content of the public water supply9
Elm May hurst wood Aurora Joliet
873
1,625
1,412
1,030 2,125
1,662 1,354
Number of 12-14 year old white chil dren whose histories on repeated questioning indicated continuity of exposure and who were examined 170
171
633
447
403
329
26.9
19.6
Watei supply
Source Permanent hardness in parts per million 323.4
1.8
Dental caries experience, permanent teeth, per 100 children examined 252
Clinical aminati
39.0
31.7
39.1
Deep wells
Meanfluoride(F) content, 1939- 1940, in parts per million
X V
Evans Oak Wau ton Park kegan
Total number of 12-14 year old chil dren present at time of sampling 633
Percentage of the total present who were examined
G O
Elgin
Dental caries experience, proximal surfaces, superior incisors, per 100 surfaces
0.60
75.0 328.5 1.2*
258
0.59
1.2
281
0.78
256 12.0
19.8
CO CN
Sampling
City or village
31.2
Lake Michigan
349.3
102.6
1.3
0.5
323
444
131.0 132.2 134.4 0.0
673
0.0
722
0.0
810
1.3
4.1
10.7
9 .0.
17.7 79.9
First permanent molar mortality, per 100 children examined
11.8
11.7
14.5
19.5
20.3
42.6
31.0
Percentage of children with no den tal caries experience
25.3
29.8
23.5
18.3
11.4
3.9
4.3
3. 1 '
Percentage incidence of endemic dental fluorosis (mottled enamel)
40.0
33.3
15.0
25.3
4.2
1.6
0.6
0.2
Percentage of those ex- Negative amined bacteriologically and < 10 0
36.7
39.6
38.2
35.6
21.2
15.4
18.2
14.9
24.7
20.9
25.6
26.6
33.2
41.8
43.8
54.6
ctf £O dophilus counts were: M
30,000 ánd>
*There is both presumptive and direct evidence that prior to a few years ago the Maywood water contained probably 1.4-1.6 p.p.m. of fluoride (F ).
32
The Journal of the American Dental Association
content of these water supplies. Attention is called to the possibility that other factors or combinations of fac tors associated with water may play a role in the carious process or in the cariesfluorine relationship. At present compara tively little information is available re garding the physiologic role of such a major part of our daily diet as our drink ing water. It is known that there is a great variation at least in the chemical composition of the different waters; fur ther study may disclose physiologic effects ,
resulting from some of these variables. In addition to indicating that fluorine is the active inhibitory factor, the obser vations in the twenty-seven cities also suggest that almost full benefit is ob tained if the fluoride content is only about i p.p.m. At this low range of fluoride concentration the dental fluorosis index is less than 0.5 (see illustration). A de scription of the method12 of establishing 12. Dean, H . T ., The Investigation of Physiological Effects of the Epidemiological Method, in Fluorine and Dental Health. Washington, D . C .: American Association for the Advancement of Science, 1942.
Table ¿.— Summary of dental caries and fluorosis findings in 8,576 selected white school children, 12 to 14 years of age, in twenty-seven cities of eight states in relation to fluoride content, total hardness and source of domestic water supply
Water supply
Clinical
City
Hereford, Texas. . . . Colorado Sprgs, Colo. Galesburg, 111.......... Elmhurst, 111........... Maywood, 111.......... East Moline, 111. . . . Joliet, 111................ Aurora, 111............... Kewanee, 111............ Vicksburg, Miss.. . . Nashville, Tenn.. . . Clarksville, Tenn.. . Pueblo, Colo........... Marion, Ohio.......... Elgin, 111.................. Lima, Ohio............. Evanston, 111........... Zanesville, Ohio. . . . Escanaba, Mich... . Portsmouth, Ohio. . Middletown, Ohio.. Oak Park, 111.......... Elkhart, Ind........... Quincy, 111............... Waukegan, 111......... Michigan City, Ind.
Caries experience Chil Total dren in permanent teeth Per cent children Fluoride, hard ex p.p.m. ness, with amined Per cent Number with per 100 fluorosis p.p.m.* children none
60 404 273 170 171 152 447 633 123 1-72 662 60 614 263 403 454 256 459 270 469 370 329 278 330 423 236 95
38.3 28.5 27.8 25.3 29.8 20.4 18.3 23.5 17.9 S.7 9.4 16.7 10.6
5.7 11.4 2.2
3.9 2.6 1.1
1.3 1.9 4.3 1.4 2.4 3.1 0.0 0.0
147 246 236 252 258 303 323 281 343 587 461 458 412 556 444 652 673 733 877 772 703 722 823 706 810 1,037 1,070
100.0
3.1
73.8 47.6 40.0 33.3 31.6 25.3 15.0
2.6
12.2 11.0 8.6
0.9
8.3 6.5 6.1
4 .2 . 2.2 1.6
1.5 1.5 1.3 1.1 0.6
0.4 0.3 0.2 0.0 0.0
1.9 1.8 1.2f 1.2t
1.3 1.2 0.2 0.0 0.2 0.6
0.4 0.5 0.3 0.0 0.2 ' 0.2 0.1 0.2 0.0 0.1 0.1 0.0 0.1 0.1Î
383 27 247 323 75 276 349 329 445 94 79 89 302 209 103 223 131 291 123 80 329 132 220 88
134 141 36Î
Source
Deep wells Surface (Pike’s Peak) Deep wells Deep wells Deep wells Deep wells Deep wells Deep wells Deep wells Yazoo and Miss. River Cumberland River Cumberland River Arkansas River Deep wells Deep wells Impounded surface Lake Michigan Deep wells Lake Michigan Ohio River Deep and shallow wells Lake Michigan Deep wells Mississippi River Lake Michigan Lake Michigan Cisterns
*T otal hardness calculated'in accordance with Standard Methods for the Examination of Water and Sewage. New York: American Public Health Association. 1936. fT here is both presumptive and direct evidence that these water supplies contained about 1.5 p.p.m. of fluoride prior to a few years ago. 4 ^ < ^ ^Average of five water samples from cisterns considered typical of those used by most K ey West inhabitants.
Arnold
J.A.D.A., Vol. 36, January 1948 . . . 33
Relation between the amount of dental caries in the perm anent teeth of 8,sy6 white children 1 2 to 14 years old in twenty-seven cities of eight states and the index of dental fluorosis
a fluorosis in d e x has been g ive n b y D e a n . taine d 0.25 p .p .m . O ck erse15 re p o rte d T h e p ro b le m of d e n ta l fluorosis in these s im ila r findings in S o u th A fric a . areas is n e gligib le as in d ica te d b y the ob T h is evidence, like m ost e p id em io lo gic servation in A u ro ra , 111., as p revio usly evid en ce, is p re s u m p tive in ch a ra cte r,16 described. In cre a s in g the flu o rid e content b u t in a d d itio n to these studies on h u m a n m u c h above i p .p .m . produces little a d beings there are reports fro m la b o ra to ry d itio n a l in h ib itio n o f caries b u t does m a research w h ic h in d ica te th a t flu o rin e in te ria lly increase the a m o u n t a n d degree fluences the carious process in e x p e ri o f fluorosis.13 m e n ta l anim a ls. A n u m b e r of authors I n studies in this a n d o th e r countries h a ve s h o w n th a t in d u c e d caries in the observations of a sim ila r ch a ra cte r h a ve m olars o f rats m a y be in h ib ite d b y the been m a d e . W e a v e r14 m a d e a stud y of a d d itio n o f fluorides to the d rin k in g w a d ental co n d itio ns in the B ritis h Isles an d ter o r the foo d a n d th a t this ca rie s-in fo u n d th a t the c h ild re n o f S o u th Shields h ib itin g actio n is related to the a m o u n t of h a d o n ly a b o u t h a lf as m u c h d ental caries flu o rid e ingested. A ls o , d ental caries in experience as d id the c h ild re n o f N o rth S y ria n ham sters has been in flu en ced in a Shields, w h ic h is ju s t across the R iv e r like m a n n e r. A re v ie w of e xp e rim e n ta l T y n e . H e a ttrib u te d the v a ria b ility in caries a n d fluorides w as m a d e b y H o d g e caries pre va le n ce to differences in the a n d Sognnaes.17 dom estic w a te r su p p ly, specifically to the A lth o u g h it is n o t k n o w n th a t d ental flu o rid e co n ten t. T h e w a te r sup ply of caries in h u m a n beings is co m p a ra b le to S o u th Shields co n taine d 1.4 p .p .m . of 15. Ockerse, T ., Fluorine and Dental Caries in South flu o rid e a n d th a t of N o rth Shields c o n Africa, in D e n t a l C a r ie s a n d F lu o r in e ,1pp. 36-42. 13. In Hereford, Texas (Table 2 and illustration) the number of continuous residents is small, and the low caries rate observed may be due to sampling error. 14. Weaver, Robert, Epidemiological Studies in the British Isles and India, in D e n t a l C a rie s a n d F l u o r in e / PP.
32- 35*
16. Although the epidemiologic evidence is pre sumptive in character, this fact does not decrease the validity of the evidence. Chlorination of water sup plies for the control of typhoid fever is based on sim ilar presumptive evidence. 17. Hodge, H. C ., and Sognnaes, R . F ., Experi mental Caries and a Discussion of the Mechanism of Caries Inhibition by Fluorine, in D e n ta l C a r ie s a n d F lu o r in e ,1pp. 53*73*
34
The Journal of the American Dental Association
induced caries in animals, such labora tory evidence lends weight to the pre sumption that fluorine is the active in gredient in the water supplies in areas of decreased caries rates.
differences were noted in the o-ioo range (inactive caries) and in those counts over the 20,000-30,000 range (active caries). The differences in the counts paralleled the differences in caries experience rates observed in these children. Mechanism Studies Cities using waters with a fluoride con • At the present time the mechanism of tent of more than i p.p.m. averaged 37.5 the action of fluoride waters on the cari per cent in the o-iop range compared ous process is not known; however, the with 17.8 per cent in the fluoride-free limited information available sheds some areas (Table 3). In the intermediate fluo light on this phenomenon. One of the ride level (0.5 p.p.m.) 21.2 per cent of more important observations made in the counts were in the low range. Con connection with the epidemiologic studies versely, the high counts (20,000 and is in regard to changes in the oral bac more) were 28.7 per cent in the fluoride tériologie flora. It is a generally accepted areas and 52.4 per cent in the fluoridebelief that dental caries activity can be free areas. These differences in the per diagnosed by determining the number centage distribution of high and low Lac of Lactobacilli in the saliva. The oral tobacillus counts suggest that the cariesflora of persons in whom caries is active fluorine relationship may be based on shows a large number of Lactobacilli per action of the fluoride waters on the oral cubic centimeter, but in persons in whom flora. In other words, they point to a de caries is inactive or absent the oral flora crease in the activating forces of the is negative for Lactobacilli, or the organ carious process. This result would seem to ism is present only sporadically and in indicate a theory of action opposed to that suggested by laboratory evidence on low numbers. The results of such determinations in chemical composition of the teeth.13 Additional information on the mecha the fluoride and nonfluoride areas, shown in Table 3, indicate that there is a definite nism of action of fluoride waters has been difference in the percentage distribution gained by studying persons who gave a of Lactobacillus counts in the saliva of history of change from fluoride to fluo children included in some of the epidemi ride-free water supplies or vice versa. In ologic studies.8’ 9' 18 The most striking Bauxite, Ark., it was observed that chil18. Jay, P h ilip, and A rn o ld , F . A ., J r., E p idem io logical A spects o f O ral Lactobacillus Counts in F lu o rid e an d N on -F lu oride Areas, in Dental Caries and Fluorine,1 p p . 43-46.
19. Arm strong, W . D ., C h em ical D ifferences o f C a ries Susceptible and Im m u ne T e e th and a Consideration o f . F o o d Sources o l F luorine, in Dental Caries and Fluorine,1 p p . 47-52.
Table 3.— Percentage distribution of oral Lactobacillus counts on 12 to 14 year old children living in fluoride and nonfiuoride areas*
Fluoride ' , ■ content of , domestic water, p.p.m. 1.0 0.5 0.0
-
L. acidophilus counts Children examined
1,052 250 1,087
0-100, per cent
20,000 and more, per cent
37.5 21.2 17.8
28.7 40.0 52.4
Dental caries experience (rate per 100) f 280 444 760
♦Includes children from the follow in g cities: Q u in cy , Evanston, O ak Park, W aukegan, E lgin, Elm hurst, M a y w ood , A u rora , Joliet and G alesburg, 111., and Escanaba, M ich . fB a sed on all children given dental exam ination in respective com m unities.
Arnold
dren whose teeth were formed on a fluo ride water and erupted into a fluoridefree environment experienced less dental caries than children of a neighboring community who had used fluoride-free water throughout life.20 This finding is in accord with the results reported by Deatherage21 on selectees in Illinois. Observations somewhat opposite in character were made in Maywood, 111., Garrettsville, Ohio and Escanaba, M ich.22’ 23 In Maywood a study was made of children who had moved to that city at varying ages from areas using fluoridefree Lake Michigan water. With the ex ception of those children who had moved to Maywood prior to 4 years of age, there was no significant evidence of any inhibi tion of caries incidence following the change from a fluoride-free to a fluoride environment. At Garrettsville, Ohio, and Escanaba, Mich., the respective water supplies were changed from fluoride-free water to that containing increased amounts of fluorine. Dental and bac tériologie examinations in Garrettsville and bacteriologic examinations in Es canaba gave no indication of a change in caries activity in children whose teeth had formed on fluoride-free water. Analysis of these results suggests the necessity of an increased ingestion of fluo ride from birth or at least during the period of tooth formation in order to ob tain beneficial effects. These results, how ever, are not entirely in accord with the findings of Deatherage.24 20. D ean, H . T . , and others, D om estic W ater and D ental Caries. I . A D ental Caries Study, In clu din g L. Acidophilus Estim ations, o f a Population Severely A ffected b y M ottled Enam el an d W h ich fo r the Past 12 Y ears H as U sed a F lu oride-F ree W ater. Pub. Health Rep. 5 6 :3 6 5 (February 28) 1941« q i. D eatherage, C . F ., F luoride D om estic W aters and D ental Caries E xperience in 2026 W h ite Illinois Selective Service M en . / . D. Res. 2 2 :1 2 9 (A p r il) 1943. 22. A rn old , F . A ., J r.; D ean, H . T . , and Elvove, E ., D om estic W ater and D ental Caries. I V . E ffect o f Increasing th e F lu oride C on ten t o f a C om m on W ater Supply o n th e Lactobacillus Acidophilus Counts o f the Saliva. Pub. Health Rep. 5 7 :7 7 3 (M a y 22) 1942. 23. A rn o ld , F . A ., Jr., T h e Possibility o f R ed u cin g D ental C aries b y Increasing F lu oride Ingestion, in Dental Caries ana Fluorine,1 p p . 99-107. sa. D eatherage, C . F ., A Study o f F lu oride D om es tic W aters an d D ental Caries E xperience in 263 W hite Illinois Selective Servicem en Living^ in F luoride Areas F ollow in g th e P eriod o f C alcification o f the Perm a nent T eeth . / . D. Res. 2 2 :1 7 3 (Ju n e) 1943-
J.A.D.A., Vol. 36, January 1948 . . . 35 Discussion
A logical assumption, in view of the evidence so far presented, is that dental caries may be controlled by fluorination of water supplies. The unusually high prevalence of this disease makes it par ticularly suited to a type of control measure which is on a communal basis and is not dependent on individual voli tion. Fluorination of water is economical. According to Harris,25 the per capita cost per year is about 7J/2 cents. This figure varies, depending on the price of the fluo ride used and the amount already present in the water supply. The results that might be expected from such a control measure were dis cussed in a previous article.26 In comparing the dental caries picture of the children of Aurora where the do mestic water contains 1.2 p.p.m. of fluo ride with that of children using fluoridefree water, it was noted that in Aurora ( 1) about six times as many children had had no dental caries experience; (2) there was about 60 per cent less dental caries experience; (3) there was a de crease of almost 75 per cent in mortality of the permanent first molar, and (4) ap proximately 95 per cent less caries oc curred on the proximal surfaces of the four upper incisors. Attention is called to the fact that the evidence to date in regard to the effect of water-borne fluoride on human dental caries is of an epidemiologic nature. At present one can only assume that the in gestion of a water which has fluoride added artificially'will produce the same effects as the ingestion of a natural fluo ride water. All evidence suggests that it will, but the final answer to this question awaits the results of research studies now 25. H arris, W . L ., E xperience in the A p plication o f F lu oride t o a P u blic W ater Supply, in Inservice Train ing Course for Waterworks Personnel Lectures. Ann A r b o r : U niversity o f M ich igan S ch ool o f Public H ealth , M a y 22-24, *94526. A rn old , F . A ., Jr., T h e R o le o f Fluorides in P reventive D entistry. J.A.D.A. 3 0 :4 9 9 (A pril 1) 1943.
36
in progress. In eight communities in the United States and Canada (Grand R ap ids, M ich; Newburgh, N. Y .; Midland, M ich.; Brantford, O nt.; Sheboygan, Wis.; Ottawa, K an .; Evanston, 111., and Marshall, Texas) fluorine is being added to the domestic water supply for the pur pose of studying the effect on dental caiies. The first of these communities to start adding the fluorine was Grand R ap ids, Mich. This study is sponsored by the U. S. Public Health Service with the co operation of the Michigan Department of Health and the University of Michi gan. Fluorination was begun in-January 1945, and to date no major mechanical difficulties have been encountered. More complete technical details of this prob lem were given by Harris.25 The influence of fluorination on the teeth of present inhabitants of these communities poses another question. Both Deatherage24 and Klein27 have reported beneficial effects in persons who had moved from supposedly fluoride-free areas to fluoride areas. However, the lim ited epidemiologic evidence in Maywood suggests that fluorination of water sup plies may haye little or no effect on the carious process in teeth formed on fluoride-free waters. In view of the latter evidence, an ab rupt beneficial change in dental caries in cidence would not be expected to follow the fluorination of a domestic water supply. Preliminary tabulation of results of annual examinations in Grand Rapids supports this contention. It appears dif ficult to explain, however, why numerous 27. K lein , H enry, D ental Caries E x perience in R e located C h ildren E xposed to W ater C on ta in in g F lu o rin e. I . In cid en ce o f N ew Caries after 2 Years o f Exposure A m on g Previously C aries-Free Perm anent T eeth . Pub. Health Rep. 6 0 :1 4 6 2 (D e ce m b e r 7) 1945.
The Journal of the American Dental Association
topical applications of fluoride, as would occur from the ingestion of fluorinated waters, would not have over a period of years a beneficial effect somewhat similar to topical application of stronger solu tions of fluoride. It may be that the diagnostic methods used were not critical enough to demonstrate the minor changes in the incidence of caries and that more time will be required for clarification of the results. Attention is called to the fact that other factors such as climatic conditions, culinary habits and unusual mineral con stituents of water supplies may play an important role. Further research studies together with the results of projects now in operation should shed light on some of these aspects of the problem. Summary
Epidemiologic evidence indicates that the ingestion during the formative period of the teeth of natural waters containing as little as i part per million of fluorine is accompanied by a decrease in the preva lence of dental caries. Presumptive evi dence points to fluorine as the active inhibitory agent and suggests the possi bility of adding controlled amounts of fluorine to water supplies as a practical method of preventing or inhibiting caries. Studies of such a procedure are now in progress, but final evaluation must be deferred during^ the time required for teeth to develop and erupt under this changed environment. At present it is deemed advisable to await the results of these research projects before an un qualified recommendation for mass fluorination of water supplies is made.
Prepaid Medical Care.-—The Council on Medical Service of the American Medical Association reports that 5,000,000 persons are covered by medical society approved prepayment medical care plans.—Illinois M. J. 92: 4, October 1947.