Journal of Investigative and Clinical Dentistry (2015), 6, 45–52
ORIGINAL ARTICLE Community Dentistry and Oral Epidemiology
Dental fluorosis in the Blue Mountains and Hawkesbury, New South Wales, Australia: policy implications Ikreet S. Bal1, Peter J. Dennison2 & R. Wendell Evans2 1 Department of Public Health Dentistry, Dr Harvansh Singh Judge Institute of Dental Sciences and Hospital, Panjab University, Chandigarh, India 2 Sydney Dental School, The University of Sydney, Sydney, NSW, Australia
Keywords caries, fluoridation, fluoride toothpaste, fluorosis, health policy. Correspondence Associate Professor R. Wendell Evans, Community Oral Health and Epidemiology, Westmead Centre for Oral Health, Population Oral Health, 1 Mons Road, Westmead, New South Wales 2145, Australia. Tel: +61-02-8821-4364 Fax: +61-02-8821-4366 Email: [email protected] Received 5 April 2014; accepted 26 October 2014. doi: 10.1111/jicd.12138
Abstract Aim: The aim of the present study was to determine whether the adjustment of the fluoride concentration to 1 ppm in the drinking water supplied to the Blue Mountains, New South Wales, Australia in 1993 was associated with fluorosis incidence. Methods: In 2003, children attending schools in the Blue Mountains and a control region (fluoridated in 1967) that had been randomly selected at baseline in 1992 were examined for dental fluorosis (maxillary central incisors only) using Dean’s index. A fluoride history for each child was obtained by questionnaire. Associations between fluorosis and 58 potential explanatory variables were explored. Results: The response rate was 63%. A total of 1138 children aged from 7 to 11 years with erupted permanent central incisors were examined for dental fluorosis. Fluorosis prevalence was the same in both regions. The Community Index of Dental Fluorosis values were slightly different, but were both above 0.6, indicative of public health concern. Conclusions: For the group as a whole, we concluded that: (a) fluorosis prevalence (0.39) in both regions was similar; and (b) the higher-than-expected prevalence and severity of fluorosis was due mainly to two factors: (a) the higher-than-optimal fluoride level in drinking water; and (b) swallowing of fluoride toothpaste in early childhood.
Introduction The element fluorine, usually found in an ionic form, is naturally ubiquitous as the 13th most abundant in the earth’s crust.1 It has been present in the environment long before the start of biological life, and all living organisms, including Homo sapiens, have been exposed to it throughout their development. Because it is completely unavoidable, the only question that need exercise the minds of scientists is how much of this tasteless element human beings are exposed to and what the effects of such exposure are. Dean demonstrated that the optimal fluoride concentration in drinking water in the USA for the prevention of dental caries and dental fluorosis was 1 ppm.2 Later, ª 2014 Wiley Publishing Asia Pty Ltd
Arnold noted that diet and climatic factors were related, and that concentrations of fluoride necessary to prevent caries might vary.3 He suggested that concentrations of 0.5–0.7 ppm in southwestern regions of the USA might be as effective as 1–1.5 ppm in the north. Subsequently, Galagan and Vermillion developed an algorithm for determining the optimal fluoride concentration based on fluid intake of US population groups in relation to ambient temperature and caries experience data.4 Variants of these considerations guide determinations of target concentrations elsewhere. It is accepted that, consequent to achieving an optimal fluoride concentration to bring about the desired reduction in dental caries experience, the mildest manifestations of fluorosis will be prevalent in approximately 15% of the population.5 Most milder forms of 45
I.S. Bal et al.
Policy implications of dental fluorosis
dental fluorosis are difficult to identify by an untrained examiner. A recent review of water fluoridation, commissioned by the British Government, confirmed the association of dental fluorosis with fluoridated water.6 At optimal concentrations, the Community Index of Dental Fluorosis (CFI) value (the mean of the fluorosis rank scores) is below 0.4, a level which Dean claimed was compatible with public health.2 It has been demonstrated that higher-than-expected levels of fluorosis might occur in association with above-optimal concentrations of fluoride in drinking water, but this might be corrected by adjusting the fluoride concentration downwards,7 or by controlling the ingestion of fluoride from known sources.8–10 In 1980, the New South Wales (NSW) Water Board (Australia) assumed responsibility for the water supply in the Blue Mountains local government area (LGA), a partly rural district on the western fringe of metropolitan Sydney, and foreshadowed that fluoridation of the reticulated water supply to this region would occur.11 This commenced following substantial infrastructure development in 1992, thereby completing the extension of the fluoridation program to the entire greater Sydney metropolitan region. The fluoride concentration was set at 1 ppm, in line with the Sydney target. The current investigation was part of a larger study which confirmed the substantial benefits of extending water fluoridation to the Blue Mountains LGA for the prevention of dental caries.12 The adjacent Hawkesbury LGA, which had been fluoridated since 1969, was selected as the reference area due to its similar sociodemographic profile and geographic proximity to the Blue Mountains. The populations of the Blue Mountains and Hawkesbury in 2006 were approximately 73 000 and 64 000, respectively. The respective land areas are 1430 and 2793 square kilometers, of which 70–85% is within the World Heritage Blue Mountains National Park. The main economic activity in both areas is tourism, and additionally in Hawkesbury, fruit farming and market gardening. The 1992 baseline survey covering children in both regions reported on dental caries experience, but fluorosis status was not assessed.11 Therefore, the purposes of this analysis were: (a) to determine whether the adjustment of the fluoride concentration to 1 ppm in the drinking water supplied to the Blue Mountains 11 years previously was associated with dental fluorosis prevalence; (b) to determine the similarity or otherwise in fluorosis prevalence in the Blue Mountains and Hawkesbury regions; and (c) to evaluate the risk of fluorosis from a range of fluoride sources. Permission for the survey was obtained from the Human Research Ethics Committees of The University of Sydney, the New South Wales Department of Education and Training, The Catholic Archdiocese, and the Wentworth Area Health Service. 46
Material and methods Sampling In this cross-sectional survey, the target population was the primary school children in the Blue Mountains and Hawkesbury LGA of NSW. The 1992 caries experience baseline survey conducted by Patterson and Weidenhofer was carried out in 18 schools.11 These were “selected randomly using comparative size as the main criteria. Additionally some smaller schools were included and attempts were made to obtain a reasonable geographic spread across the LGAs”.11 The schools, as selected for the 1992 survey, became the target schools for the 2003 survey, except that two schools having enrolments of 107 and 16 pupils, respectively, in the Hawkesbury LGA were excluded as they had come under the purview of another Area Health Service. Recruitment of sample The principals of the selected schools were visited by PJD and given letters of invitation to include their schools in the survey. All agreed to participate, and school staff assisted by distributing information packages to the parents/guardians of the children, and also by collecting the responses that were returned in sealed envelopes. The information packages included an information statement, a request for consent to participate in the survey, a consent form, and a questionnaire. The information statement explained the purpose of the study and assured parents of confidentiality. Parents were requested to return the completed questionnaire and the consent form to the school before the date set for examining the children at that particular school. A reminder notice was sent to the parents 1 week after forwarding the information packages. Only children whose parents/guardians gave consent for inclusion in the survey were examined. Children born before 1992 were excluded from this analysis, which is based on those aged 7–11 years for the 2003 survey. Fluoride history A questionnaire (available on request) developed by the Australian Research Centre for Population Oral Health (ARCPOH) was used in this survey to seek demographic information, as well as a detailed fluoride history regarding exposure to fluoridated water and use of toothpastes, fluoride supplements, fluoride mouth rinses, and professionally-applied fluoride gel. Some rural households within the Blue Mountains and Hawkesbury LGA did not receive mains supply water; instead, their water source was from springs or rain water. Children from ª 2014 Wiley Publishing Asia Pty Ltd
I.S. Bal et al.
Policy implications of dental fluorosis
other communities, fluoridated or otherwise, who became resident in the area, were also included in the survey. Measurement of fluorosis Dental fluorosis was assessed according to the CFI criteria developed by Dean,2 as recommended by the World Health Organization (WHO),13 but restricted to the examination of the maxillary central incisors, as the control of dental fluorosis with reference to this tooth type is synonymous with the control of fluorosis in the dentition as a whole.14 The WHO reference photographs were used as the diagnostic standards. Russell’s differential diagnosis criteria for fluorosis were followed.15 Clinical examination All consenting children with erupted permanent maxillary central incisors were examined. Examinations of the wet maxillary central incisors were conducted in shaded natural daylight, either outdoors or indoors beside a window, and the score for each child was recorded on a class list obtained from the school. Although fluorosis manifestation is bilaterally symmetrical, in rare cases when this did not hold, diagnosis was made in relation to the more affected incisor. The examinations were carried out by ISB, who had been calibrated before the survey. Data quality was monitored daily during the examinations, and formal interand intra-examiner calibration (with RWE) was periodically carried out during the survey. Data analysis Fluorosis prevalence and CFI values were computed. Associations between fluorosis and 58 potential explana-
tory variables were explored in a series of bivariate analyses. Associations between categorical variables were evaluated using the chi-squared-test or Fisher’s exact test, while regression analysis was used to evaluate associations with continuous variables. Associations that were significant at the P < 0.2 level were evaluated further using logistic regression. Generalized estimating equations were used to assess the effect of potential risk factors on individuals’ fluorosis status (“very mild or more” vs “normal or questionable”). In these binomial models, an exchangeable correlation structure was used to allow for clustering by school.16 The unadjusted intracluster correlation coefficient over all the schools was 0.016. The statistical software package S-PLUS version 8 (Insightful, Seattle, WA, USA) was used to analyze the data. Two-tailed tests with a significance level of 5% were used throughout. Results The response rate was 63%. A total of 1138 children (16 clusters, number range: 34–195, median number: 63) aged from 7 to 11 years with erupted permanent central incisors were examined for dental fluorosis (Table 1). Weighted kappa values for monitoring examiner reliability ranged from 0.92 initially (59 duplicate recordings) to 0.96 finally (60 duplicates). A fluorosis prevalence of 39% was observed in the Blue Mountains and Hawkesbury regions inclusive of 16 cases of moderate or severe (1.4%) (Table 1). CFI values were slightly different (P < 0.01), but both were above the 0.6 level nominated by Dean as indicative of a public health concern.2 Sixty-four percent had been exposed to fluoridated water from birth. As shown in Table 2, children were exposed to other fluorosis risk factors, including the use of fluoridated water for infant formula reconstitution, toothbrushing with fluoride toothpaste before the age of
Table 1. Distribution of enamel fluorosis and the Index of Dental Fluorosis based on maxillary central incisors alone by local government area Blue Mountains
Hawkesbury
Both
Enamel fluorosis
n
%
n
%
n
%
95% CI
Normal Questionable Very mild Mild Moderate Severe Total Fluorosis prevalence† (SE) Index of Dental Fluorosis
110 233 176 33 8 4 564 0.39 0.71*
19.5 41.3 31.2 5.9 1.4 0.7 100 (0.01)
245 220 241 48 5 3 762 0.39 0.62
32.2 28.9 31.6 6.3 0.7 0.4 100 (0.01)
355 453 417 81 13 7 1326 0.39 0.66
26.3 34.2 31.4 6.1 1.0 0.5 100 (0.01)
24.4, 31.6, 29.0, 4.9, 0.5, 0.2,
29.3 36.8 34.0 7.6 1.7 1.1
CI, confidence interval; SE, standard error. *P < 0.001. † Very mild or more.
ª 2014 Wiley Publishing Asia Pty Ltd
47
I.S. Bal et al.
Policy implications of dental fluorosis
Table 2. Sample distribution by potential high-risk fluorosis factors and its degree of severity Normal or questionable Potential high-risk factor for dental fluorosis Infant-feeding practices Fluoridated mains supply as formula-reconstituting agent Oral hygiene practices when toothbrushing commenced Commenced brushing before age 2 years Once or more times daily toothbrushing Used brush length of toothpaste Used standard concentration toothpaste (1000 ppm) Just swallowed without rinsing after brushing Eating/licking toothpaste often Use of fluoride rinse before age 5 years Use of fluoride tablets or drops before age 5 years
Very mild or mild
n
%†
n
%†
352
58
250
41
305 583 53 149 4 21 43 25
59 60 58 64 27 55 51 54
202 366 37 81 10 17 41 21
39 38 41 35 67 45 49 46
Moderate or severe
Total
%†
n
%‡
9
2
611
54
9 16 1 4 1 0 0 0
2 2 1 2 7 0 0 0
516 965 91 234 15 38 84 46
45 85 8 21 1 3 7 4
n
†
% of row total. % of sample total.
‡
2 years, early use of fluoride rinse, and fluoride supplements. Some of these factors, including related toothbrushing habits (rinsing practice and toothpaste swallowing), were associated with higher than average prevalances of very mild or more severe fluorosis (Table 2). For example, 45% of children who often licked or ate toothpaste were in this category. Of the 58 potential explanatory variables that were assessed in bivariate associations with fluorosis, only five were statistically significant and entered into logistic regression modeling. These included frequency of toothbrushing, rinsing habit after brushing, eating or licking toothpaste (these behaviors relate to when toothbrushing commenced as a habit), exposure to fluoridated water, and type of water used for the reconstitution of infant formula. As exposure to fluoridated water and water from various sources used for reconstitution of infant formula were highly-correlated variables, they were modeled separately with the three oral hygiene habit variables. In each case, all variables remained in the model as significant independent explanations of very mild or more severe fluorosis (Tables 3 and 4). In both models, the adjusted odds ratios (OR) that were associated with the oral hygiene habits were almost identical. Compared with referent groups, swallowing with or without rinsing following brushing (OR = 2.3) or licking or eating toothpaste often (OR = 1.8) had elevated odds of dental fluorosis. Compared with those who brushed daily with a fluoride toothpaste, those brushing less frequently (OR = 0.58) had reduced odds of fluorosis. Compared with children who had no exposure to fluoridated mains supply water (i.e. those whose drinking water was from spring or rain water sources), those so exposed had elevated odds (OR = 1.6) of very mild or more severe fluo48
Table 3. Logistic regression analysis of potential explanatory variables for very mild or more severe fluorosis on maxillary central incisors (model 1), in which infant formula-reconstituting agent is omitted Potential explanatory variable
n
Unadjusted OR
Adjusted OR
OR
OR
95% CI
Exposure to fluoridated water No (0%) 145 1.00 1.00 Some (1–99%) 263 1.44 0.94, 2.22 1.46 Lifelong (100%) 730 1.50 1.07, 2.09 1.55 Frequency of toothbrushing when habit first started Once daily 529 1.00 1.00 Twice daily or more 436 0.84 0.66, 1.09 0.83 Less than daily 137 0.60 0.38, 0.94 0.58 Not reported 36 1.25 0.75, 2.10 1.97 Rinsing habit after toothbrushing Rinse and spit 949 1.00 1.00 Spit only 91 0.99 0.62, 1.57 0.95 Swallow with or 53 2.41 1.38, 4.20 2.34 without rinsing Not reported 45 1.35 0.92, 1.97 1.04 Licked or ate toothpaste Never 582 1.00 1.00 Sometimes 431 1.01 0.76, 1.36 1.00 Often 87 1.87 1.37, 2.56 1.83 Not reported 38 1.36 0.76, 2.43 0.63
95% CI
0.98, 2.18 1.21, 2.13
0.63, 1.08 0.38, 0.95 0.24, 16.02
0.59, 1.52 1.32, 4.13 0.38, 2.83
0.75, 1.31 1.30, 2.60 0.13, 3.13
CI, confidence interval; OR, odds ratio. Bold indicates significance (P < 0.05).
rosis, as did those whose infant formula had been reconstituted with fluoridated mains supply (OR = 1.7). In a separate analysis, it was shown that exposure to increasing numbers of the risk factors, shown in Tables 3 and 4, significantly increased the risk of very mild or more severe fluorosis (OR = 1.33, 95% confidence interval = 1.17, 1.52). ª 2014 Wiley Publishing Asia Pty Ltd
I.S. Bal et al.
Policy implications of dental fluorosis
Table 4. Logistic regression analysis of potential explanatory variables for very mild or more severe fluorosis on maxillary central incisors (model 2), in which exposure to fluoridated water is omitted Potential explanatory variable
n
Unadjusted OR
Adjusted OR
OR
OR
95% CI
Infant formula-reconstituting agent Spring/rain water 145 1.00 1.00 Mains fluoridated 611 1.64 1.14, 2.37 1.69 water No formula/not 382 1.28 0.89, 1.84 1.30 reported Frequency of toothbrushing when habit first started Once daily 529 1.00 1.00 Twice daily or more 436 0.84 0.66, 1.09 0.83 Less than daily 137 0.60 0.38, 0.94 0.58 Not reported 36 1.25 0.75, 2.10 2.01 Rinsing habit after toothbrushing Rinse and spit 949 1.00 1.00 Spit only 91 0.99 0.62, 1.57 0.93 Swallow with or 53 2.41 1.38, 4.20 2.30 without rinsing Not reported 45 1.35 0.92, 1.97 1.08 Licked or ate toothpaste Never 582 1.00 1.00 Sometimes 431 1.01 0.76, 1.36 0.98 Often 87 1.87 1.37, 2.56 1.81 Not reported 38 1.36 0.76, 2.43 0.62
95% CI
1.21, 2.37 0.91, 1.87
0.63, 1.09 0.36, 0.94 0.24, 16.87
0.59, 1.48 1.30, 4.08 0.40, 2.91
0.74, 1.30 1.29, 2.56 0.12, 3.16
CI, confidence interval; OR, odds ratio. Bold indicates significance (P < 0.05).
Discussion The principal limitation of this study, which is common to most other cross-sectional studies, is that the quality of data on toothbrushing experience and all factors associated with it cannot be guaranteed because it was gathered retrospectively, often many years after the event. Thus, the separation of the fluoride toothpaste and fluoridated water effects on fluorosis variation is subject to uncertainty. A higher-than-expected prevalence of fluorosis was observed. In fact, fluorosis prevalence was approaching the level that might be expected in areas where approximately 2 ppm of fluoride is naturally present in drinking water, at which concentration the appearance of moderate fluorosis begins to emerge.2 It is noteworthy, therefore, that the children in this study who were not exposed to fluoridation had a low fluorosis prevalence, but their primary dentition caries experience was highly elevated (OR = 2.7).17 However, in Dean’s time, fluoride in drinking water was the only source of fluoride to which populations were exposed, and the so-called optimum fluoride concentration of 1 ppm that he set to minimize the risk of fluorosis and maximize the caries-preventive effect continues to be followed with some variation related to ª 2014 Wiley Publishing Asia Pty Ltd
climatic temperature.2 Today, however, most children are exposed to other sources of fluoride besides that in water, mainly from dental products containing fluoride, such as toothpaste; topically-applied rinses, gels, and varnish; or from fluoride supplements. In the present study, none of the most severe cases was associated with reported prior use of fluoride supplements. The much higher-than-expected level of fluorosis might have been in part due to the fact that, during a 5-year period, the mean fluoride concentration maintained in Sydney water supplies between 1998 and 2003 was 0.1 ppm above the target concentration of 1 ppm.18 This is 32% above the level of 0.83 that is reckoned as optimal (Sydney mean maximum daily temperature of 22.3°C) according to the Galagan and Vermillion algorithm.4 It was reported in the 2007 NSW Child Dental Health Survey that the prevalence of moderate or more severe fluorosis in fluoridated regions was 3%; twice the level found in the Blue Mountains and Hawkesbury.19 The use of the upper central incisor data to estimate the CFI might have resulted in an underestimate, as laterforming teeth are often more likely to be more severely affected than incisors.20 In the context of this discussion, however, the esthetics of the canine and premolar teeth are somewhat peripheral; the condition of the central incisors is more visually striking and relevant to any determination of personal or community approval of dental appearance.21 The proportion of children having very mild or more severe fluorosis in both regions was almost identical, but the proportions categorized as normal and questionable in each region were substantially different. This might have reflected a true difference, or might have been due to a shift in diagnostic standard as the study progressed, as all schools in one region were surveyed before moving to the other. Variation in the distribution of fluorosis manifestation is apparent in regions where water supplies are fluoridated or have naturally low fluoride levels. For example, in Ireland, where fluoridation is mandatory, the prevalence of higher grades of fluorosis was nearly identical to that in our study. However, the prevalence of very mild fluorosis in our study was 31% compared with only 10% in Ireland.22 In Ulster, a non-fluoridated city in New York state, moderate and severe fluorosis prevalence was comparable to that observed in the Blue Mountains and Hawkesbury.23 Yet in fluoridated Newburgh, New York, the prevalence of moderate and severe fluorosis was negligible.23 Similarly, in an optimally-fluoridated region of Illinois, fluorosis prevalence was at a low level, corresponding to the situation in the 1930s.24 The high level of fluorosis in Ulster was linked to fluoride toothpaste use. Osuji et al. were the first authors to 49
I.S. Bal et al.
Policy implications of dental fluorosis
show a strong link between fluorosis and early use of fluoride toothpaste, and these observations were confirmed by Pendrys et al.25,26 More recently, results from the Iowa Fluoride Study indicated that the ingestion of fluoride toothpaste by children before the age of 3 years continues to be a fluorosis risk factor.27 Currently, in view of the risk of fluorosis from fluoridated drinking water and other fluoride sources, uncertainty over the variation in water intake by climatic temperature, and that significant caries preventive benefits can be achieved at lower than 1 ppm, the US Department of Health and Human Services is considering a proposal to set the optimal concentration for fluoridated drinking water at 0.7 ppm.28 The 2002 South Australian Fluoride Study (cross-sectional design), using the questionnaire developed by ARCPOH, found that fluorosis was significantly associated with residence in a fluoridated area, supplement use, toothbrushing frequency, age when toothbrushing first started, amount of toothpaste used when first started, and eating or licking toothpaste when first started.9 However, in our study, fluorosis was associated with only two of these factors: brushing frequency when first started and eating or licking toothpaste. In addition, our observation that swallowing toothpaste without rinsing was strongly associated with elevated odds of fluorosis indicates that health promotion on toothpaste use should be implemented. In our study also, contrary to findings in South Australia,9 fluorosis was associated with infant formula use when reconstituted with fluoridated mains supply. Those whose formula had been reconstituted with rain water were only half as likely to experience fluorosis. Levy pointed out that if fluoridated water was added to powdered concentrate, it could result in ingestion of high levels of fluoride if infants were ingesting many ounces of reconstituted formula per day.29 On the basis of an evaluation of the fluoride content in infant formula, Siew et al. stated: “When powdered or liquid concentrate infant formulas are the primary source of nutrition, some infants are likely to exceed the recommended fluoride upper limit if the formula is reconstituted with water containing 1 ppm fluoride” (p. 1233).30 Further, in a systematic review of infant formula use and fluorosis, Hujoel et al.31 reported that the risk of fluorosis increased by 5% for each 0.1 ppm increase in reported levels of fluoride in the reconstituting water. However, the authors cautioned that this estimate might be inflated due to probable non-publication bias and to recall bias associated with most of the included studies for which the study designs were crosssectional. A majority (64%) of children in our study had reportedly been exposed to fluoridated water since birth. 50
Their elevated odds of fluorosis, compared with those not as exposed, are consistent with worldwide experience. Similarly, children whose formula had been reconstituted with fluoridated mains supply experienced a significantly higher degree of fluorosis than those whose formula had been reconstituted with rain or spring water. It is well known that preschool children cannot reliably avoid swallowing toothpaste or other products, such as fluoride rinses. Thus, it was of concern that 7% of children claimed to have used a fluoride rinse when they were under the age of 5 years. The severity of fluorosis observed in the Blue Mountains and Hawkesbury regions is of some concern. While fluoridation is a known risk factor for low-level fluorosis, it is hailed as one of the 10 great public health achievements of the 20th century.32 Its pre-eminence is its power to reduce both caries incidence and the burden of caries experience. The 72% reduction in decayed, missing, and filled teeth in children aged 8–11 years in the Blue Mountains following 11 years of fluoridation was among the best outcomes achieved worldwide.12 However, this benefit can be delivered by fluoridated water without an associated risk of moderate or severe fluorosis, as was demonstrated by Dean and his successors.2,4,5,7,23,24 This risk is diminished by moderating the target concentration for fluoride in relation to climatic temperature, and by controlling exposures to fluoride-containing dental products. Although the Galagan and Vermillion guideline was established for US conditions, these are likely to be similar for conditions across Australia.4 The application of the guideline to the Sydney metropolitan region would indicate that a target dose of 0.8 mg F/L of water would be appropriate. In Hong Kong, a reduction in the target fluoride concentration by 0.2 mg resulted in a substantial reduction in fluorosis risk without any corresponding increase in caries prevalence or experience.7 In view of the use of other fluoride-containing products advocated for caries prevention nowadays, especially toothpaste, it is important to ensure that target fluoride concentrations for fluoridated water are justified in terms of minimizing caries and fluorosis risks, and importantly, to safeguard against unjust criticism of fluoridation. Conclusions For the group as a whole, it is concluded that: (a) fluorosis prevalence in the Blue Mountains and Hawkesbury was similar; and (b) the higher-than-expected prevalence and severity of fluorosis was due mainly to two factors: (a) the higher-than-optimal fluoride concentration in drinking water; and (b) swallowing of fluoride toothpaste in early childhood. ª 2014 Wiley Publishing Asia Pty Ltd
I.S. Bal et al.
Policy implications of dental fluorosis
Recommendations
Acknowledgments
In terms of NSW oral health policy, overall exposure to fluoride should be reduced, in line with the 2006 guidelines on use of fluoride in Australia,33 through an oral health-promotion program that addresses a combination of controls on the: (a) fluoride concentration in Sydney water supplies; and (b) inappropriate use by preschool children of fluoride toothpaste, fluoride rinse, and fluoride dietary supplements.
We are grateful for the administrative support and other help given by Ms Debbie Gibbons (Wentworth Area Health Service), Ms Ramona Grimm, and Dr Andy Hsiau and Dr Shanti Sivaneswaran. We thank Dr Karen Byth and Dr Manish Arora for statistical advice and assistance. The school principals, teachers, parents, and children are also gratefully acknowledged. The study was funded through a grant from the Centre for Oral Health Strategy, New South Wales Health.
References 1 WHO. Environmental health criteria 36; Fluorine and fluorides. Geneva: World Health Organization, 1984. 2 Dean HT. The investigation of physiological effects by the epidemiological method. In: Moulton FR, ed. Fluorine and dental health. Washington DC: American Association for the Advancement of Science, 1942: 23–31. 3 Arnold FA. Role of fluorides in preventive dentistry. J Am Dent Assoc 1943; 30: 499–508. 4 Galagan DJ, Vermillion JR. Determining optimum fluoride concentrations. Public Health Rep 1957; 72: 491–3. 5 Russell AL. Dental fluorosis in Grand Rapids during the seventeenth year of fluoridation. J Am Dent Assoc 1962; 65: 608–12. 6 McDonagh MS, Whiting PF, Wilson PM et al. Systematic review of water fluoridation. Br Med J 2000; 321: 855. 7 Evans RW, Stamm JW. Dental fluorosis following downward adjustment to fluoride in drinking water. J Public Health Dent 1991; 51: 91–8. 8 Riordan PJ. Dental fluorosis decline after changes to supplement and toothpaste regimens. Community Dent Oral Epidemiol 2002; 30: 233–40. 9 Spencer AJ, Do LG. Changing risk factors for fluorosis among South Australian children. Community Dent Oral Epidemiol 2008; 36: 210–8. 10 Do LG, Spencer AJ. Decline in the prevalence of dental fluorosis among South Australian children. Community Dent Oral Epidemiol 2007; 35: 282–91. 11 Patterson AF, Weidenhofer RNG. A study of the dental health of primary
ª 2014 Wiley Publishing Asia Pty Ltd
12
13
14
15
16
17
18
19
school children in the local government areas of the Blue Mountains and Hawkesbury NSW, 1993. Sydney: Dental Health Unit, NSW Department of Health, 1993. Evans RW, Hsiau ACY, Dennison PJ, Patterson A, Jalaludin B. Water fluoridation in the Blue Mountains reduces risk of tooth decay. Aust Dent J 2009; 54: 368–73. WHO. Oral health surveys basic methods. Geneva: World Health Organization, 1997. Evans RW, Stamm JW. An epidemiologic estimate of the critical period during which human maxillary central incisors are most susceptible to fluorosis. J Public Health Dent 1991; 51: 251–9. Russell AL. The differential diagnosis of fluoride and nonfluoride enamel opacities. J Public Health Dent 1961; 21: 143–6. Cuming RG, Sherrington C, Lord RS et al. Cluster randomized trial of a targeted multifactorial intervention to prevent falls among older people in hospital. BMJ 2008; 336: 758–60. Hsiau ACY. Dental caries in primary school children residing in the Blue Mountains and Hawkesbury regions, NSW. MDSc thesis. University of Sydney, 2004. Public Health Division. The health of the people of NSW. Report of the NSW Chief Health Officer 2005. Sydney: NSW Department of Health, 2006. Centre for Oral Health Strategy NSW. The New South Wales Child Dental Health Survey 2007. Wentworthville, NSW: Centre for Oral Health Strategy, 2009: 28–9.
20 Larson MJ, Kirkegaard E, Poulsen S. Patterns of dental fluorosis in a European country in relation to the fluoride concentration of drinking water. J Dent Res 1987; 66: 10–2. 21 Edwards M, Macpherson LMD, Simmons DR, Gilmore WH, Stephen KW. An assessment of teenagers’ perceptions of dental fluorosis using digital simulation and Web-based testing. Community Dent Oral Epidemiol 2005; 33: 298–306. 22 Whelton H, Crowley E, O’Mullane D, Donaldson M, Cronin M, Kelleher V. Dental caries and enamel fluorosis among the fluoridated population in the Republic of Ireland and non fluoridated population in Northern Ireland in 2002. Community Dent Health 2006; 23: 37–43. 23 Kumar JV, Swango PA. Fluoride exposure and dental fluorosis in Newburgh and Kingston, New York: policy-implications. Community Dent Oral Epidemiol 1999; 27: 171–80. 24 Selwitz RH, Nowjack-Raymer RE, Kingman A, Driscoll WS. Prevalence of dental caries and dental fluorosis in areas with optimal and above-optimal water fluoride concentrations: a 10-year follow-up survey. J Public Health Dent 1995; 55: 85–93. 25 Osuji OO, Leake JL, Chipman ML, Nikiforuk G, Locker D, Levine N. Risk factors for dental fluorosis in a fluoridated community. J Dent Res 1988; 67: 1488–92. 26 Pendrys DG, Katz RV, Morse DE. Risk factors for enamel fluorosis in a fluoridated population. Am J Epidemiol 1994; 140: 461–71. 27 Franzman MR, Levy SM, Warren JJ, Broffit B. Fluoride dentifrice ingestion
51
I.S. Bal et al.
Policy implications of dental fluorosis
and fluorosis on the permanent incisors. J Am Dent Assoc 2006; 137: 645–52. 28 US Department of Health and Human Services. Proposed HHS Recommendation for Fluoride Concentration in Drinking Water for Prevention of Dental Caries. Washington DC: Office of Secretary. Available from: http://www.hhs.gov/news/press/ 2011pres/01/pre_pub_frn_fluoride. html (Accessed 20 June 2011).
52
29 Levy SM. An update on fluorides and fluorosis. J Can Dent Assoc 2003; 69: 286–91. 30 Siew C, Strock S, Ristic H et al. Assessing a potential risk factor for enamel fluorosis: a preliminary evaluation of fluoride content in infant formulas. J Am Dent Assoc 2009; 140: 1228–36. 31 Hujoel PP, Zina LG, Moimaz SAS, Cunha-Cruz J. Infant formula and
enamel fluorosis: a systematic review. J Am Dent Assoc 2009; 140: 841–54. 32 Centers for Disease Control and Prevention. Ten great public health achievements – United States, 1900– 1999. J Am Med Assoc 1999; 281: 1481. 33 Australian Research Centre for Population Oral Health. The use of fluorides in Australia: guidelines. Aust Dent J 2006; 51: 195–9.
ª 2014 Wiley Publishing Asia Pty Ltd
ORIGINAL ARTICLE Community Dentistry and Oral Epidemiology
Dental fluorosis in the Blue Mountains and Hawkesbury, New South Wales, Australia: policy implications Ikreet S. Bal1, Peter J. Dennison2 & R. Wendell Evans2 1 Department of Public Health Dentistry, Dr Harvansh Singh Judge Institute of Dental Sciences and Hospital, Panjab University, Chandigarh, India 2 Sydney Dental School, The University of Sydney, Sydney, NSW, Australia
Keywords caries, fluoridation, fluoride toothpaste, fluorosis, health policy. Correspondence Associate Professor R. Wendell Evans, Community Oral Health and Epidemiology, Westmead Centre for Oral Health, Population Oral Health, 1 Mons Road, Westmead, New South Wales 2145, Australia. Tel: +61-02-8821-4364 Fax: +61-02-8821-4366 Email: [email protected] Received 5 April 2014; accepted 26 October 2014. doi: 10.1111/jicd.12138
Abstract Aim: The aim of the present study was to determine whether the adjustment of the fluoride concentration to 1 ppm in the drinking water supplied to the Blue Mountains, New South Wales, Australia in 1993 was associated with fluorosis incidence. Methods: In 2003, children attending schools in the Blue Mountains and a control region (fluoridated in 1967) that had been randomly selected at baseline in 1992 were examined for dental fluorosis (maxillary central incisors only) using Dean’s index. A fluoride history for each child was obtained by questionnaire. Associations between fluorosis and 58 potential explanatory variables were explored. Results: The response rate was 63%. A total of 1138 children aged from 7 to 11 years with erupted permanent central incisors were examined for dental fluorosis. Fluorosis prevalence was the same in both regions. The Community Index of Dental Fluorosis values were slightly different, but were both above 0.6, indicative of public health concern. Conclusions: For the group as a whole, we concluded that: (a) fluorosis prevalence (0.39) in both regions was similar; and (b) the higher-than-expected prevalence and severity of fluorosis was due mainly to two factors: (a) the higher-than-optimal fluoride level in drinking water; and (b) swallowing of fluoride toothpaste in early childhood.
Introduction The element fluorine, usually found in an ionic form, is naturally ubiquitous as the 13th most abundant in the earth’s crust.1 It has been present in the environment long before the start of biological life, and all living organisms, including Homo sapiens, have been exposed to it throughout their development. Because it is completely unavoidable, the only question that need exercise the minds of scientists is how much of this tasteless element human beings are exposed to and what the effects of such exposure are. Dean demonstrated that the optimal fluoride concentration in drinking water in the USA for the prevention of dental caries and dental fluorosis was 1 ppm.2 Later, ª 2014 Wiley Publishing Asia Pty Ltd
Arnold noted that diet and climatic factors were related, and that concentrations of fluoride necessary to prevent caries might vary.3 He suggested that concentrations of 0.5–0.7 ppm in southwestern regions of the USA might be as effective as 1–1.5 ppm in the north. Subsequently, Galagan and Vermillion developed an algorithm for determining the optimal fluoride concentration based on fluid intake of US population groups in relation to ambient temperature and caries experience data.4 Variants of these considerations guide determinations of target concentrations elsewhere. It is accepted that, consequent to achieving an optimal fluoride concentration to bring about the desired reduction in dental caries experience, the mildest manifestations of fluorosis will be prevalent in approximately 15% of the population.5 Most milder forms of 45
I.S. Bal et al.
Policy implications of dental fluorosis
dental fluorosis are difficult to identify by an untrained examiner. A recent review of water fluoridation, commissioned by the British Government, confirmed the association of dental fluorosis with fluoridated water.6 At optimal concentrations, the Community Index of Dental Fluorosis (CFI) value (the mean of the fluorosis rank scores) is below 0.4, a level which Dean claimed was compatible with public health.2 It has been demonstrated that higher-than-expected levels of fluorosis might occur in association with above-optimal concentrations of fluoride in drinking water, but this might be corrected by adjusting the fluoride concentration downwards,7 or by controlling the ingestion of fluoride from known sources.8–10 In 1980, the New South Wales (NSW) Water Board (Australia) assumed responsibility for the water supply in the Blue Mountains local government area (LGA), a partly rural district on the western fringe of metropolitan Sydney, and foreshadowed that fluoridation of the reticulated water supply to this region would occur.11 This commenced following substantial infrastructure development in 1992, thereby completing the extension of the fluoridation program to the entire greater Sydney metropolitan region. The fluoride concentration was set at 1 ppm, in line with the Sydney target. The current investigation was part of a larger study which confirmed the substantial benefits of extending water fluoridation to the Blue Mountains LGA for the prevention of dental caries.12 The adjacent Hawkesbury LGA, which had been fluoridated since 1969, was selected as the reference area due to its similar sociodemographic profile and geographic proximity to the Blue Mountains. The populations of the Blue Mountains and Hawkesbury in 2006 were approximately 73 000 and 64 000, respectively. The respective land areas are 1430 and 2793 square kilometers, of which 70–85% is within the World Heritage Blue Mountains National Park. The main economic activity in both areas is tourism, and additionally in Hawkesbury, fruit farming and market gardening. The 1992 baseline survey covering children in both regions reported on dental caries experience, but fluorosis status was not assessed.11 Therefore, the purposes of this analysis were: (a) to determine whether the adjustment of the fluoride concentration to 1 ppm in the drinking water supplied to the Blue Mountains 11 years previously was associated with dental fluorosis prevalence; (b) to determine the similarity or otherwise in fluorosis prevalence in the Blue Mountains and Hawkesbury regions; and (c) to evaluate the risk of fluorosis from a range of fluoride sources. Permission for the survey was obtained from the Human Research Ethics Committees of The University of Sydney, the New South Wales Department of Education and Training, The Catholic Archdiocese, and the Wentworth Area Health Service. 46
Material and methods Sampling In this cross-sectional survey, the target population was the primary school children in the Blue Mountains and Hawkesbury LGA of NSW. The 1992 caries experience baseline survey conducted by Patterson and Weidenhofer was carried out in 18 schools.11 These were “selected randomly using comparative size as the main criteria. Additionally some smaller schools were included and attempts were made to obtain a reasonable geographic spread across the LGAs”.11 The schools, as selected for the 1992 survey, became the target schools for the 2003 survey, except that two schools having enrolments of 107 and 16 pupils, respectively, in the Hawkesbury LGA were excluded as they had come under the purview of another Area Health Service. Recruitment of sample The principals of the selected schools were visited by PJD and given letters of invitation to include their schools in the survey. All agreed to participate, and school staff assisted by distributing information packages to the parents/guardians of the children, and also by collecting the responses that were returned in sealed envelopes. The information packages included an information statement, a request for consent to participate in the survey, a consent form, and a questionnaire. The information statement explained the purpose of the study and assured parents of confidentiality. Parents were requested to return the completed questionnaire and the consent form to the school before the date set for examining the children at that particular school. A reminder notice was sent to the parents 1 week after forwarding the information packages. Only children whose parents/guardians gave consent for inclusion in the survey were examined. Children born before 1992 were excluded from this analysis, which is based on those aged 7–11 years for the 2003 survey. Fluoride history A questionnaire (available on request) developed by the Australian Research Centre for Population Oral Health (ARCPOH) was used in this survey to seek demographic information, as well as a detailed fluoride history regarding exposure to fluoridated water and use of toothpastes, fluoride supplements, fluoride mouth rinses, and professionally-applied fluoride gel. Some rural households within the Blue Mountains and Hawkesbury LGA did not receive mains supply water; instead, their water source was from springs or rain water. Children from ª 2014 Wiley Publishing Asia Pty Ltd
I.S. Bal et al.
Policy implications of dental fluorosis
other communities, fluoridated or otherwise, who became resident in the area, were also included in the survey. Measurement of fluorosis Dental fluorosis was assessed according to the CFI criteria developed by Dean,2 as recommended by the World Health Organization (WHO),13 but restricted to the examination of the maxillary central incisors, as the control of dental fluorosis with reference to this tooth type is synonymous with the control of fluorosis in the dentition as a whole.14 The WHO reference photographs were used as the diagnostic standards. Russell’s differential diagnosis criteria for fluorosis were followed.15 Clinical examination All consenting children with erupted permanent maxillary central incisors were examined. Examinations of the wet maxillary central incisors were conducted in shaded natural daylight, either outdoors or indoors beside a window, and the score for each child was recorded on a class list obtained from the school. Although fluorosis manifestation is bilaterally symmetrical, in rare cases when this did not hold, diagnosis was made in relation to the more affected incisor. The examinations were carried out by ISB, who had been calibrated before the survey. Data quality was monitored daily during the examinations, and formal interand intra-examiner calibration (with RWE) was periodically carried out during the survey. Data analysis Fluorosis prevalence and CFI values were computed. Associations between fluorosis and 58 potential explana-
tory variables were explored in a series of bivariate analyses. Associations between categorical variables were evaluated using the chi-squared-test or Fisher’s exact test, while regression analysis was used to evaluate associations with continuous variables. Associations that were significant at the P < 0.2 level were evaluated further using logistic regression. Generalized estimating equations were used to assess the effect of potential risk factors on individuals’ fluorosis status (“very mild or more” vs “normal or questionable”). In these binomial models, an exchangeable correlation structure was used to allow for clustering by school.16 The unadjusted intracluster correlation coefficient over all the schools was 0.016. The statistical software package S-PLUS version 8 (Insightful, Seattle, WA, USA) was used to analyze the data. Two-tailed tests with a significance level of 5% were used throughout. Results The response rate was 63%. A total of 1138 children (16 clusters, number range: 34–195, median number: 63) aged from 7 to 11 years with erupted permanent central incisors were examined for dental fluorosis (Table 1). Weighted kappa values for monitoring examiner reliability ranged from 0.92 initially (59 duplicate recordings) to 0.96 finally (60 duplicates). A fluorosis prevalence of 39% was observed in the Blue Mountains and Hawkesbury regions inclusive of 16 cases of moderate or severe (1.4%) (Table 1). CFI values were slightly different (P < 0.01), but both were above the 0.6 level nominated by Dean as indicative of a public health concern.2 Sixty-four percent had been exposed to fluoridated water from birth. As shown in Table 2, children were exposed to other fluorosis risk factors, including the use of fluoridated water for infant formula reconstitution, toothbrushing with fluoride toothpaste before the age of
Table 1. Distribution of enamel fluorosis and the Index of Dental Fluorosis based on maxillary central incisors alone by local government area Blue Mountains
Hawkesbury
Both
Enamel fluorosis
n
%
n
%
n
%
95% CI
Normal Questionable Very mild Mild Moderate Severe Total Fluorosis prevalence† (SE) Index of Dental Fluorosis
110 233 176 33 8 4 564 0.39 0.71*
19.5 41.3 31.2 5.9 1.4 0.7 100 (0.01)
245 220 241 48 5 3 762 0.39 0.62
32.2 28.9 31.6 6.3 0.7 0.4 100 (0.01)
355 453 417 81 13 7 1326 0.39 0.66
26.3 34.2 31.4 6.1 1.0 0.5 100 (0.01)
24.4, 31.6, 29.0, 4.9, 0.5, 0.2,
29.3 36.8 34.0 7.6 1.7 1.1
CI, confidence interval; SE, standard error. *P < 0.001. † Very mild or more.
ª 2014 Wiley Publishing Asia Pty Ltd
47
I.S. Bal et al.
Policy implications of dental fluorosis
Table 2. Sample distribution by potential high-risk fluorosis factors and its degree of severity Normal or questionable Potential high-risk factor for dental fluorosis Infant-feeding practices Fluoridated mains supply as formula-reconstituting agent Oral hygiene practices when toothbrushing commenced Commenced brushing before age 2 years Once or more times daily toothbrushing Used brush length of toothpaste Used standard concentration toothpaste (1000 ppm) Just swallowed without rinsing after brushing Eating/licking toothpaste often Use of fluoride rinse before age 5 years Use of fluoride tablets or drops before age 5 years
Very mild or mild
n
%†
n
%†
352
58
250
41
305 583 53 149 4 21 43 25
59 60 58 64 27 55 51 54
202 366 37 81 10 17 41 21
39 38 41 35 67 45 49 46
Moderate or severe
Total
%†
n
%‡
9
2
611
54
9 16 1 4 1 0 0 0
2 2 1 2 7 0 0 0
516 965 91 234 15 38 84 46
45 85 8 21 1 3 7 4
n
†
% of row total. % of sample total.
‡
2 years, early use of fluoride rinse, and fluoride supplements. Some of these factors, including related toothbrushing habits (rinsing practice and toothpaste swallowing), were associated with higher than average prevalances of very mild or more severe fluorosis (Table 2). For example, 45% of children who often licked or ate toothpaste were in this category. Of the 58 potential explanatory variables that were assessed in bivariate associations with fluorosis, only five were statistically significant and entered into logistic regression modeling. These included frequency of toothbrushing, rinsing habit after brushing, eating or licking toothpaste (these behaviors relate to when toothbrushing commenced as a habit), exposure to fluoridated water, and type of water used for the reconstitution of infant formula. As exposure to fluoridated water and water from various sources used for reconstitution of infant formula were highly-correlated variables, they were modeled separately with the three oral hygiene habit variables. In each case, all variables remained in the model as significant independent explanations of very mild or more severe fluorosis (Tables 3 and 4). In both models, the adjusted odds ratios (OR) that were associated with the oral hygiene habits were almost identical. Compared with referent groups, swallowing with or without rinsing following brushing (OR = 2.3) or licking or eating toothpaste often (OR = 1.8) had elevated odds of dental fluorosis. Compared with those who brushed daily with a fluoride toothpaste, those brushing less frequently (OR = 0.58) had reduced odds of fluorosis. Compared with children who had no exposure to fluoridated mains supply water (i.e. those whose drinking water was from spring or rain water sources), those so exposed had elevated odds (OR = 1.6) of very mild or more severe fluo48
Table 3. Logistic regression analysis of potential explanatory variables for very mild or more severe fluorosis on maxillary central incisors (model 1), in which infant formula-reconstituting agent is omitted Potential explanatory variable
n
Unadjusted OR
Adjusted OR
OR
OR
95% CI
Exposure to fluoridated water No (0%) 145 1.00 1.00 Some (1–99%) 263 1.44 0.94, 2.22 1.46 Lifelong (100%) 730 1.50 1.07, 2.09 1.55 Frequency of toothbrushing when habit first started Once daily 529 1.00 1.00 Twice daily or more 436 0.84 0.66, 1.09 0.83 Less than daily 137 0.60 0.38, 0.94 0.58 Not reported 36 1.25 0.75, 2.10 1.97 Rinsing habit after toothbrushing Rinse and spit 949 1.00 1.00 Spit only 91 0.99 0.62, 1.57 0.95 Swallow with or 53 2.41 1.38, 4.20 2.34 without rinsing Not reported 45 1.35 0.92, 1.97 1.04 Licked or ate toothpaste Never 582 1.00 1.00 Sometimes 431 1.01 0.76, 1.36 1.00 Often 87 1.87 1.37, 2.56 1.83 Not reported 38 1.36 0.76, 2.43 0.63
95% CI
0.98, 2.18 1.21, 2.13
0.63, 1.08 0.38, 0.95 0.24, 16.02
0.59, 1.52 1.32, 4.13 0.38, 2.83
0.75, 1.31 1.30, 2.60 0.13, 3.13
CI, confidence interval; OR, odds ratio. Bold indicates significance (P < 0.05).
rosis, as did those whose infant formula had been reconstituted with fluoridated mains supply (OR = 1.7). In a separate analysis, it was shown that exposure to increasing numbers of the risk factors, shown in Tables 3 and 4, significantly increased the risk of very mild or more severe fluorosis (OR = 1.33, 95% confidence interval = 1.17, 1.52). ª 2014 Wiley Publishing Asia Pty Ltd
I.S. Bal et al.
Policy implications of dental fluorosis
Table 4. Logistic regression analysis of potential explanatory variables for very mild or more severe fluorosis on maxillary central incisors (model 2), in which exposure to fluoridated water is omitted Potential explanatory variable
n
Unadjusted OR
Adjusted OR
OR
OR
95% CI
Infant formula-reconstituting agent Spring/rain water 145 1.00 1.00 Mains fluoridated 611 1.64 1.14, 2.37 1.69 water No formula/not 382 1.28 0.89, 1.84 1.30 reported Frequency of toothbrushing when habit first started Once daily 529 1.00 1.00 Twice daily or more 436 0.84 0.66, 1.09 0.83 Less than daily 137 0.60 0.38, 0.94 0.58 Not reported 36 1.25 0.75, 2.10 2.01 Rinsing habit after toothbrushing Rinse and spit 949 1.00 1.00 Spit only 91 0.99 0.62, 1.57 0.93 Swallow with or 53 2.41 1.38, 4.20 2.30 without rinsing Not reported 45 1.35 0.92, 1.97 1.08 Licked or ate toothpaste Never 582 1.00 1.00 Sometimes 431 1.01 0.76, 1.36 0.98 Often 87 1.87 1.37, 2.56 1.81 Not reported 38 1.36 0.76, 2.43 0.62
95% CI
1.21, 2.37 0.91, 1.87
0.63, 1.09 0.36, 0.94 0.24, 16.87
0.59, 1.48 1.30, 4.08 0.40, 2.91
0.74, 1.30 1.29, 2.56 0.12, 3.16
CI, confidence interval; OR, odds ratio. Bold indicates significance (P < 0.05).
Discussion The principal limitation of this study, which is common to most other cross-sectional studies, is that the quality of data on toothbrushing experience and all factors associated with it cannot be guaranteed because it was gathered retrospectively, often many years after the event. Thus, the separation of the fluoride toothpaste and fluoridated water effects on fluorosis variation is subject to uncertainty. A higher-than-expected prevalence of fluorosis was observed. In fact, fluorosis prevalence was approaching the level that might be expected in areas where approximately 2 ppm of fluoride is naturally present in drinking water, at which concentration the appearance of moderate fluorosis begins to emerge.2 It is noteworthy, therefore, that the children in this study who were not exposed to fluoridation had a low fluorosis prevalence, but their primary dentition caries experience was highly elevated (OR = 2.7).17 However, in Dean’s time, fluoride in drinking water was the only source of fluoride to which populations were exposed, and the so-called optimum fluoride concentration of 1 ppm that he set to minimize the risk of fluorosis and maximize the caries-preventive effect continues to be followed with some variation related to ª 2014 Wiley Publishing Asia Pty Ltd
climatic temperature.2 Today, however, most children are exposed to other sources of fluoride besides that in water, mainly from dental products containing fluoride, such as toothpaste; topically-applied rinses, gels, and varnish; or from fluoride supplements. In the present study, none of the most severe cases was associated with reported prior use of fluoride supplements. The much higher-than-expected level of fluorosis might have been in part due to the fact that, during a 5-year period, the mean fluoride concentration maintained in Sydney water supplies between 1998 and 2003 was 0.1 ppm above the target concentration of 1 ppm.18 This is 32% above the level of 0.83 that is reckoned as optimal (Sydney mean maximum daily temperature of 22.3°C) according to the Galagan and Vermillion algorithm.4 It was reported in the 2007 NSW Child Dental Health Survey that the prevalence of moderate or more severe fluorosis in fluoridated regions was 3%; twice the level found in the Blue Mountains and Hawkesbury.19 The use of the upper central incisor data to estimate the CFI might have resulted in an underestimate, as laterforming teeth are often more likely to be more severely affected than incisors.20 In the context of this discussion, however, the esthetics of the canine and premolar teeth are somewhat peripheral; the condition of the central incisors is more visually striking and relevant to any determination of personal or community approval of dental appearance.21 The proportion of children having very mild or more severe fluorosis in both regions was almost identical, but the proportions categorized as normal and questionable in each region were substantially different. This might have reflected a true difference, or might have been due to a shift in diagnostic standard as the study progressed, as all schools in one region were surveyed before moving to the other. Variation in the distribution of fluorosis manifestation is apparent in regions where water supplies are fluoridated or have naturally low fluoride levels. For example, in Ireland, where fluoridation is mandatory, the prevalence of higher grades of fluorosis was nearly identical to that in our study. However, the prevalence of very mild fluorosis in our study was 31% compared with only 10% in Ireland.22 In Ulster, a non-fluoridated city in New York state, moderate and severe fluorosis prevalence was comparable to that observed in the Blue Mountains and Hawkesbury.23 Yet in fluoridated Newburgh, New York, the prevalence of moderate and severe fluorosis was negligible.23 Similarly, in an optimally-fluoridated region of Illinois, fluorosis prevalence was at a low level, corresponding to the situation in the 1930s.24 The high level of fluorosis in Ulster was linked to fluoride toothpaste use. Osuji et al. were the first authors to 49
I.S. Bal et al.
Policy implications of dental fluorosis
show a strong link between fluorosis and early use of fluoride toothpaste, and these observations were confirmed by Pendrys et al.25,26 More recently, results from the Iowa Fluoride Study indicated that the ingestion of fluoride toothpaste by children before the age of 3 years continues to be a fluorosis risk factor.27 Currently, in view of the risk of fluorosis from fluoridated drinking water and other fluoride sources, uncertainty over the variation in water intake by climatic temperature, and that significant caries preventive benefits can be achieved at lower than 1 ppm, the US Department of Health and Human Services is considering a proposal to set the optimal concentration for fluoridated drinking water at 0.7 ppm.28 The 2002 South Australian Fluoride Study (cross-sectional design), using the questionnaire developed by ARCPOH, found that fluorosis was significantly associated with residence in a fluoridated area, supplement use, toothbrushing frequency, age when toothbrushing first started, amount of toothpaste used when first started, and eating or licking toothpaste when first started.9 However, in our study, fluorosis was associated with only two of these factors: brushing frequency when first started and eating or licking toothpaste. In addition, our observation that swallowing toothpaste without rinsing was strongly associated with elevated odds of fluorosis indicates that health promotion on toothpaste use should be implemented. In our study also, contrary to findings in South Australia,9 fluorosis was associated with infant formula use when reconstituted with fluoridated mains supply. Those whose formula had been reconstituted with rain water were only half as likely to experience fluorosis. Levy pointed out that if fluoridated water was added to powdered concentrate, it could result in ingestion of high levels of fluoride if infants were ingesting many ounces of reconstituted formula per day.29 On the basis of an evaluation of the fluoride content in infant formula, Siew et al. stated: “When powdered or liquid concentrate infant formulas are the primary source of nutrition, some infants are likely to exceed the recommended fluoride upper limit if the formula is reconstituted with water containing 1 ppm fluoride” (p. 1233).30 Further, in a systematic review of infant formula use and fluorosis, Hujoel et al.31 reported that the risk of fluorosis increased by 5% for each 0.1 ppm increase in reported levels of fluoride in the reconstituting water. However, the authors cautioned that this estimate might be inflated due to probable non-publication bias and to recall bias associated with most of the included studies for which the study designs were crosssectional. A majority (64%) of children in our study had reportedly been exposed to fluoridated water since birth. 50
Their elevated odds of fluorosis, compared with those not as exposed, are consistent with worldwide experience. Similarly, children whose formula had been reconstituted with fluoridated mains supply experienced a significantly higher degree of fluorosis than those whose formula had been reconstituted with rain or spring water. It is well known that preschool children cannot reliably avoid swallowing toothpaste or other products, such as fluoride rinses. Thus, it was of concern that 7% of children claimed to have used a fluoride rinse when they were under the age of 5 years. The severity of fluorosis observed in the Blue Mountains and Hawkesbury regions is of some concern. While fluoridation is a known risk factor for low-level fluorosis, it is hailed as one of the 10 great public health achievements of the 20th century.32 Its pre-eminence is its power to reduce both caries incidence and the burden of caries experience. The 72% reduction in decayed, missing, and filled teeth in children aged 8–11 years in the Blue Mountains following 11 years of fluoridation was among the best outcomes achieved worldwide.12 However, this benefit can be delivered by fluoridated water without an associated risk of moderate or severe fluorosis, as was demonstrated by Dean and his successors.2,4,5,7,23,24 This risk is diminished by moderating the target concentration for fluoride in relation to climatic temperature, and by controlling exposures to fluoride-containing dental products. Although the Galagan and Vermillion guideline was established for US conditions, these are likely to be similar for conditions across Australia.4 The application of the guideline to the Sydney metropolitan region would indicate that a target dose of 0.8 mg F/L of water would be appropriate. In Hong Kong, a reduction in the target fluoride concentration by 0.2 mg resulted in a substantial reduction in fluorosis risk without any corresponding increase in caries prevalence or experience.7 In view of the use of other fluoride-containing products advocated for caries prevention nowadays, especially toothpaste, it is important to ensure that target fluoride concentrations for fluoridated water are justified in terms of minimizing caries and fluorosis risks, and importantly, to safeguard against unjust criticism of fluoridation. Conclusions For the group as a whole, it is concluded that: (a) fluorosis prevalence in the Blue Mountains and Hawkesbury was similar; and (b) the higher-than-expected prevalence and severity of fluorosis was due mainly to two factors: (a) the higher-than-optimal fluoride concentration in drinking water; and (b) swallowing of fluoride toothpaste in early childhood. ª 2014 Wiley Publishing Asia Pty Ltd
I.S. Bal et al.
Policy implications of dental fluorosis
Recommendations
Acknowledgments
In terms of NSW oral health policy, overall exposure to fluoride should be reduced, in line with the 2006 guidelines on use of fluoride in Australia,33 through an oral health-promotion program that addresses a combination of controls on the: (a) fluoride concentration in Sydney water supplies; and (b) inappropriate use by preschool children of fluoride toothpaste, fluoride rinse, and fluoride dietary supplements.
We are grateful for the administrative support and other help given by Ms Debbie Gibbons (Wentworth Area Health Service), Ms Ramona Grimm, and Dr Andy Hsiau and Dr Shanti Sivaneswaran. We thank Dr Karen Byth and Dr Manish Arora for statistical advice and assistance. The school principals, teachers, parents, and children are also gratefully acknowledged. The study was funded through a grant from the Centre for Oral Health Strategy, New South Wales Health.
References 1 WHO. Environmental health criteria 36; Fluorine and fluorides. Geneva: World Health Organization, 1984. 2 Dean HT. The investigation of physiological effects by the epidemiological method. In: Moulton FR, ed. Fluorine and dental health. Washington DC: American Association for the Advancement of Science, 1942: 23–31. 3 Arnold FA. Role of fluorides in preventive dentistry. J Am Dent Assoc 1943; 30: 499–508. 4 Galagan DJ, Vermillion JR. Determining optimum fluoride concentrations. Public Health Rep 1957; 72: 491–3. 5 Russell AL. Dental fluorosis in Grand Rapids during the seventeenth year of fluoridation. J Am Dent Assoc 1962; 65: 608–12. 6 McDonagh MS, Whiting PF, Wilson PM et al. Systematic review of water fluoridation. Br Med J 2000; 321: 855. 7 Evans RW, Stamm JW. Dental fluorosis following downward adjustment to fluoride in drinking water. J Public Health Dent 1991; 51: 91–8. 8 Riordan PJ. Dental fluorosis decline after changes to supplement and toothpaste regimens. Community Dent Oral Epidemiol 2002; 30: 233–40. 9 Spencer AJ, Do LG. Changing risk factors for fluorosis among South Australian children. Community Dent Oral Epidemiol 2008; 36: 210–8. 10 Do LG, Spencer AJ. Decline in the prevalence of dental fluorosis among South Australian children. Community Dent Oral Epidemiol 2007; 35: 282–91. 11 Patterson AF, Weidenhofer RNG. A study of the dental health of primary
ª 2014 Wiley Publishing Asia Pty Ltd
12
13
14
15
16
17
18
19
school children in the local government areas of the Blue Mountains and Hawkesbury NSW, 1993. Sydney: Dental Health Unit, NSW Department of Health, 1993. Evans RW, Hsiau ACY, Dennison PJ, Patterson A, Jalaludin B. Water fluoridation in the Blue Mountains reduces risk of tooth decay. Aust Dent J 2009; 54: 368–73. WHO. Oral health surveys basic methods. Geneva: World Health Organization, 1997. Evans RW, Stamm JW. An epidemiologic estimate of the critical period during which human maxillary central incisors are most susceptible to fluorosis. J Public Health Dent 1991; 51: 251–9. Russell AL. The differential diagnosis of fluoride and nonfluoride enamel opacities. J Public Health Dent 1961; 21: 143–6. Cuming RG, Sherrington C, Lord RS et al. Cluster randomized trial of a targeted multifactorial intervention to prevent falls among older people in hospital. BMJ 2008; 336: 758–60. Hsiau ACY. Dental caries in primary school children residing in the Blue Mountains and Hawkesbury regions, NSW. MDSc thesis. University of Sydney, 2004. Public Health Division. The health of the people of NSW. Report of the NSW Chief Health Officer 2005. Sydney: NSW Department of Health, 2006. Centre for Oral Health Strategy NSW. The New South Wales Child Dental Health Survey 2007. Wentworthville, NSW: Centre for Oral Health Strategy, 2009: 28–9.
20 Larson MJ, Kirkegaard E, Poulsen S. Patterns of dental fluorosis in a European country in relation to the fluoride concentration of drinking water. J Dent Res 1987; 66: 10–2. 21 Edwards M, Macpherson LMD, Simmons DR, Gilmore WH, Stephen KW. An assessment of teenagers’ perceptions of dental fluorosis using digital simulation and Web-based testing. Community Dent Oral Epidemiol 2005; 33: 298–306. 22 Whelton H, Crowley E, O’Mullane D, Donaldson M, Cronin M, Kelleher V. Dental caries and enamel fluorosis among the fluoridated population in the Republic of Ireland and non fluoridated population in Northern Ireland in 2002. Community Dent Health 2006; 23: 37–43. 23 Kumar JV, Swango PA. Fluoride exposure and dental fluorosis in Newburgh and Kingston, New York: policy-implications. Community Dent Oral Epidemiol 1999; 27: 171–80. 24 Selwitz RH, Nowjack-Raymer RE, Kingman A, Driscoll WS. Prevalence of dental caries and dental fluorosis in areas with optimal and above-optimal water fluoride concentrations: a 10-year follow-up survey. J Public Health Dent 1995; 55: 85–93. 25 Osuji OO, Leake JL, Chipman ML, Nikiforuk G, Locker D, Levine N. Risk factors for dental fluorosis in a fluoridated community. J Dent Res 1988; 67: 1488–92. 26 Pendrys DG, Katz RV, Morse DE. Risk factors for enamel fluorosis in a fluoridated population. Am J Epidemiol 1994; 140: 461–71. 27 Franzman MR, Levy SM, Warren JJ, Broffit B. Fluoride dentifrice ingestion
51
I.S. Bal et al.
Policy implications of dental fluorosis
and fluorosis on the permanent incisors. J Am Dent Assoc 2006; 137: 645–52. 28 US Department of Health and Human Services. Proposed HHS Recommendation for Fluoride Concentration in Drinking Water for Prevention of Dental Caries. Washington DC: Office of Secretary. Available from: http://www.hhs.gov/news/press/ 2011pres/01/pre_pub_frn_fluoride. html (Accessed 20 June 2011).
52
29 Levy SM. An update on fluorides and fluorosis. J Can Dent Assoc 2003; 69: 286–91. 30 Siew C, Strock S, Ristic H et al. Assessing a potential risk factor for enamel fluorosis: a preliminary evaluation of fluoride content in infant formulas. J Am Dent Assoc 2009; 140: 1228–36. 31 Hujoel PP, Zina LG, Moimaz SAS, Cunha-Cruz J. Infant formula and
enamel fluorosis: a systematic review. J Am Dent Assoc 2009; 140: 841–54. 32 Centers for Disease Control and Prevention. Ten great public health achievements – United States, 1900– 1999. J Am Med Assoc 1999; 281: 1481. 33 Australian Research Centre for Population Oral Health. The use of fluorides in Australia: guidelines. Aust Dent J 2006; 51: 195–9.
ª 2014 Wiley Publishing Asia Pty Ltd
