Original Paper Received: July 11, 2014 Accepted: January 23, 2015 Published online: March 25, 2015
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Association between Dairy Intake and Caries among Children and Adolescents. Results from the Danish EYHS Follow-Up Study Susanne M. Lempert a, c, e Lisa B. Christensen b Karsten Froberg d Kyle Raymond e Berit L. Heitmann a, e, f
a
National Institute of Public Health, University of Southern Denmark, and b Institute of Odontology and c School of Oral Health Care, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, d Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, and e Institute of Preventive Medicine, Research Unit for Dietary Studies, Bispebjerg and Frederiksberg Hospital, Frederiksberg, Denmark; f Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney, Sydney, N.S.W., Australia
Abstract The aim of this paper was to investigate associations between the intake of dairy products and the development in caries (DMFS, decayed, missing and filled surfaces) among children/adolescents over a period of 3 and 6 years, and to investigate whether dairy intake protects against caries incidence. A total of 68.9% of the children were caries free at the age of 9 compared with 34.0% of the adolescents at the age of 15 (measured as DMFS = 0). A larger percentage of children/adolescents with a dairy intake above the mean were caries free compared with the group of children/adolescents with an intake below the mean (72.8 vs. 65.8% at age 9 and 41.1 vs. 30.7% at age 15). The results from the generalized estimation equation showed that dairy and milk intake, as well as intakes of components of dairy such as dairy calcium, whey and casein, was generally inversely associated with childhood/adolescent caries experience (measured as DMFS). With regard to caries incidence, the same inverse association was found for incidence over a period of 3 years and for incidence over 6 years, but the results were only sta-
© 2015 S. Karger AG, Basel 0008–6568/15/0493–0251$39.50/0 E-Mail [email protected] www.karger.com/cre
tistically significant for the 3-year incidence and for the unadjusted models of the 6-year incidence. This study found that previous dairy intake, as well as milk intake or intake of dairy components, may be a predictor of future risk of caries measured by the DMFS count level. This relationship was inverse, meaning that a high intake of dairy products was associated with less future caries development. However, more studies on larger cohorts are needed to confirm these findings. © 2015 S. Karger AG, Basel
Oral diseases are multifactorial diseases. With regard to the prevention of oral diseases such as caries and periodontal diseases, optimal oral hygiene is identified as the main preventive factor [Sheiham, 2001; Mobley, 2003; Petersen, 2003]. In Denmark, there are practical and good opportunities to deliver preventive dental care services at an individual level in public and private dental practice through the dental care system, which is free of charge for all up to the age of 17 years [Christensen et al., 2010a]. Other factors influencing oral disease are diet and socioeconomic factors, which have been found to play an important role in caries incidence in children and adolescents [Nicolau et al., 2003; Cameron et al., 2006; Marshall et al., Susanne M. Lempert National Institute of Public Health, University of Southern Denmark Øster Farimagsgade 5A DK–1353 Copenhagen (Denmark) E-Mail susanne.lempert @ sund.ku.dk
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Key Words Dairy · Milk · Calcium · Whey · Casein · Childhood · Caries · DMFS · European Youth Heart Study
252
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
The following hypotheses were examined: – A high intake of dairy products generally has a beneficial effect on dental caries in children and adolescents, and children with the highest intake of dairy have the lowest caries prevalence and incidence. – A high intake of milk has a beneficial effect on dental caries in children and adolescents, and children with the highest intake of milk have the lowest caries incidence. – A high intake of dairy calcium, whey or casein has a beneficial effect in children and adolescents, and children with the highest intake of dairy calcium, dairy whey or dairy casein have the lowest caries prevalence and incidence. Materials and Methods Study Design The European Youth Heart Study (EYHS) is a multicenter, crosscultural and longitudinal study of associations between lifestyle and cardiovascular disease among children and adolescents. The present study used data from the Danish part of the EYHS and data from the National Board of Health’s Recording System for the Danish Child Dental Health Services. The study was school based, and schools in the Danish municipality of Odense were stratified according to school type, location (urban, suburban and rural) and socioeconomic status of the uptake area [Riddoch et al., 2005]. Each school was allocated a weighted equivalent to the number of children enrolled in the school who were eligible for selection into the study; 3 replacement schools were sampled to substitute schools refusing to participate and to be used in case of low response rates. Children within the age groups of 8–10 and 14–16 years were allocated code numbers and randomly selected using random number tables [Riddoch et al., 2005]. At each study location, a sampling frame of schools using official lists was compiled. A total of 35 schools were sampled, of which 25 (71%) agreed to participate. Of these, 3 schools where lost to follow-up – one rural and 1 urban from a middle-class area and 1 urban school from a low-income area. All 3 schools gave interference with the educational process as the reason for dropout. The study was approved by the Danish Data Protection Agency and the local ethics committee and was performed according to the Declaration of Helsinki. All children gave verbal consent, and their parents gave written consent [Riddoch et al., 2005]. Study Population In the first EYHS study (EYHS I), a total of 586 3rd-grade students were sampled (310 girls and 276 boys). In the second EYHS study (EYHS II), a total of 448 3rd-grade students were sampled (257 girls and 191 boys). For this paper, both of these 3rd-grade populations were merged and gave a study population of 1,034 children from the 3rd grade (fig. 1). In the second EYHS study (EYHS II), a total of 444 9th-grade students (of the 586 3rd-graders) were sampled (251 girls and 193 boys). In the third EYHS study (EYHS III), a total of 399 students from the 9th grade (of the original 448 3rd-graders) were sampled (220 girls and 179 boys). For this paper, both of these 9th-grade populations were merged and gave a study population of 843 children from the 9th grade
Lempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
2007; Christensen et al., 2010b; Cinar et al., 2011]. However, previous studies examining dietary influences on the incidence of dental disease have focused mostly on caries development in relation to dietary sugars and fluorides [Sheiham, 2001; Mobley, 2003] or on deficiency disorders such as malnutrition and vitamin deficiencies. Two recent review studies pointed out that milk contains a variety of biologically active substances which may have particular beneficial effects on dental health and mineral malabsorption as well as an antimicrobial effect [Séverin and Wenshui, 2005, Johansson and Lif Holgerson, 2011]. Indeed, a few recent studies have suggested that dairy products may inhibit the progression and development of caries [Lingström et al., 2003; Caglar et al., 2005, 2008; Twetman and Stecksén-Blicks, 2008; Twetman and Keller, 2012]. This has, however, generally been examined in cross-sectional studies, and not yet among children, which is surprising since children and adolescents often belong to the groups of the population with the highest intake of dairy products and milk. Hence, studies are needed that examine the relationship between the intake of dairy products and oral health in longitudinal study designs among children. The biological rationale for a link between the consumption of dairy products and an oral health-promoting effect is based on several ideas. Firstly, lactobacilli from dairy products may contribute to maintain the microecological balance, including the balance of microflora in the oral cavity [KõllKlais et al., 2005; Stecksén-Blicks et al., 2009; Twetman et al., 2009; Lexner et al., 2010; Glavina et al., 2012]. Secondly, a caries inhibiting effect may be seen to be related to the intake of calcium, whey or casein from dairy products, since whey and casein may prevent bacterial adherence to the tooth surface, and calcium is important for the remineralization process in the oral cavity [Azarpazhooh and Limeback, 2008; Llena et al., 2009; Twetman et al., 2009; Tanaka et al., 2010]. Therefore, it seems relevant to examine whether a high dairy intake, including a high intake of milk, may contribute to prevent the development of caries and to assess whether a high intake of calcium, whey or casein from dairy products may have a preventive effect on dental caries. Furthermore, even if an effect of dairy may have limited influence on caries development at the individual level, influencing oral health through the intake of dairy products might have profound effects at the population level and thereby contribute to an increase in public health. The present project aims to examine the consumption of dairy products of children and young people in relation to caries experience as well as the subsequent change in caries experience (DMFS, decayed, missing and filled surfaces), with a follow-up period of 3 and 6 years.
EYHS I, 3rd grade
586
EYHS II, 3rd grade
448
Sum EYHS I/II, 3rd grade
1,034
Cohort of 3rd grade children
749
EYHS II, 9th grade
number of children. The cohort of 749 children from for whom we had background, dairy and dental information at baseline and the cohort of 343 children for whom we had background, dairy and dental information at baseline and follow-up.
(fig. 1) Data used for this paper contained details of 749 children (417 girls and 332 boys) from the 3rd grade on whom we had complete information on dietary intake and dental measures at baseline (referred to as 9-year-olds) and 340 adolescents (203 girls and 137 boys) from the 9th grade with complete follow-up data on dietary intake and dental measures (referred to as 15-year-olds; fig. 1). The study design, sampling and methods have been described in more detail elsewhere [Riddoch et al., 2000; Nielsen et al., 2005]. EYHS I was conducted in 1997–1998, EYHS II in 2003– 2004 and EYHS III in 2009–2010. Measurements Caries Status Data on caries experience was retrieved from the Danish Child Dental Health Services, which contains a range of information on oral health status, including caries status [Helm, 1973; Hausen et al., 2001]. Caries data used for this investigation was collected in 1997 (EYHS I), 2003 (EYHS II) and 2009 (EYHS III). In addition, data on caries experience were collected from all participants the year they turned 12 years. The data were collected by dentists in the municipality of Odense. The children were recalled for visits regularly, and standardized dental examinations were performed for registration of oral health status; these examinations included X-rays when the dentists found it was indicated. Caries was recorded at cavity level. Caries experience was calculated as the sum of DMFT (decayed, missing and filled teeth) or DMFS in the permanent dentition. Caries-free children/adolescents were defined as children/adolescents with no prior caries experience (DMFT/DMFS = 0). DMFT and DMFS were used as continuous count variables. Test variables (outcome) were DMFS or changes in DMFS (ΔDMFS), where changes were defined as the differences in DMFS between age 15 and age 9 or the difference between the examination at 12 and 9 years of age. Test variables were used as continuous count variables.
Dairy Intake and Caries among Children and Adolescents
EYHS III, 9th grade
399
Sum EYHS II/III, 9th grade
843
Cohort of 9th grade children
340
Explanatory variables (exposure) were total dairy intake, total milk intake, total dairy calcium intake, total whey intake, and total casein intake. Dietary Intake At all three EYHS studies, information on habitual dietary intake was obtained by a 24-hour dietary recall method, which was supported by a qualitative food record and Food Frequency Questionnaire. Standard measures were visualized with glasses, plates and spoonfuls. The interviews were conducted from Monday to Friday and lasted 20–30 min. The 3rd-graders were assisted by a parent. Foods were converted into nutrients using the Dankost 3000 software program (Danish Catering Center, Copenhagen, Denmark), from which nutrient intake from single food items, whole meals or the entire diet was calculated using national food composition tables. The data were analyzed with a specific emphasis on nutritional components from dairy products. Dairy products were classified as milk, cheese, butter, fermented dairy products, and ice cream. Dietary records have been validated in a subsample against a 4-day weighed food intake, and a validation with doubly labeled water was performed on a smaller subsample [Riddoch et al., 2000]. Covariates The covariates were determined a priori from analyzing a directed acyclic graph [Merchant and Pitiphat, 2002] and included as possible confounders or mediators. Gender was used as a categorical binary variable. The total intake of added sugar was defined as all monosaccharaides and disaccharides in the diet. Intake of sugar was used as a continuous variable. Parental education was used as a categorical binary variable. Age was used as a continuous variable in the analysis. The socioeconomic status of the parents was based on information on the educational level obtained through questionnaires in
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
253
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Fig. 1. Flow diagram showing the sampled
444
Caries-free children
68.9 69.5 68.1 53.8 57.5
Caries-free girls 48.9
Caries-free boys 34.0 34.3 33.7
Fig. 2. Diagram showing the percentage of caries-free children in total by age and by age and gender (n = 749).
12 years
9 years
15 years
Table 1. Distribution of selected characteristics for children at age 9 (baseline data; n = 749) and for adolescents at age 15 (follow-up
data; n = 340)
mean Age, years 9.7 Caries (DMFS) 0.7 Caries (DMFT) 0.5 Total Energy intake, MJ/day 8.3 Add sugar intake, MJ/day 0.8 Total energy from dairy, MJ/day 1.5 Total energy from milk, MJ/day 1.0 Total calcium from dairy, g/day 0.7 Total whey from dairy, g/day 3.7 Total casein from dairy, g/day 17.9
boys (n = 137)
SD
min. max. mean
SD
min. max.
mean
SD
min. max.
mean
SD
min.
max.
0.4 1.4 1.0 2.2 0.6 0.9 0.7 0.4 2.1 10.2
8.4 0 0 2.4 0.0 0.0 0.0 0.0 0.0 0.0
0.4 1.5 1.0 2.6 0.7 0.9 0.7 0.4 2.2 10.5
8.7 0 0 2.2 0.0 0.0 0.0 0.0 0.0 0.0
15.7 3.7 2.6 8.0 0.6 2.0 0.8 1.1 4.8 19.7
0.4 4.7 3.1 3.0 0.7 2.5 0.6 1.5 12.7 52.0
14.7 0 0 2.4 0.0 0.0 0.0 0.0 0.0 0.0
15.7 3.5 2.7 10.7 1.0 2.2 1.1 1.2 3.9 15.9
0.3 4.4 3.3 4.7 1.1 2.7 0.9 1.6 13.5 55.3
14.9 0 0 4.0 0.0 0.0 0.0 0.0 0.0 0.0
16.6 22 14 39.1 8.2 19.3 4.1 11.5 91.5 373.7
10.8 9 7 17.1 3.1 4.9 3.4 2.0 11.0 56.7
9.7 0.8 0.6 9.3 1.0 1.7 1.2 0.8 4.1 19.7
one of the three EYHS rounds. The variables of the mother’s and father’s educational level were recoded into a binary parental education variable, with one category including a maximum of 10 years of elementary school or/and a vocational education and the second category including high school diploma or education above this level. These variables were used as categorical binary variables. Statistics The statistical analysis was performed in SPSS 20.0 for Windows and StataSE 12. All variables were tested by the Shapiro-Wilk test for normality. Differences between groups including categorical variables were assessed using the nonparametric WilcoxonMann-Whitney and χ2 tests [Altman, 1991]. A relative risk analysis was performed and the differences in relative risk were tested by means of a χ2 test. Data were analyzed using the generalized estimation equation (GEE) approach in order to account for the complex sampling design of the EYHS and to obtain population level estimates of the association between dairy intake and caries experience. Failure to account for the cluster randomized design of the EYHS study could result in misleading inference [Liang and Zenger, 1986; Lawless, 1987;]. Firstly, a basic raw model was analyzed including only the test and explanatory variables in the model. Secondly, the models were adjusted for gender, total added sugar and parental education.
254
Follow-up girls (n = 203)
boys (n = 332)
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
11.2 8 7 20.1 5.6 5.8 5.6 2.3 13.1 61.2
17.1 28 15 18.6 5.5 14.9 3.4 8.9 70.3 287.4
Offset variables were included in the model. The variable corresponded to the logarithm of the duration of exposure and adjusted for the differences in a systematic way. Only 16 persons had an ethnicity other than Caucasian. These 16 participants were excluded from all analysis and, therefore, ethnicity was not included in the models. The significance level was set at 5%.
Results
Of the cohort of 749 children, 68.9% were caries free (DMFS = 0) at the age of 9 years (baseline), and the mean number of caries experience as DMFS at age 9 was 0.7 (SD 1.5). Of the 749 children, 53.8% were caries free at the age of 12 years and 34.0% were caries free at age 15 (fig. 2). In the same cohort, the mean number of caries as DMFS at the age of 15 years was 3.6 (SD 3.2; n = 340). Characteristics of the study population are given in table 1. No significant differences between girls and boys regarding age or mean caries experience (DMFS/DMFT) were obLempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Baseline girls (n = 417)
72.8
65.8
69.3
58.9
30.7 33.2
41.1
27.2
Caries Caries free
Fig. 3. Caries experience (DMFS) in two Low dairy
and 15 years (n = 340) by dichotomous exposure variables (above and below the mean) with the corresponding p values
Total dairy intake 0.795 Total milk intake 0.777
p
95% CI
Low dairy
DMFS 15 years
p
95% CI
0.04 0.637–0.991
0.850
0.05 0.691–0.995
0.03 0.618–0.976
1.013
0.29 0.932–1.283
Exposure variables: dairy, milk, added sugar (expressed as kJ/day) and fizzy drinks (expressed as daily/not daily).
High dairy
15 years
9 years
Table 2. Relative risk estimates of having caries at age 9 (n = 749)
DMFS 9 years
High dairy
exposed to a sugary fizzy drink intake above the mean, but these results did not reach significance (data not shown). The results from the GEE, comparing individuals from the study population for their diet intakes, showed that dairy intake at 9 years of age was inversely associated with caries experience (DMFS) at 9 years of age (table 3). Also, a statistically significant inverse association was found with regard to milk intake, as well as between calcium, whey and casein intake at 9 years of age and caries experience at age 9. The results maintained significance after adjustment for outcome-specific covariates. This was not the case for dairy intake at 15 years of age and caries experience at age 15; these results were only significant in the raw models and not with regard to milk intake (table 3). Dairy intake in children was also inversely associated with subsequent changes in caries experience from the age of 9 to the age of 12, as well as with changes in caries experience from the age of 9 to the age of 15. In addition, a statistically significant inverse association was found with regard to milk intake, as well as between calcium, whey and casein intake at 9 years of age and changes in caries experience between the age of 9 and 12 as well as from 9 to 15 years. The results maintained significance after adjustment for outcome-specific covariates, with the exception of total dairy and milk as an exposure variable (table 3).
served. An inverse association between parental educational level and children’s mean caries experience (DMFS/ DMFT) was found (data not shown). Among the adolescents, approximately 17% of the total daily energy intake came from dairy products, which was 2 % lower than the total daily energy intake from dairy among the 9-year-old children examined in this study. At the ages of 9 and 15 years, the percentage of cariesfree children/adolescents was higher in the group with a high dairy intake compared with the group with a low dairy intake, as illustrated in figure 3; however these differences were only statistically significant at the age of 9 years. The results from the relative risk assessment of four exposure variables on caries experience, as measured by DMFS at age 9 and 15, showed that the risk of caries in the group with dairy and milk intake above the mean intake was approximately 20% less relative to the group at age 9 with an intake below the mean (p = 0.04 and p = 0.03 respectively; table 2). At the age of 15 years, this was only the case with regard to the total dairy intake (p = 0.005). In both age groups, the risk of having caries was greater in the group exposed to added sugar as well as in the group
This paper found an inverse association between dairy intake and caries experience (measured as DMFS) at baseline as well as caries incidence over 3 and 6 years. It was shown that a high intake of dairy foods, including milk, may contribute to prevent caries development during childhood and adolescence. Such inverse associations have previously been found only in cross-sectional stud-
Dairy Intake and Caries among Children and Adolescents
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Discussion
255
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
groups – caries and caries free. The percentage of children in each group by age and dairy intake. Low dairy was defined as dairy intake below the mean and high intake as dairy intake above the mean. Analyses were cross-sectional.
Lempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Longitudinal GEE analysis of association between intake of dairy, or components of dairy, at age 9 and caries increment over 3 and 6 years ΔDMFS. Raw includes only the exposure and outcome variable in the analysis. Adjusted includes exposure and outcome variable adjusted for gender, total intake of added sugar and parental education as a dichotomous variable. Offset variable was set as Δ age between dental examination.
0.003 0.002 0.03 0.006 0.003 –0.013 –0.024 –0.013 –0.008 –0.005 –0.044 –0.080 –0.072 –0.019 –0.017 0.97 0.95 0.96 0.99 0.99 –0.029 –0.052 –0.042 –0.013 –0.011 0.001 0.02 0.02 0.001 0.003 –0.016 –0.009 –0.010 –0.006 –0.005 –0.067 –0.099 –0.103 –0.025 –0.025 0.96 0.95 0.95 0.99 0.99 –0.041 –0.054 –0.056 –0.016 –0.015 0.59 0.01 0.81 0.02 0.74 0.004 –0.002 0.006 –0.000 0.001 –0.007 –0.014 –0.007 –0.003 –0.002 0.99 0.99 0.99 0.99 1.00 –0.002 –0.008 –0.001 –0.002 –0.000 0.005 0.01 0.01 0.001 0.003 –0.010 –0.012 –0.013 –0.006 –0.005 0.97 0.95 0.95 0.99 0.99 –0.033 –0.054 –0.056 –0.015 –0.014
–0.056 –0.096 –0.099 –0.024 –0.023
–0.041 0.96 –0.065 –0.017 0.001 –0.034 0.97 –0.049 –0.019 0.002 –0.002 0.01 –0.012 0.99 0.002 –0.007 –0.013 0.97 –0.035
–0.058
–0.019 –0.005 0.001 –0.017 0.001 0.08 –0.020 –0.002 0.03 –0.015 0.004 0.69 0.99 0.99 0.99 0.99 –0.012 –0.009 –0.011 –0.006 0.009 0.03 0.03 0.02 –0.004 –0.001 –0.002 –0.003 –0.027 –0.025 –0.030 –0.032 0.99 0.99 0.99 0.99 –0.015 –0.013 –0.016 –0.017 0.01 0.29 0.07 0.10 –0.001 –0.002 0.001 0.002 –0.007 –0.005 –0.019 –0.019 0.99 0.99 0.99 0.99 –0.004 –0.002 –0.009 –0.009 0.02 0.01 0.001 0.000 –0.003 –0.004 –0.010 –0.013 –0.024 –0.026 –0.039 –0.043 0.99 0.99 0.98 0.97 –0.013 –0.015 –0.024 –0.028
Dairy raw, kJ/day Dairy adjusted, kJ/day Milk raw, kJ/day Milk adjusted, kJ/day Dairy calcium raw, mg/day Dairy calcium adjusted, mg/day Whey raw, mg/day Whey adjusted, mg/day Casein raw, mg/day Casein adjusted, mg/day
95% CI 95% CI p β lnβ 95% CI 95% CI p β β
95% CI
95% CI
p
β lnβ lnβ
95% CI 95% CI p
lnβ
ΔDMFS 9–15 years ΔDMFS 9–12 years DMFS 15 years DMFS 9 years
Table 3. Cross-sectional GEE analysis of association between intake of dairy, or components of dairy, at age 9 and 15 and caries experience (DMFS) at age 9 and 15 years
256
ies, [Lingström et al., 2003; Caglar et al., 2005, 2008; Twetman and Stecksén-Blicks, 2008; Twetman and Keller, 2012]. However, in earlier cross-sectional data, the inverse association with DMFS at age 15 was found to be statistically significant only for dairy intake – and not for milk intake. Also, for analysis examining associations with dairy components such as calcium, whey and casein, the results of the present study were significant only in the raw models – but not in the adjusted model. Since the cohort in the present study was reduced from a total of 749 9-yearolds to 340 15-year-olds, the lack of significance may be the consequence of lack of power in the analyses including the 15-year-olds. This is supported by the fact that the direction and the magnitude of the association were similar for the adjusted and unadjusted models. The analysis from the longitudinal data showed a statistically significant inverse association between total dairy, as well as milk intake and other components of dairy, and the DMFS incidence over 3 years. However, for the DMFS incidence over a 6-year period, the results were only significant in the raw models. With regard to the adjusted models, the direction of the association was the same as in the raw models but did not reach significance, possibly for the same reasons related to lack of power, as explained above. An explanation for the inverse relation might be that, in general, dairy intake not only contains cariostatic components such as whey and casein [Kõll-Klais et al., 2005; Azarpazhooh and Limeback, 2008; Llena et al., 2009; Tanaka et al., 2010] but may also be a marker for a lifestyle that contains less cariogenic challenges [Bachman et al., 2006]. The fact that estimates were somewhat weakened in the adjusted models supports this notion, as some of the associations between a high dairy intake and caries development may be explained via healthy lifestyle behaviors accompanying the high dairy intake, which may also influence the caries development. In addition, an alleged beneficial effect may be related to the combined effects of probiotics, calcium, whey, and casein. However, in the present study we could not separate these, and future studies should address this further. There are most likely cultural differences in dairy intake, and different ethnic groups may display different intakes. In the present study, however, only 16 individuals had an ethnicity other than Caucasian, and hence subgroup analyses could not be performed. A few strengths and limitations should be mentioned. For instance, in the present study there were no differences in age or mean caries experience between responders and nonresponders. This indicates that the dropout in relation to the analysis performed for this paper was not
selective, which strengthens the results and generalizability of the present results. On the other hand, even if adjusted for age, gender, total intake of added sugar, and parental education, residual confounding for unmeasured factors such as dietary intake frequency, toothbrushing habits and fluoride exposure may still be present. In addition, the 24-hour recall method used to obtain dietary information is limited by the fact that a single day of intake may not be representative of usual daily intake. However, such bias would most likely have attenuated our results, suggesting even stronger associations than those reported. The Nordic nutrition recommendations were 7.7 MJ/ day for total energy intake and 0.8 MJ/day from dairy products at age 9; the recommendations at age 15 were 9.6 MJ/day for girls and 12.3 MJ/day for boys, with still 0.8 MJ/day from dairy products [NNR, 2004; Beck et al., 2010]. Hence, the energy intake of our study population at age 9 was above the intake recommendations, which indicates that general underreporting does not seem to be a major issue. However, at age 15 years, the mean energy intake was below the Nordic recommendations for both girls and boys, suggesting some underreporting, which might have attenuated our results. Despite this, we still found significant results, which support the validity of the present findings. Since total dairy intake in the cohorts at 9 and 15 years of age were above the Nordic recommendations for dairy intake, the potential general underreporting might not be due to underreporting of dairy intake but rather to underreporting of other dietary components, which were not investigated in this paper. Also, the intake of added sugar was low in the study population at age 9 as well as at age 15, and the variation in sugar intake may have been too low to find significant results. Caries scores were obtained at regular dental checkups, which indicate that no systematic selection was made, which will have limited the risk of selection bias. The criteria for the recording of caries is defined in detail by the National Board of Health, so the DMFS scores were obtained using the same guidelines limiting systematic bias [Helm, 1973; Hausen et al., 2001]. Furthermore, it has been found that even if the Danish Child Dental Health Services data registration is made by a large number of different dentists, and even if experienced dentists may have underestimated caries diagnoses, the Danish Child Dental Health Services data are considered to be of acceptable reliability [Hausen et al., 2001]. However, there may still be some limitations related to the measuring of caries. For example, DMF indices are inadequate for measuring new decay in teeth or surfaces which have already been restored; this
might have diluted the results obtained for this paper. Also, the fact that only aggregated data for the permanent dentition were used might have resulted in dilution of the examined relationship [Christensen et al., 2010a]. Hence, stronger, not weaker, associations would have been expected had less crude information been available, thus speaking in favor of real associations in the present study. Hence, given the evidence presented in this paper, the association of dental caries with dairy intake is biologically plausible [Azarpazhooh and Limeback, 2008; Llena et al., 2009; Twetman et al., 2009; Tanaka et al., 2010] and supported by the results obtained in the analysis presented.
Dairy Intake and Caries among Children and Adolescents
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Conclusions
A high dairy intake, as well as milk intake or intake of dairy components, may be a new and unrecognized predictor for low caries risk on DMFS level and may contribute to prevent caries development during childhood and adolescence. Hence, a potentially relevant public health initiative may be to recommend a sufficient dairy or milk intake to prevent caries development. Also, dental care practitioners may recommend high-risk children to raise their dairy intake. To our knowledge, this is the first study to report associations of dairy intake with caries development in children. Therefore, more studies are needed before clinical practices or public health recommendations are changed.
Acknowledgments This study was supported by the TrygFond Foundation and the University of Southern Denmark. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Role of authors: S.M. Lempert wrote the manuscript, conducted analyses and was responsible for merging with SCOR data and collection of dietary data. K. Raymond contributed to statistical analyses, reviewed the manuscript and was responsible for merging with SCOR data. L.B. Christensen contributed to the discussion, reviewed the manuscript and was responsible for merging with SCOR data. B.L. Heitmann contributed to the discussion, reviewed the manuscript and was responsible for collection of dietary data. K. Froberg was responsible for physical and questionnaire data collection and reviewed the manuscript.
Disclosure Statement
257
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
The authors declare that there are no conflicts of interest relating to the data and opinions presented in this paper.
References
258
Hausen H, Karkkainen S, Seppa L: Caries data collected from public health records compared with data based on examinations by trained examiners. Caries Res 2001; 35: 360– 365. Helm S: Recording system for the Danish Child Dental Health Services. Community Dent Oral Epidemiol 1973;1:3–8. Johansson I, Lif Holgerson P: Milk and oral health. Nestle Nutr Workshop Ser Pediatr Program 2011;67:55–66. Kõll-Klais P, Mändar R, Leibur E, Marcotte H, Hammarström L, et al: Oral lactobacilli in chronic periodontitis and periodontal health: species composition and antimicrobial activity. Oral Microbiol Immunol 2005; 20: 354– 361. Lawless J: Negative binomial and mixed Poisson regression. Can J Stat 1987;15:209–225. Lexner MO, Blomqvist S, Dahlén G, Twetman S: Microbiological profiles in saliva and supragingival plaque from caries-active adolescents before and after a short-term daily intake of milk supplemented with probiotic bacteria – a pilot study. Oral Health Prev Dent 2010; 8: 383–388. Liang KY, Zenger SL: Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13–22. Lingström P, Holm A, Mejàre I, Twetman S, Söder B, et al: Dietary factors in the prevention of dental caries: a systematic review. Acta Odontol Scand 2003;61:331–340. Llena C, Forner L, Baca P: Anticariogenicity of casein phosphopeptide-amorphous calcium phosphate: a review of the literature. J Contemp Dent Pract 2009;10:1–9. Marshall TA, Eichenberger-Gilmore JM, Broffitt BA, Warren JJ, Levy SM: Dental caries and childhood obesity: roles of diet and socioeconomic status. Community Dent Oral Epidemiol 2007;35:449–458. Merchant AT, Pitiphat W: Directed acyclic graphs (DAGs): an aid to assess confounding in dental research. Community Dent Oral Epidemiol 2002;30:399–404. Mobley CC: Nutrition and dental caries. Dent Clin North Am 2003;47:319–336. Nicolau B, Marcenes W, Bartley M, Sheiham A: A life course approach to assessing causes of dental caries experience: the relationship between biological, behavioural, socio-economic and psychological conditions and caries in adolescents. Caries Res 2003;37:319–326.
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Nielsen BM, Bjornsbo KS, Tetens I, Heitmann BL: Dietary glycaemic index and glycaemic load in Danish children in relation to body fatness. Br J Nutr 2005;94:992–997. NNR: Nordic Nutrition Recommendations, ed 4. 2004. Petersen PE: The World Oral Health Report 2003: continuous improvement of oral health in the 21st century – the approach of the WHO Global Oral Health Programme. Community Dent Oral Epidemiol 2003;31(suppl 1):3–23. Riddoch C, Edwards D, Page A, Froberg K, Anderssen SA, et al: European Youth Heart Study background. World Wide Web 2000, pp 1–23. http://www.enshe.lu/committees/ children/background.pdf (accessed May 4, 2003). Riddoch C, Edwards D, Page A, Froberg K, Anderssen S, et al: The European Youth Heart Study – cardiovascular disease risk factors in children: rationale, aims, study design, and validation of methods. J Phys Act Health 2005;2:115–129. Séverin S, Wenshui X: Milk biologically active components as nutraceuticals: review. Crit Rev Food Sci Nutr 2005;45:645–656. Sheiham A: Dietary effects on dental diseases. Public Health Nutr 2001;4:569–591. Stecksén-Blicks C, Sjöström I, Twetman S: Effect of long-term consumption of milk supplemented with probiotic lactobacilli and fluoride on dental caries and general health in preschool children: a cluster-randomized study. Caries Res 2009;43:374–381. Tanaka K, Miyake Y, Sasaki S: Intake of dairy products and the prevalence of dental caries in young children. J Dent 2010;38:579–583. Twetman L, Larsen U, Fiehn N, Stecksen-Blicks C, Twetman S: Coaggregation between probiotic bacteria and caries-associated strains: an in vitro study. Acta Odontol Scand 2009; 67: 284–288. Twetman S, Keller MK: Probiotics for caries prevention and control. Adv Dent Res 2012; 24: 98–102. Twetman S, Stecksén-Blicks C: Probiotics and oral health effects in children. Int J Paediatr Dent 2008;18:3–10.
Lempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Altman DG: Practical Statistics for Medical Research. London, Capman & Hall 1991, vol 247–48, pp 316–318, 337. Azarpazhooh A, Limeback H: Clinical efficacy of casein derivatives: a systematic review of the literature. J Am Dent Assoc 2008; 139: 915– 924, quiz 994–925. Bachman CM, Baranowski T, Nicklas TA: Is there an association between sweetened beverages and adiposity? Nutr Rev 2006;64:153–174. Beck AM, Hoppe C, Ygil KH, Andersen NL, Pedersen AN: Vidensgrundlag for rådgivning om indtag af mælk, mælkeprodukter og ost i Danmark. Technical University of Denmark, DTU, 2010. Caglar E, Kuscu OO, Selvi Kuvvetli S, Kavaloglu Cildir S, Sandalli N, et al: Short-term effect of ice-cream containing Bifidobacterium lactis Bb-12 on the number of salivary mutans streptococci and lactobacilli. Acta Odontol Scand 2008;66:154–158. Caglar E, Sandalli N, Twetman S, Kavaloglu S, Ergeneli S, et al: Effect of yogurt with Bifidobacterium DN-173 010 on salivary mutans streptococci and lactobacilli in young adults. Acta Odontol Scand 2005;63:317–320. Cameron FL, Weaver LT, Wright CM, Welbury RR: Dietary and social characteristics of children with severe tooth decay. Scott Med J 2006;51:26–29. Christensen LB, Petersen PE, Hede B: Oral health in children in Denmark under different public dental health care schemes. Community Dent Health 2010a;27:94–101. Christensen LB, Twetman S, Sundby A: Oral health in children and adolescents with different socio-cultural and socio-economic backgrounds. Acta Odontol Scand 2010b;68: 34– 42. Cinar AB, Christensen LB, Hede B: Clustering of obesity and dental caries with lifestyle factors among Danish adolescents. Oral Health Prev Dent 2011;9:123–130. Glavina D, Gorseta K, Skrinjaric I, Vranic DN, Mehulic K, et al: Effect of LGG yoghurt on Streptococcus mutans and Lactobacillus spp. salivary counts in children. Coll Antropol 2012;36:129–132.
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Association between Dairy Intake and Caries among Children and Adolescents. Results from the Danish EYHS Follow-Up Study Susanne M. Lempert a, c, e Lisa B. Christensen b Karsten Froberg d Kyle Raymond e Berit L. Heitmann a, e, f
a
National Institute of Public Health, University of Southern Denmark, and b Institute of Odontology and c School of Oral Health Care, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, d Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, and e Institute of Preventive Medicine, Research Unit for Dietary Studies, Bispebjerg and Frederiksberg Hospital, Frederiksberg, Denmark; f Boden Institute of Obesity, Nutrition, Exercise and Eating Disorders, University of Sydney, Sydney, N.S.W., Australia
Abstract The aim of this paper was to investigate associations between the intake of dairy products and the development in caries (DMFS, decayed, missing and filled surfaces) among children/adolescents over a period of 3 and 6 years, and to investigate whether dairy intake protects against caries incidence. A total of 68.9% of the children were caries free at the age of 9 compared with 34.0% of the adolescents at the age of 15 (measured as DMFS = 0). A larger percentage of children/adolescents with a dairy intake above the mean were caries free compared with the group of children/adolescents with an intake below the mean (72.8 vs. 65.8% at age 9 and 41.1 vs. 30.7% at age 15). The results from the generalized estimation equation showed that dairy and milk intake, as well as intakes of components of dairy such as dairy calcium, whey and casein, was generally inversely associated with childhood/adolescent caries experience (measured as DMFS). With regard to caries incidence, the same inverse association was found for incidence over a period of 3 years and for incidence over 6 years, but the results were only sta-
© 2015 S. Karger AG, Basel 0008–6568/15/0493–0251$39.50/0 E-Mail [email protected] www.karger.com/cre
tistically significant for the 3-year incidence and for the unadjusted models of the 6-year incidence. This study found that previous dairy intake, as well as milk intake or intake of dairy components, may be a predictor of future risk of caries measured by the DMFS count level. This relationship was inverse, meaning that a high intake of dairy products was associated with less future caries development. However, more studies on larger cohorts are needed to confirm these findings. © 2015 S. Karger AG, Basel
Oral diseases are multifactorial diseases. With regard to the prevention of oral diseases such as caries and periodontal diseases, optimal oral hygiene is identified as the main preventive factor [Sheiham, 2001; Mobley, 2003; Petersen, 2003]. In Denmark, there are practical and good opportunities to deliver preventive dental care services at an individual level in public and private dental practice through the dental care system, which is free of charge for all up to the age of 17 years [Christensen et al., 2010a]. Other factors influencing oral disease are diet and socioeconomic factors, which have been found to play an important role in caries incidence in children and adolescents [Nicolau et al., 2003; Cameron et al., 2006; Marshall et al., Susanne M. Lempert National Institute of Public Health, University of Southern Denmark Øster Farimagsgade 5A DK–1353 Copenhagen (Denmark) E-Mail susanne.lempert @ sund.ku.dk
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Key Words Dairy · Milk · Calcium · Whey · Casein · Childhood · Caries · DMFS · European Youth Heart Study
252
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
The following hypotheses were examined: – A high intake of dairy products generally has a beneficial effect on dental caries in children and adolescents, and children with the highest intake of dairy have the lowest caries prevalence and incidence. – A high intake of milk has a beneficial effect on dental caries in children and adolescents, and children with the highest intake of milk have the lowest caries incidence. – A high intake of dairy calcium, whey or casein has a beneficial effect in children and adolescents, and children with the highest intake of dairy calcium, dairy whey or dairy casein have the lowest caries prevalence and incidence. Materials and Methods Study Design The European Youth Heart Study (EYHS) is a multicenter, crosscultural and longitudinal study of associations between lifestyle and cardiovascular disease among children and adolescents. The present study used data from the Danish part of the EYHS and data from the National Board of Health’s Recording System for the Danish Child Dental Health Services. The study was school based, and schools in the Danish municipality of Odense were stratified according to school type, location (urban, suburban and rural) and socioeconomic status of the uptake area [Riddoch et al., 2005]. Each school was allocated a weighted equivalent to the number of children enrolled in the school who were eligible for selection into the study; 3 replacement schools were sampled to substitute schools refusing to participate and to be used in case of low response rates. Children within the age groups of 8–10 and 14–16 years were allocated code numbers and randomly selected using random number tables [Riddoch et al., 2005]. At each study location, a sampling frame of schools using official lists was compiled. A total of 35 schools were sampled, of which 25 (71%) agreed to participate. Of these, 3 schools where lost to follow-up – one rural and 1 urban from a middle-class area and 1 urban school from a low-income area. All 3 schools gave interference with the educational process as the reason for dropout. The study was approved by the Danish Data Protection Agency and the local ethics committee and was performed according to the Declaration of Helsinki. All children gave verbal consent, and their parents gave written consent [Riddoch et al., 2005]. Study Population In the first EYHS study (EYHS I), a total of 586 3rd-grade students were sampled (310 girls and 276 boys). In the second EYHS study (EYHS II), a total of 448 3rd-grade students were sampled (257 girls and 191 boys). For this paper, both of these 3rd-grade populations were merged and gave a study population of 1,034 children from the 3rd grade (fig. 1). In the second EYHS study (EYHS II), a total of 444 9th-grade students (of the 586 3rd-graders) were sampled (251 girls and 193 boys). In the third EYHS study (EYHS III), a total of 399 students from the 9th grade (of the original 448 3rd-graders) were sampled (220 girls and 179 boys). For this paper, both of these 9th-grade populations were merged and gave a study population of 843 children from the 9th grade
Lempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
2007; Christensen et al., 2010b; Cinar et al., 2011]. However, previous studies examining dietary influences on the incidence of dental disease have focused mostly on caries development in relation to dietary sugars and fluorides [Sheiham, 2001; Mobley, 2003] or on deficiency disorders such as malnutrition and vitamin deficiencies. Two recent review studies pointed out that milk contains a variety of biologically active substances which may have particular beneficial effects on dental health and mineral malabsorption as well as an antimicrobial effect [Séverin and Wenshui, 2005, Johansson and Lif Holgerson, 2011]. Indeed, a few recent studies have suggested that dairy products may inhibit the progression and development of caries [Lingström et al., 2003; Caglar et al., 2005, 2008; Twetman and Stecksén-Blicks, 2008; Twetman and Keller, 2012]. This has, however, generally been examined in cross-sectional studies, and not yet among children, which is surprising since children and adolescents often belong to the groups of the population with the highest intake of dairy products and milk. Hence, studies are needed that examine the relationship between the intake of dairy products and oral health in longitudinal study designs among children. The biological rationale for a link between the consumption of dairy products and an oral health-promoting effect is based on several ideas. Firstly, lactobacilli from dairy products may contribute to maintain the microecological balance, including the balance of microflora in the oral cavity [KõllKlais et al., 2005; Stecksén-Blicks et al., 2009; Twetman et al., 2009; Lexner et al., 2010; Glavina et al., 2012]. Secondly, a caries inhibiting effect may be seen to be related to the intake of calcium, whey or casein from dairy products, since whey and casein may prevent bacterial adherence to the tooth surface, and calcium is important for the remineralization process in the oral cavity [Azarpazhooh and Limeback, 2008; Llena et al., 2009; Twetman et al., 2009; Tanaka et al., 2010]. Therefore, it seems relevant to examine whether a high dairy intake, including a high intake of milk, may contribute to prevent the development of caries and to assess whether a high intake of calcium, whey or casein from dairy products may have a preventive effect on dental caries. Furthermore, even if an effect of dairy may have limited influence on caries development at the individual level, influencing oral health through the intake of dairy products might have profound effects at the population level and thereby contribute to an increase in public health. The present project aims to examine the consumption of dairy products of children and young people in relation to caries experience as well as the subsequent change in caries experience (DMFS, decayed, missing and filled surfaces), with a follow-up period of 3 and 6 years.
EYHS I, 3rd grade
586
EYHS II, 3rd grade
448
Sum EYHS I/II, 3rd grade
1,034
Cohort of 3rd grade children
749
EYHS II, 9th grade
number of children. The cohort of 749 children from for whom we had background, dairy and dental information at baseline and the cohort of 343 children for whom we had background, dairy and dental information at baseline and follow-up.
(fig. 1) Data used for this paper contained details of 749 children (417 girls and 332 boys) from the 3rd grade on whom we had complete information on dietary intake and dental measures at baseline (referred to as 9-year-olds) and 340 adolescents (203 girls and 137 boys) from the 9th grade with complete follow-up data on dietary intake and dental measures (referred to as 15-year-olds; fig. 1). The study design, sampling and methods have been described in more detail elsewhere [Riddoch et al., 2000; Nielsen et al., 2005]. EYHS I was conducted in 1997–1998, EYHS II in 2003– 2004 and EYHS III in 2009–2010. Measurements Caries Status Data on caries experience was retrieved from the Danish Child Dental Health Services, which contains a range of information on oral health status, including caries status [Helm, 1973; Hausen et al., 2001]. Caries data used for this investigation was collected in 1997 (EYHS I), 2003 (EYHS II) and 2009 (EYHS III). In addition, data on caries experience were collected from all participants the year they turned 12 years. The data were collected by dentists in the municipality of Odense. The children were recalled for visits regularly, and standardized dental examinations were performed for registration of oral health status; these examinations included X-rays when the dentists found it was indicated. Caries was recorded at cavity level. Caries experience was calculated as the sum of DMFT (decayed, missing and filled teeth) or DMFS in the permanent dentition. Caries-free children/adolescents were defined as children/adolescents with no prior caries experience (DMFT/DMFS = 0). DMFT and DMFS were used as continuous count variables. Test variables (outcome) were DMFS or changes in DMFS (ΔDMFS), where changes were defined as the differences in DMFS between age 15 and age 9 or the difference between the examination at 12 and 9 years of age. Test variables were used as continuous count variables.
Dairy Intake and Caries among Children and Adolescents
EYHS III, 9th grade
399
Sum EYHS II/III, 9th grade
843
Cohort of 9th grade children
340
Explanatory variables (exposure) were total dairy intake, total milk intake, total dairy calcium intake, total whey intake, and total casein intake. Dietary Intake At all three EYHS studies, information on habitual dietary intake was obtained by a 24-hour dietary recall method, which was supported by a qualitative food record and Food Frequency Questionnaire. Standard measures were visualized with glasses, plates and spoonfuls. The interviews were conducted from Monday to Friday and lasted 20–30 min. The 3rd-graders were assisted by a parent. Foods were converted into nutrients using the Dankost 3000 software program (Danish Catering Center, Copenhagen, Denmark), from which nutrient intake from single food items, whole meals or the entire diet was calculated using national food composition tables. The data were analyzed with a specific emphasis on nutritional components from dairy products. Dairy products were classified as milk, cheese, butter, fermented dairy products, and ice cream. Dietary records have been validated in a subsample against a 4-day weighed food intake, and a validation with doubly labeled water was performed on a smaller subsample [Riddoch et al., 2000]. Covariates The covariates were determined a priori from analyzing a directed acyclic graph [Merchant and Pitiphat, 2002] and included as possible confounders or mediators. Gender was used as a categorical binary variable. The total intake of added sugar was defined as all monosaccharaides and disaccharides in the diet. Intake of sugar was used as a continuous variable. Parental education was used as a categorical binary variable. Age was used as a continuous variable in the analysis. The socioeconomic status of the parents was based on information on the educational level obtained through questionnaires in
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
253
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Fig. 1. Flow diagram showing the sampled
444
Caries-free children
68.9 69.5 68.1 53.8 57.5
Caries-free girls 48.9
Caries-free boys 34.0 34.3 33.7
Fig. 2. Diagram showing the percentage of caries-free children in total by age and by age and gender (n = 749).
12 years
9 years
15 years
Table 1. Distribution of selected characteristics for children at age 9 (baseline data; n = 749) and for adolescents at age 15 (follow-up
data; n = 340)
mean Age, years 9.7 Caries (DMFS) 0.7 Caries (DMFT) 0.5 Total Energy intake, MJ/day 8.3 Add sugar intake, MJ/day 0.8 Total energy from dairy, MJ/day 1.5 Total energy from milk, MJ/day 1.0 Total calcium from dairy, g/day 0.7 Total whey from dairy, g/day 3.7 Total casein from dairy, g/day 17.9
boys (n = 137)
SD
min. max. mean
SD
min. max.
mean
SD
min. max.
mean
SD
min.
max.
0.4 1.4 1.0 2.2 0.6 0.9 0.7 0.4 2.1 10.2
8.4 0 0 2.4 0.0 0.0 0.0 0.0 0.0 0.0
0.4 1.5 1.0 2.6 0.7 0.9 0.7 0.4 2.2 10.5
8.7 0 0 2.2 0.0 0.0 0.0 0.0 0.0 0.0
15.7 3.7 2.6 8.0 0.6 2.0 0.8 1.1 4.8 19.7
0.4 4.7 3.1 3.0 0.7 2.5 0.6 1.5 12.7 52.0
14.7 0 0 2.4 0.0 0.0 0.0 0.0 0.0 0.0
15.7 3.5 2.7 10.7 1.0 2.2 1.1 1.2 3.9 15.9
0.3 4.4 3.3 4.7 1.1 2.7 0.9 1.6 13.5 55.3
14.9 0 0 4.0 0.0 0.0 0.0 0.0 0.0 0.0
16.6 22 14 39.1 8.2 19.3 4.1 11.5 91.5 373.7
10.8 9 7 17.1 3.1 4.9 3.4 2.0 11.0 56.7
9.7 0.8 0.6 9.3 1.0 1.7 1.2 0.8 4.1 19.7
one of the three EYHS rounds. The variables of the mother’s and father’s educational level were recoded into a binary parental education variable, with one category including a maximum of 10 years of elementary school or/and a vocational education and the second category including high school diploma or education above this level. These variables were used as categorical binary variables. Statistics The statistical analysis was performed in SPSS 20.0 for Windows and StataSE 12. All variables were tested by the Shapiro-Wilk test for normality. Differences between groups including categorical variables were assessed using the nonparametric WilcoxonMann-Whitney and χ2 tests [Altman, 1991]. A relative risk analysis was performed and the differences in relative risk were tested by means of a χ2 test. Data were analyzed using the generalized estimation equation (GEE) approach in order to account for the complex sampling design of the EYHS and to obtain population level estimates of the association between dairy intake and caries experience. Failure to account for the cluster randomized design of the EYHS study could result in misleading inference [Liang and Zenger, 1986; Lawless, 1987;]. Firstly, a basic raw model was analyzed including only the test and explanatory variables in the model. Secondly, the models were adjusted for gender, total added sugar and parental education.
254
Follow-up girls (n = 203)
boys (n = 332)
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
11.2 8 7 20.1 5.6 5.8 5.6 2.3 13.1 61.2
17.1 28 15 18.6 5.5 14.9 3.4 8.9 70.3 287.4
Offset variables were included in the model. The variable corresponded to the logarithm of the duration of exposure and adjusted for the differences in a systematic way. Only 16 persons had an ethnicity other than Caucasian. These 16 participants were excluded from all analysis and, therefore, ethnicity was not included in the models. The significance level was set at 5%.
Results
Of the cohort of 749 children, 68.9% were caries free (DMFS = 0) at the age of 9 years (baseline), and the mean number of caries experience as DMFS at age 9 was 0.7 (SD 1.5). Of the 749 children, 53.8% were caries free at the age of 12 years and 34.0% were caries free at age 15 (fig. 2). In the same cohort, the mean number of caries as DMFS at the age of 15 years was 3.6 (SD 3.2; n = 340). Characteristics of the study population are given in table 1. No significant differences between girls and boys regarding age or mean caries experience (DMFS/DMFT) were obLempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Baseline girls (n = 417)
72.8
65.8
69.3
58.9
30.7 33.2
41.1
27.2
Caries Caries free
Fig. 3. Caries experience (DMFS) in two Low dairy
and 15 years (n = 340) by dichotomous exposure variables (above and below the mean) with the corresponding p values
Total dairy intake 0.795 Total milk intake 0.777
p
95% CI
Low dairy
DMFS 15 years
p
95% CI
0.04 0.637–0.991
0.850
0.05 0.691–0.995
0.03 0.618–0.976
1.013
0.29 0.932–1.283
Exposure variables: dairy, milk, added sugar (expressed as kJ/day) and fizzy drinks (expressed as daily/not daily).
High dairy
15 years
9 years
Table 2. Relative risk estimates of having caries at age 9 (n = 749)
DMFS 9 years
High dairy
exposed to a sugary fizzy drink intake above the mean, but these results did not reach significance (data not shown). The results from the GEE, comparing individuals from the study population for their diet intakes, showed that dairy intake at 9 years of age was inversely associated with caries experience (DMFS) at 9 years of age (table 3). Also, a statistically significant inverse association was found with regard to milk intake, as well as between calcium, whey and casein intake at 9 years of age and caries experience at age 9. The results maintained significance after adjustment for outcome-specific covariates. This was not the case for dairy intake at 15 years of age and caries experience at age 15; these results were only significant in the raw models and not with regard to milk intake (table 3). Dairy intake in children was also inversely associated with subsequent changes in caries experience from the age of 9 to the age of 12, as well as with changes in caries experience from the age of 9 to the age of 15. In addition, a statistically significant inverse association was found with regard to milk intake, as well as between calcium, whey and casein intake at 9 years of age and changes in caries experience between the age of 9 and 12 as well as from 9 to 15 years. The results maintained significance after adjustment for outcome-specific covariates, with the exception of total dairy and milk as an exposure variable (table 3).
served. An inverse association between parental educational level and children’s mean caries experience (DMFS/ DMFT) was found (data not shown). Among the adolescents, approximately 17% of the total daily energy intake came from dairy products, which was 2 % lower than the total daily energy intake from dairy among the 9-year-old children examined in this study. At the ages of 9 and 15 years, the percentage of cariesfree children/adolescents was higher in the group with a high dairy intake compared with the group with a low dairy intake, as illustrated in figure 3; however these differences were only statistically significant at the age of 9 years. The results from the relative risk assessment of four exposure variables on caries experience, as measured by DMFS at age 9 and 15, showed that the risk of caries in the group with dairy and milk intake above the mean intake was approximately 20% less relative to the group at age 9 with an intake below the mean (p = 0.04 and p = 0.03 respectively; table 2). At the age of 15 years, this was only the case with regard to the total dairy intake (p = 0.005). In both age groups, the risk of having caries was greater in the group exposed to added sugar as well as in the group
This paper found an inverse association between dairy intake and caries experience (measured as DMFS) at baseline as well as caries incidence over 3 and 6 years. It was shown that a high intake of dairy foods, including milk, may contribute to prevent caries development during childhood and adolescence. Such inverse associations have previously been found only in cross-sectional stud-
Dairy Intake and Caries among Children and Adolescents
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Discussion
255
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
groups – caries and caries free. The percentage of children in each group by age and dairy intake. Low dairy was defined as dairy intake below the mean and high intake as dairy intake above the mean. Analyses were cross-sectional.
Lempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Longitudinal GEE analysis of association between intake of dairy, or components of dairy, at age 9 and caries increment over 3 and 6 years ΔDMFS. Raw includes only the exposure and outcome variable in the analysis. Adjusted includes exposure and outcome variable adjusted for gender, total intake of added sugar and parental education as a dichotomous variable. Offset variable was set as Δ age between dental examination.
0.003 0.002 0.03 0.006 0.003 –0.013 –0.024 –0.013 –0.008 –0.005 –0.044 –0.080 –0.072 –0.019 –0.017 0.97 0.95 0.96 0.99 0.99 –0.029 –0.052 –0.042 –0.013 –0.011 0.001 0.02 0.02 0.001 0.003 –0.016 –0.009 –0.010 –0.006 –0.005 –0.067 –0.099 –0.103 –0.025 –0.025 0.96 0.95 0.95 0.99 0.99 –0.041 –0.054 –0.056 –0.016 –0.015 0.59 0.01 0.81 0.02 0.74 0.004 –0.002 0.006 –0.000 0.001 –0.007 –0.014 –0.007 –0.003 –0.002 0.99 0.99 0.99 0.99 1.00 –0.002 –0.008 –0.001 –0.002 –0.000 0.005 0.01 0.01 0.001 0.003 –0.010 –0.012 –0.013 –0.006 –0.005 0.97 0.95 0.95 0.99 0.99 –0.033 –0.054 –0.056 –0.015 –0.014
–0.056 –0.096 –0.099 –0.024 –0.023
–0.041 0.96 –0.065 –0.017 0.001 –0.034 0.97 –0.049 –0.019 0.002 –0.002 0.01 –0.012 0.99 0.002 –0.007 –0.013 0.97 –0.035
–0.058
–0.019 –0.005 0.001 –0.017 0.001 0.08 –0.020 –0.002 0.03 –0.015 0.004 0.69 0.99 0.99 0.99 0.99 –0.012 –0.009 –0.011 –0.006 0.009 0.03 0.03 0.02 –0.004 –0.001 –0.002 –0.003 –0.027 –0.025 –0.030 –0.032 0.99 0.99 0.99 0.99 –0.015 –0.013 –0.016 –0.017 0.01 0.29 0.07 0.10 –0.001 –0.002 0.001 0.002 –0.007 –0.005 –0.019 –0.019 0.99 0.99 0.99 0.99 –0.004 –0.002 –0.009 –0.009 0.02 0.01 0.001 0.000 –0.003 –0.004 –0.010 –0.013 –0.024 –0.026 –0.039 –0.043 0.99 0.99 0.98 0.97 –0.013 –0.015 –0.024 –0.028
Dairy raw, kJ/day Dairy adjusted, kJ/day Milk raw, kJ/day Milk adjusted, kJ/day Dairy calcium raw, mg/day Dairy calcium adjusted, mg/day Whey raw, mg/day Whey adjusted, mg/day Casein raw, mg/day Casein adjusted, mg/day
95% CI 95% CI p β lnβ 95% CI 95% CI p β β
95% CI
95% CI
p
β lnβ lnβ
95% CI 95% CI p
lnβ
ΔDMFS 9–15 years ΔDMFS 9–12 years DMFS 15 years DMFS 9 years
Table 3. Cross-sectional GEE analysis of association between intake of dairy, or components of dairy, at age 9 and 15 and caries experience (DMFS) at age 9 and 15 years
256
ies, [Lingström et al., 2003; Caglar et al., 2005, 2008; Twetman and Stecksén-Blicks, 2008; Twetman and Keller, 2012]. However, in earlier cross-sectional data, the inverse association with DMFS at age 15 was found to be statistically significant only for dairy intake – and not for milk intake. Also, for analysis examining associations with dairy components such as calcium, whey and casein, the results of the present study were significant only in the raw models – but not in the adjusted model. Since the cohort in the present study was reduced from a total of 749 9-yearolds to 340 15-year-olds, the lack of significance may be the consequence of lack of power in the analyses including the 15-year-olds. This is supported by the fact that the direction and the magnitude of the association were similar for the adjusted and unadjusted models. The analysis from the longitudinal data showed a statistically significant inverse association between total dairy, as well as milk intake and other components of dairy, and the DMFS incidence over 3 years. However, for the DMFS incidence over a 6-year period, the results were only significant in the raw models. With regard to the adjusted models, the direction of the association was the same as in the raw models but did not reach significance, possibly for the same reasons related to lack of power, as explained above. An explanation for the inverse relation might be that, in general, dairy intake not only contains cariostatic components such as whey and casein [Kõll-Klais et al., 2005; Azarpazhooh and Limeback, 2008; Llena et al., 2009; Tanaka et al., 2010] but may also be a marker for a lifestyle that contains less cariogenic challenges [Bachman et al., 2006]. The fact that estimates were somewhat weakened in the adjusted models supports this notion, as some of the associations between a high dairy intake and caries development may be explained via healthy lifestyle behaviors accompanying the high dairy intake, which may also influence the caries development. In addition, an alleged beneficial effect may be related to the combined effects of probiotics, calcium, whey, and casein. However, in the present study we could not separate these, and future studies should address this further. There are most likely cultural differences in dairy intake, and different ethnic groups may display different intakes. In the present study, however, only 16 individuals had an ethnicity other than Caucasian, and hence subgroup analyses could not be performed. A few strengths and limitations should be mentioned. For instance, in the present study there were no differences in age or mean caries experience between responders and nonresponders. This indicates that the dropout in relation to the analysis performed for this paper was not
selective, which strengthens the results and generalizability of the present results. On the other hand, even if adjusted for age, gender, total intake of added sugar, and parental education, residual confounding for unmeasured factors such as dietary intake frequency, toothbrushing habits and fluoride exposure may still be present. In addition, the 24-hour recall method used to obtain dietary information is limited by the fact that a single day of intake may not be representative of usual daily intake. However, such bias would most likely have attenuated our results, suggesting even stronger associations than those reported. The Nordic nutrition recommendations were 7.7 MJ/ day for total energy intake and 0.8 MJ/day from dairy products at age 9; the recommendations at age 15 were 9.6 MJ/day for girls and 12.3 MJ/day for boys, with still 0.8 MJ/day from dairy products [NNR, 2004; Beck et al., 2010]. Hence, the energy intake of our study population at age 9 was above the intake recommendations, which indicates that general underreporting does not seem to be a major issue. However, at age 15 years, the mean energy intake was below the Nordic recommendations for both girls and boys, suggesting some underreporting, which might have attenuated our results. Despite this, we still found significant results, which support the validity of the present findings. Since total dairy intake in the cohorts at 9 and 15 years of age were above the Nordic recommendations for dairy intake, the potential general underreporting might not be due to underreporting of dairy intake but rather to underreporting of other dietary components, which were not investigated in this paper. Also, the intake of added sugar was low in the study population at age 9 as well as at age 15, and the variation in sugar intake may have been too low to find significant results. Caries scores were obtained at regular dental checkups, which indicate that no systematic selection was made, which will have limited the risk of selection bias. The criteria for the recording of caries is defined in detail by the National Board of Health, so the DMFS scores were obtained using the same guidelines limiting systematic bias [Helm, 1973; Hausen et al., 2001]. Furthermore, it has been found that even if the Danish Child Dental Health Services data registration is made by a large number of different dentists, and even if experienced dentists may have underestimated caries diagnoses, the Danish Child Dental Health Services data are considered to be of acceptable reliability [Hausen et al., 2001]. However, there may still be some limitations related to the measuring of caries. For example, DMF indices are inadequate for measuring new decay in teeth or surfaces which have already been restored; this
might have diluted the results obtained for this paper. Also, the fact that only aggregated data for the permanent dentition were used might have resulted in dilution of the examined relationship [Christensen et al., 2010a]. Hence, stronger, not weaker, associations would have been expected had less crude information been available, thus speaking in favor of real associations in the present study. Hence, given the evidence presented in this paper, the association of dental caries with dairy intake is biologically plausible [Azarpazhooh and Limeback, 2008; Llena et al., 2009; Twetman et al., 2009; Tanaka et al., 2010] and supported by the results obtained in the analysis presented.
Dairy Intake and Caries among Children and Adolescents
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Conclusions
A high dairy intake, as well as milk intake or intake of dairy components, may be a new and unrecognized predictor for low caries risk on DMFS level and may contribute to prevent caries development during childhood and adolescence. Hence, a potentially relevant public health initiative may be to recommend a sufficient dairy or milk intake to prevent caries development. Also, dental care practitioners may recommend high-risk children to raise their dairy intake. To our knowledge, this is the first study to report associations of dairy intake with caries development in children. Therefore, more studies are needed before clinical practices or public health recommendations are changed.
Acknowledgments This study was supported by the TrygFond Foundation and the University of Southern Denmark. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Role of authors: S.M. Lempert wrote the manuscript, conducted analyses and was responsible for merging with SCOR data and collection of dietary data. K. Raymond contributed to statistical analyses, reviewed the manuscript and was responsible for merging with SCOR data. L.B. Christensen contributed to the discussion, reviewed the manuscript and was responsible for merging with SCOR data. B.L. Heitmann contributed to the discussion, reviewed the manuscript and was responsible for collection of dietary data. K. Froberg was responsible for physical and questionnaire data collection and reviewed the manuscript.
Disclosure Statement
257
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
The authors declare that there are no conflicts of interest relating to the data and opinions presented in this paper.
References
258
Hausen H, Karkkainen S, Seppa L: Caries data collected from public health records compared with data based on examinations by trained examiners. Caries Res 2001; 35: 360– 365. Helm S: Recording system for the Danish Child Dental Health Services. Community Dent Oral Epidemiol 1973;1:3–8. Johansson I, Lif Holgerson P: Milk and oral health. Nestle Nutr Workshop Ser Pediatr Program 2011;67:55–66. Kõll-Klais P, Mändar R, Leibur E, Marcotte H, Hammarström L, et al: Oral lactobacilli in chronic periodontitis and periodontal health: species composition and antimicrobial activity. Oral Microbiol Immunol 2005; 20: 354– 361. Lawless J: Negative binomial and mixed Poisson regression. Can J Stat 1987;15:209–225. Lexner MO, Blomqvist S, Dahlén G, Twetman S: Microbiological profiles in saliva and supragingival plaque from caries-active adolescents before and after a short-term daily intake of milk supplemented with probiotic bacteria – a pilot study. Oral Health Prev Dent 2010; 8: 383–388. Liang KY, Zenger SL: Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13–22. Lingström P, Holm A, Mejàre I, Twetman S, Söder B, et al: Dietary factors in the prevention of dental caries: a systematic review. Acta Odontol Scand 2003;61:331–340. Llena C, Forner L, Baca P: Anticariogenicity of casein phosphopeptide-amorphous calcium phosphate: a review of the literature. J Contemp Dent Pract 2009;10:1–9. Marshall TA, Eichenberger-Gilmore JM, Broffitt BA, Warren JJ, Levy SM: Dental caries and childhood obesity: roles of diet and socioeconomic status. Community Dent Oral Epidemiol 2007;35:449–458. Merchant AT, Pitiphat W: Directed acyclic graphs (DAGs): an aid to assess confounding in dental research. Community Dent Oral Epidemiol 2002;30:399–404. Mobley CC: Nutrition and dental caries. Dent Clin North Am 2003;47:319–336. Nicolau B, Marcenes W, Bartley M, Sheiham A: A life course approach to assessing causes of dental caries experience: the relationship between biological, behavioural, socio-economic and psychological conditions and caries in adolescents. Caries Res 2003;37:319–326.
Caries Res 2015;49:251–258 DOI: 10.1159/000375505
Nielsen BM, Bjornsbo KS, Tetens I, Heitmann BL: Dietary glycaemic index and glycaemic load in Danish children in relation to body fatness. Br J Nutr 2005;94:992–997. NNR: Nordic Nutrition Recommendations, ed 4. 2004. Petersen PE: The World Oral Health Report 2003: continuous improvement of oral health in the 21st century – the approach of the WHO Global Oral Health Programme. Community Dent Oral Epidemiol 2003;31(suppl 1):3–23. Riddoch C, Edwards D, Page A, Froberg K, Anderssen SA, et al: European Youth Heart Study background. World Wide Web 2000, pp 1–23. http://www.enshe.lu/committees/ children/background.pdf (accessed May 4, 2003). Riddoch C, Edwards D, Page A, Froberg K, Anderssen S, et al: The European Youth Heart Study – cardiovascular disease risk factors in children: rationale, aims, study design, and validation of methods. J Phys Act Health 2005;2:115–129. Séverin S, Wenshui X: Milk biologically active components as nutraceuticals: review. Crit Rev Food Sci Nutr 2005;45:645–656. Sheiham A: Dietary effects on dental diseases. Public Health Nutr 2001;4:569–591. Stecksén-Blicks C, Sjöström I, Twetman S: Effect of long-term consumption of milk supplemented with probiotic lactobacilli and fluoride on dental caries and general health in preschool children: a cluster-randomized study. Caries Res 2009;43:374–381. Tanaka K, Miyake Y, Sasaki S: Intake of dairy products and the prevalence of dental caries in young children. J Dent 2010;38:579–583. Twetman L, Larsen U, Fiehn N, Stecksen-Blicks C, Twetman S: Coaggregation between probiotic bacteria and caries-associated strains: an in vitro study. Acta Odontol Scand 2009; 67: 284–288. Twetman S, Keller MK: Probiotics for caries prevention and control. Adv Dent Res 2012; 24: 98–102. Twetman S, Stecksén-Blicks C: Probiotics and oral health effects in children. Int J Paediatr Dent 2008;18:3–10.
Lempert/Christensen/Froberg/Raymond/ Heitmann
Downloaded by: Yale Medical Library 198.143.38.65 - 6/28/2015 9:11:47 PM
Altman DG: Practical Statistics for Medical Research. London, Capman & Hall 1991, vol 247–48, pp 316–318, 337. Azarpazhooh A, Limeback H: Clinical efficacy of casein derivatives: a systematic review of the literature. J Am Dent Assoc 2008; 139: 915– 924, quiz 994–925. Bachman CM, Baranowski T, Nicklas TA: Is there an association between sweetened beverages and adiposity? Nutr Rev 2006;64:153–174. Beck AM, Hoppe C, Ygil KH, Andersen NL, Pedersen AN: Vidensgrundlag for rådgivning om indtag af mælk, mælkeprodukter og ost i Danmark. Technical University of Denmark, DTU, 2010. Caglar E, Kuscu OO, Selvi Kuvvetli S, Kavaloglu Cildir S, Sandalli N, et al: Short-term effect of ice-cream containing Bifidobacterium lactis Bb-12 on the number of salivary mutans streptococci and lactobacilli. Acta Odontol Scand 2008;66:154–158. Caglar E, Sandalli N, Twetman S, Kavaloglu S, Ergeneli S, et al: Effect of yogurt with Bifidobacterium DN-173 010 on salivary mutans streptococci and lactobacilli in young adults. Acta Odontol Scand 2005;63:317–320. Cameron FL, Weaver LT, Wright CM, Welbury RR: Dietary and social characteristics of children with severe tooth decay. Scott Med J 2006;51:26–29. Christensen LB, Petersen PE, Hede B: Oral health in children in Denmark under different public dental health care schemes. Community Dent Health 2010a;27:94–101. Christensen LB, Twetman S, Sundby A: Oral health in children and adolescents with different socio-cultural and socio-economic backgrounds. Acta Odontol Scand 2010b;68: 34– 42. Cinar AB, Christensen LB, Hede B: Clustering of obesity and dental caries with lifestyle factors among Danish adolescents. Oral Health Prev Dent 2011;9:123–130. Glavina D, Gorseta K, Skrinjaric I, Vranic DN, Mehulic K, et al: Effect of LGG yoghurt on Streptococcus mutans and Lactobacillus spp. salivary counts in children. Coll Antropol 2012;36:129–132.
![[The association between dietary macronutrients intake and obesity among children and adolescents; a case-control study].](/assets/img/hold.png)
