1-s2.0-s0002817714602981-main

7/26/2019 1-s2.0-S0002817714602981-main

1/8

1254 JADA 145(12) http://jada.ada.org December 2014

ORIGINAL CONTRIBUTIONSORIGINAL CONTRIBUTIONS

Early childhood caries and intake

of 100 percent fruit juiceData from NHANES, 1999-2004

Clemencia M. Vargas, DDS, PhD; Bruce A. Dye, DDS, MPH;

Catherine R. Kolasny, BS; Dennis W. Buckman, PhD;

Timothy S. McNeel, BA; Norman Tinanoff, DDS, MS;

Teresa A. Marshall, PhD; Steven M. Levy, DDS, MPH

Dental caries in young children oten is reerredto as early childhood caries (ECC). he Ameri-can Academy o Pediatric Dentistry deinedECC as the presence o or more decayed

(noncavitated or cavitated lesions), missing (due tocaries), or illed tooth suraces in any primary toothin a child under the age o six.ECC is more likely toaect the most socially vulnerable children: those liv-ing in poverty and those rom ethnic minority groups.However, a recent increase in dental caries prevalenceamong children younger than years shows that a higherproportion o traditionally low-risk children, such as

those living in higher-income amilies, is being aectedby dental caries.No national data have been reportedrecently regarding the distribution o caries among chil-dren aged through years on the basis o poverty andrace/ethnicity.

he U.S. Food and Drug Administrationdeines the

ARTICLE 4

ABSTRACT

Background. Te results o several studies conduct-ed in the United States show no association betweenintake o 100 percent ruit juice and early childhoodcaries (ECC). Te authors examined this association

according to poverty and race/ethnicity among U.S.preschool children.Methods.Te authors analyzed data rom the 1999-2004 National Health and Nutrition ExaminationSurvey (NHANES) or 2,290 children aged 2 through 5years. Tey used logistic models or caries (yes or no)to assess the association between caries and intake o100 percent ruit juice, dened as consumption (yes orno), ounces (categories) consumed in the previous 24hours or usual intake (by means o a statistical methodrom the National Cancer Institute).

Results. Te association between caries and con-

sumption o 100 percent ruit juice (yes or no) wasnot statistically signicant in an unadjusted logisticmodel (odds ratio [OR], 0.76; 95 percent condenceinterval [CI], 0.57-1.01), and it remained nonsigni-cant afer covariate adjustment (OR, 0.89; 95 percentCI, 0.63-1.24). Similarly, models in which we evaluatedcategorical consumption o 100 percent juice (thatis, 0 oz; > 0 and 6 oz; and > 6 oz), unadjusted andadjusted by covariates, did not indicate an associationwith ECC.Conclusions. Our study ndings are consistent withthose o other studies that show consumption o 100percent ruit juice is not associated with ECC.

Practical Implications. Dental practitioners shouldeducate their patients and communities about the lowrisk o developing caries associated with consumptiono 100 percent ruit juice. Limiting consumption o 100percent ruit juice to 4 to 6 oz per day among children1 through 5 years o age should be taught as part ogeneral health education.

Key Words.Caries; eating habits; pediatric dentistry;epidemiology; National Health and Nutrition Exami-nation Survey.JADA 2014;145(12):1254-1261.

doi:./jada..

Dr. Vargas is an associate professor, School of Dentistry, University of

Maryland, 650 W. Baltimore St., Room 2217, Baltimore, Md. 21201, e-

mail [email protected]. Address correspondence to Dr. Vargas.

Dr. Dye is a dental epidemiology officer, National Center for Health

Statistics, Centers for Disease Control and Prevention, Hyattsville, Md.

Ms. Kolasny is a dental student, School of Dentistry, University of

Maryland, Baltimore.Dr. Buckman is a statistician, Information Management Services,

Calverton, Md.

Mr. McNeel is a senior systems analyst, Information Management

Services, Calverton, Md.

Dr. Tinanoff is a professor and chief, Division of Pediatric Dentistry,

School of Dentistry, University of Maryland, Baltimore.

Dr. Marshall is an associate professor, Department of Preventive and

Community Dentistry, College of Dentistry, TheUniversity of Iowa, Iowa

City.

Dr. Levy is Wright-Bush Shreves Professor of Research and graduate

program associate director, Dental Public Health, College of Dentistry,

and a professor, Department of Epidemiology, College of Public Health,

The University of Iowa, Iowa City.

7/26/2019 1-s2.0-S0002817714602981-main

2/8

JADA 145(12) http://jada.ada.org December 2014 1255


term percent fruit juice as a beverage made fromfruit; percent fruit juice differs from juice drinks,which are diluted with water and may contain addedsugars or other ingredients. Researchers in a studyabout dental caries and beverage consumption foundthat percent fruit juice was associated minimallywith caries in participants from the Iowa Fluoride Study

(IFS).In , an analysis of additional data from theIFS participants showed that consumption of percentfruit juice was not associated with caries.Similarly,researchers in a study focused on low-income African-American children in Detroit found that percentfruit juice was protective against caries,,and data fromthe same study showed that milk and juice could be usedtogether as the noncariogenic reference.One importantfactor regarding the studies with findings that showedno association between consumption of percent fruitjuice and caries is that they were conducted in specificpopulations: white middle-income children in Iowa,

and low-income black children in Detroit.

Investigatorsin a studyinvolving the use of nationally representativedata (National Health and Nutrition Examination Survey[NHANES] III)to ascertain the association betweendental caries and consumption of percent fruit juicepresented results for all children with primary teeth andnot for preschool-aged children specifically.They foundthat consuming milk, percent fruit juice or water wasless likely to be associated with dental caries than wasconsuming other beverages.The results of a small studyof patients at childrens medical centers showed that percent fruit juice was not associated with severe ECC.

Since the s, the beverage consumption patternsof children have changed. More children are drinkingbeverages with added sugar, such as sodas and soda pop,juice drinks, and other sugary drinks instead of milkand water.,One hundred percent fruit juice is anotherbeverage that has had a large increase in consumption.Some of the factors linked to the increase in consump-tion of percent fruit juice include the following: it isa convenient snack for children, it is considered a healthybeverage choice, it often costs less than milk, and mar-keting efforts to increase fruit and vegetable intake maypromote juice as an alternative.

Because of the increased consumption of juice bever-ages, the American Academy of Pediatricshas recom-mended that young children limit their consumption of percent fruit juice to to ounces per day. Despitethis recommendation, intake of percent fruit juiceamong young children remains high, with the averagepreschooler consuming twice the recommended amountof percent fruit juice per day.,With passage of the Healthy Hunger-Free Kids Act,there has been a callfor the U.S. Department of Agriculture to begin promot-ing whole fruit instead of percent fruit juice in chil-drens school meals to help prevent childhood obesity.

Information regarding the association between con-

sumption of percent fruit juice and dental cariesin young children is lacking at the national level. Thepurpose of this study was to assess the relationship be-tween consumption of percent fruit juice and cariesamong U.S. preschool children, adjusting for sociodemo-graphic characteristics. In addition, we provide updatedinformation about caries prevalence according to pov-

erty level and race/ethnicity among U.S. children through years of age.

METHODS

Study population.Data for this study come from the- NHANES.NHANES is a series of cross-sectional, nationally representative health examinationstudies conducted by the Centers for Disease Controland Preventions (CDC) National Center for HealthStatistics (NCHS) to evaluate the health of the U.S. civil-ian noninstitutionalized population. Each survey periodinvolves the use of a stratified, multistage probability

sampling design to select participants. NHANES overs-amples some population groups to provide an adequatesample size to improve estimate precision by reducingvariances. Relevant to the children included in this study,the - NHANES oversampled non-Hispanicblack children, Mexican-American children and low-income white children. The - NHANES is themost current data set that includes all of the variablesof interest with sufficient sample size of children aged through years to allow multivariate analysis.

The study sample consisted of a total of , childrenaged through years for whom we had complete dental

examination and dietary recall data. NHANES staffmembers collected data by means of a personal inter-view at the participants home, a health examinationat a mobile examination center (MEC) and laboratoryanalyses. Interviewers collected oral health data fromparticipants years and older during the home interviewand by means of a detailed oral examination in the MECby dentists who underwent calibration. Additional in-formation detailing examination procedures is providedelsewhere.,A trained NHANES interviewer, usingdietary interview software and visual aids, collectedinformation about beverage consumption as part of the-hour dietary recall interview conducted at the MEC.Incomplete interviews were completed by telephone, andthe - public-use data sets include dietary datafrom a second -hour recall interview conducted via

ABBREVIATION KEY. CDC:Centers for Disease Controland Prevention. dft:Decayed or filled teeth (primary). ECC:Early childhood caries. FPG:Federal poverty guideline. IFS:Iowa Florida Study. MEC:Mobile examination center. NA:Not applicable. NCHS:National Center for Health Statistics.NHANES:National Health and Nutrition Examination Survey.PIR:Poverty income ratio. ZINB:Zero-inflated negative bino-mial. ZIP:Zero-inflated poisson.

7/26/2019 1-s2.0-S0002817714602981-main

3/8



telephone. All NHANES study protocols were approvedby the CDC/NCHS Institutional/Ethics Review Board,and written informed consent was obtained from partici-pants or their legal guardians. The University of Mary-land at Baltimoreinstitutional review board consideredour study to be exempt from IRB review because thestudy involved the use of previously collected data with

no personal identifiers (exemption no. ).Variables. We analyzed dental caries experience as

the prevalence (yes or no) of untreated caries or filledteeth by means of the dft index, which is the count ofaffected teeth, whereby d represents the number ofprimary teeth with carious lesions and f representsthe number of filled (restored) primary teeth (t). Thepresence of caries in our study group was similar to thatreported previously on the basis of NHANES data; themean dft for children through years of age was ..

We defined consumption of percent fruit juice ashaving a reported intake of percent juice made from

any fruit in the previous hours. We analyzed juiceconsumption as the prevalence of consumption (yes orno) and the ounces of consumption recoded into threecategories: oz, more than oz but less than or equal to oz, and more than oz.Although a responsible adultreported beverage consumption for the child, we use thephrase child reported throughout this article.

Another independent variable was age (in years).Sex was self-reported as male or female. We includedrace/ethnicity and poverty because many study findingshave shown that they are social determinants of healthassociated strongly with caries in preschool children.,,

Race/ethnicity was self-reported, and we recoded it asnon-Hispanic white, non-Hispanic black, Mexican-American and others. We based poverty level on thepoverty income ratio (PIR); a PIR of . is consideredrepresentative of a poverty level at percent of the fed-eral poverty guideline (FPG).We grouped participantsinto three categories: less than . PIR (< percentFPG [poverty]), .-. PIR (- percent FPG) or. or higher PIR ( percent FPG). We defined anannual dental visit (yes or no) as having had at least onedental visit within the previous months.

Data analysis. We performed all statistical analy-ses by using statistical software (SAS Version ..,SAS Institute, Cary, N.C.; SUDAAN Version ., RTIInternational, Research Triangle Park, N.C.; and Stata,Version ., StataCorp, College Station, Texas). Weused dietary sample weights to account for oversam-pling, nonresponse and other survey design features.SUDAAN and Stata account for the complex surveydesign, including design effects. We used SUDAAN toestimate weighted percentages and standard errors, andwe used Taylor series linearization to calculate standarderrors. We evaluated differences between percentages byusing two-sided significance tests at the . level, withall comparisons made to the reference group within the

selected variable. We also tested for trend within the ageand poverty covariates by using the stratum-adjustedCochran-Mantel-Haenszel test.,If we did not observea linear relationship, we presented findings only for thecomparisons to the reference group.

We assessed the relationship between dental cariesand percent fruit juice consumption by using two

distinct modeling approaches. We fit logistic regressionmodels to analyze caries prevalence (yes or no) by usingSUDAAN software to account for the complex sampledesign. Caries prevalence among preschool childrenin the United States is below percent,,resulting ina sizeable number of preschool children who have notexperienced caries. Consequently, we fit zero-inflatedpoisson (ZIP) models and zero-inflated negative bi-nomial (ZINB) models to analyze the number of teethwith caries (dft) and to allow for excess zeros in thiscount variable. We fit ZIP and ZINB models by usingStata software with appropriate options to account for

the complex survey design. Because results from bothZIP and ZINB models essentially were the same as thoseobtained from logistic regression only, we present resultsfrom the logistic regression models. We used logisticregression models to test the associations between ECCand sociodemographic indicators (age, sex, race/ethnicity and poverty status).

We also assessed the relationship between dental car-ies and usual intake of percent fruit juice by using astatistical method from the National Cancer Institute de-signed for -hour recall data for episodically consumeddietary components that corrects for the measurement

error occurring in dietary variables.

An episodicallyconsumed dietary component, such as percentfruit juice, may or may not be consumed during a given-hour period; consequently, the -hour recall dataconsist of an excess number of zeros owing to noncon-sumption days, and they consist of positive consumptionamounts from consumption days. We used SAS macrosto fit the measurement error model and to predict usualintake for use in the logistic model to assess the relation-ship between caries and usual intake of percent fruitjuice, with correction for dietary measurement error. Weused the jackknife method to estimate variance.

RESULTS

Children were distributed evenly, according to age andsex (Table ,). Almost two-thirds were non-Hispanicwhite, percent were non-Hispanic black, percentwere Mexican-American and less than percent werefrom other groups. More than one-quarter of childrenlived in poverty (< percent FPG) and one-third livedin families at or above percent FPG. Just under one-half (. percent) of the children reported having had adental visit in the previous year.

Table ,(page ) shows the consumption of percent fruit juice in the previous hours. Nearly

7/26/2019 1-s2.0-S0002817714602981-main

4/8



percent of children consumed percent fruit juiceduring that period. Consumption decreased accordingto age. We found no significant difference in consump-tion of percent fruit juice between children livingin poverty and children living in families at or above percent FPG; however, fruit juice consumption wassignificantly lower (P .) among children living infamilies at - percent FPG than it was among chil-dren living at higher levels of the FPG.

Approximately percent of all children consumedmore than oz of percent fruit juice in the previous hours. This high level of consumption decreased withage (P .). We found no significant difference in thepercentage of children consuming more than oz of percent fruit juice according to poverty level. In addi-tion, we observed no significant difference in childrensconsumption of percent fruit juice according tosex or race/ethnicity. Finally, we found no statisticallysignificant difference between children who had dentalcaries (dft > ) and reported consumption of percentfruit juice ( percent) and those who did not have caries(dft = ) and reported consumption of percent fruitjuice ( percent).

Table ,(page ) shows the prevalence of ECCaccording to sociodemographic indicators. The preva-lence of dental caries (dft) among - to -year-olds was. percent. Dental caries prevalence increased inmagnitude as age increased (P .), ranging from percent for -year-olds to nearly percent for -year-olds. We found no difference in ECC prevalence accord-

ing to sex. Mexican-American children had the highestprevalence of dental caries (. percent) (P .)compared with that among non-Hispanic white children,non-Hispanic black children and children from otherracial/ethnic groups, all of whom had a caries prevalencebelow percent. Overall, as poverty increased, den-tal caries prevalence increased, from approximately percent ( percent FPG) to about percent ( and oz, and > oz), neither the

unadjusted nor the adjusted modelcontrolling for age,sex, race/ethnicity, poverty and dental visitsindicatedan association with dental caries.

Results from modeling that involved the use of atechnique to correct for potential measurement error forintake of percent fruit juice and that was adjusted forage, sex, race/ethnicity, poverty and dental visits also didnot show an association between intake of percentfruit juice and dental caries (Table ). Likewise, whencomparing the results of an adjusted model in which wecorrected for potential dietary measurement error forjuice intake with the results of the other two adjustedmodels that involved the use of the more typical model-ing approach, we observed little difference across thecovariates and in the association of those covariates withdental caries.

DISCUSSION

Consistent with the findings of previous studies,-wefound no association between consumption of per-cent fruit juice in the previous hours and dental cariesamong young children in NHANES -. Thecrude ORs for the two variables representing consump-tion of percent fruit juiceprevalence of consump-tion (yes or no) and categories for ounces consumed

TABLE 1

Sociodemographic characteristicsamong children 2-5 years of age,according to NHANES* 1999-2004.CHARACTERISTICS PERCENTAGE OF CHILDREN (SE)

Age, in Years

2 25.8 (1.1)

3 24.2 (1.1)

4 24.1 (1.6)

5 26.0 (1.1)

Sex

Male 49.3 (1.6)

Female 50.7 (1.6)

Race/Ethnicity

Non-Hispanic white 62.0 (2.3)

Non-Hispanic black 13.9 (1.4)

Mexican-American 13.6 (1.4)

Other 10.5 (1.4)

Poverty Level, in PercentageFPG

< 100 26.8 (1.2)

100-299 41.0 (2.0)

300 32.2 (2.0)

Dental Visit Within PreviousYear

Yes 45.8 (1.6)

No 54.2 (1.6)

* NHANES: National Health and Nutrition Examination Survey. Source: Centers for Disease Control and Prevention.17

SE: Standard error. FPG: Federal poverty guideline.22

7/26/2019 1-s2.0-S0002817714602981-main

5/8



remained nonsignificant after we enteredthem in the adjusted models.

We used two different modeling ap-proaches in analyzing dietary data from-hour recalls. As demonstrated by thereported zero values in Table , per-cent fruit juice was not consumed daily

by everyone in our study. Furthermore,dietary data from self-reported assess-ment methods, such as the -hour recall,may not accurately reflect intake, so weused an established statistical approach toevaluate the relationship between a healthoutcome and dietary data with excesszeros and measurement error.

Our findings show no evidence of anassociation between dental caries andusual intake of percent fruit juicewhen comparing the th and th per-

centiles of the usual intake distribution.Likewise, we observed little differenceamong the covariates evaluated betweenthe other two adjusted models by usingthe more typical modeling approach andan adjusted model corrected for measure-ment error.

The risk of caries (that is, OR) ad-justed for the sociodemographic factorsdid not vary considerably in the modelsthat included one of the indicators (thatis, prevalence or ounces consumed) of

consumption of percent fruit juice(data not shown). This is consistent withthe lack of association between consump-tion of percent fruit juice and ECC,as well as the overall lack of difference inconsumption of percent fruit juiceaccording to sociodemographic indica-tors. Therefore, we presented the logisticmodels for juice consumption withoutpresenting the sociodemographic factors.

Limitations of this study are relatedclosely to limitations in the dietary datacollection methods and cross-sectionalnature of the data. There are severalmethods for collecting dietary data, andall of them have their limitations.Analyses based on asingle -hour recall period can be subject to recall andsocial desirability biases and measurement error. Onealso could argue that the -hour dietary recall may bean inadequate risk indicator for a disease that developsover many months. Therefore, the results presented inthis report showing a lack of association between con-sumption of percent fruit juice and ECC should beconsidered preliminary and indicate the need for furtherresearch. Furthermore, dietary measurement error often

attenuates relationships between diet and health out-comes.However, Tooze and colleaguesdemonstratedthat food intakes reported for two or more -hour recallperiods can be used to estimate the usual intake distribu-tion of an episodically consumed dietary component,such as percent fruit juice. Kipnis and colleaguesextended this method to evaluate relationships be-tween usual intake and health outcomes. As part of ouranalyses, we used this established statistical approach tocorrect for potential measurement error. Another limita-

TABLE 2

Consumption of 100 percent fruit juice inprevious 24 hours by children aged 2-5 years,NHANES* 1999-2004.

CHARACTERISTICS PERCENTAGE (SE) OF ALL CHILDREN

Prevalence of

100 PercentFruit Juice

Consumption

Consumption of 100 Percent Fruit Juice,

in Ounces

Yes 0 > 0 and 6 > 6

Total Percentage(SE)

53.6 (2.2) 46.4 (2.2) 15.1 (1.5) 38.5 (1.8)

Age, in Years

2 (reference) 62.0 (2.9) 38.0 (2.9) 16.3 (1.8) 45.8 (2 .6)

3 55.6 (2.7) 44.4 (2.7) 14.6 (2.3) 41.0 (2.7)

4 53.8 (3.1) 46.2 (3.1) 16.6 (2.5) 37.2 (3.2)

5 43.0 (4.3) 57.0 (4.3) 13.0 (2.6) 30.0 (3.8)

Sex

Male 54.7 (2.8) 45.3 (2.8) 16.3 (2.2) 38.4 (2.1)

Female (reference) 52.4 (2.7) 47.6 (2.7) 13.9 (1.8) 38.5 (2.5)

Race/Ethnicity

Non-Hispanic white(reference)

53.0 (3.1) 47.0 (3.1) 14.8 (1.9) 38.2 (2.5)

Non-Hispanic black 53.3 (2.4) 46.7 (2.4) 14.3 (1.9) 39.0 (2.4)

Mexican-American 52.7 (2.5) 47.3 (2.5) 14.7 (1.4) 38.1 (2.2)

Other 58.0 (5.5) 42.0 (5.5) 18.4 (3.8) 39.6 (5.3)

Poverty Level, inPercentage FPG

< 100 52.7 (2.8) 47.3 (2.8) 14.2 (1.8) 38.5 (2.9)

100-299 46.8 (3.2)# 53.2 (3.2)# 13.1 (1.6) 33.6 (3.0)

300 (reference) 60.3 (4.2) 39.7 (4.2) 18.5 (3.3) 41.8 (3.6)

Dental Visit WithinPrevious Year

Yes 52.1 (3.2) 47.9 (3.2) 15.3 (2.1) 36.9 (2.5)

No (reference) 54.8 (2.4) 45.2 (2.4) 15.0 (1.7) 39.8 (2.2)

Dental Caries

Yes 48.0 (3.2) 52.0 (3.2) (3.0) 13.0 (1.9) 35.0 (2.7)

No (reference) 54.8 (2.6) 45.2 (2.6) 16.0 (1.9) 38.9 (2.6)


SE: Standard error. P.05 (linear trend test) for age and poverty only. FPG: Federal poverty guideline. FPG based on the poverty income ratio.22

# P.05 for comparison with reference group.

7/26/2019 1-s2.0-S0002817714602981-main

6/8



tion associated with the NHANES design is the lack ofinformation about fluoride exposures.

The finding of no association between consumptionof percent fruit juice and dental caries in youngerchildren may seem counterintuitive given the generalunderstanding of sugars and their influence on dentalcaries etiology. However, several factors may contributeto this lack of association, including biological, behav-ioral and social factors.

One of the major biological factors that could explainthe lack of association between consumption of percent fruit juice and caries is fluoride protectionagainst caries owing to its ubiquitousness.For example,Campain and colleaguesreported that low-caries-riskadolescents exhibited a very weak association betweensugar consumption and caries, which was attributed tofrequent exposure to fluoride, both in the water sup-ply and in toothpastes. Gibson and Williamsreportedthat the association between caries and sugar consump-tion was present only among preschool-aged childrenwho had their teeth brushed less than twice a day. They

concluded that regular toothbrushing twice a day withfluoridated toothpaste may have a greater effect on cariesprevention among young children than does limiting theconsumption of sugary foods.Duggal and colleaguesdemonstrated that demineralization after consump-tion of foods high in carbohydrates can be preventedby brushing with fluoridated toothpaste compared with

brushing with nonfluoridated toothpaste. In addition, percent fruit juices contain numerous phytochemi-cals with known antibacterial activities; thus, percentfruit juice could inhibit oral bacterias growth and meta-bolic activities.-

Other factors related to the timing and frequency ofsugar consumption also could explain the lack of asso-ciation found between intake of percent fruit juiceand ECC. However, Burt and colleaguesfound thatthe frequency and time of sugar consumption were notassociated with caries increments among children to years old. In , Marshall and colleaguesreported

similar results. Their study findings also showed thatthe timing of consumption of percent fruit juice hadminimal impact on caries risk among children aged through years.We also found that consuming per-cent fruit juice with meals or with snacks did not haveany influence on the association between consumptionof percent fruit juice and ECC (data not shown).

The findings of the Vipehlm study, the most influen-tial study on the topic of sugar consumption frequency,are questionable in relation to the population studied(patients in a mental institution) and the extreme natureof the sugar ingested ( sticky toffees that, because of

the large size, had to be sucked).

It is unlikely that theseconditions are similar to real-world sugar consump-tion.,Moreover, the widespread availability of fluo-ride today makes that studys results even less applicable.Another epidemiological factor relates to the overallhigh consumption of added sugar among U.S. children,which makes it difficult to find children who have notbeen exposed to added sugars to serve as control partici-pants in studies.

Behavioral factors also may contribute to the lackof association between dental caries in young childrenand consumption of percent fruit juice. One hun-dred percent fruit juice, as opposed to fruit drinks withadded sugar, is viewed by many parents as a healthydietary option for children, particularly young chil-dren.Consequently, more health-conscious and well-informed parents or caregivers might be more inclinedto provide percent fruit juice to their children alongwith other healthy foods and to instill other healthybehaviors.,Lim and colleaguesfound that childrenwho had high consumption of percent fruit juice alsohad high consumption of milk. Therefore, consumptionof percent fruit juice by young children could be anindicator of overall healthy behaviors, which, consider-ing the multifactorial etiology of caries, would result

TABLE 3

Caries prevalence, according tosociodemographic characteristicsamong children aged 2-5 years,NHANES* 1999-2004.

VARIABLES PERCENTAGE (SE) OFCHILDREN WITH CARIES

TOTAL 28.5 (1.5)

Age, in Years

2 (reference) 11.9 (1.8)

3 25.0 (2.7)

4 35.2 (4.0)

5 39.5 (3.6)

Sex

Male 29.8 (2.4)

Female (reference) 27.3 (2.1)

Race/Ethnicity

Non-Hispanic white (reference) 25.7 (2.2)

Non-Hispanic black 29.2 (2.2)

Mexican-American 42.6 (2.5)

Other 26.3 (4.2)

Poverty Level, in PercentageFPG#

< 100 43.4 (3.3)

100-299 29.4 (2.4)

300 13.2 (2.4)


SE: Standard error. P.05 (linear trend test) for age and poverty only for caries

prevalence.

P.05 for comparison with reference group for caries prevalence.# FPG: Federal poverty guideline.22

7/26/2019 1-s2.0-S0002817714602981-main

7/8



in caries prevention.

Our finding of a lack of association between ECC andconsumption of percent fruit juice suggests that oralhealth care providers and educators should recommend

limiting consumption of percent fruit juice to to oz per day for children to years of agefor its overallhealth benefit (that is, nutritional value)

rather than as ameans to prevent caries.

TABLE 4

Logistic regression results for the association between consumptionof 100 percent fruit juice in the previous 24 hours and caries prevalence,NHANES*1999-2004.

VARIABLES ODDS RATIO (95 PERCENT CONFIDENCE INTERVAL)

Crude Odds Ratios Adjusted ModelWith Prevalence ofConsumption of 100Percent Fruit Juice

Adjusted Model WithOunces of Consumption

of 100 Percent FruitJuice

Adjusted Model,Measurement Error

Correction for Intakeof 100 Percent Fruit

Juice#

Consumption of 100 PercentFruit Juice

Yes 0.76 (0.57-1.01) 0.89 (0.63-1.24) NA** NA

No (reference) Reference Reference NA NA

Consumption of 100 PercentFruit Juice

>6 ounces 0.78 (0.56-1.09) NA 0.90 (0.62-1.30) NA

>0 and 6 oz 0.71 (0.46-1.08) NA 0.85 (0.51-1.42) NA

0 oz (reference) Reference NA Reference NA

Consumption of 100 PercentFruit Juice: 12.6 oz Versus0.9 oz

NA NA NA 0.97 (0.44-2.12)

Childs Age, in Years

2 (reference) Reference Reference Reference Reference

3 2.47 (1.54-3.97) 2.27 (1.43-3.61) 2.27 (1.43-3.62) 2.29 (1.44-3.63)

4 4.04 (2.48-6.58) 3.79 (2.26-6.36) 3.79 (2.26-6.35) 3.80 (2.24-6.45)

5 4.85 (2.98-7.90) 4.14 (2.44-7.02) 4.14 (2.45-7.02) 4.21 (2.37-7.48)

Sex

Male 1.13 (0.81-1.58) 1.23 (0.85-1.77) 1.23 (0.85-1.77) 1.22 (0.85-1.75)

Female (reference) Reference Reference Reference Reference

Race/Ethnicity

Non-Hispanic white(reference) Reference Reference Reference Reference

Non-Hispanic black 1.19 (0.87-1.63) 0.79 (0.55-1.15) 0.79 (0.55-1.15) 0.79 (0.55-1.14)

Mexican-American 2.15 (1.57-2.95) 1.56 (1.13-2.16) 1.56 (1.13-2.17) 1.56 (1.15-2.11)

Other 1.03 (0.63-1.69) 0.93 (0.56-1.55) 0.93 (0.56-1.56) 0.92 (0.56-1.52)

Poverty Level, in PercentageFPG

7/26/2019 1-s2.0-S0002817714602981-main

8/8



CONCLUSION

Our findings are consistent with those of other studiesthat show consumption of percent fruit juice is notassociated with ECC.

Disclosure. None of the authors reported any disclosures.

This research was funded by grant R DE - from the

National Institute of Dental and Craniofacial Research, National Insti-tutes of Health, Bethesda, Md., to Dr. Vargas.

. American Academy of Pediatric Dentistry. Policy on early childhoodcaries (ECC): classifications, consequences, and preventive strategies.Revised . ww w.aapd.org/media/Policies_Guidelines/P_ECCClassifications.pdf. Accessed Oct. , .

. Vargas CM, Crall JJ, Schneider DA. Sociodemographic distribution ofpediatric dental car ies: NHANES III, -. JADA ;():-.

. Dye BA, Arevalo O, Vargas CM. Trends in paediatric dental cariesby poverty status in the United States, - and -. Int JPaediatr Dent ;():-.

. U.S. Food and Drug Administ ration. CFR: Code of FederalRegulations Title . www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=.. Accessed Nov. , .

. Marshall TA, Levy SM, Broffitt B, et a l. Dental caries and beverageconsumption in young chi ldren. Pediatrics ;(, pt ):e-e.

. Marshall TA, Broffitt B, Eichenberger-Gilmore J, Warren JJ, Cun-ningham MA, L evy SM. The roles of meal, snack, and daily total foodand beverage exposures on caries ex perience in young children. J PublicHealth Dent ;():-.

. Kolker JL, Yuan Y, Burt BA, et al. Dental caries and dietar y patternsin low-income African American children. Pediatr Dent ;():-.

. Lim S, Sohn W, Burt BA, et al. Cariogenicity of soft drinks , milkand fruit ju ice in low-income African-American children: a longitudinalstudy. JADA ;():-.

. Sohn W, Burt BA, Sowers MR. Carbonated soft drinks and dentalcaries in the primary dentition. J Dent Res ;():-.

. Evans EW, Hayes C, Palmer CA, Bermudez OI, Cohen SA, Must A.Dietary intake and severe early childhood caries in low-income, young

children. J Acad Nutr Diet ;():-.. Wang YC, Bleich SN, Gortmaker SL. Increasing caloric contribution

from sugar-sweetened beverages and fruit juices among US childrenand adolescents, -. Pediatrics ;():e-e.

. Harnack L, Stang J, Story M. Soft drink consumption among USchildren and adolescents: nutritional consequences. J Am Diet Assoc;():-.

. Dennison BA. Fruit juice consumption by infants a nd children: areview. J Am Coll Nutr ;()(suppl):S-S.

. American Academy of Pediatrics: the use a nd misuse of fruit juice inpediatricsCommittee on Nutrition. Pediatrics ;():-.

. Gidding SS, Dennison BA, Birch LL, et al; American Heart Associa-tion. Dietary recommendations for children and adolescents: a guide forpractitioners (published correction appears in Pediatrics ;[]:.Gilman, Matthew W [corrected to Gillman, Matthew W]). Pediatrics;():-.

. Wojcicki JM, Heyman MB. Reducing childhood obesity by eliminat-ing fruit juice. Am J Public Health ;():-.

. Centers for Disease C ontrol and Prevention. National Health andNutrition Examination Survey. Questionnaires, datasets, and relateddocumentation. www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm.Accessed Nov. , .

. Dye BA, Barker LK, Selwitz RH, et al. Overv iew and quality as-surance for the National Health and Nutrition Exami nation Survey(NHANES) oral health component, -. Community Dent OralEpidemiol ;():-.

. Dye BA, Nowjack-Raymer R, Barker LK, et a l. Overview and qualityassurance for the oral health component of the National Health and Nu-trition Examination Survey (NHANES), -. J Public Health Dent;():-.

. Dye BA, Tan S, Smith V, et al. Trends in oral health status: UnitedStates, - and -. Vital Health Stat ;():-.

. Fisher-Owens SA, Gansky SA, Platt LJ, et a l. Inf luences on childrensoral health: a conceptual model. Pediatrics ;():e-e.

. U.S. Census Bureau. Poverty. Definitions. w ww.census.gov/hhes/www/poverty/methods/definitions.html. Accessed Nov. , .

. Landis JR, Heyman ER, Koch GG. Average partial association inthree-way contingency tables: a review and discussion of alternative tests.Int Stat Rev ;:-.

. Agresti A. Categorical Data Analysis. nd ed. New York City: Wileyand Sons; .

. Kipnis V, Midthune D, Buckman DW, et al. Modeling data with ex-cess zeros and measurement error: application to evaluating relationshipsbetween episodically consumed foods and health outcomes. Biometrics;():-.

. National Cancer Institute. Usual dietar y intakes: SAS macros for theNCI method. http://appliedresearch.cancer.gov/diet/usualintakes/

macros.html. Accessed Oct. , .. Burt BA, Pai S. Sugar consumption and car ies risk: a systematicreview. J Dent Educ ;():-.

. Tooze JA, Midthune D, Dodd KW, et al. A new statistical method forestimating the usual inta ke of episodically consumed foods with applica-tion to t heir distribution. J Am Diet Assoc ;():-.

. Burt BA, Eklund SA. Dentistry, Dental Practice, and the Commu-nity. th ed. St. Louis: Elsevier Saunders; .

. Campain AC, Morgan MV, Evans RW, et al. Sugar-starch combina-tions in food and the relationship to dental caries in low-risk adolescents.Eur J Oral Sci ;():-.

. Gibson S, Williams S . Dental caries in pre-school children: associa-tions with social class, toothbrushing habit and consumption of sugarsand sugar-containing foodsfurther analysis of data from the NationalDiet and Nutrition Survey of children aged .-. years. Caries Res;():-.

. Duggal MS, Toumba KJ, Amaechi BT, Kowash MB, Higham SM.Enamel demineralization in situ with various frequencies of carbohy-drate consumption with and without f luoride toothpaste. J Dent Res;():-.

. Tabasco R, Snchez-Patn F, Monagas M, et al. Effect of grape poly-phenols on lactic acid bacteria and bifidobacteria growth: resistance andmetabolism. Food Microbiol ;():-.

. Russell W, Duthie G. Plant secondary metabolites and g ut health:the case for phenolic acids. Proc Nutr Soc ;():-.

. Del Rio D, Borges G, Crozier A. Berr y flavonoids and phenolics:bioavailability and ev idence of protective effects. Br J Nutr ;(suppl):S-S.

. Burt BA, Eklund SA, Morgan KJ, et al. The effects of sugars intakeand frequency of ingestion on dental caries increment in a three-yearlongitudinal study. J Dent Res ;():-.

. Gustafsson BE, Quensel CE, Lanke LS, et al. The Vipehlm dental

caries study: the effect of different levels of carbohydrate intake on cariesactivity in indiv iduals observed for five years. Acta Odontol Scand;(-):-.

. Ervin RB, Kit BK, Carroll MD, Ogden CL. Consumption of addedsugar among U.S. chi ldren and adolescents, -. NCHS Data Brief;():-.

1-s2.0-s0002817714602981-main

Documents