JDR JDR Most Read Articles
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sohn, W.
Right arrow Articles by Sowers, M.R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sohn, W.
Right arrow Articles by Sowers, M.R.
J Dent Res 85(3):262-266, 2006
© 2006 International and American Associations for Dental Research

RESEARCH REPORT
Clinical

Carbonated Soft Drinks and Dental Caries in the Primary Dentition

W. Sohn1,*, B.A. Burt2, and M.R. Sowers2

1 Department of Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, 1101 N. University, Ann Arbor, MI 48109-1078, USA; and
2 Department of Epidemiology, School of Public Health, University of Michigan

* corresponding author, woosung{at}umich.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We analyzed fluid intake data among children aged 2–10 years from a 24-hour dietary recall interview in the NHANES III (1988–94) to investigate the effect of high consumption of carbonated soft drinks on caries in the primary dentition. We used cluster analysis to determine fluid consumption patterns. Four distinct fluid consumption patterns were identified: high carbonated soft drinks, high juice, high milk, and high water. About 13% of children had a high carbonated soft drink consumption pattern; they also had a significantly higher dental caries experience in the primary dentition than did children with other fluid consumption patterns. A fluid intake pattern comprised mainly of milk, water, or juice was less likely to be associated with dental caries. Findings of this study suggest that high consumption of carbonated soft drinks by young children is a risk indicator for dental caries in the primary dentition and should be discouraged.

KEY WORDS: fluid consumption pattern • carbonated • soft drinks • NHANES III • cluster analysis • dental caries • primary dentition


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Patterns of fluid consumption by children are of public health interest, since they may be related to both oral health and general health conditions such as obesity and diabetes (Marshall, 2003). It has been reported that contemporary fluid consumption patterns of children are now more diverse than in past years, since carbonated soft drinks and fruit juices have replaced much of the previous consumption of water and milk among children (Harnack et al., 1999; Heller et al., 1999). The implications of these changes in fluid consumption have not been well-studied, especially the association between carbonated soft drink consumption and dental caries. While the association between the consumption of sugars (all mono-and disaccharides) and dental caries experience in permanent teeth has been well-documented (Ismail et al., 1984; Rugg-Gunn, 1996; Jones et al., 1999), the association between carbonated soft drink consumption and dental caries in the primary dentition is less clear. Some studies have reported significant associations (Grindefjord et al., 1995; Moynihan and Holt, 1996; Levy et al., 2003), while others did not (Heller et al., 2001; Sayegh et al., 2002).

These inconsistencies may arise because fluid consumption patterns are complex and methodologically challenging to ascertain. For example, like other diet and nutrition data, fluid consumption sources are highly correlated, which precludes testing the association between fluid intake from a single source and a disease outcome without considering the effect of fluid intake from other sources (Wirfalt and Jeffery, 1997). To develop a more simple measure of complex patterns, some investigators have used cluster analysis (Aldenderfer and Blashfield, 1984) to aggregate dietary patterns and then associate the patterns with disease outcomes (Akin et al., 1986; Tucker et al., 1992; Wirfalt and Jeffery, 1997).

The aims of this analysis were: (1) to identify, by cluster analysis, distinct fluid consumption patterns, with a focus on high consumption of carbonated soft drinks, among children aged 2 to 10 yrs; and (2) to assess the association between high consumption of carbonated soft drinks and dental caries in the primary dentition among children in the United States.


   MATERIALS & METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
The NHANES III Survey
    Data for Fluid Consumption
This analysis used fluid intake data from a 24-hour dietary recall interview in the Third National Health and Nutrition Examination Survey (NHANES III, 1988–94) in the United States (National Center for Health Statistics, 1994). Detailed descriptions of the data preparation and variables used in this analysis can be found elsewhere (Sohn et al., 2001). We defined major fluid sources as milk (and milk drinks), juice (fruit and vegetable juices and other non-carbonated drinks), carbonated soft drinks (sugared and non-sugared), plain water (tap and spring water), and coffee and tea. Fluid intake from sources other than these major sources—such as soup, homemade beverages, and water used for cooking—was all grouped into ‘other foods and beverages’. This definition is consistent with previous studies from other researchers (Ershow and Cantor, 1989; Heller et al., 1999). In this analysis, sugared and non-sugared carbonated soft drinks were not separated; natural juice and fruit-based juice drinks were also not separated.

    Dental Caries Data
Dental caries data were obtained in the form of decayed and filled tooth surfaces (dfs) in the primary dentition through dental examinations conducted at the Mobile Examination Centers (MECs) during the NHANES III. Details of dental caries examinations and diagnostic criteria have been described (Kaste et al., 1996). The sample for this analysis consisted of the 5985 children aged 2–10 yrs who completed both a 24-hour dietary interview and dental examination during the NHANES III. Due to a skewed distribution of dental caries among children, a dichotomized dental caries outcome was used in the analysis, i.e., children with at least one decayed or filled primary tooth surface vs. those with no decayed or filled primary tooth surface.

    Sociodemographic Information
Race and ethnicity classifications were non-Hispanic whites, non-Hispanic Blacks (African-Americans), Mexican-Americans, and Others. The ‘Others’ category included all Hispanics, regardless of race, who were not Mexican-American and also all non-Hispanics from racial groups other than whites or African-Americans. Socioeconomic status (SES) was categorized on the basis of the poverty income ratio (PIR), which is a ratio of reported annual family income to the Federal poverty threshold. The categories of SES in this analysis were: low SES (0.000–1.300 PIR), middle SES (1.301–3.500 PIR), and high SES (3.501 and above PIR).

Statistical Analysis
We used cluster analysis to group children based on similarity of fluid consumption patterns. The FASTCLUS procedure in the Statistical Analysis System® Software (version 9.1, SAS Institute, Inc., Cary, NC, USA) was used for this cluster analysis. FASTCLUS performs a disjoint cluster analysis on the basis of Euclidean distances computed from one or more quantitative variables. The observations are divided into clusters such that every observation belongs to a unique cluster (SAS Institute Inc., 1989–1996). The dataset is first scanned for initial cluster ‘seeds’. The procedure then makes repeated comparisons between the variable means of initial clusters with subsequent updates of cluster groupings and means. Subjects are moved between clusters and new means are computed until the distances between observations within a cluster are smaller than the distances between cluster means.

We conducted the cluster analysis based on the proportions of total fluid intake represented by each of the four primary sources defined above, rather than the absolute amount of fluid intake. With the FASTCLUS procedure, the number of clusters as an outcome must be pre-determined. We did this by evaluating various cluster numbers (2 to 10) by comparing the approximate R-squared values and the within-cluster standard deviations. The R-squared value increased with more clusters, but the within-cluster standard deviations no longer decreased after four clusters, which was the criterion for determining the number of clusters to be used in our analyses.

Drinking habits and amounts varied substantially by children’s age in our preliminary analysis. Hence, the cluster analysis procedures were applied to data that were stratified by children’s age groups: 2-year-olds (toddlers), 3- to 5-year-olds (pre-school children), and 6- to 10-year-olds (school-aged children).

The crude associations between dental caries and fluid consumption groups, as well as with other sociodemographic characteristics, were analyzed with the chi-square test. Subsequently, a multivariate logistic regression model was constructed to predict dental caries in the primary dentition with fluid consumption patterns and sociodemographic factors.

All analyses incorporated sampling weights to adjust for unequal sampling probabilities and non-response bias (National Center for Health Statistics, 1994). We used SUDAAN (Release 9.0.0 SAS Callable, Research Triangle Institute, Research Triangle Park, NC, USA) to estimate variances adjusted for the design effect from the complex, multi-stage cluster sample design of NHANES III (Shah et al., 1997).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Fluid Consumption Patterns
Carbonated soft drinks comprised 8.5% of total fluid consumption among children aged 2–10 yrs. Milk and juice each comprised less than 20%, whereas plain water constituted about 32% of total fluid consumption (Table 1Go).


View this table:
[in this window]
[in a new window]
  		Table 1. Fluid Consumption Clusters and Fluid Consumption Profile among Children Aged 2–10 Years Old: NHANES III (1988–1994)
 
Cluster analysis identified the four groupings with distinct fluid consumption profiles (Table 1Go). There was a clear pattern in each fluid consumption grouping, based on the proportion of major fluid sources. For example, the first (left-most column) grouping constituted, on average, 31.8% of the total fluid intake from carbonated soft drinks, and was labeled as the high carbonated soft drink consumption cluster. The other groupings were similarly named, according to their largest contributing fluid sources: high juice consumption cluster, high milk consumption cluster, and high plain water consumption cluster, respectively. Overall, 619 children (12.7%) were grouped in the ‘high carbonated soft drink consumption’ cluster. Similarly, 21.1% were grouped in the ‘high juice’ cluster, 24.5% in the ‘high milk’ cluster, and 41.7% in the ‘high plain water’ cluster (Table 1Go).

There were significant associations between the fluid consumption pattern and sociodemographic factors (Table 2Go). High carbonated soft drink consumption and high water intake appeared to be more characteristic of older children (aged 6–10 yrs). Boys showed slightly higher carbonated soft drink consumption than girls. African-American children showed higher water consumption and lower milk consumption than children from other race/ethnic groups. White children showed a higher tendency toward high carbonated soft drink consumption than children from other race/ethnic groups. There was a positive association between SES and high carbonated soft drink and high juice consumption, and an inverse relationship between SES and the ‘high water consumption’ pattern.


View this table:
[in this window]
[in a new window]
  		Table 2. Socio-demographic Characteristics, Dental Caries Status, and Fluid Consumption Clusters among Children Aged 2–10 Years Old: NHANES III (1988–1994)
 
Carbonated Soft Drink Consumption and Dental Caries in the Primary Dentition
About 38% of children had at least one decayed or filled tooth surface in the primary dentition; the mean number of decayed and filled primary tooth surfaces (dfs) was 2.94. Only 52% of children with high carbonated soft drink consumption were caries-free, which was from 10% to 15% lower than children in the other groups (Table 2Go). In a logistic regression model, these children were about 1.8 times more likely to experience dental caries in the primary dentition than those in the ‘high milk’ and ‘high water’ clusters, after adjustment for age, race/ethnicity, sex, and SES (Table 3Go). Children in the ‘high carbonated soft drink consumption’ cluster also showed a significantly higher tendency toward caries experience than those in the ‘high juice’ clusters. While children of the ‘high juice’ cluster showed a higher tendency toward caries experience than those children in the ‘high milk’ and ‘high water’ clusters, the differences were not statistically significant. Children with a high milk consumption pattern had a tendency toward lowest caries experience. The model explained about 15% of variability of dental caries prevalence among children.


View this table:
[in this window]
[in a new window]
  		Table 3. Logistic Regression Model of Dental Cariesa in the Primary Dentition by Fluid Consumption Pattern and Socio-demographic Factors
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Using the cluster analysis method, we analyzed children’s fluid consumption patterns. To our knowledge, this is the first cluster analysis of fluid consumption patterns among children in the United States. We identified four distinct fluid consumption patterns. Because different clustering methods can generate different solutions in the same dataset, several trials with higher numbers of groupings (5 to 10 clusters) were evaluated to provide more detailed fluid consumption patterns (i.e., high-milk-high-juice; high-juice-high-water; and so on). However, more complicated patterns were not necessarily always clearly discernible, or useful in subsequent analyses.

Children with high carbonated soft drink consumption had a significantly higher prevalence of dental caries in the primary dentition than did children with any other fluid consumption pattern. We did not separately analyze sugared and non-sugared carbonated soft drinks, due to a small number of children (only 6.7%) who reported consuming non-sugared carbonated soft drinks. A preliminary analysis showed that cluster analysis with non-sugared carbonated soft drinks as a separate entity yielded results that did not differ from those in the combined category. Children with a high carbonated soft drink consumption pattern showed significantly higher caries experience, even compared with those children with a high juice consumption pattern.

The findings of this analysis agree with the results from some previous studies (Grindefjord et al., 1995; Moynihan and Holt, 1996; Marshall et al., 2003). In contrast, a previous analysis of the NHANES III data reported no significant association between sugared soda consumption and dental caries in the primary dentition (Heller et al., 2001). The disparity between these results could be explained, in part, by differences in data analytic methods, for the previous analysis (Heller et al., 2001) used individual fluids rather than clusters. Cluster analysis has the advantage of dealing with a greater complexity of fluid consumption patterns when compared with single-fluid-source approaches; it examines all fluid sources, especially those that are consumed in high quantities (Akin et al., 1986).

Children who consumed high amounts of carbonated soft drinks may also have undesirable eating patterns and eat high amounts of sugars from other dietary sources. If so, the high carbonated soft drink consumption pattern might serve more as a marker for unhealthy diet patterns, such as high consumption of sugars. The complexity of the modern diet makes the analysis of diet patterns and fluid consumption patterns challenging. Nevertheless, further analysis, such as cluster analysis including both solid food and fluid sources, should investigate more detailed relationships between diet and fluid intake patterns, and their implications for health and disease outcomes such as dental caries and obesity.

Fluid consumption patterns of children varied significantly by sociodemographic factors. This finding showed some consistency with results from previous reports (Ershow and Cantor, 1989; Harnack et al., 1999; Sohn et al., 2001). In our results, white children showed a higher tendency toward a high carbonated soft drink consumption pattern, and African-American children showed a tendency toward higher plain water and lower milk consumption patterns than did children from other race/ethnic groups. Unfortunately, the implication of various fluid consumption patterns among different population groups in relation to their health outcomes has received only limited attention. Conducting a large-scale study on fluid consumption and dental caries may not be practical; however, small-scale studies of beverage and diet intake, focusing on specific population groups, could widen our understanding of the mechanism behind these interactions.

A few limitations of this analysis should be noted. The NHANES III used a cross-sectional design. Therefore, the results from this analysis should not be interpreted as a cause-effect relationship between and among variables. A fluid consumption pattern from a 24-hour recall interview may not be representative of an individual’s general fluid consumption pattern over time. Although the 24-hour recall interview has been reported to provide reliable group mean estimations (Persson and Carlgren, 1984), it was also reported to be prone to reporting of extreme values, non-reporting, and under-reporting (Yetley et al., 1992). In the NHANES III dietary recall, proxy respondents (caregivers) were allowed for young children (all children until 6 yrs old, and some up to 11 yrs old). For children 1 to 2 yrs old, the caregivers might be well aware of their children’s fluid consumption, but less likely so when the children are older and can obtain drinks for themselves.

In summary, we found that a high consumption of carbonated soft drinks in early childhood was significantly associated with an increased risk of dental caries in the primary dentition after adjustment for age, sex, race/ethnicity, and SES, as well as consumption of fluid from other sources. Findings from this study suggest that the high consumption of carbonated soft drinks by young children is a risk indicator for dental caries in the primary dentition, and should be discouraged.


   ACKNOWLEDGMENTS
 
This research was funded independently by the lead author while a student at the University of Michigan School of Public Health.

Received August 12, 2004; Last revision August 26, 2005; Accepted October 27, 2005


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Akin JS, Guilkey DK, Popkin BM, Fanelli MT (1986). Cluster analysis of food consumption patterns of older Americans. J Am Diet Assoc 86:616–624.[ISI][Medline]

Aldenderfer MS, Blashfield RK (1984). Cluster analysis. Newbury Park, CA: Sage Publications.

Ershow AG, Cantor KP (1989). Total water and tapwater intake in the United States: population-based estimates of quantities and sources. Report #263-MD-810264, National Cancer Institute. Bethesda, MD: Life Sciences Research Office, Federation of American Societies for Experimental Biology.

Grindefjord M, Dahllof G, Nilsson B, Modéer T (1995). Prediction of dental caries development in 1-year-old children. Caries Res 29:343–348.[ISI][Medline]

Harnack L, Stang J, Story M (1999). Soft drink consumption among US children and adolescents: nutritional consequences. J Am Diet Assoc 99:436–441.[ISI][Medline]

Heller KE, Sohn W, Burt BA, Eklund SA (1999). Water consumption in the United States in 1994–96 and implications for water fluoridation policy. J Public Health Dent 59:3–11.[ISI][Medline]

Heller KE, Burt BA, Eklund SA (2001). Sugared soda consumption and dental caries in the United States. J Dent Res 80:1949–1953.[Abstract/Free Full Text]

Ismail AI, Burt BA, Eklund SA (1984). The cariogenicity of soft drinks in the United States. J Am Dent Assoc 109:241–245.[Abstract]

Jones C, Woods K, Whittle G, Worthington H, Taylor G (1999). Sugar, drinks, deprivation and dental caries in 14-year-old children in the north west of England in 1995. Community Dent Health 16:68–71.[ISI][Medline]

Kaste LM, Selwitz RH, Oldakowski RJ, Brunelle JA, Winn DM, Brown LJ (1996). Coronal caries in the primary and permanent dentition of children and adolescents 1–17 years of age: United States, 1988–1991. J Dent Res 75(Spec Iss):631–641.

Levy SM, Warren JJ, Broffitt B, Hillis SL, Kanellis MJ (2003). Fluoride, beverages and dental caries in the primary dentition. Caries Res 37:157–165.[ISI][Medline]

Marshall TA (2003). Diet and nutrition in pediatric dentistry. Dent Clin North Am 47:279–303.[Medline]

Marshall TA, Levy SM, Broffitt B, Warren JJ, Eichenberger-Gilmore JM, Burns TL, et al. (2003). Dental caries and beverage consumption in young children. Pediatrics 112(3 Pt 1):e184–e191.[Abstract/Free Full Text]

Moynihan PJ, Holt RD (1996). The national diet and nutrition survey of 1.5 to 4.5 year old children: summary of the findings of the dental survey. Br Dent J 181:328–332.[ISI][Medline]

National Center for Health Statistics (1994). Plan and operation of the Third National Health and Nutrition Examination Survey, 1988–94. Hyattsville, MD: National Center for Health Statistics.

Persson LA, Carlgren G (1984). Measuring children’s diets: evaluation of dietary assessment techniques in infancy and childhood. Int J Epidemiol 13:506–517.[Abstract/Free Full Text]

Rugg-Gunn AJ (1996). Diet and dental caries. In: Prevention of oral disease. Murray JJ, editor. Oxford: Oxford University Press, pp. 3–31.

SAS Institute Inc (1989–1996). SAS for personal computers. Version 6.12. Cary, NC: SAS Institute Inc.

Sayegh A, Dini EL, Holt RD, Bedi R (2002). Food and drink consumption, sociodemographic factors and dental caries in 4–5-year-old children in Amman, Jordan. Br Dent J 193:37–42.[ISI][Medline]

Shah BV, Barnwell BG, Bieler GS (1997). SUDAAN user’s manual. Rel. 7.5. Research Triangle Park, NC: Research Triangle Institute.

Sohn W, Heller KE, Burt BA (2001). Fluid consumption related to climate among children in the United States. J Public Health Dent 61:99–106.[ISI][Medline]

Tucker KL, Dallal GE, Rush D (1992). Dietary patterns of elderly Boston-area residents defined by cluster analysis. J Am Diet Assoc 92:1487–1491.[ISI][Medline]

Wirfalt AK, Jeffery RW (1997). Using cluster analysis to examine dietary patterns: nutrient intakes, gender, and weight status differ across food pattern clusters. J Am Diet Assoc 97:272–279.[ISI][Medline]

Yetley EA, Beloian AM, Lewis CJ (1992). Dietary methodologies for food and nutrition monitoring. In: Vital & health statistics. Series 4: Documents & Committee Reports, pp. 58–67.




This article has been cited by other articles:


Home page
 Journal of the American Dental AssociationHome page
S. Lim, W. Sohn, B. A. Burt, A. M. Sandretto, J. L. Kolker, T. A. Marshall, and A. I. Ismail
Cariogenicity of Soft Drinks, Milk and Fruit Juice in Low-Income African-American Children: A Longitudinal Study
J Am Dent Assoc, July 1, 2008; 139(7): 959 - 967.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Public HealthHome page
A. E. Sanders, S. Lim, and W. Sohn
Resilience to Urban Poverty: Theoretical and Empirical Considerations for Population Health
Am J Public Health, June 1, 2008; 98(6): 1101 - 1106.
[Abstract] [Full Text] [PDF]

Home page
 CLIN PEDIATRHome page
L. S. Nield, J. P. Stenger, and D. Kamat
Common Pediatric Dental Dilemmas
Clinical Pediatrics, March 1, 2008; 47(2): 99 - 105.
[PDF]

Home page
 Journal of the American Dental AssociationHome page
T. A. Marshall, J. M. Eichenberger-Gilmore, M. A. Larson, J. J. Warren, and S. M. Levy
Comparison of the intakes of sugars by young children with and without dental caries experience
J Am Dent Assoc, January 1, 2007; 138(1): 39 - 46.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sohn, W.
Right arrow Articles by Sowers, M.R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sohn, W.
Right arrow Articles by Sowers, M.R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
IADR Journals Advances in Dental Research ®
Journal of Dental Research ® Critical Reviews (1990-2004)