The Journal of the American Dental Association
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Am Dent Assoc, Vol 139, No 7, 959-967.
© 2008 American Dental Association
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lim, S.
Right arrow Articles by Ismail, A. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lim, S.
Right arrow Articles by Ismail, A. I.

RESEARCH

JADA Continuing Education

Cariogenicity of Soft Drinks, Milk and Fruit Juice in Low-Income African-American Children

A Longitudinal Study



Sungwoo Lim, MA, MS, Woosung Sohn, DDS, PhD, DrPH, Brian A. Burt, BDS, MPH, PhD, Anita M. Sandretto, PhD, Justine L. Kolker, MS, PhD, DDS, Teresa A. Marshall, PhD, RD and Amid I. Ismail, BDS, MPH, MBA, DrPH


   ABSTRACT
TOP
 ABSTRACT
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
Background. The authors conducted a study to test the hypothesis that high consumption of soft drinks, relative to milk and 100 percent fruit juice, is a risk factor for dental caries in low-income African-American children in Detroit.

Methods. Trained dentists and interviewers examined a representative sample of 369 children, aged 3 to 5 years, in 2002–2003 and again two years later. The authors used the 2000 Block Kids Food Frequency Questionnaire (NutritionQuest, Berkeley, Calif.) to collect dietary information. They assessed caries by using the International Caries Detection and Assessment System.

Results. Soft drinks, 100 percent fruit juice and milk represented the sugared beverages consumed by the cohort. A cluster analysis of the relative proportion of each drink at baseline and follow-up revealed four consumption patterns. Using zero-inflated negative binomial models, the authors found that children who changed from being low consumers of soft drinks at baseline to high consumers after two years had a 1.75 times higher mean number of new decayed, missing and filled tooth surfaces compared with low consumers of soft drinks at both time points.

Conclusion. Children who consumed more soft drinks, relative to milk and 100 percent fruit juice, as they grew older were at a greater risk of developing dental caries.

Clinical Implications. Health promotion programs and health care providers should emphasize to patients and caregivers the caries risk associated with consumption of soft drinks.

Key Words: Cariogenicity; soft drinks; low-income children; caries

Abbreviations: d1: Noncavitated surfaces • d2: Cavitated/dentinal surfaces • dtmfs: Noncavitated surfaces, cavitated/dentinal surfaces, missing owing to caries, filled surfaces • f: Filled surfaces • FFQ: Food Frequency Questionnaire • IRR: Incidence rate ratio • m: Missing surfaces owing to caries • PAM: Partitioning around medoids • s: Surfaces • ZINB: Zero-inflated negative binomial regression

Beverage intake in children in the United States has become more diverse and extensive during the last several decades,1,2 and the increasing consumption of sugared drinks among children has raised concerns in the public health community.1,3,4 Some of these concerns center on the risk of dental caries. Cariogenic bacteria need simple carbohydrates to produce the organic acids that initiate the carious process in enamel,5 so the concern is that an increased intake of drinks containing sugar may increase the risk of developing dental caries.

Kolker and colleagues6 found that the three sugared drinks most consumed by low-income African-American children in Detroit were soft drinks (defined here as soda plus fruit drinks), milk and 100 percent fruit juice. Using a cross-sectional analysis, they demonstrated that high consumption of soft drinks in the presence of gingival plaque deposits was significantly related to caries in low-income children6 and adults.7 These findings confirmed those of previous cross-sectional studies of the cariogenic potential of soft drinks.1,3 Milk is considered to have low cariogenic potential,8 while 100 percent fruit juice, which contains about the same amount of sugar as soft drinks,3 can be cariogenic if consumed frequently.9,10 The cariogenicity of sugared drinks is determined not only by their potential to induce demineralization of enamel, but also by the frequency of consumption.11

The aims of this longitudinal study were to identify the intake patterns of soft drinks, milk and 100 percent fruit juice among low-income black children over a two-year period and to investigate the association between these patterns and the development of new carious lesions. The study tested the hypothesis that children with a high consumption of soft drinks, relative to milk and 100 percent fruit juice, over two years were at a higher risk of developing new carious lesions.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 
Study population. Data used in this analysis were collected as part of the Detroit Dental Health Project, a longitudinal study designed to investigate why some low-income black children and their caregivers have better oral health than others, even when socioeconomic and environmental conditions appear similar. 6,7,12–14 The rationale for this study is based on the hypothesis that understanding the determinants of intracommunity disparities provides better insights for designing interventions to promote oral health in the communities in which oral or dental disease levels are high.6,7,12–14

The sampling design consisted of a stratified two-stage area probability sample of households from the 39 census tracts with the highest proportion of low-income residents in the city of Detroit. To be eligible for the study, households needed to have a black child aged up to 5 years, and the family’s income needed to be lower than the 250th percentile of the federal poverty guidelines. Power calculations showed that we needed 994 families to complete data collection for the project. Among contacted households, 1,386 families were eligible. Of these families, 1,021 dyads of child and caregiver completed the baseline dental examination and interviews in 2002–2003 (response rate = 74.0 percent). Details of the sampling and data collection procedures are described elsewhere.7,14 At the two-year follow-up visit in 2004–2005, 790 dyads of child and caregiver (77.0 percent) returned to complete an interview and a clinical examination.

This study focused on children 3 years and older at baseline (n = 522) for whom follow-up data were collected (n = 403). We omitted from these analyses children who were 2 years and younger at baseline because no data regarding their beverage consumption were collected. We excluded 34 additional children because of implausible dietary intakes, leaving a final sample of 369 children.

The Institutional Review Board for Health Sciences at the University of Michigan, Ann Arbor, reviewed and approved the study each year. All caregivers or legal guardians signed consent forms for all children in this study.

Data collection. One of four dentists (W.S. and three others) examined the children in this study at the Dental Assessment Center in Detroit. They used a portable dental chair and halogen light.7 They cleaned the teeth of loose debris and dried them using an air syringe. The dentists assessed the caries status of all primary tooth surfaces using the International Caries Detection and Assessment System.15 The reliability of the examiners ranged from good to excellent ({kappa} coefficients = 0.59–0.82).16 In particular, the interexaminer reliability for the decision regarding the stage of dental caries was 0.63 to 0.78, indicating good reliability of the examiners in terms of identifying noncavitated surfaces.16

Caries severity. We counted the numbers of noncavitated surfac es (d1), cavitated/dentinal surfaces (d2), filled surfaces (f) and missing surfaces owing to caries (m) for each child. We computed the following four indexes to evaluate caries severity in the primary dentition: d1, d2, d2mfs (d2 + f + m) and dtmfs (d1 + d2 + f + m).14

We classified the few sealed surfaces detected as sound. We compared the caries status of each tooth surface at follow-up with the status at baseline and assessed caries increments for each child. We adjusted the increments of d2, f and m surfaces for reversals due to examiner misclassification by using the prevalence-based adjustment formula.17 We did not carry out any adjustment for new d1 scores, because noncavitated surfaces at baseline could regress naturally. The resulting outcomes included the increments of d1, d2, f and m surfaces, as well as d2mfs and dtmfs.

We collected dietary information for children 3 years and older by using the 2000 Block Kids Food Frequency Questionnaire (Kids FFQ) (NutritionQuest, Berkeley, Calif. [formerly Block Dietary Data Systems]). Trained interviewers administered the questionnaire in person. The reliability of the FFQ instrument is accepted widely as a result of its use in field studies.18,19

The Kids FFQ asked caregivers about the frequencies of consumption of a wide variety of food and drink items during the previous week and the usual amount of each item consumed in one day. We used the FFQ data to generate estimates of usual dietary intake, which is more representative of an individual’s general diet when compared with 24-hour recalls that provide information for only one day.20 Using an algorithm, the company that supplied the questionnaire quantified the collected data in grams (or milliliters for liquids). From these dietary outcomes, which were correlated highly with the frequencies of consumption (r = 0.77–0.94), we selected the beverage items and combined them to create three beverage variables (soft drinks, milk and 100 percent fruit juice) based on the American Diabetes Association and American Dietetic Association’s Exchange Lists for Meal Planning (Table 1Go).21


View this table:
[in this window]
[in a new window]

  		TABLE 1 Beverage items in Block Kids Food Frequency Questionnaire (FFQ).*

 
Solid foods such as candies also are known to be cariogenic,22 and we computed the total sugar intake for each child to control for the cariogenic potential of solid foods. Public water supplies in the city of Detroit are fluoridated at optimal levels (0.97 parts per million).23 Plain nonbottled water and soft drinks sweetened with low-calorie sweetener were not measured in the Kids FFQ.

Definition of clusters according to beverage intake. We used cluster analysis to define and evaluate the best grouping of children on the basis of their similarities in beverage intake. We determined the total beverage intake for each child by aggregating soft drinks, milk and 100 percent juice. The primary variable used in the cluster analysis was the proportion of intake of each beverage relative to the total beverage intake. This method was data-driven, so given the correlation among the intakes of the three beverages, cluster analysis should be more valid than a subjective classification.

In this study, we used a statistical technique known as "partitioning around medoids" (PAM) in R software.24 This method is more robust than the conventional k-means method because it does not require an assumption that the data are normally distributed.25 In addition, PAM, unlike the k-means method, does not require initial guesses of the cluster center, and it is not affected by outliers.25 The statistical program identified a median point (medoid) where a sum of the unsquared distance between each data point of the three beverages and that point is minimized. The software assigned each child, on the basis of his or her beverage intake, to the cluster corresponding to its nearest medoid. This analysis identified three clusters of children at baseline and two clusters of children at follow-up.

We found that the clusters defined by PAM had some overlap on the basis of the silhouette coefficient, an estimate of the quality of the clustering structure.25 This finding was not unexpected given the sociodemographic homogeneity of the sampled children.

Four intake patterns. Using the clusters at both time points, we defined four intake patterns across two years. Children were classified with regard to whether they had high intake of milk or 100 percent fruit juice at both waves (that is, baseline and follow-up) (high-high milk-juice), had high intake of soft drinks at both waves (high-high soft drinks) or switched between high intake of milk or 100 percent juice and that of soft drinks (low-high soft drinks, high-low soft drinks).

Utilization of dental services. We received Medicaid dental records for children whose care-givers consented to the release of these data (Medicaid records were not available for 31 children). Using the Medicaid data, we identified three types of children’s dental visits during the two-year period:

– no dental visit (n = 118; 32.0 percent) or no record in the Medicaid database (n = 31; 8.4 percent);
– preventive care–only visit (n = 129; 35.0 percent);
– restorative/surgical visit (n = 91; 24.7 percent).

Statistical analysis. We used the {chi}2 test to evaluate unadjusted bivariate associations between the longitudinal intake patterns and demographic characteristics, as well as oral health behaviors. We used zero-inflated negative binomial regression (ZINB) to model the association between the longitudinal intake pattern and the increments of decayed (that is, d1 and d2), f or m tooth surfaces; this is the appropriate model for analysis of an outcome measure with a high proportion of zeros (that is, no new carious surfaces in this study).26 We evaluated the association after controlling for children’s age, total sugar intake, baseline caries and caregiver’s education and income, as well as for children’s oral health behaviors. We considered differences to be statistically significant at P < .05.

We conducted statistical analyses using software (STATA version 10) to account for the clustering effect from the complex sample design.27 We adjusted all analyses by using sample weights to account for unequal selection probabilities and nonresponse. We substituted imputed values for the small number of missing values (< 4 percent) on predictor variables by using the IVEware software.28


   RESULTS
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSION
 REFERENCES
 
At baseline, the cluster analysis identified three groupings of children on the basis of their consumption of milk, 100 percent fruit juice and soft drinks (Table 2Go). We labeled clusters as "milk," "100 percent fruit juice" and "soft drinks" on the basis of the largest percentage of total beverage intake. At baseline, 107 (29.0 percent) of the 369 children had 72.0 percent or more of their beverage consumption from soft drinks. At follow-up, the analysis identified two clusters labeled as "milk-100 percent fruit juice" and "soft drinks" (Table 3Go). Unlike the baseline pattern, children with high milk intake at follow-up also were high consumers of 100 percent fruit juice. The proportion of children with high consumption of soft drinks increased to 48.8 percent (180 of 369).


View this table:
[in this window]
[in a new window]

  		TABLE 2 Daily beverage intake for different clusters at baseline.

 

View this table:
[in this window]
[in a new window]

  		TABLE 3 Daily beverage intake for different clusters at two-year follow-up.

 
Table 4Go presents the distribution of children according to the four longitudinal intake patterns defined earlier. Of 369 children, 158 (42.8 percent) had a high-high milk-juice pattern, and 104 (28.2 percent) had a low-high soft drink pattern. In addition, 31 children (8.4 percent) had a high-low soft drink pattern and 76 (20.6 percent) had a high-high soft drink pattern.


View this table:
[in this window]
[in a new window]

  		TABLE 4 Daily beverage consumption for longitudinal intake patterns.

 
Table 5Go (page 964) presents the associations between the longitudinal intake pattern, demographic characteristics, total sugar intake, oral health behaviors and dental caries. We found significant differences between the four intake patterns with regard to the number of new f surfaces, dtmfs and d2mfs, as well as toothbrushing frequency. According to the questionnaire findings, children with high-high milk-juice and low-high soft drink patterns had a higher frequency of toothbrushing than did children with a high-high soft drink pattern. The results showed a higher prevalence of baseline caries in the high-high soft drink group compared with the other consumption groups.


View this table:
[in this window]
[in a new window]

  		TABLE 5 Associations between longitudinal intake patterns and covariates.

 
Of the 369 children, 160 (43.4 percent) had no new d2mfs. onsequently, the most appropriate statistical model for these data (that is, number of new carious lesions or dtmfs) is the ZINB, as described previously. Table 6Go (page 965) presents the outcome of the ZINB models for the four longitudinal patterns associated with caries increments after we adjusted for a large number of zero counts (such as no new carious lesions between baseline and follow-up), age, baseline caries, total sugar intake, caregiver’s income and education, and children’s oral health behaviors.


View this table:
[in this window]
[in a new window]

  		TABLE 6 Adjusted incidence rate ratio (IRR) for caries increment (new d2mfs* and new filled surfaces).{dagger}{ddagger}

 
We chose children with a high-high milk-juice pattern to be the reference group in the models. After adjusting for all other confounding factors, we found that children with a low-high soft drink pattern (low consumption at baseline but high consumption at follow-up) had a 1.75 times greater risk of developing new d2mfs and a 2.67 times greater risk of developing new f surfaces compared with children in the reference group. In contrast, children with a high-high soft drink pattern had a similar significant risk of developing new f surfaces, but not new d2mfs, as did children in the reference group. The differences associated with the longitudinal intake pattern with regard to the number of new d1, d2 and m surfaces were not statistically significant (data not shown).


   DISCUSSION
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSION
 REFERENCES
 
This study presents findings regarding the cariogenicity of sugared drinks from a two-year follow-up of a cohort of low-income black children in Detroit. Cariogenicity of sugared drinks is difficult to determine in human studies, and Burt and Ismail11 proposed definitions to guide human studies in this area. They defined "cariogenic potential" as the ability of foods to cause a significant drop in plaque pH, to cause demineralization of enamel or to result in more caries (than that without the food) in animals under controlled experimental conditions. They suggested that researchers should consider the frequency and manner of consumption along with the cariogenic potential to determine the "effective cariogenicity" of foods. Thus, effective cariogenicity of foods can be determined only in longitudinal studies such as the one described in this report.11

Study limitations. This study had several limitations. First, there were not enough children to achieve statistical significance for all of the outcome measures. Second, the Kids FFQ did not collect data regarding detailed eating patterns (for example, between meals) or intake of water. This absence of information prevented a detailed analysis of frequency of consumption and fluoride exposure. Accordingly, it also meant that our results could not be compared directly with those from other studies that reported the importance of frequent exposure to cariogenic foods. Third, to obtain data regarding the dietary intake of these young children, interviewers questioned the caregivers. Consequently, reporting accuracy could have been compromised as the children grew up and spent more time out of the home. To adjust for this potential reporting error, we used relative measures (for example, percentage of total beverage intake) instead of absolute amounts of beverages consumed, as an earlier study suggested.29

Another methodological difficulty is the large number of zero counts in measuring caries incidence; 57.0 percent of the children developed no new cavitated surfaces between baseline and follow-up. If we had used a Poisson model, the observed dispersion likely would have been underestimated, which would have resulted in large discrepancies between observed and expected counts.30 To model this skewed distribution, some authors have suggested the use of a zero-inflated Poisson or negative binomial model (ZINB).26,30 We also found that the ZINB model provided a better fit when the nonzero counts were overdispersed with respect to a Poisson distribution.31 After examining the distribution of nonzero counts, we decided to use a ZINB model.

Study strengths. The study also had a number of strengths. It consisted of a longitudinal design and included a clinical assessment, interviews and use of Medicaid dental care records. To test our hypothesis, we combined the data from examinations and interviews at baseline and follow-up with data from Medicaid dental records. The cluster analysis, which yielded some intake patterns of sugared drinks over time, has construct validity on the basis of the findings of the ZINB models. However, the intake patterns were not associated with demographic characteristics, which is not consistent with results of other studies of children’s beverage intake patterns.1,32 This finding probably reflects the sociodemographic homogeneity of the study population.

This study found that low-income black children with a high intake of soft drinks over time, or those who changed from low to high consumption of soft drinks, had a higher risk of receiving new fillings than did those with a relatively higher intake of milk and 100 percent fruit juice at baseline and follow-up. We did not find any significant difference in the numbers of new noncavitated (d1) or cavitated (d2) tooth surfaces. Given that a majority of children (59.0 percent) had visited a dentist during the two years of the study, we interpret the significant increase in f surfaces in children in the low-high soft drink group as a rapid progression of caries to a stage at which restorative care is necessary. Caries progresses rapidly under conditions of high sugar exposure5; thus, symptoms of sensitivity and pain may have led caregivers to seek dental care for these children.

These results support the American Dietetic Association’s dietary guideline for children that discourages a high intake of soft drinks.33 While total elimination of soft drinks is unrealistic for most children, clinicians should recommend to caregivers that they limit children’s exposure to a few times per week, followed by toothbrushing with fluoridated dentifrice, in the interest of caries prevention.34

Milk. Milk is a good alternative to soft drinks because of its potential protective factors: calcium, phosphorus and proteose-peptone fractions 3 and 5.35 Milk has many of the physical and chemical properties of saliva, which could make it a suitable substitute for saliva.36 Studies with animals have concluded that bovine milk is noncariogenic,37,38 and several authors also have found evidence of the noncariogenic potential of bovine milk in human studies.1,3,8,9,39 One study contains a cluster analysis of data from the Third National Health and Nutrition Examination Survey, in which children with a high intake pattern of milk were found to have lower caries severity in the primary dentition.1 Another study is the longitudinal Iowa Fluoride Study, which found that routine milk consumption was protective in the primary dentition.39

Fruit juice. Kolker and colleagues6 conducted a cross-sectional study of the same population in this study, and they found that consumption of 100 percent fruit juice (excluding orange juice) was associated with a low prevalence and severity of caries. Other studies have reported that consumption of 100 percent fruit juice by young children was not associated with dental caries.3,22 It may be that high consumption of 100 percent fruit juice is a marker for more health-conscious caregivers in this population.6 In this study, we found a parallel between high consumption of milk and 100 percent fruit juice in the follow-up assessment (Table 3Go).


   CONCLUSION
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
REFERENCES
 
This two-year longitudinal study of low-income black children supports the hypothesis that children with a high consumption of soft drinks over time, or those who increase their consumption of soft drinks relative to consumption of milk and 100 percent fruit juice, are at a higher risk of developing dental caries. Soft drinks are potentially cariogenic; however, young children still can consume them more safely if the exposure is reduced and dental plaque is removed regularly with a fluoridated dentifrice. It is important for health care providers to emphasize in health promotion programs the need to maintain a balance between exposure to soft drinks and preventive oral hygiene behaviors.


   FOOTNOTES
 

At the time this study was conducted, Mr. Lim was a research analyst, University of Michigan, School of Dentistry, Department of Cariology, Restorative Sciences and Endodontics, Ann Arbor. He now is a city research scientist, Bureau of Epidemiology Services, New York City Department of Health and Mental Hygiene, 125 Worth St., Room 315, CN-6, New York, N.Y. 10013, e-mail "lsw1227{at}gmail.com". Address reprint requests to Mr. Lim.


Dr. Sohn is an assistant professor, Department of Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, Ann Arbor.


Dr. Burt is a professor emeritus, Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor.


Dr. Sandretto is a lecturer, Department of Environmental Health Sciences, School of Public Health, University of Michigan, Ann Arbor.


Dr. Kolker is an assistant professor, Department of Operative Dentistry, College of Dentistry, University of Iowa, Iowa City.


Dr. Marshall is an assistant professor, Department of Preventive and Community Dentistry, College of Dentistry, University of Iowa, Iowa City.


Dr. Ismail is a professor, Department of Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, Ann Arbor.


Disclosure. The authors did not report any disclosures.


This study was supported by grant number U-54 DE 14261 from the National Institute of Dental and Craniofacial Research, Bethesda, Md., the University of Michigan, Ann Arbor, and the Delta Dental Fund of Michigan.


   REFERENCES
TOP
 ABSTRACT
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSION
 REFERENCES
 

  1. Sohn W, Burt BA, Sowers MR. Carbonated soft drinks and dental caries in the primary dentition. J Dent Res 2006;85(3):262–266.[Abstract/Free Full Text]

  2. Marshall TA, Eichenberger-Gilmore JM, Broffitt B, Stumbo PJ, Levy SM. Diet quality in young children is influenced by beverage consumption. J Am Coll Nutr 2005;24(1):65–75.[Abstract/Free Full Text]

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

  4. French SA, Lin BH, Guthrie JF. National trends in soft drink consumption among children and adolescents age 6 to 17 years: prevalence, amounts, and sources, 1977/1978 to 1994/1998. J Am Diet Assoc 2003;103(10):1326–1331.[Medline]

  5. Selwitz RH, Ismail AI, Pitts NB. Dental caries. Lancet 2007; 369(9555):51–59.[Medline]

  6. Kolker JL, Yuan Y, Burt BA, et al. Dental caries and dietary patterns in low-income African-American children. Pediatr Dent 2007;29(6):457–464.[Medline]

  7. Burt BA, Kolker JL, Sandretto AM, Yuan Y, Sohn W, Ismail AI. Dietary patterns related to caries in a low-income adult population. Caries Res 2006;40(6):473–480.[Medline]

  8. Levine RS. Milk, flavoured milk products and caries. Br Dent J 2001;191(1):20.[Medline]

  9. Sheiham A. Dietary effects on dental diseases. Public Health Nutr 2001;4(2B):569–591.[Medline]

  10. Tinanoff N, Palmer CA. Dietary determinants of dental caries and dietary recommendations for preschool children. J Public Health Dent 2000;60(3):197–206.[Medline]

  11. Burt BA, Ismail AI. Diet, nutrition, and food cariogenicity. J Dent Res 1986;65(special issue):1475–1484.

  12. Finlayson TL, Siefert K, Ismail AI, Sohn W. Maternal self-efficacy and 1–5-year-old children’s brushing habits. Community Dent Oral Epidemiol 2007;35(4):272–281.[Medline]

  13. Sohn W, Ismail A, Amaya A, Lepkowski J. Determinants of dental care visits among low-income African-American children. JADA 2007;138(3):309–318.[Abstract/Free Full Text]

  14. Ismail AI, Sohn W, Tellez M, Willem J, Betz J, Lepkowski J. Risk indicators for dental caries using the International Caries Detection and Assessment System (ICDAS). Community Dent Oral Epidemiol 2007;36(1):55–68.

  15. Pitts N. ICDAS: an international system for caries detection and assessment being developed to facilitate caries epidemiology, research and appropriate clinical management. Community Dent Health 2004;21(3):193–198.[Medline]

  16. Ismail AI, Sohn W, Tellez M, et al. The international caries detection and assessment system (ICDAS): an integrated system for measuring dental caries. Community Dent Oral Epidemiol 2006;345(3): 170–178.

  17. Beck JD, Lawrence HP, Koch GG. A method for adjusting caries increments for reversals due to examiner misclassification. Community Dent Oral Epidemiol 1995;23(6):321–330.[Medline]

  18. Block G, Thompson FE, Hartman AM, Larkin FA, Guire KE. Comparison of two dietary questionnaires validated against multiple dietary records collected during a 1-year period. J Am Diet Assoc 1992;92(6):686–693.[Medline]

  19. Hartman AM, Block G, Chan W, et al. Reproducibility of a self-administered diet history questionnaire administered three times over three different seasons. Nutr Cancer 1996;25(3):305–315.[Medline]

  20. Lee RD, Nieman DC. Nutritional assessment. 2nd ed. St. Louis: Mosby; 1996:91–141.

  21. American Diabetes Association, American Dietetic Association. Exchange Lists for Meal Planning. Alexandria, Va. and Chicago: American Dietetic Association and American Diabetes Association; 1995.

  22. Marshall TA, Broffitt B, Eichenberger-Gilmore J, Warren JJ, Cunningham MA, Levy SM. The roles of meal, snack, and daily total food and beverage exposures on caries experience in young children. J Public Health Dent 2005;65(3):166–173.[Medline]

  23. 2006 Water Quality Report, Detroit Water and Sewerage Department; 2007. "www.dwsd.org/cust/wqr_2006.pdf". Accessed May 13, 2008.

  24. R Development Core Team. A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2005.

  25. Struyf A, Hubert M, Rousseeuw PJ. Integrating robust clustering techniques in S-PLUS. Computational Statistics Data Analysis 1997;26(1):17–37.

  26. Bulsara MK, Holman CD, Davis EA, Jones TW. Evaluating risk factors associated with severe hypoglycaemia in epidemiology studies: what method should we use? Diabet Med 2004;21(8):914–919.[Medline]

  27. Stata Statistical Software. Version 10. College Station, Texas: StataCorp; 2007.

  28. Raghunathan TE, Lepkowski, JM, Van Hoewyk J, Solenberger P. A multivariate technique for multiply imputing missing values using a sequence of regression models. Survey Method 2001;27(1):85–95.

  29. Wang Y, Bentley ME, Zhai F, Popkin B. Tracking of dietary intake patterns of Chinese from childhood to adolescence over a six-year follow-up period. J Nutr 2002;132(3):430–438.[Abstract/Free Full Text]

  30. Böhning D, Ekkehart D, Schlattmann P, Mendonça L, Kircher U. The zero-inflated Poisson model and the decayed, missing and filled teeth index in dental epidemiology. J Royal Statistical Soc 1999;162(series A):195–209.

  31. Lewsey JD, Thomson WM. The utility of the zero-inflated Poisson and zero-inflated negative binomial models: a case study of cross-sectional and longitudinal DMF data examining the effect of socioeconomic status. Community Dent Oral Epidemiol 2004;32(3):183–189.[Medline]

  32. LaRowe TL, Moeller SM, Adams AK. Beverage patterns, diet quality, and body mass index of US preschool and school-aged children. J Am Diet Assoc 2007;107(7):1124–1133.[Medline]

  33. Nicklas T, Johnson R, American Dietetic Association. Position of the American Dietetic Association: dietary guidance for healthy children ages 2 to 11 years [published correction appears in J Am Diet Assoc 2004;104(7):1075]. J Am Diet Assoc 2004;104(4):660–677.[Medline]

  34. Featherstone JD. The caries balance: the basis for caries management by risk assessment. Oral Health Prev Dent 2004;2(suppl 1): 259–264.[Medline]

  35. Grenby TH, Andrews AT, Mistry M, Williams RJ. Dental caries-protective agents in milk and milk products: investigations in vitro. J Dent 2001;29(2):83–92.[Medline]

  36. Samarawickrama DY. Saliva substitutes: how effective and safe are they? Oral Dis 2002;8(4):177–179.[Medline]

  37. Bowen WH, Lawrence RA. Comparison of the cariogenicity of cola, honey, cow milk, human milk, and sucrose. Pediatrics 2005;116(4):921–926.[Abstract/Free Full Text]

  38. Peres RC, Coppi LC, Franco EM,Volpato MC, Groppo FC, Rosalen PL. Cariogenicity of different types of milk: an experimental study using animal model. Braz Dent J 2002;13(1):27–32.[Medline]

  39. Levy SM, Warren JJ, Broffitt B, Hillis SL, Kanellis MJ. Fluoride, beverages and dental caries in the primary dentition. Caries Res 2003;37(3):157–165.[Medline]





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lim, S.
Right arrow Articles by Ismail, A. I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lim, S.
Right arrow Articles by Ismail, A. I.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS