JDR Woodhead Publishing
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
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 (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Griffin, S.O.
Right arrow Articles by Zlobin, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Griffin, S.O.
Right arrow Articles by Zlobin, N.
J Dent Res 84(8):715-720, 2005
© 2005 International and American Associations for Dental Research

RESEARCH REPORT
Clinical

New Coronal Caries in Older Adults: Implications for Prevention

S.O. Griffin1,*, P.M. Griffin2, J.L. Swann2, and N. Zlobin2

1 Centers for Disease Control and Prevention/Division of Oral Health/Surveillance, Investigations, and Research Branch, 4770 Buford Highway, MSF10, Chamblee, GA 30341, USA; and
2 School of Industrial & Systems Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0205, USA;

* corresponding author, sig1{at}cdc.gov


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
To characterize the extent and severity of coronal caries among older US adults and document their need for prevention interventions, we systematically reviewed studies on coronal caries incidence, increment, and attack rate. We abstracted six studies and calculated summary measures using a random-effects model (95% confidence interval [95%CI]). We tested for heterogeneity and identified associated factors by examining the correlation between outcome measures and baseline population risk and study characteristics. We re-calculated summary measures after adjusting outcomes that netted out examiner reversals. Incidence and increment varied significantly by study. Adjusting studies for netting out examiner reversals reduced heterogeneity significantly. Annual attack rate among adjusted North American studies was 1.4 surfaces per 100 surfaces (95%CI = 1.0–1.9), or approximately 1 new carious surface per person per year. These rates are equal to or higher than those in children and indicate a need for caries-prevention services.

KEY WORDS: coronal caries • incidence • increment • attack rate • adults


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The purpose of this research was to derive estimates of new dental disease in older adults (aged 60+ yrs) using standard tools of meta-analysis. Documenting new caries among older adults is necessary to determine both the potential impact and cost-effectiveness of caries-prevention efforts. We have good evidence that interventions to avert dental caries are effective (Truman et al., 2002; Marinho et al., 2003). Although most trials measuring effectiveness have been conducted among younger populations, the available but limited evidence for adults suggests that effects are consistent with those expected effects based on experience with children (US Department of Health and Human Services, 2000).

Currently, water fluoridation is the main caries-prevention program that affects older adults. Other common primary prevention programs delivered at the community level—school-based/-linked sealant programs and delivery of fluoride via mouthrinse, supplement, or topical applications—primarily target children (ASTDD, 2002). Medicaid (a Federally funded, state-operated and -administered program that provides medical benefits to indigent or low-income persons) frequently covers only emergency dental care for adults (Oral Health America, 2003), and Medicare (a Federal health insurance program for the elderly) does not cover routine dental services.

These interventions may be able to prevent significant amounts of dental caries in older adults. While national surveys provide cross-sectional data to estimate the prevalence and severity of caries, at present, we have no longitudinal estimates of new caries at the national level. However, at the state and community levels, longitudinal data for older adults suggest that they may have more new caries than children (Hand et al., 1988a). Two previous articles estimated new dental caries among older adults by averaging the findings from some of these studies (Garcia, 1989; Leake, 2001). Neither article, however, took study variance into account in deriving summary measures, or examined if outcome measures were homogenous across studies.

Finally, demographic trends suggest that the number of older adults will continue to increase relative to the general population (Administration on Aging, 2000), and that these older adults are more likely to retain their teeth than were previous generations (US Department of Health and Human Services, 2000).


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Search for Studies
We searched MEDLINE and EMBASE for studies on new dental caries conducted after 1980 among adult populations, published in English (Appendix Table 1). We also searched bibliographies of retrieved articles; we did not, however, hand-search journals, search for unpublished studies, or contact authors of published studies for additional data. We used the following to measure new caries: (1) incidence (proportion of individuals developing caries); (2) increment (average number of tooth surfaces per person developing caries); and (3) attack rate (proportion of at-risk surfaces developing caries). We reviewed titles and abstracts and ordered articles with study populations older than 40 yrs living in established market economies (Appendix Fig.). The search strategy and data abstraction process (Appendix Table 2) have been described previously (Griffin et al., 2004).

Inclusion Criteria
We included studies with populations whose mean age was greater than 60 yrs, conducted among populations with characteristics similar to those of the aggregate US population (i.e., we excluded findings from studies conducted exclusively among populations with other diseases such as cancer, who were institutionalized, or were treatment groups in caries-prevention trials). Included studies had to be of sufficient design quality—prospective, cohort studies that used probability sampling of a general population. Examiner reliability had to be reported, and diagnostic criteria had to be specified. We excluded studies that did not report sufficient information for a variance to be calculated. Six studies consisting of 3113 older adults (mean age = 66.5 yrs) met our inclusion criteria (TableGo).


View this table:
[in this window]
[in a new window]
  		Table. Characteristics of Six Studies Used to Estimate Rates of New Caries among Older Adults
 
Data Adjustment
Some studies reported outcome measures for the same population for different time intervals (i.e., 18 and 36 mos). Because, in some of the studies, individuals who dropped out had significantly higher levels of disease at baseline (Hand et al., 1988b; Hawkins et al., 1997; Gilbert et al., 2000; Thomson et al., 2002), we reported outcome measures for the shortest study interval, to minimize potential bias resulting from the attrition of study subjects. For studies reporting caries increment or attack rate for periods greater than a year, we assumed that the outcome measure was independently and identically distributed for each year; thus, we calculated the annual measure by dividing the reported measure by the number of years in the reported interval, and estimated the annual standard error by dividing the reported standard error for the interval by the square root of the interval years.

To estimate annual incidence, we assumed that the probability of experiencing caries was the same for each study year. We first calculated the probability that no disease was incurred over the study interval, and then took the nth root of this value (where n represents number of yrs in study) to calculate the probability that no disease was incurred in a given year. We then subtracted this value from 1 to obtain the annual incidence.

Some studies reported separate measures for specific groups. For these studies, we estimated the outcome measure (standard error) for the population by taking the weighted average (standard error = {surd} na*sea2 + nb*seb2/na + nb) of the reported groups.

We excluded studies that reported increment or attack rate but did not report a variance, standard deviation, or standard error. For studies reporting incidence but not a standard error, we calculated the standard error with the following formula: (standard error = {surd} incidence*(1 – incidence)/n).

Most studies did not include crowns in their outcome measures for disease. For studies that reported increment or attack rate including and excluding crowns, we used the outcome measure, which excluded crowns.

To make the findings of studies that adjusted for examiner reversals more comparable with the results of studies that did not, we reported, when possible, crude estimates of new disease that did not net out negative reversals. Netting out negative reversals may underestimate new disease (Burt and Eklund, 1999). Two studies comparing the two methods found that the net method resulted in more biased estimates of new caries in older adults than did the crude method. One study found that using the crude instead of the net estimation method increased estimates of incidence, increment, and attack rate by 44%, 119%, and 124%, respectively (Slade and Caplan, 2000), while, similarly, the other found that crude increment estimates were 121% higher (Beck et al., 1995). We also calculated estimates of new disease adjusted for analytic convention by inflating net measures by the above percentages.

Analysis
We calculated the 95% confidence intervals for each outcome measure reported in each study. To estimate summary measures across studies, we used a random-effects model, which assumes that each study was randomly selected from a hypothetical population of studies [DerSimonian and Laird method (Normand, 1999)]. We calculated summary measures (unadjusted and adjusted for netting out examiner reversals) for all studies and for North American studies.

To determine if the dental health status of the study populations was similar to that of the US population as a whole, we compared the mean number of teeth present and mean number of decayed and filled coronal surfaces among the study populations at baseline (1981 to 1997) against the national estimates from the National Health and Nutrition Examination Survey (NHANES) III (1988–1991) (Winn et al., 1996) for similar age groups. Using a chi-square test (QW) (Normand, 1999), we tested for homogeneity of effect size (new disease) among studies. For effect sizes failing the heterogeneity test, we estimated the quantity I2, which describes the percentage of total variation, across studies, due to heterogeneity vs. chance (Higgins et al., 2003). Thus, if study populations at baseline had dental health status similar to that of the US population, and if new disease outcomes are homogenous across studies, our findings may be applicable to the US population.

Finally, to identify potential sources of heterogeneity, we examined the correlation coefficients between each outcome measure and: (1) population baseline risk factors [measured by age and number of decayed, missing, and filled tooth surfaces (DMFS)] and (2) study design factors (measured by study length, whether the examination included 3rd molars, and whether it netted out examiner reversals) (Appendix Table 3).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Among all studies, the annual incidence estimate was 31.9% (95%CI, 23.3.0%–40.5%; heterogeneity present) (Fig. 1Go). Among North American studies (n = 4), the estimate was 33.4 (95%CI, 25.1%–41.7%; heterogeneity present). Among all studies, incidence was negatively correlated with study length (p = 0.05) and netting out negative reversals (p < 0.09) (Appendix Table 4). When we regressed incidence onto study length and on whether the study netted out examiner reversal, both variables were significant at p < 0.03 (Adjusted R2 = 0.90). After adjustment, the estimates were 39.8% (95%CI, 33.1.0%–46.5%; heterogeneity present) for all studies and 43.3% (95%CI, 36.9%–49.7%; heterogeneity present) for North American studies. Although heterogeneity remained after adjustment for netting out examiner reversals, QW for the adjusted studies was about one-third that of the unadjusted studies.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 1. Mean annual coronal caries incidence and 95% confidence interval (not adjusted for analytic convention criteria) for six studies used in analysis. Vertical lines indicate summary estimate and 95% confidence interval.

 
Among all studies, the estimate of annual coronal caries increment was 0.86 surfaces (95%CI, 0.66–1.07; heterogeneity present) (Fig. 2Go). Summary annual coronal caries increment among North American studies was quite similar: 0.84 surfaces (95%CI, 0.57–1.11; heterogeneity present). Heterogeneity was much lower for estimates of increment than for incidence; QW for unadjusted incidence in all studies was about 5 times higher. Increment was not significantly associated with either length or netting out examiner reversals. Adjusting studies that netted out examiner reversals decreased heterogeneity only slightly; QW was about 1.2 times higher for unadjusted studies compared with those adjusted for netting out examiner reversals for both all and North American studies. The summary measures for adjusted values from all and North American studies were 1.16 surfaces (95%CI, 0.91–1.41) and 1.22 surfaces (0.86–1.58), respectively.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Mean annual coronal caries increment and 95% confidence interval (not adjusted for analytic convention) for six studies used in analysis. Vertical lines indicate summary estimate and 95% confidence interval.

 
Among all studies, the annual caries attack rate equaled 1.1% (95%CI, 0.6%–1.6%; heterogeneity present; Fig. 3Go). Among North American studies, the estimate equaled 0.9% (95%CI, 0.5%–1.3%; heterogeneity present). Among adjusted studies, the estimate was 1.6% (95%CI, 1.2%–2.1%; heterogeneity present) for all studies and 1.4% for North American studies (95%CI, 1.0%–1.9%). Although attack rate was not associated with any of the baseline risk or study design variables (Appendix Table 4), QW was approximately 2.7 times higher for the unadjusted studies than for the studies adjusted for netting out examiner reversals, for both all and North American studies. In fact, for the adjusted North American studies, the Q statistic indicated that we could not reject the null hypothesis of homogenous outcome measures; the I2 statistic, however, indicated that 67% of the variation across studies was attributable to heterogeneity vs. chance. Among all adjusted studies, Australia accounted for a substantial portion of the heterogeneity (about 50% of the variation). Among studies defining what constituted an at-risk surface, Australia was the only study to exclude decayed or filled uncrowned coronal surfaces.



View larger version (14K):
[in this window]
[in a new window]
  		Figure 3. Mean annual coronal caries attack rate and 95% confidence interval for six studies used in analysis. Attack rates shown are not adjusted for analytic convention, and only sound surfaces were designated as being at risk when attack rate was calculated. Vertical lines indicate summary estimate and 95% confidence interval.

 
It appears that, among the study populations included in this review, baseline mean DMFS was slightly lower than levels reported for similar age groups in NHANES III (Appendix Table 5). The baseline number of retained teeth among the studies included in our review, excepting Florida and perhaps Australia (because the study did not report the standard error for this value, we know only that the reported value is either the same or lower), was similar to estimates reported for the US among similar age groups. Baseline mean DFS was similar in Florida, less in Iowa and Australia, and either less than or similar in Canada and North Carolina.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Because of heterogeneity, summary effect measures should be interpreted with caution. The I2 statistic exceeded 70% for all summary effect measures, with the exception of attack rate among adjusted North American studies. For these studies, we found that, on average, 1.43% of at-risk surfaces become carious each year. The mean DMFS for these studies was 59.75; assuming no recurrent caries, the average number of new carious surfaces after one year would thus be 0.98 [= 0.0143*(128-59.75)]. These findings suggest that new decay among older adults is similar to or higher than that in children. A review of studies estimating annual caries increment among children found that children living in fluoridated and fluoride-deficient communities experienced, on average, 0.8 and 1.4 new decayed tooth surfaces per year, respectively (Garcia, 1989). The review included controls from fluoridated toothpaste clinical trials and studies from the late 1970s, and thus may not be directly comparable with our analysis of more recent data from general populations. Using published values of DMFS among children aged 12 to 18 yrs as reported in the NHANES III (Kaste et al., 1996), and assuming a constant caries increment, we imputed an annual caries increment of 0.50 surfaces (Appendix Table 6).

Caries among the elderly is also more likely to remain untreated; their mean number of untreated decayed surfaces (1.6) (Winn et al., 1996) is about four times that among US schoolchildren (0.4) (Kaste et al., 1996). This may be partly attributable to the limited ‘safety net’ available to provide routine dental care to adults—lower-income adults are about 6 times more likely to have untreated decay than their higher-income counterparts (Drury et al., 1999). Untreated decay reflects a need not only for treatment services, but also for preventive interventions.

We found that study design characteristics were more likely to contribute to heterogeneity than were differences in baseline risk. None of the outcome measures was correlated with baseline risk variables, suggesting that risk differences did not contribute to heterogeneity among outcomes. This may be because neither baseline age nor DMFS differed greatly by study; the respective coefficients of variation were 0.07 and 0.09. Only two study design variables were correlated with outcome measures and thus may have contributed to heterogeneity—whether the study netted out examiner reversals, and the study duration (time to first follow-up).

We adjusted the data to control for netting out examiner reversals; however, we could not do so completely for study duration. Study duration influences the outcome measure in at least two ways. First, subjects with higher levels of baseline disease may be more likely to drop out as study duration increases. To control for this effect, we used the shortest reported study interval. Second, the probability of recurrent caries should increase as duration increases; thus, the probability that the outcome measure is truly cumulative decreases. Incidence would be especially sensitive to increased study duration, since it is measured at the person, not the tooth, level. This may be why Sweden, where the study lasted two years longer than any of the other studies, accounted for 60% of the variation in mean incidence when the Q statistic was calculated for all adjusted studies. Our hypothesis—that confounding due to study duration will be greater for incidence than for increment—is also supported by the higher Q statistic for the adjusted summary measure for all incidence studies than for all increment studies (2.5 times higher).

Our findings suggest that older adults experience caries at rates equal to or greater than those in children, the primary recipients of caries-prevention services. For quantification of the benefit of prevention among older adults, further research is needed on the effectiveness of prevention—two recent reviews found few studies conducted among adult populations (Rozier, 2001; Truman et al., 2002)—and treatment costs, which may be substantially higher if adults with more affected surfaces require procedures such as crowns to treat new decay.


   ACKNOWLEDGMENTS
 
We thank Brenda Mazzocchi, CDC Reference Librarian, who conducted all literature searches, Dr. Barbara Gooch for comments made during the preparation of this manuscript, and the National Science Foundation (ESS Award #0223364).


   FOOTNOTES
 
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org.

Received June 6, 2004; Last revision February 18, 2005; Accepted April 25, 2005


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Administration on Aging (2000). Statistical information on older persons [accessed July, 2003]. Available from http://www.aoa.gov/prof/Statistics/statistics.asp

Association of State and Territorial Dental Directors and Centers for Disease Control and Prevention (2002). Synopses of state and territorial dental public health programs [accessed August, 2003]. Available from http://www2.cdc.gov/nccdphp/doh/synopses/index.asp

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

Burt BA, Eklund SA (1999). Dentistry, dental practice, and the community. 5th rev. ed. Philadelphia, PA: W.B. Saunders Company.

Drake CW, Hunt RJ, Beck JD, Koch GG (1994). Eighteen month coronal caries incidence in North Carolina older adults. J Public Health Dent 54:24–30.[Medline]

Drury TF, Redford M, Garcia AI, Adesanya M (1999). Identifying and estimating oral health disparities among U.S. adults [accessed July, 2003]. Available from http://www.nidr.nih.gov/research/healthDisp/disp-estimation.pdf.

Fure S (1997). Five-year incidence of coronal and root caries in 60-, 70-, and 80-year old Swedish individuals. Caries Res 31:249–258.[ISI][Medline]

Fure S, Zickert I (1997). Incidence of tooth loss and dental caries in 60-, 70-, and 80-year-old Swedish individuals. Community Dent Oral Epidemiol 25:137–142.[ISI][Medline]

Garcia AI (1989). Caries incidence and costs of prevention programs. J Public Health Dent 49(5 Suppl):259–271.[ISI][Medline]

Gilbert GH, Foerster U, Dolan TA, Duncan RP, Ringelberg ML (2000). Twenty-four month coronal caries incidence: the role of dental care and race. Caries Res 34:367–379.[Medline]

Griffin SO, Griffin PM, Swann JL, Zlobin N (2004). Estimating rates of new root caries in older adults. J Dent Res 83:634–638.[Abstract/Free Full Text]

Hand JS, Hunt RJ, Beck JD (1988a). Incidence of coronal and root caries in an older adult population. J Public Health Dent 48:14–19.[ISI][Medline]

Hand JS, Hunt RJ, Beck JD (1988b). Coronal and root caries in older Iowans: 36-month incidence. Gerodontics 4:136–139.[ISI][Medline]

Hand JS, Hunt RJ, Kohout FJ (1991). Five-year incidence of tooth loss in Iowans aged 65 and older. Commmunity Dent Oral Epidemiol 19:48–51.

Hawkins RJ, Jutai DK, Brothwell DJ, Locker D (1997). Three-year coronal caries incidence in older Canadian adults. Caries Res 31:405–410.[Medline]

Heft MW, Gilbert GH (1991). Tooth loss and caries prevalence in older Floridians attending senior activity centers. Commmunity Dent Oral Epidemiol 19:228–232.

Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003). Measuring inconsistency in meta-analyses. BMJ 327:557–560.[Free Full Text]

Hunt RJ, Eldredge JB, Beck JD (1989). Effect of residence in a fluoridated community on the incidence of coronal and root caries in an older adult population. J Public Health Dent 49:138–141.[ISI][Medline]

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

Leake JL (2001). Clinical decision-making for caries management in root surfaces. J Dent Educ 65:1147–1153.[Abstract]

Locker D, Leake JL (1993). Coronal and root decay experience in older adults in Ontario Canada. J Public Health Dent 53:158–164.[ISI][Medline]

Marinho VCC, Higgins JPT, Logan S, Sheiham A (2003). Fluoride varnishes for preventing dental caries in children and adolescents. The Cochrane Library 2 [accessed Aug 2003]. Available from http://www.cochranelibrary.com.

Normand SL (1999). Meta-analysis: formulating, evaluating, combining, and reporting. Stat Med 18:321–359.[ISI][Medline]

Oral Health America (2003). Keep America Smiling: 2003 Oral Health Report Card. 2003, Chicago IL [accessed July 2003]. Available from http://www.oralhealthamerica.org/pdf/2003ReportCard.pdf

Rozier RG (2001). Effectiveness of methods used by dental professionals for the primary prevention of dental caries. J Dent Educ 65:1063–1072.[Abstract]

Slade GD, Caplan DJ (2000). Impact of analytic conventions on outcome measures in two longitudinal studies of dental caries. Community Dent Oral Epidemiol 28:202–210.[ISI][Medline]

Slade GD, Spencer AJ (1997). Distribution of coronal and root caries experience among persons aged 60+ in South Australia. Aust Dent J 42:178–184.[ISI][Medline]

Thomson WM, Spencer AJ, Slade GD, Chalmers JM (2002). Is medication a risk factor for dental caries among older people. Community Dent Oral Epidemiol 30:224–232.[Medline]

Truman BI, Gooch BF, Sulemana I, Gift HC, Horowitz AM, Evans CA, et al. (2002). Reviews of evidence on interventions to prevent dental caries, oral and pharyngeal cancers, and sports-related craniofacial injuries. Am J Prev Med 23:21–54.[ISI][Medline]

US Department of Health and Human Services (2000). Oral health in America: a report of the Surgeon General. Rockville, MD: US Department of Health and Human Services, National Institute of Dental and Craniofacial Research, National Institutes of Health, p.308.

Winn DM, Brunelle JA, Selwitz RH, Kaste LM, Oldakowski RJ, Kingman A, et al. (1996). Coronal and root caries in the dentition of adults in the United States, 1988–1991. J Dent Res 75(Spec Iss):642–651.




This article has been cited by other articles:


Home page
 J. Dent. Res.Home page
S.O. Griffin, E. Regnier, P.M. Griffin, and V. Huntley
Effectiveness of Fluoride in Preventing Caries in Adults
J. Dent. Res., May 1, 2007; 86(5): 410 - 415.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
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 (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Griffin, S.O.
Right arrow Articles by Zlobin, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Griffin, S.O.
Right arrow Articles by Zlobin, N.

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