Printer Friendly

Can caries in the primary dentition be used to predict caries in the permanent dentition? An analysis of longitudinal individual data from 3-19 years of age in Sweden.


The most common oral disease in childhood and adolescence is dental caries. This is a significant public health concern because this disease, and its effects, has an impact on most people as early restorations have to be replaced regularly. This means that there is still a great need to find factors that can help predict future caries disease. Due to the large number of teeth which have to be treated for decay, there should be important health economic benefits from such discoveries. The Swedish society's total dental expenses in 2005 amounted to approximately 21 billion SEK (approximately 2.5 billion [euro]) [Regeringskansliet, 2007]. This cost included all dental treatments for all groups of patients, not only for caries, but also for orthodontics, periodontal treatment, oral surgery, etc.

Registered cases of caries have decreased dramatically in the Swedish population over the last 50 years, particularly up to the turn of the millennium, according to the data reported to the National Board of Health and Welfare (NBHW) [Socials-tyrelsen, 2010]. The reason for this change is not fully clear and remains a question. Is dental care mainly responsible for this improvement or are there other causes, e.g. general lifestyle changes, social improvements or changes in the diagnostic criteria [Petersen, 2003a]? A summary of the number of fillings in schoolchildren in Gothenburg illustrates these questions (Fig. 1). It shows a clear relationship between dental caries and other factors such as sugar rationing during both world wars. The continual decline from the 1960s can be attributed to several factors beyond dentistry, such as limiting candy consumption by children, fluoridated toothpastes and fluoride rinses in schools [Watt and Sheiham, 1999]. Moreover, oral hygiene has improved over the years, and the frequency and methods of brushing have changed [Hugoson et al., 2005].


There are still many gaps in our knowledge, and much of our knowledge of caries applies only to particular populations [Reisine and Psoter, 2001]. Harris et al. [2004] also noted in a literature review of caries-risk factors in children that most longitudinal studies that exist were performed in a Scandinavian setting, i.e. in a population with organised paediatric dentistry. A setting with organised care is a prerequisite to ensure universality in a study population. A longitudinal study design is also necessary if risk factors are to be identified or predictors for caries are to be implemented on a solid basis [Harris et al., 2004].

We know from other longitudinal studies that caries are stable from adolescence up to early adulthood and also further up to the age of 41 years [Mejare et al., 2004; Crossner and Unell, 2007]. Data also suggests that those groups of individuals who have caries lesions also are at increased risk for future caries, as several studies show that the single best indicator for the risk of tooth decay is pre-existing caries [Powell, 1998; Zero et al., 2001]. However, there is limited knowledge about the longterm correlation between caries in the primary dentition and caries in the permanent dentition on an individual level. The literature does not contain any studies with a large population that was followed from caries registration at an individual level over a long period of time, i.e. between three and 19 years of age. Based on these gaps of knowledge, there is good reason to study the relationship between the frequency of active caries disease in the primary and permanent dentitions.

The aim of this study was to investigate if active caries disease in the primary dentition can be used as a long-term predictor of active caries disease in the permanent dentition. It was hypothesised that those with active caries disease in the primary dentition at three and six years of age would have active caries disease in the permanent dentition at 19 years of age.

Material and methods

Ethical approval The study was approved by the Ethics Committee in Uppsala in 2012 (Dnr 2012/203).

Population and response rate Swedish dental care provides regular and comprehensive care to all citizens up to the age 20 years. The care is free of charge and includes preventive and specialist care [Socialstyrelsen, 2011a]. Caries data used herein were retrieved from the Orebro County Council. All clinical examinations were conducted by uncalibrated dentists in clinics with modern equipment and in accordance with the County Council's written procedures. Radiographic examinations were performed for individual indications. Until 2006, the data were manually registered, but after 2006, were collected automatically from digital charts. The participants in the study were born in 1987. Individual data used here were collected at three, six and 19 years of age.

This study included only those living in Orebro county who came to the clinic for an examination at all three measuring points in 1990, 1993 and 2006 (n = 1,985, panel). The rest were termed non-panel subjects (n = 2,568). The panel consisted of 77.3% of the baseline group (1990) and represented 60.0% of the three-year-olds in the population in 1990. In 1990 and 1993, all dental care was provided within the public dental health care system. This changed in 2002, so the examination of the 19-year-olds may have been performed by private dentists working from the same written procedures.

Measures. Caries data were compared at the individual level and broken down into components of decayed, filled primary teeth (dft) and decayed surfaces (ds) (three years and six years of age) and decayed filled permanent teeth (DFT) and decayed surfaces (DS) (19 years of age). DFT (dft) was used as an indicator of caries experience and DS (ds) as an indicator of active caries disease. The missing (m, M) component in the index was not used, as extractions in the permanent dentition were usually due to orthodontic reasons in this age group. In the three-year-olds and six-year-olds, only the primary dentition was registered and from seven-year-olds, only the permanent teeth were evaluated; this is in accordance with the instructions from the NBHW [Socialstyrelsen, 2011b]. As a measure of the predictive ability of this measure, the sensitivity and specificity were calculated. Sensitivity is the ability to correctly detect the disease by a test and specificity indicates the ability to correctly detect healthy subjects using a test. Here, caries in the primary dentition was treated as the test and the outcome at 19 years of age was compared. A frequently-used limit for this type of test is that the sum of sensitivity and specificity should be greater than 160 [Zero et al., 2001].

Statistical methods. All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS, Release version 15). Bivariate analyses were conducted by cross-tabulation and Chi-squared statistics. Multivariate analyses were conducted using binary multiple logistic regression with categorical data.


When the panel group was compared with the non-panel group, the panel group had a lower percentage of active dental caries disease (ds/DS) (Table 2). There were no significant differences in terms of gender between the groups.

Table 3 shows the relationship between active caries disease (ds/DS) and caries experience (dft/DFT) at different ages. All correlations were weak or very weak. There was a very low correlation between active caries disease at six years of age and active caries disease at 19 years of age. A stronger correlation, but still weak, could be measured between dental caries experience at six years of age (primary dentition) and caries experience at 19 years of age (permanent dentition).

To investigate the relationships between active caries disease in the primary dentition and active caries disease at 19 years of age, a logistic regression was conducted (Table 4) with gender and ds at three and six years as the independent variables. The DS value at 19 years was used as the dependent variable. The result showed a very low level of correlation [r=0.017], even though there was a statistically significant OR for ds at six years of age.

A receiver operating curve (ROC curve) can be used to show the correlation between sensitivity and specificity. An ROC curve should have an area under the curve of at least 0.7 to be considered acceptably predictive [Vanobbergen et al., 2001]. The curve for these data had a value of 0.6.

Active caries disease in six-year-olds was used as a test to predict active caries disease in 19-year-olds. The calculations for the sensitivity and specificity of active caries disease (ds/DS) are shown in Table 5. Table 6 shows these values for caries experience (dft/DFT).


Earlier caries experience has been used and is still used as an important indicator of future caries activity and for the assessment of risk. The fact that there was a poor correlation in the long term could be related to unsuccessfully treated, still active caries disease. The frequently used measure dft/ DFT does not differentiate between active disease (D) and caries experience in the form of filled teeth (F).

This study showed a very weak individual correlation between active caries disease in the primary dentition at three and six years of age and active caries disease at 19 years of age (Table 3). It is possible that this poor correlation depends on adequate work by dentists, but it may also reflect individual changes over this rather long time period.

Using the individual data on the group level by treating active caries disease or caries experience at three years of age as a test for active caries disease or caries experience at 19 years of age showed low predictive power.

The frequently used and reported definitions of caries disease thus reflect the presence of signs of hard tissue damage to the teeth. What is recorded is not by definition caries disease, but rather varying degrees of demineralisation and the presence of restorations. Instead of caries disease, we have rather reported dental injuries. It is worth noting that the Swedish Council on Health Technology Assessment [Statens beredning for medicinsk utvardering, SBU 2007] in its caries report points out that "caries is a disease you can be free from". This is done without providing any clinically useful definition of when a sick person is free from disease. Neither caries reports from the NBHW or from SBU address the question of who is "sick" with the disease "caries" [SBU, 2007]. One problem with the NBHW definition, and the commonly used indices (DFMT or DFMS), is that previous injuries (F, M) are part of the disease measure. This means that an individual's previous disease will always be added to the present-day value for that individual. Health care data as the basis for reporting is also based on dentists' registrations. Registration is principally done for therapy and not for epidemiological purposes, which may compromise quality [Davies et al., 1996]. In addition, the registration of caries is mixed with other hard tissue defects such as erosive lesions and injuries from trauma. We have not found any analysis of how many people develop caries or how many people recover from this disease, as there are no uniform and accepted definitions of the disease on a personal level, only in terms of decay at the tooth or surface levels.

A weakness of this study is a lack of more background data on the study individuals. However, the material is unique in that it contains caries data from all individuals in this age group (born in 1987) that have been living permanently in this region from three to 19 years of age. In this way, the data provided an opportunity to follow an entire age group longitudinally over time and at an individual level. As always in a longitudinal study, there are also many unknown factors that may be significant to the results. One reason for the poor correlation between caries in three-year-olds and caries in 19-year-olds could be that dental care has actually been successful.

Previously published material has shown that socio-economical risk grouping by the cluster technique may be an effective way to predict caries [Ordell and Aronsson, 2002; Ekback and Persson, 2012]. This creates an opportunity to use interventions at a group level, while individual risk grouping may mainly target individual-related efforts. This study shows that the individual risk groups still lack good indicators to assess caries risk in the longer term. This supports WHO recommendations regarding intersectional action on common risk factors, at least in a Scandinavian context [Petersen 2003b].


The lack of a uniform definition of caries disease at the individual level is disturbing, and complicates any discussion of when an individual is free from active caries disease. The correlations between active caries disease in the primary dentition and active caries disease at 19 years of age were very low at an individual level. Moreover, using early disease as a predictive test for later disease showed low sensitivity and low specificity over a long time period. As caries appears to be easier to predict at the group level compared to the individual level, there is good reason to place added focus on general efforts in groups rather than individuals to improve the situation of those with this disease.


Crossner C, Unell L. A longitudinal study of dental health from the age of 14 to 41. Swed Dent J 2007; 31:65-74.

Davies JA, Smith PA, Pitts NB. Monitoring the Dental Health of Scottish Children Using Routinely Collected Data. Health Bulletin 1996; 54:348-354.

Ekback G, Persson C. Caries in five different socio-economic clusters in Orebro county: w 2012 ; 29:229-32

Harris R, Nicoll A, Adair P, Pine C. Risk factors for dental caries in young children: a systematic review. Comm Dent Health 2004; 21:71-85.

Hugoson A, Koch G, Gothberg C et al. Oral health of individuals aged 3-80 years in Jonkoping, Sweden during 30 years (1973-2003). Part II. Review of clinical and radiological findings. Swed Dent J 2005; 29:139-155.

Mejare I, Stenlund H, Zelezny-Holmlund C. Caries incidence and lesion progression from adolescence to young adulthood: a prospective 15-year cohort study in Sweden. Caries Res 2004; 38:130-141.

Ordell S, Aronsson K. Identifying deprived areas with populations at higher risk for oral disease Comm Dent Health 2002; 19:195.

Petersen PE. (2003a) The World Oral Health Report 2003: Continuous improvement of oral health in the 21st century--the approach of the WHO Global Oral Health Programme. Community Dent Oral Epidemiol; 31(Suppl 1), 3-24.

Petersen PE. (2003b) Continuous improvement of oral health in the 21st century- the approach of the WHO Global Oral Health Programme. The World Oral Health Report 2003.

Powell L. Caries prediction: a review of the literature. Community Dent Oral Epidemiol 1998; 26:361-371.

Regeringskansliet Prop 2007/08:49 Statligt tandvardsstod. [National support for dental care]. Stockholm 2007.

Reisine S, Psoter W. Socioeconomic status and selected behavioural determinants as risk factors for dental caries. J Dent Educ 2001; 65:1009-1016.

SBU Karies-diagnostik, riskbedomning och icke invasiv behandling SBU, [Dental caries, diagnostics, risks and non-invasive treatment]. Stockholm 2007.

Socialstyrelsen (2010) Karies hos barn och ungdomar. En lagesrapport for ar 2008. [Dental caries among children and adolescents. A statement for the year 2008]. Stockholm.

Socialstyrelsen. (2011a) Nationella riktlinjer for vuxentandvard 2011 (National Guidelines for Adult Dental Care--summary. National Board of Health and Welfare). Edita Vastra Aros, Vasteras.

Socialstyrelsen (2011b) Tandvardslagen.[ Dental Act]. SFS 1985:125. Stockholm Torell P. Karies hos skolbarnen i Goteborg 1914-1978. [Dental caries among schoolchildren in Gothenburg 1914-1978]. Tandlakartidningen 1980; 72:230-232.

Vanobbergen J, Martens L, Lesaffre E, Bogaerts K, Declerck D. Assessing risk indicators for dental caries in the primary dentition. Community Dent Oral Epidemol 2001; 29:424-434.

Watt R, Sheiham A. Inequalities in oral health: a review of the evidence and recommendations for action. BDJ 1999; 187:6-12.

Zero D, Fontana M, Lennon A. Clinical applications and outcomes of using indicators of risk in caries management. J Dent Educ 2001; 65:1126-1132.

G. Ekback *,**, S. Ordell ***,****, L. Unell *****

* Department of Dentistry, Orebro County Council, Orebro.

** School of Health and Medical Sciences, Orebro University.

*** Department of Oral Public Health, Faculty of Odontology, Malmo University.

**** Dental Commissioning Unit, Ostergotland County Council.

***** Postgraduate Dental Education Centre, Dental Public Health, Orebro County Council, Orebro, Sweden

Postal address: G. Ekback, Dental Department, Orebro County Council, Box 1613, SE-70116 Orebro, Sweden.

Table 1. Proportions and numbers of Swedish children examined and
numbers by population in different years.

Age (examination year)   Examined (n)   Population (n)

    3 years (1990)          2,568           3,311
    6 years (1993)          2,773           3,420
   19 years (2006)          3,291           3,544

Age (examination year)   Proportion examined (%)

    3 years (1990)                 78%
    6 years (1993)                 81%
   19 years (2006)                 93%

Table 2. Average ds/DS values for panel and non-panel at each
assessment point

Age (years)     Panel       Non-panel
     3           0.16         0.22 *
     6           0.57         0.69
    19           0.49         0.68 ***

* p<0.05 *** p<0.001

Table 3. Correlation between caries at different ages

ds 3 years age    ds 6 years age     0.21 ***
ds 3 years age    DS 19 years age    0.02
ds 6 years age    DS 19 years age    0.12 ***

dft 3 years age   ds 6 years age     0.20 ***
dft 3 years age   DS 19 years age    0.04 ***
dft 6 years age   DS 19 years age    0.17 ***

dft 3 years age   dft 6 years age    0.39 ***
dft 3 years age   DFT 19 years age   0.10 ***
dft 6 years age   DFT 19years age    0.32 ***

*** p<0.001

Table 4. DS=0 (19 year age) regressed on gender and ds
(3 and 6 year age)

                     Sig.   Adjusted        (95% CI)
Gender       male    0.54     1.07       0.87      1.31
            female            ref
ds 3          0      0.88     0.94       0.42      2.12
years old     1      0.58     1.16       0.68      2.00
              >1              ref
ds 6          0      0.01     1.66       1.16      2.38
years old     1      0.00     1.89       1.40      2.55
              >1              ref

Table 5. Sensitivity and specificity of active caries disease when
active caries disease at age 6 is used to predict active caries
disease in 19 years old.

                    19 years old DS=0   19 years old DS>0

6 years old ds=0          1,252                348
6 years old ds>0           254                 127
sensitivity               26.7%
specificity               83.1%

Table 6. Sensitivity and specificity of caries when caries experience
at age 6 is used to predict caries experience in 19 years old.

                      19 years old        19 years old
                          DFT=0               DFT>0
6 years old dft=0          433                 871
6 years old dft>0          126                 555
   sensitivity           38.90%
   specificity            83.1%
COPYRIGHT 2012 European Academy of Paediatric Dentistry
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ekback, G.; Ordell, S.; Unell, L.
Publication:European Archives of Paediatric Dentistry
Article Type:Report
Geographic Code:4EUSW
Date:Dec 1, 2012
Previous Article:Long-term effect of maternal xylitol exposure on their children's caries prevalence.
Next Article:Caries incidence in a cohort of primary school students in Lagos State, Nigeria followed up over a 3 years period.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters