Main

Carious primary teeth are a well recognised source of pain and discomfort, and when it strikes, toothache is distressing for the child and is disruptive for other family members.1 Dental caries, with its unfortunate sequaelae, is an avoidable disease and the cornerstone of care for the child population should be prevention,2 through interventions at the individual patient and public health level. Nevertheless, once caries is present in the primary dentition, interventive treatment, including restorative care is regarded as a necessity.2,3

The raison d'etre for treating carious primary teeth is that such procedures are thought likely to reduce the risk of the children developing toothache.2 Given that 40% of 5-year-olds in the United Kingdom have active caries,4 it is clear that the scope for such restorative care is great. Early intervention is felt to be desirable, as once the caries has progressed and a large cavity has developed, simple restorations are considered to be inadequate. Badly decayed primary teeth may well require more complex treatment including pulp therapy, followed by the placement of preformed crowns.2,3 This more extensive treatment may be somewhat daunting for young children, yet supporters of this treatment philosophy argue that studies demonstrate that these treatments are highly effective5,6,7 and that behavioural management techniques exist to ensure young children can comply with treatment and that any pain and discomfort associated with the procedure can be controlled.8 This approach has been successful in the hands of specialist practitioners, possibly prompting some commentators to describe failure to restore the carious primary teeth of children as cruel and unfeeling.9

General dental practitioners (GDPs) make up more than 80 % of the dental profession and are responsible for the care of the majority of children in the UK. From a public health view it is therefore very important to understand how GDPs approach the care of children with carious primary teeth, as the care they provide will have the greatest impact on the health and well being of the child population. A recent paper10 reported that amongst children with carious primary molars who regularly attended their dentist, over 80% of carious primary molars received simple fillings at some stage. The 677 children in the study had 3,145 primary teeth that had a documented history of caries, but not one preformed crown was fitted and only 120 pulp treatments were provided. A sizeable proportion of carious teeth were left unrestored and these teeth had similar outcomes to restored teeth after controlling for various confounding factors. This study, and an earlier pilot project11 which reported similar findings, have highlighted an under-researched area; namely what are the outcomes of unrestored carious primary teeth and do these unrestored teeth have a bigger impact on the quality of life of children than restored primary teeth?

A recent paper10 and the pilot study11 analysed the outcomes of care at the tooth level. In an attempt to clarify the relationship between dental caries experience in primary molars, its treatment and recorded episodes of pain experienced by children who are regular attenders, a series of patient level analyses were undertaken with the following objectives :

1. To measure the proportion of children regularly attending a GDP who: Experienced dental pain associated with primary molars Required extraction of deciduous molars because of pain or sepsis Required a prescription of one or more courses of antibiotics for their primary molars.

2. At the patient level to examine the relationship between three outcome measures: Episodes of pain Number of extractions due to pain or sepsis Number of courses of antibiotics prescribed.

These objectives were undertaken after controlling for the proportion of carious molar teeth filled, the age that caries was first recorded, gender and total number of teeth affected by caries.

Method

The study took place across four health authorities in the North West Region: Bury and Rochdale, Salford and Trafford, North Cheshire and South Cheshire. These districts were selected to provide a wide variety of urban, rural, affluent and deprived communities. All dentists working in each area were asked if they would like to participate in the study. However, the number of dentists in the study was greatly restricted by a detailed set of inclusion criteria. Only those dentists who could supply the clinical records of between 10 and 20 children who met the following criteria were included in the study:

  • Date of birth 1/1/84 to 31/12/85

  • Care overseen by the same dentist from or before 31/12/90

  • Must have a history of interproximal caries experience in their primary molar teeth

  • Must be a regular attender, which was defined as a child who attended at least once every 18 months from 31/12/90 until September 1999 when the data collection took place.

These criteria enabled a complete history of the dental disease and treatment patterns of the primary dentition to be captured for each subject. Those dentists with more than 20 patients that satisfied the criteria had 20 cases selected at random for inclusion.

Trained and calibrated data abstractors collected the following data relating to the primary dentition of each child from their case notes:

  • Total number of carious teeth

  • Total number of carious teeth that had been filled

  • The total number of episodes of pain

  • The total number of extractions prescribed to relieve pain or eradicate sepsis

  • The total number of courses of antibiotics prescribed

  • The age caries was first recorded.

For this study data were aggregated to, and analysed at the patient level. Frequency distributions were constructed for three outcome measures:

  • Children for whom there was a note in the record card indicating that they had experienced pain associated with their primary molars

  • Children who had undergone extraction of one or more primary molar because of pain or sepsis

  • Children who had received one or more courses of antibiotics.

Logistic regression models were fitted to each of the three outcome measures, taking into account the clustering of children within dentists. Dependent variables included never/ever having:

Pain recorded

Extraction of a primary molar due to pain or sepsis

Antibiotics prescribed.

The independent variables used in the models included age when caries was first recorded, gender, total number of teeth affected by caries and the proportion of carious primary molars that had been filled.

Results

The records of 677 children registered with 50 GDPs were examined. Of the 50 dentists included in the study only 3 had more than 20 patients which met the criteria, requiring random selection of 20 patients. The mean number of patients per GDP was 13.5 (SD 3.3). Figure 1 demonstrates the extent of restorative care in primary molars amongst those regularly attending children with caries experience. Each child had on average 4.6 teeth affected by caries (S.D = 2.64). Over half of the children (55.4%, N=375) had all their carious primary molars restored at some stage, whereas four in five children (85.8%, N= 581) had 50% or more of their carious primary molars restored. Only 41 children (6.1%) received no restorative care at any time for their carious primary molar teeth.

Figure 1
figure 1

The extent of restorative care amongst regularly attending children with caries in their primary molars

Full size image

Table 1 demonstrates the frequency with which children with caries experience in their primary molar teeth experienced pain, required extraction of primary molar teeth for pain or sepsis, or were treated with antibiotics for conditions associated with their primary molars. Slightly more than half of the children (52.1%, N=353) had no record of dental pain. A similar distribution was found in the pattern of primary molar tooth extraction because of pain or sepsis, with just over half of the children, (56.7%, N=384) never requiring extraction because of pain or sepsis. The pattern of prescription of antibiotics shows that two thirds of the children (66.3%, N=449) had no record of receiving a course of antibiotics.

Table 1 Table 1
Full size table

Table 2 displays the results of multiple logistic regression analyses for three dependent variables:

  • never/ever had recorded pain associated with primary molar teeth

  • never/ever had a primary molar extracted because of pain or sepsis

  • never /ever having received antibiotics.

Table 2 Table 2
Full size table

When recorded pain was considered as the dependent variable, only total caries experience was a significant predictor (p<0.01, odds ratio 1.1) after controlling for age when caries was first recorded, proportion of carious primary molars restored and gender. For each tooth affected by caries, the odds of experiencing pain increased 1.1 times. The same was true when having a primary molar extracted due to pain or sepsis was used as the dependent variable (p<0.01, odds ratio 1.16). The final logistic regression analysis again showed that total caries experience, (p<0.01, odds ratio 1.23) and also gender (p<0.05, odds ratio 0.69) were significant predictors of children having one or more courses of antibiotics. Boys were more likely to be prescribed antibiotics after controlling for the other independent variables.

Discussion

Only those children who had approximal caries in their primary molar teeth were included in this study. These children represent a problematical group of patients for GDPs; their regular attendance implies that parents have confidence in the dentist to look after their children's teeth and yet these children still experienced high levels of caries (on average 4.6 carious teeth per child). The results describe the treatment experience that young children with caries in their primary molars could have expected if they attended their GDP on a regular basis over the last 10 years. They also give us an insight into the extent to which GDPs are successful in coping with the results of caries in the primary dentition. Over 90% of the children received restorative care at some point in time. The results showed that almost half of the children in the study experienced pain associated with these affected teeth, and although this population was made up of regular attenders with caries and recorded (not reported) pain was measured, the results agree with other studies1 which demonstrate that pain resulting from tooth decay is a sizeable problem. The children not only had pain but also had treatment as a consequence; almost half had at least one carious primary molar tooth extracted due to pain or sepsis and one third of the children were prescribed at least one course of antibiotics by their GDP.

It appears that when other factors are controlled for, the risk of young children experiencing pain associated with carious primary molar teeth, or having a primary molar extracted due to pain or sepsis, is predicted only by the child's total caries experience. No association was found between the proportion of primary molars that were restored and either the presence of recorded pain or the need for extraction. Increased levels of restorative care did not lead either to reduced levels of recorded pain or reduced levels of extraction therapy. Given that the placement of restorations in primary molar teeth is often not without some form of discomfort for the child, these observations are disappointing. A similar story emerged when the prescription of antibiotics to address the problem of acute infection associated with primary molars was considered. Whilst total caries experience was a predictor of the need for antibiotic prescription, and the likelihood of requiring a course of antibiotics increased 1.2 times for each extra carious molar tooth, the proportion of carious teeth which were restored did not have a significant effect on the prescription of antibiotics. This finding suggests that antibiotics are not being used by general practitioners to control pain arising from unrestored carious primary molar teeth.

Total caries experience is an important factor in predicting pain and its sequaelae in young children, yet restorative care aimed at dealing with the condition is not. Therefore it would seem logical for the dental profession to concentrate its efforts on the prevention of primary tooth decay, rather than its repair. The dental preventive messages are well understood12 and their universal adoption would radically improve dental health. In the UK, 60 % of 5-year-olds are free from active decay,4 suggesting that the disease can be prevented. However, for GDPs to play a substantial role in prevention of caries they must be able to see children on a regular basis. For prevention to work in general practice, we must find effective ways of increasing attendance and also establish an evidence base for clinicians to guide their approach to preventive care.

At the public health level, there is much that can be achieved for the child population in general, whether they attend or not. The recent review of the risks and benefits of water fluoridation undertaken by the University of York13 demonstrates that optimal water fluoridation is effective and safe. In the light of these findings, we should once again place large scale water fluoridation schemes on the public health agenda. However, their implementation requires considerable political support and new schemes may only be achievable in the long term. Because of this we may have to resort to other population based fluoride preventive programmes for communities with high caries rates in young children, such as the distribution of fluoride toothpaste and fluoride milk or the sale of fluoride salt. All of these fluoride vehicles when used at the public health level require pilot research programmes to ensure that only cost-effective schemes are implemented.

The results of this research show that for regularly attending children with caries in their primary molars, pain and its sequaelae are common findings. GDPs, to a great extent, addressed the problem of caries with a restorative approach, yet in so doing were unable to reduce the impact of the disease on the children as measured by the number of episodes of pain, number of extractions and number of courses of antibiotics recorded. Increased restorative care did not lead to better outcomes. However the failure of this study to demonstrate restorative care (or the lack of) in primary molar teeth as a predictor of pain, extraction due to pain or sepsis, or the prescription of antibiotics, is difficult to explain at this stage, although possible explanations have been suggested.10

These subject level findings are in agreement with the findings of the tooth level analyses10 and those of the pilot study.11 This fact, plus similar results whichever outcome variables were examined, give a degree of content and concurrent validity to the findings. Nevertheless the design of this study had limitations, mainly in the selection of GDPs and their patients for inclusion in the study. The selection of GDPs was principally determined by criteria relating to their patients. They were required to have at least 10 children who had been coming to see them on a regular basis for at least 10 years, so this automatically excluded dentists who had been practising for less than 10 years. Selection of the dentists who could meet the study inclusion criteria could not be truly random, as some dentists chose not to participate even though all dentists were offered a fee to join in the research project. This was their prerogative as independent practitioners, but it does illustrate one of the major problems in undertaking research in primary dental care, that is obtaining a truly random sample of practitioners to participate in research. This will remain a difficult problem for the foreseeable future. The other flaw in the study design is in using case notes as a means of collecting data. All measures of data collection have positive and negative benefits,14 but it is difficult to see how the data reported here could be collected in any other way. Certainly the Dental Practice Board could not provide the necessary data and a prospective design could produce a large Hawthorn effect, as GDPs may change their treatment regime in response to being subjects of research.

It must be remembered that these are the first steps in an unresearched area. These primary studies highlight an anomaly, and lay the foundations to move from observational research, to studies designed to examine the effectiveness of restorative care within the primary care setting.