|Home | About | Journals | Submit | Contact Us | Français|
To date, the evidence supporting the benefits of dental visiting comes from cross-sectional studies. We investigated whether long-term routine dental visiting was associated with lower experience of dental caries and missing teeth, and better self-rated oral health, by age 32. A prospective cohort study in New Zealand examined 932 participants’ use of dentistry at ages 15, 18, 26, and 32. At each age, routine attenders (RAs) were identified as those who (a) usually visited for a check-up, and (b) had made a dental visit during the previous 12 months. Routine attending prevalence fell from 82% at age 15 to 28% by 32. At any given age, routine attenders had better-than-average oral health, fewer had teeth missing due to caries, and they had lower mean DS and DMFS scores. By age 32, routine attenders had better self-reported oral health and less tooth loss and caries. The longer routine attendance was maintained, the stronger the effect. Routine dental attendance is associated with better oral health.
The adult users of dental services can be categorized into routine attenders and problem-oriented attenders (Gilbert et al., 2000). Promoting regular dental visits is one of the cornerstones of preventive dentistry (Axelsson et al., 1991; Murray, 1996; Richards and Ameen, 2002), but, typically, only about half of the adult population in most Western countries are routine attenders (Roberts-Thomson et al., 1995; Jamieson and Thomson, 2002), with rates being lower among men and in particular social, ethnic, or age groups (Roberts-Thomson et al., 1995; Dixon et al., 1999; Green et al., 2003), and higher in Scandinavia (Hjern et al., 2001). There is epidemiological evidence showing that problem-oriented attenders have poorer oral health than routine attenders, even after adjustment for putative confounders such as social class, age, gender, and ethnicity, but almost all of that comes from cross-sectional studies. One exception, a cohort study of young New Zealand adults, found that problem-oriented attenders were three times more likely to experience caries-associated tooth loss over an eight-year period (Thomson et al., 2000). Such differences are not confined to clinical measures: A recent UK study of a representative sample of adults found that problem-oriented attenders had poorer oral-health-related quality of life (McGrath and Bedi, 2001). A US study of older adults reported that the prevalence of “oral disadvantage” (defined according to a range of self-reported measures) was greater among problem-oriented attenders, even after adjustment for clinical measures (Gilbert et al., 1997).
It is currently unclear whether the difference between problem-oriented and routine attenders is due to the routine visiting itself—that is, that dental attendance and the associated preventive (and interceptive) care and advice are efficacious—or whether it is because of a “healthy user” effect (Posthuma et al., 1994): Routine attenders have better oral health and health behaviors anyway. A recent systematic review was inconclusive (Davenport et al., 2003), owing to a shortage of appropriate, high-quality studies. It would be an unusual society which permitted random allocation of its citizens to particular dental attendance patterns for the purposes of a randomized control trial. Thus, the most useful higher-level source of information on the issue is likely to be a prospective cohort study where a (preferably representative) sample is followed for long enough to determine whether routine dental attendance is contributing to routine attenders’ better oral health over and above their better oral health behavior.
The aim of this study was to determine whether long-term routine dental attenders had (a) better self-rated oral health and (b) lower experience of dental caries and missing teeth by age 32.
The Dunedin Multidisciplinary Health and Development Study is a longitudinal study of a birth cohort born in Dunedin (New Zealand) between 1 April 1972 and 31 March 1973 (Silva and Stanton, 1996). The sample that formed the basis for the longitudinal study was 1037 children assessed within a month of their third birthdays and is considered to be broadly representative of its age group in the South Island population. Periodic collections of health and developmental data have since been undertaken, and this study uses data collected from assessments conducted at ages 15, 18, 26, and 32. Over 90% of the cohort self-identified as European. Ethical approval for the study was obtained from the Otago Ethics Committee, and informed consent was obtained from each participant (and also from parents at the assessments conducted during adolescence).
Information on use of dental services was collected at ages 15, 18, 26, and 32, and was determined differently as participants aged. The assumption was made that all were routine attenders before age 15, since the New Zealand School Dental Service provided routine care to almost all children at that time (and the small number who had opted out are believed to have routinely sought private dental care) (Thomson, 2001). At ages 15 and 18, participants were asked whether they were enrolled with the General Dental Benefit scheme (whereby NZ adolescents were entitled to receive free routine dental care) and about the time since their last dental visit (and the reason for it). At ages 26 and 32, use of dental services was determined by asking participants whether they usually visited the dentist for a check-up or only when a dental problem arose, together with the number of months since the last visit. For each of ages 15, 18, 26, and 32, routine attenders were identified as those who (a) usually visited for a check-up, and (b) had made a dental visit during the previous 12 months.
At each age, dental examinations for caries (collected as surface-level data) and missing teeth were conducted by calibrated dental examiners, who obtained an estimate of accumulated tooth loss due to caries by observing the presence or absence of each tooth, and ascertaining the reason for its absence. In this study, third molars were not included in the computation of tooth loss; only those teeth which had been lost because of caries were included. Self-rated oral health was measured by asking participants to rate their oral health in comparison with that of other persons their age (with response options: ‘among the nicest’, ‘better than average’, ‘worse than average’, or ‘among the worst’). The dependent variables used in the current study were mean DMFS and mean DS, the prevalence of 1+ teeth missing due to caries, and self-reported oral health (as a binary variable) by age 32.
We measured socio-economic status (SES) by using data collected on parental socio-economic status, using standard New Zealand occupationally based indices (Irving and Elley, 1977; Elley and Irving, 1985), which involve a 6-category classification (where, for example, a doctor scores ‘1’ and a laborer scores ‘6’). Childhood SES was calculated as the average of the highest SES level of either parent, assessed repeatedly from birth to 15 years. Participants were classified as having low (groups 5 and 6), medium (groups 3 and 4), or high (groups 1 and 2) childhood SES. SES in adulthood was classified the same way, but using the participant’s occupation at age 32.
Dental plaque accumulation was measured at ages 15, 18, 26, and 32 according to the Simplified Oral Hygiene Index (Greene and Vermillion, 1964). Because we had no direct measure of self-care at any age (other than self-reported toothbrushing frequency), plaque scores were used to represent self-care (as one directly observable measure of the efficacy of that self-care) to determine whether a “healthy user” effect existed.
Following the computation of descriptive statistics, we fitted models using generalized estimating equations (GEE) methods in Stata, according to a recently described approach (Pepe et al., 1999). The aim was to estimate the strength of the association between regular dental visiting and the outcome variable of interest (the “univariate association”) at each assessment. A non-standard GEE with an independent correlation matrix provided an appropriate covariance matrix for these comparisons to be made. Further analyses adjusted for sex and SES (and for plaque score in subsequent modeling) at each assessment. Because pairwise comparisons of the 4 assessments involved 6 statistical tests, the Bonferroni inequality test was used to adjust the type 1 error rate, so comparisons in the models were regarded as statistically significant for P < 0.008.
Participation rates in the Dunedin Study are high, with 972 participants (96%) taking part in the age-32 assessment; 932 (96%) of those were dentally examined. The 40 who were not dentally examined did not differ from the latter in terms of gender, but there was a slight difference by childhood SES group (high 98.1%; medium 96.3%; low 92.9%; P = 0.04). The current analyses involved those 932 individuals (51.1% of whom were male). Complete service-use data were available for 739 individuals at age 15, 823 at age 18, 904 at age 26, and 916 at age 32.
The prevalence of routine attending fell from just over four-fifths at age 15 to about one in four by age 32 (Table 1). Just over one in ten were routine attenders at each of ages 15, 18, 26, and 32 (“long-term routine attenders”). An age-associated divergence was apparent between men and women, and more of the latter were long-term routine attenders. All three SES groups showed a decline in routine attendance with age: An apparent lack of a SES difference at age 15 had become quite a gradient by age 18 (whereby it was lowest among low-SES individuals and highest among those of high SES), and it was marked by ages 26 and 32.
Those who were routine attenders at a given age had a higher proportion reporting better-than-average oral health and a lower proportion with caries-associated tooth loss (Table 2). For example, at age 32, 67.4% of those who were routine attenders at age 26—but only 45.4% of those who were not—rated their oral health as better than average. Similarly, routine attenders had fewer untreated decayed surfaces (on average) and lower mean DMFS scores by age 32. When the patterns from ages 15 to 32 were examined, long-term routine attenders had more favorable scores on all indicators, particularly in comparison with those who were never routine attenders. For example, the latter’s mean number of untreated decayed surfaces by age 32 was more than 4 times that of the long-term routine attenders; the overall mean DMFS difference was just over 5 surfaces, and there were also marked gradients in tooth-loss experience and self-rated oral health.
Some 147 (15.8%) were routine attenders at none of the 4 ages, 220 (23.6%) at 1 age, 308 (33.0%) at 2 ages, 155 (16.6%) at 3 ages, and 102 (10.9%) at all 4 ages. This categorization by the number of ages of routine attendance was strongly associated with self-reported oral health and caries-associated missing teeth (Fig., a) and with mean DMFS and DS scores by age 32 (Fig., b): There were generally less-favorable scores among those who had been routine attenders at fewer ages.
The models comparing the associations between the 4 oral health outcomes and routine attendance at the 4 ages (Table 3) first controlled for sex and SES, after which plaque scores for each of ages 15, 18, 26, and 32 were also entered (as proxies for self-care, in an attempt to adjust for a “healthy user” effect). In interpreting the model outcomes, it is important to bear in mind that the ORs and IRRs for each age represent the univariate estimate, adjusted for the covariates. Thus, someone who was a routine attender at age 15 had 1.70 times the odds of an age-15 non-routine attender of reporting better-than-average oral health by age 32. Moreover, with the exception of age 18, the closer the age was to 32, the greater the OR, with an age-32 routine attender having 3.36 times the odds. Similarly, the odds of having caries-associated tooth loss by age 32 were lower for routine attenders regardless of age, but they were lowest for age-26 routine attenders. The models for age-32 DS and DMFS confirmed that routine attendance predicted lower scores, but there were no clear patterns with respect to age, other than age-32 routine attenders having the lowest IRR for mean DS.
This study aimed to determine whether long-term routine dental attenders had better self-rated oral health and lower experience of dental caries and missing teeth by age 32 than those with less-favorable visiting patterns. We found that routine attenders have better self-reported oral health and less tooth loss and dental caries. We also observed that the differential was greater with longer exposure to routine attendance, with long-term routine attenders having the best oral health by age 32. Analysis of the data also suggested that, for decayed surfaces and self-rated oral health, more proximal exposure to routine attendance is beneficial.
It is appropriate to consider first the weaknesses and strengths of the study. For example, the dental attendance data were self-reported, and we were unable to verify each participant’s dental utilization independently. Moreover, the collected information pertained to the individual’s usual dental visiting pattern at that age, and it is possible that some may have changed their pattern (e.g., from non-routine to routine and back to non-routine between assessments); thus, there is likely to be a degree of inaccuracy in exposure measurement. Underlying the entire approach is the assumption (unable to be tested here) that the benefits of routine attendance are similar regardless of who the dentist is. The study’s strengths lie in its prospective design, high retention rate after three decades, and its mix of clinical and self-report outcome measures. Analyses such as this are rare because of the scarcity of prospective oral health studies of birth cohorts (particularly through adulthood).
Turning to the research question, what are the effects of long-term routine dental attendance? Does repeated exposure to dental check-ups and advice (and whatever one-on-one prevention is available) over many years actually have an effect, or are there tenable alternative hypotheses which might explain the findings? The strength of the observed association between long-term routine use and better oral health is indisputable, particularly with missing teeth, self-rated oral health, and untreated caries. The findings with respect to the overall caries experience represented by age-32 DMFS are not quite as straightforward, with little difference among those who were routine attenders at 2, 3, or 4 ages, a finding which is most likely attributable to routine attenders’ higher likelihood of receiving restorative treatment (Baelum, 2008).
However, the crux of the issue is the extent to which the “healthy user” effect was responsible for at least some of the observed differences between routine attenders and the remainder: They probably differ in ways that are hard to measure but likely to result in better oral health. Sex and SES are two such factors, and we controlled for those accordingly. It might also be expected that (all other factors being equal) routine attenders are likely to have cleaner teeth than problem-oriented attenders, and so we controlled for that using plaque scores measured at each of the 4 assessment ages. After adjustment for those effects, the beneficial effect of routine dental attendance persisted; that the difference made by adjustment for plaque scores (as directly observable measures of dental self-care at each of the 4 ages) was not a great one suggests perhaps that dental visiting has a relatively strong effect. It may, of course, be that the routine attenders’ regular exposure to the dental care environment and associated oral health messages influenced their self-care behavior (and consequently their plaque scores); however, investigating this issue is beyond the scope of the current study.
In conclusion, this prospective study supports the notion that routine dental attendance is associated with better oral health outcomes. It is therefore appropriate for current oral health messages to strongly promote regular dental visiting.
This work was supported by: Grant R01 DE-015260 from the National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD 20892, USA; and by a program grant from the Health Research Council of New Zealand.