Journal of

Oral Rehabilitation

Journal of Oral Rehabilitation 2016 43; 190–197

Impact of shortened dental arches on oral health-related quality of life J. L. F. ANTUNES*, H. TAN†, K. G. PERES† & M. A. PERES†

*School of Public Health, Univer-

sity of Sao Paulo, Sao Paulo, SP, Brazil and †Australian Research Centre for Population Oral Health, School of Dentistry, The University of Adelaide, Adelaide, SA, Australia

This study described the prevalence of adults with shortened dental arches (SDA) in Brazil, specifically assessing the differences of oral healthrelated quality of life [the prevalence and severity of oral impacts on daily performance (OIDP)] by dentition status. We analysed data from the 2010 National Survey of Oral Health in Brazil, including home interviews and oral examinations. The assessment of SDA used two alternative definitions: having 3–5 natural occlusal units (OUs) in posterior teeth or having 4 OUs in posterior teeth. Both definitions included having intact anterior region and no dental prosthesis. The analysis was weighted, and a complex sampling design was used. Negative binomial regression models assessed associations as adjusted for socio-demographic conditions and dental outcomes. A total of 9779 adults (35–44 years old) participated in the study. A non-negligible proportion had SDA: 9
9% and 3
8% for the first and second definition, respectively. Individuals with SDA (first definition) ranked SUMMARY

Introduction Tooth loss affects oral health-related quality of life (OHRQoL) and may increase the demand for prosthodontic treatment. There has been conflicting evidence with regard to the effect of dentures on OHRQoL (1). More than thirty years ago, K€ ayser first proposed the concept of ‘shortened dental arches’ (SDA) for a dentition with an intact anterior region and loss of posterior teeth (2). However, this concept has not been widely used in daily clinical practice (3). A systematic review of randomised controlled trials © 2015 John Wiley & Sons Ltd

higher in OIDP prevalence [count ratio (CR) 1
22; 1
09–1
36, 95% confidence interval (CI)] and severity (CR = 1
43; 1
19–1
72, 95% CI) than those with more natural teeth. This difference was not statistically significant when adjusted for socio-demographic and dental covariates: OIDP prevalence (CR = 1
04; 0
92–1
17, 95% CI) and severity (CR = 1
09; 0
91–1
30, 95% CI). Analogous results were obtained when the second definition of SDA was adopted. These findings suggest that a considerable contingent of adults may function well without dental prostheses, despite having several missing teeth. This conclusion challenges the traditional approach of replacing any missing tooth and instructs the allocation of more dental resources to preventive, diagnostic and restorative services. KEYWORDS: dental health survey, oral health, quality of life, tooth loss, dental prosthesis Accepted for publication 23 September 2015

compared the functional outcomes of prosthodontic interventions in partially dentate adult patients and concluded that SDA is an encouraging treatment option, in terms of functioning, patient satisfaction and cost-effectiveness (4). However, only a few clinical trials with small sample sizes investigated the relationship between the presence of SDA and OHRQoL among older people (5). Only one population-based study, conducted in Australia, assessed OHRQoL in people with SDA and concluded that there was no noticeable effect on OHRQoL relative to people with more number of doi: 10.1111/joor.12364

SHORTENED DENTAL ARCHES AND QUALITY OF LIFE teeth (6). This finding has important implications for oral health care at a population level – the complex treatment plan involving removable dental prostheses for people with SDA should move towards preventive and restorative treatment plans to maintain their remaining natural dentition healthy and functional. On the other hand, one research question arises: ‘Can one size fit all?’ Population-based studies of OHRQoL in other countries are needed to determine the direction of oral treatment plans for people with SDA. The objective of this study was to investigate the prevalence of adults with SDA at the population level in Brazil, specifically assessing differences of OHRQoL by dentition status. The null hypothesis to be tested was that individuals with SDA have no poor OHRQoL compared with those with more natural teeth among the Brazilian middle-aged population, including those with complete dental arches.

Methods Subjects and study design Data on oral health conditions were gathered in a cross-sectional, nationwide dental survey performed by the Brazilian health authority. Sampling followed a complex design and included individuals aged 5, 12, 15–19, 35–44 and 65–74 years and living in 26 cities (state Capitals), the Federal District, and 150 towns (30 randomly selected towns in each of the five Brazilian geographic regions). The primary sampling units were towns in the interior of the regions and census tracts for each of the state capitals and the Federal District. The secondary sampling units were households, randomly selected in each municipality. All adults (35–44 years old) living in these households were eligible for the study. Further information on the sampling procedure is reported elsewhere (7). Questionnaires Interviews were conducted within the homes by administering a structured questionnaire on sociodemographic information. Skin colour was classified according to the Brazilian official agency of Geography and Statistics: white, brown-skinned, black, Asian and indigenous. Household income was collected in Reais, the country’s official currency, and converted to Brazilian minimum wages, a standard for the © 2015 John Wiley & Sons Ltd

measurement of income that broadly corresponded to US$300 during the period of data collection. Education was assessed by the number of years of study successfully completed and was subsequently categorised as less than 4 years (insufficient education), 4–7 (incomplete fundamental education), 8–10 (complete fundamental, incomplete secondary education), 11–14 (complete secondary, incomplete university education) and 15 or more (university). Respondents self-reported whether they experienced dental pain was in the last 6 months. OHRQoL was assessed by the modified version of the oral impacts on daily performance (OIDP) index (8, 9), which was validated in Brazil (10). This index gathers information on difficulties due to dental problems on the following daily activities: eating food; speaking clearly; cleaning teeth or dentures; performing light physical activities; going out, for example to study or work; sleeping; relaxing; smiling, laughing and showing teeth without embarrassment; and becoming more emotional or more easily upset than usual (mood affected). Each item was preceded by the following question: ‘Some people have problems which may have been caused by their teeth. Out of the following concerns, which apply to you in the last 6 months?’ The questions were answered dichotomously, thus allowing the assessment of OIDP prevalence as the proportion of individuals who answered ‘yes’ to at least one of these questions and OIDP severity as the sum of impacts (‘yes’ answers). Oral examinations Oral examinations were performed at households by a dental surgeon and a recording clerk, who were specifically trained to apply the basic methods standardised by the World Health Organization for oral health surveys (11). These professionals were calibrated in 32-h regional workshops (12), encompassing the different dental conditions that the examiners would encounter when assessing the different population groups. The weighted kappa statistics were calculated for each examiner, age group and oral health problem, with the acceptable limit being 0
80 for dental caries and 0
65 for the remaining dental conditions. Decayed, missing and filled teeth (DMFT) and untreated caries (D ≥ 1) were quantitatively assessed and subsequently classified. The assessment of periodontal conditions included bleeding on probing,

191

192

J . L . F . A N T U N E S et al. dental calculus, shallow (4–5 mm) and deep (≥6 mm) pockets. These conditions were registered separately, according to a modified version of the Community Periodontal Index (CPI) criteria, which allowed registering all concurrent periodontal conditions (bleeding at probing, calculus, deep and shallow pockets) in each sextant. The main point of interest was the dentition status, comparing individuals who had SDA with those who did not. We hypothesised that SDA individuals had no poorer OHRQoL status than those with more natural teeth, which would reduce the need of prosthetic treatment at the population level and reduce costs to the public health system. To test this hypothesis, we used two definitions of SDA previously considered in the literature: (1) having an intact anterior region, no dental prosthesis and 3–5 natural occlusal units (OUs) (13) and (2) having an intact anterior region, no dental prosthesis and 4 natural OUs (2). Both definitions consider an occlusal pair of premolar teeth as one OU, and an occlusal pair of molar teeth as two OUs (14). Figure 1 illustrates some possible conditions of SDA in contrast with the chart for the full dentition. According to the first definition, participants with intact anterior teeth were categorised as (i) more than five OUs and no prosthesis, (ii) more than five OUs with prosthesis, (iii) 3–5 OUs and no prosthesis (SDA), (iv) less than three OUs and no prosthesis and (v) five or less OUs with prosthesis. According to the second definition, the same participants were categorised as (i) more than four OUs and no prosthesis, (ii) more than four OUs with prosthesis, (iii) four OUs and no prosthesis (SDA), (iv) less than four OUs and no prosthesis and (v) four or less OUs with prosthesis Tan et al. (6).

Statistical modelling and analysis OIDP prevalence is a dichotomous variable and OIDP severity is a count variable, which is skewed, overdispersed and non-normally distributed (P < 0
001 at the Shapiro–Wilk W-test). Therefore, we used negative binomial regression (15) to assess associations between these variables and socio-demographic and dental factors. Unadjusted and multivariable models for each outcome (OIDP prevalence and OIDP severity) and for each of the main exposures (first and second definitions of dentition status) were run separately. For the dichotomous variable, the prevalence ratio (PR) and its 95% confidence interval (95%CI) were the measurements of effect resulting from regression models. For the count variable, the count ratio (CR) and its 95% CI were the measurement of effect resulting from regression analysis. Statistical modelling followed a sequential analysis, from distal to proximate determinants of OHRQoL. Model 1 assessed the association of dentition status with both outcomes, as adjusted for demographic variables (age, sex and skin colour); model 2 was additionally adjusted for socio-economic variables (income and education); and model 3 was additionally adjusted for the remaining dental conditions (dental pain, DMFT, prevalence of untreated caries, gingival bleeding, dental calculus and shallow and deep periodontal pockets). All analyses were conducted using Stata 13 (2013)*, taking into account the complex survey design and sampling weights. The survey followed Brazilian and international guidelines on ethics in research involving human beings; all participants signed a term of informed consent, and the study protocol was approved by the Brazilian National Council of Ethics in Research, file No. 15,498, 7 January 2010.

Results

Fig. 1. Some possibilities of shortened dental arches in relation to the full dentition: intact anterior region, no prosthesis and a definite number of occlusal units (OU). One unit corresponds to a pair of occluding premolars; a pair of occluding molars corresponds to two units.

All the 9779 adult participants of this study answered the OIDP questionnaire and provided socio-demographic information. They also underwent an oral examination. A non-negligible proportion of adults in Brazil have SDA; that is, they maintain an intact anterior region and a definite number of OUs in posterior teeth. For adults aged 35–44 years, this proportion reached

*StataCorp, College Station, TX, USA. © 2015 John Wiley & Sons Ltd

SHORTENED DENTAL ARCHES AND QUALITY OF LIFE 9
9% (8
7–11
2%, 95% CI) when the first definition of SDA is considered and 3
8% (3
1–4
6%, 95% CI) when the second definition of SDA is considered. Table 1 shows the distribution of adults according to demographic, socio-economic and dental covariates, which were subsequently used to adjust the association between having SDA and OHRQoL. The comparative analysis of OIDP prevalence and severity considered the population group with all anterior teeth, no prostheses and more than five OUs as the reference for comparisons involving the first definition of SDA. The group with all anterior teeth, no prostheses and more than four OUs was the reference for comparisons involving the second definition of SDA (Tables 2 and 3). The prevalence of oral impact in adults with SDA (first definition) ranked higher than that in the reference group (Table 2): PR 1
22 (1
09–1
36, 95% CI). However, this poorer condition was not statistically significant after having been adjusted for socio-demographic characteristics and dental outcomes: PR 1
04 (0
92–1
17, 95% CI). A very close result was obtained when SDA was assessed by the second definition. OIDP prevalence in adults with SDA ranked higher than that in the reference group (Table 2): PR 1
21 (1
03–1
41, 95% CI). This poorer condition was not statistically significant after having been adjusted for socio-demographic characteristics and dental outcomes: PR 1
05 (0
91–1
21, 95% CI). With regard to OIDP severity (Table 3), adults with SDA (first definition) had a poorer score than the reference group: CR 1
43 (1
19–1
72, 95% CI). However, this poor condition was not statistically significant after having been adjusted for socio-demographic characteristics and dental outcomes: CR 1
09 (0
91– 1
30, 95% CI). Analogous results were obtained when SDA was assessed by the second definition. OIDP severity ranked higher among adults with SDA than in the reference group: CR 1
50 (1
07–2
10, 95% CI). Once again, no statistically significant difference remained after the analysis was adjusted for socio-demographic and dental covariates: CR 1
13 (0
91–1
40, 95% CI) (Table 3).

Discussion The main finding of this study was that Brazilian adults with SDA had no difference in the prevalence © 2015 John Wiley & Sons Ltd

and severity of OIDP compared with those with more natural teeth, after adjustment for socio-demographic and dental outcomes. This result suggests that differences in OHRQoL between adults with SDA and those with more teeth are attributable to differences in other dental conditions such as periodontal disease and untreated caries. The same conclusion holds when assessing SDA by the first or second definition. Results reported here are relevant to dental public health and must be taken into account in the planning of dental services to adults, mainly in the provision and access to prosthodontic treatment. These findings reinforce results reported in another nationwide population-based study conducted among Australian adults (6), which adopted the same two SDA definitions used in our study. To the best of our knowledge, these are the only two population-based studies comparing OHRQoL between adults having SDA and those with more teeth. The prevalence of adults with SDA was similar in both countries: 9
9% in Brazil and 11
4% in Australia (6), when assessed by the first definition, and 3
8% in Brazil and 2
5% in Australia (6), when SDA was assessed by the second definition. Despite similarities between these two studies, it is important to highlight some differences. Brazilian adults are less educated and poorer than Australian adults; they also have a higher prevalence of unfavourable dental outcomes, such as inadequate dentition (16). The poorer overall socio-economic status and the higher prevalence of unfavourable dental outcomes in Brazil reinforce the need for assessing OHRQoL as a strategy for instructing dental services how to effectively direct prosthodontic treatments to those effectively in need, in a context in which resources are indeed scarce. These two studies also had some methodological differences. Australian data were gathered in 2004– 2006, whereas information from Brazil was collected in 2010. The age bracket of the study in Australia was wider (15 or more years old) than that in the study in Brazil (35–44 years old). Furthermore, OHIP14 was used in Australia, whereas a short version of OIDP was applied in Brazil. However, we cannot speculate any particularly relevant implication of such differences in the interpretation of the findings. This study presented evidence indicating that adults with SDA have no poorer OHRQOL than those with more teeth. This finding suggests that individuals with SDA can do without treatment, which may influence

193

194

J . L . F . A N T U N E S et al. Table 1. OIDP prevalence and severity by explanatory variables OIDP Frequency* prevalence† Demographic variables Age group 35–39 40–44

52
4% 47
6%

Sex Male

36
9%

Female

63
1%

Race/skin colour White

50
0%

Brown

37
6%

Black

10
7%

Other

1
7%

Socio-economic status Income 4 occlusal units with prosthesis 4 occlusal units and no prosthesis (SDA)