| Sign In to gain access to subscriptions and/or personal tools. |
Does Removable Partial Denture Quality Affect Individuals’ Oral Health?
1 Department of Removable Partial Denture Prosthodontics, Tokyo Medical and Dental University, Japan; Correspondence: * corresponding author, kazu{at}senzoku.showa-u.ac.jp
The impact of oral disorders and interventions on individuals’ perceived oral health and oral-health-related quality of life (OHRQoL) is being increasingly recognized as an important health component. This study examined the association between denture quality and OHRQoL in individuals wearing removable partial dentures (RPDs). The study participants were 245 consecutive patients (mean age: 63.3 ± 8.7 yrs) at a university-based prosthodontic clinic who wore RPDs for more than one month. RPD quality and OHRQoL were determined by means of a 100-mm visual analog scale (VAS) and the 49-item Oral Health Impact Profile-Japanese version (OHIP-J49), respectively. Linear regression analysis between RPD quality and OHRQoL revealed that a 10-mm VAS increase in RPD quality rating was related to –2.8 OHIP-J49 units (95% confidence interval: –4.5 to –1.1, p = 0.001), which represents an improvement in OHRQoL. The results suggest that RPD quality influences individuals’ OHRQoL to a clinically significant extent.
Key Words: denture quality • oral-health-related quality of life • removable partial denture Abbreviations: oral-health-related quality of life (OHRQoL) • removable partial denture (RPD) • visual analog scale (VAS) • Oral Health Impact Profile (OHIP) • OHIP Japanese version (49-item version: OHIP-J49, 54-item version: OHIP-J54) • intraclass correlation coefficient (ICC) • confidence interval (CI) • standard deviation (SD)
The impact of oral disorders and interventions on an individual’s perceived oral health state and oral-health-related quality of life (OHRQoL) is increasingly recognized as an important component of health (Reisine and Locker, 1995). Demands for prosthodontic treatments are expected to rise in developed countries, due to a rapid increase in their elderly population (Douglass et al., 2002). Because various treatment options to replace missing teeth exist, the choice of treatment option needs to take the multidimensional nature of health and the option’s possible outcomes into consideration. Since the limitations of the "biomedical" paradigm of health have been recognized, and the importance of quality of life in elderly persons has been emphasized, we clearly need to understand the impact of therapeutic intervention on OHRQoL (Allen, 2003), which is impaired due to tooth loss (Locker and Slade, 1994; Locker, 1995; John et al., 2004a; Steele et al., 2004; Åstrom et al., 2006). Several OHRQoL instruments have been used to assess the impact of prosthodontic treatments (Bouma et al., 1997; Awad et al., 2000, 2003; Heydecke et al., 2003). These OHRQoL instruments have been shown to work well in describing the impact of edentulism and of treatment on several domains (Atchison and Dolan, 1990; Slade and Spencer, 1994; Awad et al., 2000; Allen and Locker, 2002). However, these studies investigated edentulous individuals or those with implant-supported dentures exclusively, and little is known about OHRQoL in partially edentulous individuals who use removable partial dentures (RPDs). Because of the higher cost of treatment with, and lack of insurance coverage for, dental implants, RPDs continue to be widely provided as a treatment option for replacing missing teeth (Petropoulos and Rashedi, 2006). A review of a US database indicated that, in spite of a decline in tooth loss, RPDs are still used in all age cohorts, including young adults (Hummel et al., 2002). However, an analysis of RPDs in the same study found that the quality of RPDs was not consistent, and that two-thirds of the investigated RPDs had defects of various types. The purpose of this study was to understand the relationship between denture quality and OHRQoL in persons using RPDs. The research hypothesis of this study was that the quality of RPD would be associated with the individuals’ OHRQoL.
Participants and Setting During the study period (October, 2005), 262 consecutive patients at Tokyo Medical and Dental University’s Prosthodontics clinic were initially recruited into this study. Two hundred fifty-three individuals (96.6%) agreed to participate, and provided written informed consent. Persons were included when they (i) had adequate general health that would not interfere with dental treatment, and (ii) had worn a removable partial denture for more than one month. Persons were excluded when they reported pain in the orofacial region that did not originate from their RPD or presented acute oral disease. After exclusions, 245 individuals remained in the study. The study protocol was approved by the Ethics Committees of Tokyo Medical and Dental University.
Oral-health-related Quality of Life
Denture Quality Because of the linear relationship and the substantial correlation between the denture quality scores (rPearson = 0.62), we created a summary score by adding the denture stability and esthetics scores and dividing them by 2, to maintain the original 0–100 metric. "RPD quality" was considered the primary exposure variable of the study. Reliability of RPD quality assessment was evaluated on 30 randomly sampled clinicians. The inter-examiner reliability was ICC: 0.70 (95% CI: 0.35–0.85), and test-retest reliability (two-week time interval) was ICC: 0.95 (95% CI: 0.88–0.98).
Statistical Analysis One person was excluded from the analyses because the missing data compromised the calculation of an OHIP summary score (> 5 missing items), and remaining missing answers (n = 64) were imputed by means of a regression technique as previously described (John et al., 2003). Therefore, data from 244 individuals were included in the analysis. All analyses were performed with the statistical software package STATA, Release 9 (StataCorp. LP, College Station, TX, USA), with the probability of a type I error set at the 0.05 level.
Characteristics of the Study Population Study participants (n = 244) had a mean age (± SD) of 63.3 ± 8.7 yrs, were 67.6% women, and had a mean number (± SD) of missing teeth of 13.0 ± 6.9 (Table 1  ). A majority of the study participants had Kennedy class I or II removable partial dentures of either the maxilla or mandible, or both (Table 2). The mean OHIP-J49 summary score for the study participants was 46.8 (range, 0–138; 95% CI, 43.1–50.5). RPD quality had a median VAS score of 63.8 (interquartile range, 50.0–80.6), with a minimum of 6.5 and a maximum of 100. When participants were divided into "bad" and "good" RPD quality based on median value, substantial differences in the proportions of women, the mean age, and the mean number of missing teeth were not observed (all p > 0.05 chi-squared or t tests, Table 3). However, individuals with a bad RPD quality had higher OHIP-J49 scores, meaning that they reported more problems, than did persons with good RPD quality (t test, p = 0.02, Table 3). When the RPD quality distribution was divided into quartiles, a dose-response relationship between poorer RPD quality and more impaired OHRQoL (mean of OHIP-J49 units for RPD quality quartile 1, 56.7; RPD quality quartile 2, 46.5; RPD quality quartile 3, 44.3; RPD quality quartile 4, 40.8; Table 4) was observed.
Regression Analyses The dose-response relationship between RPD quality and OHRQoL was confirmed in the regression analyses, where the quality score quartiles were modeled by indicator variables (Table 4  ). Coefficients describing the differences between the first and the subsequent quartiles became increasingly larger in absolute value, and the p-values became increasingly smaller, whereas the precision of the results, based on the standard errors of the 3 indicator variable coefficients, remained relatively constant (Table 4). The regression coefficient for the grouped linear variable was –5.0 (95% CI: –8.3 to –1.7, p = 0.003). This indicated that, from one quality score quartile to the next better one, OHIP-J49 scores decreased by 5 OHIP units, meaning that the OHRQoL improved. When the 0- to 100-mm VAS quality score was used in the analysis, a 10-mm increase in quality was associated with –2.8 OHIP-J49 units (95% CI: –4.5 to –1.1, p = 0.001), i.e., OHRQoL improved. This result did not substantially change when the influence of age, gender, or number of missing teeth was explored. For example, when the number of missing teeth was included in the model, a 10-mm increase in quality was associated with the identical OHIP-J49 unit, only the confidence interval changed slightly to –4.6 to –1.0. The sensitivity analysis result with an OHIP-J54 summary score was also similar (a coefficient of –3.1 OHIP-J54 units, 95% CI: –4.9 to –1.2, p = 0.001). Regression diagnostics did not reveal any substantial discrepancies from the assumptions of the regression analysis (statistically non-significant test for heteroscedasticity, visual inspection of a residual plot, and a normal probability plot of the residuals).
The study results demonstrated that denture quality was substantially associated with OHIP-J49 summary scores. Better denture quality was related to better OHRQoL status, and this finding was independent of age, gender, or number of missing teeth. If the difference in OHRQoL score between two distinct groups might be considered clinically relevant, a 30-mm difference on the denture quality scale was clinically relevant. OHRQoL impairment relative to this difference in denture quality, and corresponding to the interquartile range of quality ratings, was 8.4 OHIP units, larger than the 7.5 OHIP units found in a population-based study for the difference in OHRQoL impairment between individuals without any denture or fixed partial dentures and those with removable dentures (John et al., 2004a). Previously, several OHRQoL instruments have been used to assess the impact of prosthodontic treatments (Bouma et al., 1997; Awad et al., 2000, 2003; Heydecke et al., 2003). These OHRQoL instruments have been shown to work well in describing the impact of edentulism and of treatment options (Atchison and Dolan, 1990; Slade and Spencer, 1994; Awad et al., 2000; Allen and Locker, 2002). Unfortunately, information about the association of OHRQoL and denture quality is limited; we found only one study in the literature. That study investigated OHIP summary scores’ correlation with self-ratings of denture-quality-related RPD characteristics (chewing function, esthetics, speech, retention, and fit) in 55- to 75-year-old individuals (Hassel et al., 2007). The authors found that all correlations were statistically significant. Our study results about clinician-rated denture quality and OHRQoL are in line with these findings. Patient-based and clinician-based assessments of the denture quality have been reported to be correlated only weakly in persons wearing complete dentures (Ettinger and Jakobsen, 1997; Heydecke et al., 2003). In our analysis of this study’s data, we also found small correlations between patients’ and clinicians’ quality ratings (correlation coefficients < 0.3, data not shown). This analysis may suggest that individuals’ opinion about partial denture quality is only poorly predictive of the actual (clinician-rated) quality in partially edentulous individuals. Therefore, although previous research reports have presented similar findings about denture quality, they provide limited information about the perceived impact of quality of denture care. Our study should be regarded as the first investigation of OHRQoL as it relates to denture quality in partially edentulous persons with RPDs. The strength of our study is that the results were robust against the influence of other important variables influencing the study outcome. Since the number of missing teeth and age have been reported to be significant factors influencing OHRQoL (Atchison and Dolan, 1990; Locker and Slade, 1994; Locker, 1995; John et al., 2004a; Steele et al., 2004; Åstrom et al., 2006), they were analyzed in this study as variables potentially influencing the analyses; however, the analyses with and without these variables were very similar. Our study also had limitations. Our study sample was not a population-based sample, but a consecutive sample of individuals who sought prosthodontic treatment at a university-based prosthodontic clinic. Denture quality in the general population might be different compared with that in university-based patient populations. OHRQoL impairment of clinical patients and persons in the general population may also vary (John et al., 2003, 2004b). In contrast, we are not aware of strong evidence that the relationship between the two variables studied here is much different in treatment centers compared with population-based settings. The mean OHIP-J49 summary score observed here is in fact similar to other published reports, indicating that our patients are probably not too much different from other prosthodontic patient populations in terms of their perceived oral health (John et al., 2004b; Bae et al., 2006). The association found in our study has a significant clinical relevance, because the RPD quality was previously reported to be not consistently acceptable (Hummel et al., 2002), indicating potential for quality improvement. In that previous report, the authors assessed the quality of 1306 RPDs in the US and found 65% to have defects. They concluded that there is a clear need to improve the quality of RPDs. If we assume a causal association between denture quality and perceived oral health, our study results suggest that improvement in RPD quality might be effective in improving the poorer OHRQoL in these individuals. This potential to have an impact on individuals’ oral health is of public health importance, because of the prevalence of RPDs in the general population (Douglass and Watson, 2002) and the level of non-optimal denture quality (Hummel et al., 2002). Our study quantifies the clinical experience of the dental profession via the metric of a commonly used OHRQoL questionnaire. We conclude that, because the quality of (partial) dentures affects individuals’ oral-health-related quality of life, the clinician’s ability to provide better RPDs may have the potential to contribute to better oral health.
The study was supported by the Department of Removable Partial Denture Prosthodontics, Tokyo Medical and Dental University. We thank Linda Raab for editorial assistance. Received for publication August 20, 2007. Revision received January 23, 2008. Accepted for publication April 15, 2008.
Journal of Dental Research, Vol. 87, No. 8,
736-739 (2008)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION