The role of tobacco use on dental care and oral disease severity within community dental clinics in Japan

Tobacco use is a modifiable risk behaviour in oral disease development. Dental professionals are able to reduce the oral and overall disease burden related to tobacco by counselling their patients. To date, however, tobacco intervention practices by community dental professionals have been limited, and dental professionals may have not fully embraced opportunities for tobacco use interventions [1], although tobacco use cessation clinical activities have been developed in both dental and dental hygiene programs [2].

Within undergraduate dental education for tobacco use interventions, the most important facilitator is evidence of the effects of tobacco use on oral health [3], while screening guides an important visual tool of the effects of tobacco use on oral health, particularly mucosal lesions [4]. These activities have been mainly conducted in the United States, where dental professionals may have been aware of the detrimental effects of smokeless tobacco in the oral cavity in their daily practice [5]. However, they are now expanding globally based on evidence of the detrimental effects of smokeless tobacco in the United States and European countries [6, 7] and in Central Asia [8, 9].

Numerous epidemiological studies have shown the detrimental effects of smoking on oral health. The Surgeon General Report in 2004 summarized evidence regarding causality of the association between smoking and oral health [10]. Evidence is sufficient to infer a causal relationship with smoking for cancers of the oral cavity and pharynx, and periodontitis. Evidence is suggestive but not sufficient to infer a causal relationship between smoking and root-surface caries and between maternal smoking and oral clefts. Furthermore, smokers have more missing teeth than non-smokers [11]. Dentist awareness of evidence for the detrimental effects of smoking during actual dental practice may promote smoking cessation activity.

Characterization of dental care for smokers is likely an important factor in determining whether community dentists can actively participate in smoking cessation interventions in countries in which smokeless tobacco has rarely been used. Because smokers appear to experience dental disease more frequently than non-smokers, smokers were more likely to have more perceived dental needs compared with non-smokers [12]. However, smokers are less likely to report dental visits than non-smokers [13]. These contradictory findings hinder the evaluation of the characteristics of dental care for smokers in community dental clinics. To our knowledge, no study has compared dental care between smokers and non-smokers in community dental clinics. The aim of present study was to clarify the characteristic of dental care of smokers in community dental clinics Japan.

Of the 1,022 dentists that were mailed, 753 (73.7%) responded to the survey. These dentists provided records for 14,187 patients, while those of 2,912 were excluded because of incomplete information. Smoking rate was measured at 25.1% (2,835/11,275). Former smokers (n = 1,590) were not included in the analyses. Finally, the data of 2,835 current smokers and 6,850 non-smokers were used to evaluate characteristics of dental care for smokers vs. non-smokers in community dental settings in Japan (Table 2).

Table 3, depicts the distribution of treated diseases and encounters by smoking status in the phase 1 analysis. CET accounted for nearly half of treatments and encounters (44.6%), followed by GPT (28.4%) and PRT (19.9%). The overall distribution of the five groups significantly differed by smoking (P = 0.001). The ad hoc test for each group showed significant difference in CET (P = 0.002) and POT (P = 0.001) by smoking, with current smokers more likely received CET than non-smokers and less likely to receive POT, though the differences were relatively small (47.1% vs. 43.6% and 1.6% vs. 2.7%, respectively). In contrast, proportions of GPT, PRT and OET was almost equivalent between portions of GPT, PRT and OET was almost equivalent between non-smokers and current smokers (28.8% vs. 27.7%, 20.3% vs. 19.0%, and 4.6% vs. 4.7%, respectively).

In phase 2 of the analysis for the GPT subpopulation, the participants age was identified as a significant factor (P = 0.005). Ad hoc analyses were then performed in the models stratified by age, at 20–39 years, 40–59 years, and more than 60 years (Table 4), while the number of patients increased with age (n = 489, 958, and 1,308, respectively). While the proportion of advanced stage disease in current smokers and non-smokers (4.1% vs. 4.4%) was similar in the younger age group, current smokers in the older two groups were more likely to receive treatment for advanced stage disease than non-smokers (24.9% vs. 15.3% and 40.8% vs. 22.1%). These differences were significant, with ORs adjusted for gender in the two older age groups of 1.67 (95% CI: 1.10–2.56, P = 0.001) and 2.25 (95% CI: 1.62–3.11, P <0.001). When the proportion was compared based on the entire patient population aged more than 40 years, GPT for advanced stage disease accounted for 6.0% (324/5,425) of non-smokers and 9.7% (195/1,946) of current smokers. Current smokers were not predominant among patients treated with advanced stage disease, at 48.3% and 31.9% in the 40–59 and older than 60 years age groups, respectively, and 37.6% for all patients aged 40 years or more (data not shown).

In the analysis to test the distribution of CET level by smoking status (Table 5), interactions with gender and age groups were not significant. Current smokers were more likely to receive treatment for advanced stage disease than non-smokers (54.1% vs. 49.2%), and the difference was significant, with an OR adjusted for age and gender of 1.39 (95% CI: 1.19–1.62, P < 0.001).

In this study, we compared treated diseases and encounters among individual patients in community dental clinics, and found that dental care for current smokers and non-smokers in this setting differed. Current smokers were significantly higher than non-smokers in distribution for CET, and lower for POT, and more likely received GPT for advanced stage disease than non-smokers in older age groups. Our hypothesis was that community dentists in countries in which smokeless tobacco products are rarely used may not realize the detrimental effects of smoking. Therefore, the principal concept to discuss the results should be focused on the characteristic of dental care provided for current smokers. Consideration should be given to health behaviour, particularly less frequent visits to the dentist, although unmet needs of dental treatment were beyond the scope of the present survey.

Our results showed that current smokers were more likely to receive CET and less likely to receive POT than non-smokers. This trend likely reflects the less effective oral hygiene behaviours of smokers [15] and their less frequent visits to dentists [9], respectively. However, since the differences in distribution between non-smokers vs. current smokers were relatively small, community dentists may not realize the significance of these differences in routine clinical settings.

Unexpectedly, we found that the distribution of GPT and PRT between current smokers and non-smokers was similar. If the effects of smoking on periodontal disease [16] and early tooth loss [17, 18] in the Japanese population are directly reflected in dental care, proportions of GPT and PRT in current smokers would be higher than those in non-smokers. Again, the less frequent visits by smokers [13] may be responsible for the similarity in proportions of GPT and PRT between current smokers and non-smokers. Because gingival bleeding is suppressed in smokers [19], smokers who were less likely aware of periodontal disease may less likely to visit dentists than non-smokers. Alternatively, certain cases receiving PRT would be attributable to a severe form of periodontal disease. Community dentists may not realize the detrimental effects of smoking from the distribution of treated diseases.

The characteristics of dental care for smokers due to the detrimental effect of smoking may be reflected within the GPT subpopulation, in which the proportions of treatment for advanced stage disease in the two older age groups were 15.3% vs. 24.9% and 22.1% vs. 40.8% for non-smokers vs. current smokers. Dental professionals may not recognize the detrimental effect of smoking since non-smokers were still dominant among patients receiving GPT for advanced stage disease, at 62.4% of patients aged more than 40 years, albeit that former smokers were excluded from analysis.

The characteristics of GPT in smokers may be less apparent when the proportion is compared among the entire patient population, 6.0% vs. 9.7%. The finding would be more clearly reflected in the practice of periodontists, who primarily treat periodontal patients. Among the 5,879 dentists who responded to the nationwide survey in 2009 (response rate, 59%), the ratio of dentists who reported items regarding smoking history in the medical questionnaire and who routinely asked about smoking status was about 30% for both; and although 90% agreed that dentists should advise their patients to quit smoking, only 20% actually do so [20]. Thus, community dentists may be unaware of the smoking status of their patients, and may not recognize the detrimental effect of smoking during GPT.

The deterioration of untreated dental caries often involves more severe destruction of tooth structure, which may require endodontic-related treatment. Smokers may be more likely to be treated for advanced stages of disease of the tooth structure, but this may also reflect the less frequent visits to dentists made by smokers [13]. However, the frequent need for CET in smokers is in part likely a reflection of the detrimental effects of smoking. The Surgeon General Report in 2004 concluded that evidence to infer the presence or absence of a causal relationship of smoking with coronal dental caries is insufficient [10], and the evidence that is available was based on epidemiological studies which did not address possible confounders, and inconclusive findings of salivary alteration. In other studies that controlled for possible confounders, however, dental caries were significantly and independently associated with smoking [21‐25]. A cohort study design may allow the inference of causality [21], and the effects of exposure to tobacco contents on the growth and metabolism of Streptococcus mutans[26] and on the expansion of the dental caries area [27] may suggest biological plausibility. A mechanism which acts via this cariogenic microorganism is conceivable, since the effects of tobacco use on periodontal microorganisms [28, 29] has been recently added to the list of potential causal pathways [30]. These reports invite further studies to examine whether the potential effect of smoking on tooth structure may influence oral health inequalities.

Although current smokers experience tooth loss more frequently than non-smokers [17, 18, 31], the proportion of prosthetic treatments between smokers and non-smokers in the present study was equivalent. The significant association of smoking with increased risk of prosthetic dental restorations was independent of sociodemographic characteristics, including education and income [32]. Our findings regarding CET and GPT for advanced stage disease in these subpopulations likely warrants further study regarding the intensity of PRT. Studies of Japanese workers revealed that current smokers were more likely to incur higher dental care costs than never or former smokers [33]. The increased proportion of patients receiving care for advanced stage conditions in the CET and GPT subgroups suggests the need for more intensive treatment in smokers than non-smokers.

Several limitations of our study warrant mention. First, a relatively high proportion of patients (20%) was excluded due to incomplete data. However, the smoking rate among the excluded patients was similar to that of the included patients (24.3% vs. 25.1%; data not shown), suggesting that bias caused by the exclusion of data would be limited with respect to smoking. Second, there is no information about diagnosis and socioeconomic status in the present study. Since we looked into treatments rendered by the questionnaire, it is thus difficult to draw concrete conclusions about the exact diagnosis and prognosis. It is possible that there is significant unmeasured confounding by socioeconomic status. These variables should be included to compare diagnosis and prognosis between current smokers and non-smokers in community dental clinics. Third, the survey was conducted on a single day and may have been influenced by daily variations due to weather conditions or weekly periodicity. Any effect of this single-day sampling on the representativeness of dental patients might have been small, however, given that patient distribution for gender and age was similar to that in a patient survey conducted by the Japanese Ministry of Health, Labour and Welfare in October 2005.

The association of smoking with type of dental care of CET and GPT severity would warrant the need for dental professionals to engage their patients smoking within clinical practice. The detrimental effects of smoking in dental care for smokers regarding the distribution of treatments and encounters and stage of disease at the time of treatment might not be clearly realized by dental professionals unless smoking status of all patients is identified.