Population
This study used data from the 8020 Promotion Foundation Study of Japanese Dental Patients. The survey was conducted in dental clinics and adult dental patients across the country were enrolled to determine the health-promoting effects of dental care. Baseline surveys were conducted at 1,216 dental clinics in 46 prefectures in Japan. During any one week in October 2014, 12,399 people aged 20 and over who visited dental clinics for a first visit or revisit more 2 months after their last visit underwent an oral examination and completed a questionnaire. Of these, we enrolled 12,150 people with complete oral examination and questionnaire data. Each year, we mailed all of them the same questionnaire and dental checkup form as at baseline and asked them to return it after they had completed it. Data on the oral examinations were collected if they visited a dental clinic during the follow-up period, but we did not encourage them to visit a dental clinic for the purpose our study. Of the baseline participants, those participating in follow-up survey were not necessarily matched in each year, and follow-up rates varied by year. For this study, we analyzed the questionnaire and oral examination at baseline and the 3-year follow-up oral examination from October 31, 2017 to March 31, 2018. Of the 12,150 patients, 7,877 (64.8%) participated in the 3-year follow-up, 7,601 (62.6%) responded to the questionnaire, and 3,038 (25.0%) underwent follow-up oral examinations at dental clinics (Figure 1).
The Ethics Committee of the Japanese Association for Dental Science, approved the study design, data-collection methods, and procedure for obtaining informed consent (Approval no. 0002); informed consent was obtained from all participants.
Oral health examination
The oral examination was performed by a dentist at dental clinics belonging to the Japan Dental Association, and teeth and periodontal tissues were evaluated. Tooth condition was recorded as sound, decayed, filled, or missing; the number of teeth was calculated as the sum of sound, decayed, and filled teeth, excluding third molars. The periodontal tissue examination evaluated probing pocket depth (PPD) and bleeding during probing (BOP) at six points (buccal mesial, buccal central, buccal distal, lingual mesial, lingual central, and lingual distal) on each tooth.
Questionnaire
The questionnaire items for patients included smoking habit (never, past, current), history of diabetes (no, yes), tooth brushing frequency (≥ 3, 2, or ≤ 1 times per day), use of interdental cleaning tools such as dental floss or interdental brush (no, yes), bleeding from gums when brushing teeth (no, yes), eating between meals (no, sometimes, everyday), reason for dental visit (treatment, both treatment and maintenance, maintenance), education level (≥ 13 or ≤ 12 years), subjective economic status (lower, middle, upper), and working status (no, yes).
Dental clinic survey
The dental clinic questionnaire at baseline asked about the gender and age of the clinic director (30s, 40s, 50s, 60s, 70s or older), number of dental hygienists, and amount of time spent on oral health instructions (0, 1–4, 5–9, 10–19, 20–29, ≥ 30 min). We did not ask for details of the oral health instructions.
Analysis
We analyzed 2,488 patients seen in 585 dental clinics, excluding 4,839 patients with only questionnaire data at follow-up and 550 patients who lacked some data. Those who had a decrease of one or more teeth between baseline and follow-up were counted as those who had lost teeth. The age of each dental clinic director was divided into three categories (30s or 40s, 50s, and 60s or older). The number of dental hygienists was classified into three categories (0, 1–3, and ≥ 4). The time spent on oral health instructions was divided into four categories (0, 1–9, 10–19, and ≥ 20 min). The mean PPD of teeth examined and proportion of BOP-positive teeth were used as indexes of periodontal condition. To analyze the differences in the mean or proportion of each variable, analysis of variance was used for continuous data and the chi-square test for categorical data. To analyze the stratified data, a multilevel logistic regression analysis based on the generalized estimation equation specifying the distribution and link function was used to examine the risk of tooth loss at the patient and clinic levels by calculating the odds ratio (OR) and 95% confidence intervals (CIs). The presence or absence of tooth loss was used as the dependent variable and the independent variables included patient (age, gender, number of teeth, number of decayed teeth, mean PPD, BOP, smoking, history of diabetes, tooth brushing frequency, use of interdental cleaning tools, bleeding from gums, eating between meals, reason for dental visit, education level, subjective economic status, and working status) and clinic (age and gender of dental clinic director, number of dental hygienists, and amount of time spent on oral health instructions) variables. All analyses were conducted using SPSS ver. 24.0 (IBM, Armonk, NY, USA). P < 0.05 was considered indicative of statistical significance.