1 INTRODUCTION

Toothache is a common consequence of dental problems such as dental caries, periodontitis and pericoronitis. It is a phenomenon commonly characterized as a major source of impairment in almost all aspects of human quality of life, specifically disturbing sleep, social interactions, performance of daily tasks and influencing care-seeking behaviour.1, 2 Previous studies reported that the prevalence of toothache ranged from 1.3% to 87.8% in different countries and populations.3 With regards to related factors, previous study showed that younger age and lower socioeconomic status, including household income and mother's education4 were the factors most consistently associated with higher likelihood to get toothache.5, 6 Worse oral status, snacks consumption,7 dental treatments,8 with untreated decayed teeth,9 anxiety, stress,10 depression11 and homeless,12 were also found to be associated to toothache.

Most previous studies reported the prevalence of toothache in children and adolescents, mainly in western countries.3 With the ageing of the population, oral health of the elderly deserves more attention. Oral health also affects the daily performance of the older adults.2 However, studies focussed on toothache prevalence in older adults and its related factors, especially in Chinese older adults are lacking.

In the present study, we aimed to investigate trends in toothache prevalence among adults aged 65 years and above and to identify factors associated with toothache using the national survey data among the Chinese older adults from 2011 to 2018, to provide latest data on toothache prevalence among older population.

2 METHODS 2.1 Data source

This is a national cross-sectional study. We used data from the Chinese Longitudinal Health Longevity Survey (CLHLS), which is a national study conducted among the older population in China. It was conducted in half of the counties and cities that were randomly selected, in 23 out of 31 provinces of China, which covers over 85% of the Chinese population. Details of this study have been published elsewhere.13 The first national survey was conducted in 1998 and then conducted in every 3 year. Survey questions have remained generally similar during different years. As information about toothache was added in the questionnaire since the 2011 national survey, we used data from 2011, 2014 and 2018 national surveys. There was a total of 32 831 participants investigated in the three national surveys (9765 participants in 2011, 7192 participants in 2014 and 15 874 participants in 2018, respectively). After exclusion respondents with missing information on toothache (n = 742), those aged less than 65 years old (n = 205) or missing other selected important covariates included in the further analyses (n = 6836), and 25 048 participants aged 65 years and above included in the study.

The CLHLS study was approved by the Biomedical Ethics Committee of Peking University. All participants or their legal representatives signed written consent forms that the study was performed in accordance with the Declaration of Helsinki. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

2.2 Toothache and oral-related variables

The primary outcome in this study was toothache prevalence, which was based on the question: ‘During the past 6 months, did you have a toothache more than once?’ Additionally, participants were asked about the number of natural teeth, the toothbrushing frequency and whether they had denture. Participants were organized into four groups according to the number of their natural teeth (0, 1-10, 11-20 or 21 and above). Frequency of toothbrushing was elicited with the response options (more than once a day, once a day, less than once a day or never).

2.3 Demographic and socioeconomic status

The following data were collected as follows: sex, age, year of birth, place of residence, years of schooling, marital status, whether the financial support enough to pay for daily expenses, whether participant could get adequate medical service and medical insurance type. Years of education were categorized into three groups: never got access to education, 1-6 years, ≥7 years and marital status were categorized into two groups according to the questionnaire: currently married and living with spouse and others. Insurance types were categorized into three groups: the urban basic medical insurance, the new rural cooperative medical insurance scheme and others.

2.4 Behavioural factors and chronic disease

Several behavioural variables were contained dietary diversity, smoking/alcohol consuming during lifetime, main flavour and sugar intake. Dietary diversity scores (DDS) were computed using a food frequency questionnaire related to the nine major food groups: meat, fish and seafood, eggs, beans, fruits, salty vegetables, tea, garlic and fresh vegetables.14 One DDS unit was defined as consumption of any food group ‘often or almost every day’ or ‘not every day, but at least once per week’. Sugar intake was measured in the same way. Participants were categorized into three groups (0-1, 2-5 and 6-9), where 9 represented highest level of dietary diversity.14 Smoking or drinking status was based on 4 questions: ‘Drink/Smoke or not at present/in the past?’, and then categorized into ‘Never/Ever smoked/Drank during lifetime’. Main flavour was based on the respondents' answer to the question: ‘Main flavour you have’ and categorized into five groups: insipidity, salty, sweet, hot and others. Based on questions containing 24 common chronic diseases, including hypertension, diabetes, heart disease, etc, people with any chronic disease were organized into the group ‘With one or more chronic medical conditions’, and people without at least 12 diseases were organized into the group ‘Without’.

2.5 Statistical analysis

We conducted all statistical analyses using R Software (3.6.3). Baseline characteristics were presented as mean (continuous variables) or on a frequency distribution (categorical variables). Toothache prevalence and its 95% CI was evaluated in different subgroups. Chi-square test was presented in the categorical data to examine the differences in prevalence among specific subgroups. To examine the associations between toothache and its potential risk indicators (including demographic variables, socioeconomic status, oral-related status and health-related behaviours), we used a multivariate modified Poisson regression analyses with robust error variances to estimate prevalence ratios (PR) using merged data (2011-2018) and stratified by survey years. Modified Poisson regression models were run with the following explanatory variables: age, sex, together with other socioeconomic, behavioural and oral-related variables mentioned above. Two-sided P-values <.05 were statistically significant.

3 RESULTS

As shown in Tables 1 and 2, a total of 25 048 participants were included in the present study. The mean age of participants was 85.2 ± 11.2 years. The proportion of young older adults (65-84 years old) and oldest older adults (≥85 years old) were 48.7% and 51.3%, respectively. Among all the participants, 55.2% were women, 53.5% were illiteracy and 63.5% were without denture. 33.2% of the older adults had no tooth, 31.3% had 1-10 teeth, while 16.7% and 18.8% of them still had more than nearly half of their natural teeth (11-20 and 21+), respectively. There was no obvious difference among 3 years' survey data.

TABLE 1. Basic characteristics of the 25 048 participants stratified by year 2011 2014 2018 Overall Number of participants, n 8087 5911 11 050 25 048 Age, mean (SD) 84.9 (10.9) 85.3 (10.3) 85.3 (11.8) 85.2 (11.2) Age, n (%) 65-74 1762 (21.8) 1007 (17.0) 2546 (23.0) 5135 (21.2) 75-84 2301 (28.5) 1936 (32.8) 2829 (25.6) 7066 (28.2) 85-94 2306 (28.5) 1732 (29.3) 2711 (24.5) 6749 (26.9) ≥95 1718 (21.2) 1236 (20.9) 2964 (26.8) 5918 (23.6) Sex, n (%) Male 3750 (46.4) 2695 (45.6) 4787 (43.3) 11 232 (44.8) Female 4337 (53.6) 3216 (54.4) 6263 (56.7) 13 816 (55.2) Year of education, n (%) 0 4571 (56.5) 3376 (57.1) 5444 (49.3) 13 391 (53.5) 1-6 2568 (31.8) 1864 (31.5) 3519 (31.8) 7951 (31.7) ≥7 948 (11.7) 671 (11.3) 2087 (18.9) 3706 (14.8) Marital status, n (%) Currently married and living with spouse 3129 (38.7) 2322 (39.3) 4434 (40.1) 9885 (39.5) Others 4958 (61.3) 3589 (60.7) 6616 (59.9) 15 163 (60.5) Getting adequate medical service, n (%) Yes 7632 (94.4) 5707 (96.5) 10 724 (97.0) 24 063 (96.1) No 455 (5.6) 204 (3.5) 326 (3.0) 985 (3.9) Enough financial support, n (%) Yes 6455 (79.8) 4889 (82.7) 9528 (86.2) 20 872 (83.3) No 1632 (20.2) 1022 (17.3) 1522 (13.8) 4176 (16.7) Current residential area, n (%) Urban 3877 (47.9) 2673 (45.2) 6382 (57.8) 12 932 (51.6) Rural 4210 (52.1) 3238 (54.8) 4668 (42.2) 12 116 (48.4) Medical insurance type, n (%) None 954 (11.8) 424 (7.2) 1193 (10.8) 2571 (10.3) Urban employee/resident 1112 (13.8) 950 (16.1) 2558 (23.1) 4620 (18.4) New rural cooperative 5354 (66.2) 4092 (69.2) 6500 (58.8) 15 946 (63.7) Others 667 (8.2) 445 (7.5) 799 (7.2) 1911 (7.6) TABLE 2. Behavioural, oral-related and health-related characteristics of the 25 048 participants stratified by year 2011 2014 2018 Overall Number of participants, n 8087 5911 11 050 25 048 Dietary diversity scores, n (%) 0-1 435 (5.4) 283 (4.9) 524 (4.7) 1242 (5.0) 2-5 4613 (57.0) 3313 (56.0) 6163 (55.8) 14 089 (56.2) 6-9 3039 (37.6) 2315 (39.2) 4363 (39.5) 9717 (38.8) Sugar intake, n (%) High 2708 (33.5) 1988 (33.6) 3341 (30.2) 8037 (32.1) Low 5379 (66.5) 3923 (66.4) 7709 (69.8) 17 011 (67.9) Main flavour, n (%) Insipidity 5072 (62.7) 3953 (66.9) 7608 (68.9) 16 633 (66.4) Salty 1571 (19.4) 1196 (20.2) 2149 (19.4) 4916 (19.6) Sweet 510 (6.3) 308 (5.2) 588 (5.3) 1406 (5.6) Hot 235 (2.9) 97 (1.6) 249 (2.3) 581 (2.3) Others 699 (8.6) 357 (6.0) 456 (4.1) 1512 (6.0) Smoking during lifetime, n (%) Ever 2835 (35.1) 1777 (30.1) 3305 (29.9) 7917 (31.6) Never 5252 (64.9) 4134 (69.9) 7745 (70.1) 17 131 (68.4) Alcohol consumption during lifetime, n (%) Ever 2608 (32.2) 1480 (25.0) 2879 (26.1) 6967 (27.8) Never 5479 (67.8) 4431 (75.0) 8171 (73.9) 18 081 (72.2) Number of natural teeth, n (%) None 2620 (32.4) 2006 (33.9) 3694 (33.4) 8320 (33.2) 1-10 2645 (32.7) 1933 (32.7) 3272 (29.6) 7850 (31.3) 11-20 1367 (16.9) 997 (16.9) 1817 (16.4) 4181 (16.7) 21+ 1455 (28.0) 975 (16.5) 2267 (20.5) 4697 (18.8) Denture, n (%) Without 2809 (34.7) 2093 (35.4) 4247 (38.4) 9149 (36.5) With 5278 (65.3) 3818 (64.6) 6803 (61.6) 15 899 (63.5) Toothbrushing frequency, n (%) Never 3593 (44.4) 2328 (39.4) 2322 (21.0) 8243 (32.9) Less than once a day 859 (10.6) 695 (11.8) 1357 (12.3) 2911 (11.6) Once a day 2562 (31.7) 2682 (45.4) 4406 (39.9) 9650 (38.5) More than once a day 1073 (13.3) 206 (3.5) 2965 (26.8) 4244 (16.9) One or more chronic medical conditions, n (%) Without 5592 (69.1) 4060 (68.7) 8267 (74.8) 17 919 (71.5) With 2495 (30.9) 1851 (31.3) 2783 (25.2) 7129 (28.5)

The overall toothache prevalence among older adults was 15.3% (95% CI: 14.9%-15.7%). Toothache prevalence rates were 16.3% (95% CI: 15.5%-17.1%), 12.8% (95% CI: 12.0%-13.7%) and 16.0% (95% CI: 15.3%-16.7%) in 2011, 2014 and 2018, respectively. The prevalence of toothache did not change over time. There were significant differences of toothache prevalence among people in different age groups (P < .001, Table 3). Moreover, participants with more years of schooling, more natural teeth and higher toothbrushing frequency were suffered from a higher toothache prevalence (P < .05). No significant difference was found in different sexes, while the toothache prevalence was 16.4%, 12.4% in male and 16.2%, 13.2% in female in 2011 and 2014, respectively. Additionally, those whose financial support was enough to pay for daily expenses or those who lived in rural area had a lower prevalence of toothache than those whose financial support was insufficient or those who lived in urban/town. As for behavioural factors, toothache prevalence was significantly higher in those who smoked or drank alcohol during the lifetime (P < .001, Table 4). We also found that toothache prevalence might be different among people with different medical insurance types and different main flavours.

TABLE 3. Toothache prevalence in Chinese older adults in 2011, 2014 and 2018 by basic characteristics Characteristics 2011 2014 2018 N Prevalence (95% CI) P N Prevalence (95% CI) P N Prevalence (95% CI) P Age 95+ 126 7.3 (6.1-8.6) <.001 68 5.5 (4.2-6.8) <.001 210 7.1 (6.2-8) <.001 85-94 268 11.6 (10.3-12.9) 172 9.9 (8.5-11.3) 332 12.2 (11-13.5) 75-84 491 21.3 (19.7-23) 329 17 (15.3-18.7) 582 20.6 (19.1-22.1) 65-74 432 24.5 (22.5-26.5) 188 18.7 (16.3-21.1) 643 25.3 (23.6-26.9) Sex Male 616 16.4 (15.2-17.6) .772 334 12.4 (11.1-13.6) .406 804 16.8 (15.7-17.9) .046 Female 701 16.2 (15.1-17.3) 423 13.2 (12-14.3) 963 15.4 (14.5-16.3) Education years 0 696 15.2 (14.2-16.3) .01 385 11.4 (10.3-12.5) <.001 694 12.7 (11.9-13.6) <.001 1-6 460 17.9 (16.4-19.4) 290 15.6 (13.9-17.2) 634 18 (16.7-19.3) 7+ 161 17 (14.6-19.4) 82 12.2 (9.7-14.7) 439 21 (19.3-22.8) Marital status Others 660 13.3 (12.4-14.3) <.001 398 11.1 (10.1-12.1) <.001 825 12.5 (11.7-13.3) <.001 Currently married 657 21 (19.6-22.4) 359 15.5 (14-16.9) 942 21.2 (20-22.4) Getting adequate medical service No 86 18.9 (15.3-22.5) .136 24 11.8 (7.3-16.2) .729 68 20.9 (16.4-25.3) .018 Yes 1231 16.1 (15.3-17) 733 12.8 (12-13.7) 1699 15.8 (15.2-16.5) Enough financial support No 370 22.7 (20.6-24.7) <.001 173 16.9 (14.6-19.2) <.001 336 22.1 (20-24.2) <.001 Yes 947 14.7 (13.8-15.5) 584 11.9 (11-12.9) 1431 15 (14.3-15.7) Current residential area Rural 642 15.2 (14.2-16.3) .009 329 10.2 (9.1-11.2) <.001 678 14.5 (13.5-15.5) <.001 Urban 675 17.4 (16.2-18.6) 428 16 (14.6-17.4) 1089 17.1 (16.1-18) Medical insurance type None 162 17 (14.6-19.4) .035 52 12.3 (9.1-15.4) <.001 203 17 (14.9-19.1) <.001 Urban employee/resident 208 18.7 (16.4-21) 161 16.9 (14.6-19.3) 467 18.3 (16.8-19.8) New rural cooperative 829 15.5 (14.5-16.5) 488 11.9 (10.9-12.9) 967 14.9 (14-15.7) Others 118 17.7 (14.8-20.6) 56 12.6 (9.5-15.7) 130 16.3 (13.7-18.8) Total 1317 16.3 (15.5-17.1) 757 12.8 (12.0-13.7) 1767 16.0 (15.3-16.7) TABLE 4. Toothache prevalence in Chinese older adults in 2011, 2014 and 2018 by behavioural, oral-related and health-related characteristics Characteristics 2011 2014 2018 N Prevalence (95% CI) P N Prevalence (95% CI) P N Prevalence (95% CI) P Dietary diversity scores 0-1 60 13.8 (10.6-17) .300 24 8.5 (5.2-11.7) .080 82 15.6 (12.5-18.8) .040 2-5 764 16.6 (15.5-17.6) 434 13.1 (12-14.2) 939 15.2 (14.3-16.1) 6-9 493 16.2 (14.9-17.5) 299 12.9 (11.5-14.3) 746 17.1 (16-18.2) Sugar intake Low 874 16.2 (15.3-17.2) .924 511