Epidemiology profile and outcomes of oral potentially malignant disorders among different geographic regions of Taiwan: A retrospective observational nationwide population database study

Sheng-Fu Chiu; Chung-Han Ho; Yi-Chen Chen; Li-Wha Wu; Yuh-Ling Chen; Jer-Horng Wu; Wei-Sheng Wu; Hui-Kun Hung; Wei-Fan Chiang

doi:10.4103/jmedsci.jmedsci_257_20

ORIGINAL ARTICLE

Year : 2021 | Volume : 41 | Issue : 6 | Page : 286-294

Epidemiology profile and outcomes of oral potentially malignant disorders among different geographic regions of Taiwan: A retrospective observational nationwide population database study

Sheng-Fu Chiu¹, Chung-Han Ho², Yi-Chen Chen³, Li-Wha Wu⁴, Yuh-Ling Chen⁵, Jer-Horng Wu⁶, Wei-Sheng Wu⁷, Hui-Kun Hung⁸, Wei-Fan Chiang⁹
¹ Department of Oral and Maxillofacial Surgery, Chi Mei Medical Center, Liouying, Taiwan
² Department of Medical Research, Chi Mei Medical Center; Department of Hospital and Health Care Administration, Chia Nan University of Pharmacy and Science, Tainan, Taiwan
³ Department of Medical Research, Chi Mei Medical Center, Tainan, Taiwan
⁴ Institute of Molecular Medicine, College of Medicine, National Cheng Kung University, Tainan; Department of Laboratory Science and Technology, College of Health Sciences, Kaohsiung Medical University, Kaohsiung, Taiwan
⁵ Institute of Oral Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan
⁶ Department of Environmental Engineering, National Cheng Kung University, Tainan, Taiwan
⁷ Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan
⁸ Department of Plastic Surgery, Chi Mei Medical Center, Liouying, Taiwan
⁹ Department of Oral and Maxillofacial Surgery, Chi Mei Medical Center, Liouying; Department of School of Dentistry, National Yang Ming University, Taipei, Taiwan

Date of Submission	12-Aug-2020
Date of Decision	01-Dec-2020
Date of Acceptance	22-Feb-2021
Date of Web Publication	24-May-2021

Correspondence Address:
Dr. Wei-Fan Chiang
No. 201 Taikang, Liouying, Tainan 73657
Taiwan

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/jmedsci.jmedsci_257_20

Abstract

Background: The malignant transformation of oral potentially malignant disorders (OPMDs) is a potential cause of oral cancers. However, epidemiological studies on malignant transformation of OPMDs are lacking. The present study is aimed to investigate the percentage of OPMD in the oral mucosal screening database and the malignant transformation (%) across different regions of Taiwan. Methods: We determined the cases of malignant transformation of OPMD to oral cancers during the 2-year follow-up by the oral mucosal screening database and the cancer registry database contained in the National Health Insurance Research Database. Patients with oral cancer history before OPMD diagnosis were excluded from the study. We collected data including clinical diagnosis, biopsy result, follow-up, and place of residence of the patient. The most widely used definition from the Council for Economic Planning and Development in Taiwan was used in this study for the classification into Southern, Northern, Eastern, and Western Taiwan regions. Differences in the malignant transformation (%) and ranking between regions were evaluated using analysis of variance test. Results: Chiayi city located in Southern Taiwan has the highest percentage of OPMD in oral screening citizen. The percentage of OPMD in the oral mucosal screening database was highest in Southern Taiwan and lowest in Eastern Taiwan; however, the malignant transformation (%) was highest in the Eastern Taiwan and lowest in Northern Taiwan. Conclusion: Dentists and otolaryngologists in Southern Taiwan, especially Chiayi city, should be cautious when screening the patient because of the high percentage of OPMD in oral mucosal screening database. The percentage of OPMD in the oral mucosal screening database and the malignant transformation (%) are different among the different regions in Taiwan. Further research is needed to clarify the reasons for the different malignant transformation (%) between regions.

Keywords: Oral cancer, oral potentially malignant disorders, oral mucosal screening, city of Taiwan, region of Taiwan, rate of OPMD per 1000 population, malignant transformation

How to cite this article:
Chiu SF, Ho CH, Chen YC, Wu LW, Chen YL, Wu JH, Wu WS, Hung HK, Chiang WF. Epidemiology profile and outcomes of oral potentially malignant disorders among different geographic regions of Taiwan: A retrospective observational nationwide population database study. J Med Sci 2021;41:286-94

How to cite this URL:
Chiu SF, Ho CH, Chen YC, Wu LW, Chen YL, Wu JH, Wu WS, Hung HK, Chiang WF. Epidemiology profile and outcomes of oral potentially malignant disorders among different geographic regions of Taiwan: A retrospective observational nationwide population database study. J Med Sci [serial online] 2021 [cited 2023 Nov 28];41:286-94. Available from: https://www.jmedscindmc.com/text.asp?2021/41/6/286/316667

Introduction

Previous studies have shown significant differences in cancer rates in different geographic regions of Taiwan with respect to environmental poisons, heavy metals, soil pollution, water quality, lifestyle, special habits, economic status, or resources of medication.^[1],[2],[3] However, epidemiological survey of oral cancer in different geographic regions has not been conducted and present study was conducted to fill this gap.

Oral cancer, one of the leading causes of cancer-related death in Taiwan,^[4] is usually derived from oral potentially malignant disorders (OPMDs). OPMD includes leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus, as defined by the World Health Organization. The process of OPMD lesions transforming to invasive oral cancer has been defined as malignant transformation.^[4] Based on the concept of preventive medicine,^[4] the Taiwanese government has started a free oral mucosal screening since 2010 for people with a history of smoking or betel nut chewing, even if the individuals have quit those habits. Dentists and otolaryngologists perform free oral mucosal screening and record the clinical data of patients in the oral mucosal screening database (oral screening database), established by the National Health Insurance Research Database.

This study is aimed to survey the epidemiology and the malignant transformation of OPMD among different geographic regions in Taiwan, providing information for further government resource allocation.

Materials and Methods

Taiwan launched the National Health Insurance program in 1995, covering 99% of Taiwan's residents by 2010.^[5],[6],[7] The National Health Insurance Research Database, encrypted residents' identification to ensure privacy, contains the government-paid oral screening database and cancer registry database.

The oral screening database, with data from 2010, contains clinical diagnosis, biopsy result, follow-up, further treatment condition, smoking or betel nut chewing history, place of residence, household register, and detailed information of the patient and the clinic/hospital. The clinical diagnosis includes oral mucosal erosion, oral ulcer, oral canker sores, leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, oral lichen planus, and highly suspect malignancy.

The cancer registry database, with data from 2002, contains the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10 codes for diagnoses, date of diagnosis, whether operation/chemotherapy/radiotherapy was performed, and detailed information of the status of cancer.

Selection of patients and variables

OPMD in this study included the clinical diagnosis of leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus, in the oral screening database. If the histological biopsy report, performed simultaneously with the clinical diagnosis of OPMD, confirmed the lesions as oral cancer, the patient was excluded from the study. If a patient had undergone oral screening more than once, during 2010–2013, we include his first OPMD record in our study only. The definition of oral cancer in the present study includes oral squamous cell carcinoma, verrucous carcinoma, and carcinoma in situ. The ICD-9-CM and ICD-10 codes of oral cancer were used in the data collection from the cancer registry database.

The oral screening database was linked with the cancer registry database, to investigate if the patient diagnosed with OPMD during 2010–2013 subsequently developed oral cancer, which means underwent the malignant transformation, at the 2-year follow-up.

Patients with a previous oral cancer history before the diagnosis date of OPMD were excluded from this study. [Figure 1]a is the flow chart of this study with details of sample selection and the number of cases.

Figure 1: Selection of patients and variables in our study. (a) The flow chart of this study with number of cases. The present study was based on data of cancer registry and oral mucosal screening database, which belongs to National Health Insurance Research Database. Data were obtained using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10 codes for diagnoses. Oral potentially malignant disorders include leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus. Oral screening: Taiwanese government has started a free oral mucosal screening since 2010. Malignant transformation: Oral cancer occurring during the follow-up of oral potentially malignant disorders is defined as malignant transformation. (b) The division of Taiwan into four geographic regions. The four geographic regions of Taiwan are based on the most widely used definition from the council for economic planning and development, agency in Executive Yuan, a branch of the Taiwan government. Details of the city/county in each region are given in Table 1.

Click here to view

We recorded the place of residence of OPMD patients from the oral screening database. The total population of each city was quoted from the Department of Household Registration, Ministry of the Interior in Taiwan government (please see Appendices). The division of Taiwan into four regions is based on the most widely used definition from the council for economic planning and development, agency in Executive Yuan, a branch of the Taiwan government [Figure 1]b. Details of the city/county and regions are given in [Table 1]. All outlying islands were classified as part of southern Taiwan.

Table 1: Oral potentially malignant disorder case number during 2010-2013 in each city/county of Taiwan

Click here to view

Ethical considerations

Because all types of personal data are encrypted by the National Health Insurance Research Database to ensure patient privacy, this study was granted an exemption from a full ethical review by the Institutional Review Board. The present study was approved by the Institutional Review Board of the Chi Mei Medical Center (IRB Serial No.:10705-E05).

Statistical analysis

Percentage of OPMD in the oral screening database is defined as the case number of OPMD divided by case number of people who underwent free oral screening, as presented in equation:

Percentage of OPMD in the oral screening database

This study aimed to survey the rates of OPMD per 1000 population in different regions of Taiwan, calculated using the case number of OPMD of each region divided by case number of the total population of each region, as presented in equation:

The total population is estimated by 2013 survey data of the department of household registration, Ministry of the Interior in Taiwan government.

Some OPMD cases acquired oral cancer in a 2-year retrospective follow-up study period; in this situation, these cases are referred to as subsequent oral cancer.

The malignant transformation (%)^[4] is calculated using the case number of subsequent oral cancer divided by case number of OPMD, as presented in equation:

The differences between the percentage of OPMD in the oral screening database, rates of OPMD per 1000 population, and malignant transformation (%) were derived from the analysis of one-way analysis of variance (ANOVA) (one-way ANOVA), with post hoc Tukey's Honestly Significant Difference (HSD) test which was used to calculated statistical difference for ranking between multiple Taiwan regions. SAS 9.4 (SAS Institute Inc., Cary, NC, USA) was used for data analyses. Statistical significance was set at P < 0.0001.

A t-test was used for determining the significance between the city with the highest percentage of OPMD in the oral screening database and the other cities in Taiwan. The t-test was also used for assessing the significance between the city with the highest rate of OPMD per 1000 population and the other cities.

Results

A total of 3,362,232 people received free oral mucosal screening between 2010 and 2013, the first 4 years of free screening policy for smokers or betel nut chewers. After excluding patients with oral squamous cell carcinoma, verrucous carcinomas, or carcinoma in situ, according to the first oral biopsy concurrent to OPMD's first diagnosis, and those with oral cancer history before receiving free oral mucosal screening, we enrolled 155,103 OPMD patients [Figure 1]a.

Percentage of OPMD in the oral screening database is defined as the case number of OPMD divided by case number of people who underwent free oral screening, as presented in equation:

Percentage of OPMD in the oral screening database

[Figure 2] shows the distribution of the percentage of OPMD in the oral screening database among different cities in Taiwan. Chiayi City has the highest percentage of OPMD in oral screening citizens (7.81%), followed by Yunlin (6.69%) and Changhua (6.46%). Chiayi city has a significantly higher percentage of OPMD in the oral screening database than other cities (P < 0.0001, t-test). Lienchiang has the lowest OPMD percentage in oral screening citizens (2.10%). Hualien and Hsinchu County also have very low percentages of OPMD in the oral screening database.

Figure 2: Distribution of percentage of oral potentially malignant disorders in oral mucosal screening database among different cities/counties in Taiwan. OPMD: Oral potentially malignant disorders, including leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus

Click here to view

According to our study design of the definition of Taiwan regions [Figure 1]b, percentages of OPMD in oral screening database among different regions in Taiwan show in [Figure 3]a. The percentages of OPMD in the oral screening database from the highest to the lowest were as follows: South Taiwan > Central Taiwan > North Taiwan > East Taiwan (one-way ANOVA, P < 0.0001; post hoc HSD test for ranking between multiple regions, P < 0.0001).

Figure 3: Percentage of oral potentially malignant disorders and malignant transformation in oral mucosal screening database among different regions in Taiwan. (a) Percentage of oral potentially malignant disorders in oral mucosal screening database among different regions in Taiwan. OPMD: Oral potentially malignant disorders, including leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus. Oral cancer: Including oral squamous cell carcinoma, verrucous carcinoma, and carcinoma in situ. (b) The malignant transformation (%) among different regions. Malignant transformation: Subsequent oral cancer occurring during the follow-up of oral potentially malignant disorders is defined as malignant transformation

Click here to view

[Table 2] demonstrates the number of malignant transformation cases, as the most important outcome, of OPMD patients during the 2-year follow-up. The subsequent oral cancer case number divided by the OPMD case number represented the malignant transformation (%) [Figure 3]b.

Table 2: The oral potentially malignant disorder and oral cancer during 2-year follow-up in each region of Taiwan

Click here to view

Percentage of OPMD in the oral screening database is defined as the case number of OPMD divided by case number of people who underwent free oral screening, as presented in equation:

Percentage of OPMD in the oral screening database

Malignant transformation: Subsequent oral cancer occurring during the follow-up of OPMD is defined as malignant transformation. The case number of subsequent oral cancer divided by case number of OPMD represented the malignant transformation (%), as presented in equation:

Subsequent oral cancer: The OPMD cases acquired oral cancer in a 2-year retrospective follow-up study period
Oral cancer: Including oral squamous cell carcinoma, verrucous carcinoma, and carcinoma in situ.

The malignant transformation rates among different regions could be ranked as follows: East Taiwan > Central Taiwan > South Taiwan > North Taiwan (one-way ANOVA, P < 0.0001; post hoc HSD test for ranking between multiple regions, P < 0.0001).

The malignant transformation rate in East Taiwan was significantly higher than average in Taiwan (3.22% vs. 2.68%; P < 0.0001).

To calculate the rate of OPMD per 1000 population in each city, the number of citizens, smokers, and betel chewers are required. First, we obtained the number of citizens in 2012 from the online release data of the Department of Household Registration, Ministry of the Interior in Taiwan government. However, neither the number of smokers or betel chewers nor their oral mucosal screening rates are available from any public source. For future research, it can be assumed that the widespread promotion of the oral screening policy by the Taiwanese government would cover almost all smoking or betel nut chewing citizens after 4 years. According to previous studies,^[4],[8],[9] the number of oral cancers in the non-OPMD population and the number of OPMDs from nonsmoker/nonbetel chewer populations are very small and could be ignored. Explanatory diagrams have been added to explain this hypothesis more clearly [Figure 4]. The orange block can be supposed to be near the size of the blue column [Figure 4]b, according to the clinical viewpoint:

Figure 4: Models of disease distribution. (a) A model of actual disease distribution of citizen. (b) Hypothesis of our study. OPMD: Oral potentially malignant disorders, including leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus. Oral cancer: Including oral squamous cell carcinoma, verrucous carcinoma, and carcinoma in situ.

Click here to view

The numbers of non-OPMD oral cancer and nonsmoker/nonbetel chewers are small, they can be ignored
The Taiwanese government launched free oral cancer screening for betel chewers or cigarette smokers starting in 2010, vigorously promoting in dental and otolaryngologic clinics and hospitals.

Based on the hypothesis in [Figure 4], the top three rates of OPMD per 1000 population^[10] in the general population of each city are in Chiayi city (1.78%), Taitung (0.99%), and Changhua (0.91%) [Figure 5]a.

Figure 5: Rates of oral potentially malignant disorders per 1000 population in Taiwan. (a) Rates of oral potentially malignant disorders per 1000 population among cities. (b) Rates of oral potentially malignant disorders per 1000 population among regions. All illustrations and investigation of Figure 5 were based on the hypothesis of Figure 4. OPMD: Oral potentially malignant disorders, including leukoplakia, erythroplakia, oral submucosal fibrosis, oral verrucous hyperplasia, and oral lichen planus

Click here to view

Chiayi city has the highest rate of OPMD per 1000 population, which is significantly higher than that in other cities (P < 0.0001, t-test). The lowest rates of OPMD per 1000 population are in Lienchiang, Hsinchu County, and Penghu, two of which are located on the outer islands.

Rates of OPMD of each city per 1000 population: The case number of OPMD of each city divided by case number of total population of each city, as presented in equation:

Total population is estimated by 2013 survey data of the department of household registration; Ministry of the Interior in Taiwan government is the household registration

Based on the hypothesis in [Figure 4], rates of OPMD per 1000 population from the highest to the lowest in different Taiwan regions were as follows: Eastern Taiwan > Southern Taiwan > Central Taiwan > Northern Taiwan (one-way ANOVA, P < 0.0001; post hoc HSD test for ranking between multiple regions, P < 0.0001) [Figure 5]b.

Rates of OPMD of each region per 1000 population: The case number of OPMD of each region divided by case number of total population of each region, as presented in equation:

Total population is estimated by 2013 survey data of the department of household registration; Ministry of the Interior in Taiwan government is the household registration.

According to the hypothesis in [Figure 4], Hualien (23.9%), Chiayi city (22.7%), and Taitung (21.9%) had the highest prevalence of smoking or betel nut chewing. Of the top three cities with the highest smoking or betel nut chewing prevalence, two are located in east Taiwan, when there are only two cities in eastern Taiwan.

Discussion

The epidemiology data and outcomes of malignant transformation from OPMD to oral cancer using an oral mucosal screening database and cancer registry database were surveyed. To the best of our knowledge, this is the first study to demonstrate the rates of OPMD per 1000 population under the support of government policy, based on the nationwide databases, and the hypothesis in [Figure 4].

Chiayi city, located in Southern Taiwan, has the highest percentage of OPMD in oral screening citizen. Furthermore, four of the top five cities with the highest percentage of OPMD in oral screening database were located in Southern Taiwan. Dentists and otolaryngologists working in Southern Taiwan, especially Chiayi city, should be cautious when screening the patient because of the high OPMD positivity rate in oral screening. The citizens in Chiayi city had the highest rate of OPMD per 1000 population according to [Figure 4] hypothesis; thus, the residents should prevent smoking or chewing betel nut, which had strong evidence of OPMD development.^[4],[8],[9] However, regardless of the higher percentage of OPMD in oral screening database and the highest rate of OPMD per 1000 population, Chiayi city only has a small population. The largest number of OPMD cases was in New Taipei, Kaohsiung, and Taichung, cities with the highest population in Taiwan. Based on the largest number of OPMD patients, local governments of New Taipei, Kaohsiung, and Taichung should obtain more resources to promote regular follow-up of OPMD.

Although the percentage of OPMD in oral screening database is lower in Eastern Taiwan than in other regions, the malignant transformation (%) is the highest. Malignant transformation detection is most important for early cancer treatment. The etiology of malignant transformation is still unclear; however, a few studies provided some suspected risk factors. For example, elevated concentrations of heavy metals in the soil reflected the exposure to the human body in Eastern Taiwan.^[11] Environmental nickel exposure may be associated with an increased risk of lung cancer^[12] and malignant transformation of OPMD.^{[13],[14],[15]} Arsenic in farm soils may be a new risk factor for oral cancer development.^[15] Chromium causes higher risk of cancer.^[16] Since Taiwanese individuals have higher blood mercury and cadmium levels than Western populations,^[17] further research of the relationship between mercury, cadmium, and cancer formation is needed. The levels of sulfate, nitrate, and ammonium in PM (2.5)^[18] vary inter-regionally, as widely reported in Taiwan and their role in cancer formation should be surveyed.

The association of human papillomavirus in the development of oral cancer had been assessed.^[19],[20] Since the rate of Klebsiella pneumoniae bacterial infections varies inter-regionally,^[21] there might be contrasting infection rate of human papillomavirus inter-regionally too. Comorbid diseases and older age were also associated with the incidence of cancer.^[22],[23] Special habits related to local culture should also be surveyed.^[24]

Further research is needed to clarify the reasons for the different malignant transformation (%) between Northern and Eastern Taiwan. Although the percentage of OPMD in oral screening database is low, Eastern Taiwanese citizens accomplished the highest screening rate among all populations. This may be the reason of the highest rate of OPMD per 1000 population, as estimated under the hypothesis of [Figure 4]. When the diagnosis of OPMD is established in Eastern Taiwan, dentists and otolaryngologists should closely follow-up on the patient and schedule biopsy more aggressively. However, regardless of the higher malignant transformation (%), Eastern Taiwan has a smaller population than other regions. The largest number of oral cancer cases was in Southern Taiwan. Due to the large number of oral cancer patients, local governments in Southern Taiwan should ensure sufficient resources for oral cancer treatment, such as the surgeons, radiotherapy, and chemotherapy. The Eastern Taiwan government should also improve the convenience of going to the medical centers, taking responsibility for formal cancer treatment.^{[23],[25],[24],[26],[27],[28]}

Limitation

One of the policies of the National Health Insurance Research Database is that the data cannot be published when the number of cases is <2; therefore, we can only demonstrate the number of cases of regional data instead of that with malignant transformation in each city.

Similarly, all outlying islands cannot be surveyed individually and can be classified as part of southern Taiwan by the author.

We linked the oral mucosal screening database with the cancer registry database and ruled out patients with oral cancer history before OPMD diagnosis from our study group. However, the cancer registry database started in 2002, so patients who had oral cancer before 2002 could not be excluded due to our study design.

We did not know the actual time of OPMD existence, and the time it took to undergo malignant transformation. The 2-year study period may be too short for studying malignant transformation.

The rates of OPMD per 1000 population were based on the assumption of [Figure 4]b. However, the reality model is shown in [Figure 4]a. The highest screening rate in all citizens among Eastern Taiwan may be due to either high smoking or betel nut chewing rates or because of vigorously promotion of the general screening policy by the government and health agencies. The lowest screening rate in all citizens among Northern Taiwan may be due to either low smoking or betel nut chewing rates or poor promotion of the screening policy.

The total population of each city quoted from the department of household registration, rather than place of residence, which was adopted in recoding variable.

Conclusion

The percentages of OPMD in oral mucosal screening database are different among different cities in Taiwan. The malignant transformation rates also showed differences between regions. Further research is required to identify risk factors for malignant transformation and to clarify the reasons for different malignant transformation (%), especially between Eastern and Northern Taiwan.

Vigorous screening policy should be adopted by government, health agencies, and dentists and otolaryngologists, to monitor patients with OPMD precisely.

Acknowledgment

This study is based on data from the Taiwan National Health Insurance Research Database provided by the Taiwan National Health Insurance, Taiwan Department of Health, Taipei, and managed by the National Health Research Institutes. Editage provided language help, including improved in grammar, sentence structure, general readability, native English usage, and English editing certificate for the manuscript.

Financial support and sponsorship

This work was supported by the Ministry of Science and Technology, Taiwan (grant no. MOST 107-2321-B-006-007), and Ministry of Science and Technology, Taiwan (grant no. MOST 108-2321-B-006-010), and supported by Chi Mei Medical Center, Liouying (grant no. CLFHR10824). The funding sources had no involvement in study design, the collection, the analysis, the interpretation of data, as well as the writing of the report.

Conflicts of interest

There are no conflicts of interest.

References

1.	Tseng KS, Lin C, Lin YS, Weng SF. Risk of head and neck cancer in patients with diabetes mellitus: A retrospective cohort study in Taiwan. JAMA Otolaryngol Head Neck Surg 2014;140:746-53.
2.	Ko YC, Lee CH, Chen MJ, Huang CC, Chang WY, Lin HJ, et al. Risk factors for primary lung cancer among non-smoking women in Taiwan. Int J Epidemiol 1997;26:24-31.
3.	Lee MC, Jones AM. Understanding differences in income-related health inequality between geographic regions in Taiwan using the SF-36. Health Policy 2007;83:186-95.
4.	Wang YY, Tail YH, Wang WC, Chen CY, Kao YH, Chen YK, et al. Malignant transformation in 5071 southern Taiwanese patients with potentially malignant oral mucosal disorders. BMC Oral Health 2014;14:99.
5.	Yeh HY, Wu JC, Haschler I, Chen TJ, Wetter T. Taiwan's National Health Insurance Research Database: Administrative health care database as study object in bibliometrics. Scientometrics 2011;86:365-80.
6.	Ho Chan WS. Taiwan's healthcare report 2010. EPMA J 2010;1:563-85.
7.	Hsing AW, Ioannidis JP. Nationwide population science: Lessons from the Taiwan national health insurance research database. JAMA Intern Med 2015;175:1527-9.
8.	Ganesh D, Sreenivasan P, Öhman J, Wallström M, Braz-Silva PH, Giglio D, et al. Potentially malignant oral disorders and cancer transformation. Anticancer Res 2018;38:3223-9.
9.	Shiu MN, Chen TH, Chang SH, Hahn LJ. Risk factors for leukoplakia and malignant transformation to oral carcinoma: A leukoplakia cohort in Taiwan. Br J Cancer 2000;82:1871-4.
10.	Creighton P, Lew JB, Clements M, Smith M, Howard K, Dyer S, et al. Cervical cancer screening in Australia: Modelled evaluation of the impact of changing the recommended interval from two to three years. BMC Public Health 2010;10:734.
11.	Chiang CT, Lian Ie B, Su CC, Tsai KY, Lin YP, Chang TK. Spatiotemporal trends in oral cancer mortality and potential risks associated with heavy metal content in Taiwan soil. Int J Environ Res Public Health. 2010;7:3916-28.
12.	Kuo CY, Wong RH, Lin JY, Lai JC, Lee H. Accumulation of chromium and nickel metals in lung tumors from lung cancer patients in Taiwan. J Toxicol Environ Health A 2006;69:1337-44.
13.	Tsai KY, Su CC, Chiang CT, Tseng YT, Lian IB. Environmental heavy metal as a potential risk factor for the progression of oral potentially malignant disorders in central Taiwan. Cancer Epidemiol 2017;47:118-24.
14.	14. Yuan TH, Lian Ie B, Tsai KY, Chang TK, Chiang CT, Su CC, et al. Possible association between nickel and chromium and oral cancer: A case-control study in central Taiwan. Sci Total Environ 2011;409:1046-52.
15.	Su CC, Lin YY, Chang TK, Chiang CT, Chung JA, Hsu YY, et al. Incidence of oral cancer in relation to nickel and arsenic concentrations in farm soils of patients' residential areas in Taiwan. BMC Public Health 2010;10:67.
16.	Yang SY, Lin JM, Lin WY, Chang CW. Cancer risk assessment for occupational exposure to chromium and nickel in welding fumes from pipeline construction, pressure container manufacturing, and shipyard building in Taiwan. J Occup Health 2018;60:515-24.
17.	Liu TY, Hung YM, Huang WC, Wu ML, Lin SL. Do people from Taiwan have higher heavy metal levels than those from Western countries? Singapore Med J 2017;58:267-71.
18.	Chen YT, Du WJ, Chen JS, Xu LL. Pollution Characteristics of Heavy Metals in PM2.5 and their human health risks among the Coastal City group along western Taiwan straits region, China. Huan Jing Ke Xue 2017;38:429-37.
19.	Rapado-González Ó, Martínez-Reglero C, Salgado-Barreira Á, Rodríguez-Fernández A, Aguín-Losada S, León-Mateos L, et al. Association of salivary human papillomavirus infection and oral and oropharyngeal cancer: A Meta-Analysis. J Clin Med 2020;9:1305.
20.	Rutkoski H, Tay DL, Dixon BL, Pinzon LM, Mooney R, Winkler JR, et al. A multi-state evaluation of oral health students' knowledge of human papillomavirus-related oropharyngeal cancer and HPV vaccination. J Cancer Educ 2019;35:1017-25.
21.	Chen HL, Bair MJ, Lin IT, Wu CH, Lee YK. Clinical manifestations and risk factors of amebic liver abscess in Southeast Taiwan compared with other regions of Taiwan. Am J Trop Med Hyg 2013;89:1214-8.
22.	Cheng BR, Chang HT, Lin MH, Chen TJ, Chou LF, Hwang SJ. Rural-urban disparities in family physician practice patterns: A nationwide survey in Taiwan. Int J Health Plann Manage 2019;34:e464-73.
23.	Liao PA, Chang HH, Yang FA. Does the universal health insurance program affect urban-rural differences in health service utilization among the elderly? Evidence from a longitudinal study in taiwan. J Rural Health 2012;28:84-91.
24.	Cheng MT, Shih FY, Tsai CL, Tsai HB, Tsai DF, Fang CC. Impact of major illnesses and geographic regions on do-not-resuscitate rate and its potential cost savings in Taiwan. PLoS One 2019;14:e0222320.
25.	Lin CY, Tseng YC, Guo HR, Lai DC. Prevalence of childhood hearing impairment of different severities in urban and rural areas: A nationwide population-based study in Taiwan. BMJ Open 2018;8:e020955.
26.	Wang LY, Chang YS, Liang FW, Lin YC, Lin YJ, Lu TH, et al. Comparing regional neonatal mortality rates: The influence of registration of births as live born for birth weight<500 g in Taiwan. BMJ Paediatr Open 2019;3:e000526.
27.	Wen PC, Lee CB, Chang YH, Ku LE, Li CY. Demographic and rural-urban variations in dental service utilization in Taiwan. Rural Remote Health 2017;17:4161.
28.	Leu HI, Chang WT, Lin MH, Chen TJ, Hwang SJ, Chou LF, et al. Urban-rural disparity in geographical and temporal availability of pediatric clinics: A nationwide survey in Taiwan. Pediatr Neonatol 2017;58:344-9.