[Skip to Content]
[Skip to Content Landing]
Table 1.  
Factor Loadings and Item Parameters for a Unidimensional Model of DSM-5 Betel-Quid Use Disorder, Combined Results From 6 Asian Populationsa
Factor Loadings and Item Parameters for a Unidimensional Model of DSM-5 Betel-Quid Use Disorder, Combined Results From 6 Asian Populationsa
Table 2.  
Prevalence and Pattern of 12-Month DSM-5 BUD Symptoms
Prevalence and Pattern of 12-Month DSM-5 BUD Symptoms
Table 3.  
Association of Use Characteristics With Pathologic Behaviors and BQ Use Disorder Among Current Users of BQ, Combined Results From 6 Asian Populationsa
Association of Use Characteristics With Pathologic Behaviors and BQ Use Disorder Among Current Users of BQ, Combined Results From 6 Asian Populationsa
Table 4.  
Odds of Oral Potentially Malignant Disorder Associated With Demographic and Risk Factors, Use Characteristics, and DSM-5 BQ Use Disordera
Odds of Oral Potentially Malignant Disorder Associated With Demographic and Risk Factors, Use Characteristics, and DSM-5 BQ Use Disordera
Table 5.  
Combined and Conditional Associations of DSM-5 Symptoms Among Users of BQ and BQ Use Disorder With OPMD, Combined Resultsa
Combined and Conditional Associations of DSM-5 Symptoms Among Users of BQ and BQ Use Disorder With OPMD, Combined Resultsa
1.
Boucher  BJ, Mannan  N.  Metabolic effects of the consumption of Areca catechu Addict Biol. 2002;7(1):103-110.PubMedGoogle ScholarCrossref
2.
Gupta  PC, Warnakulasuriya  S.  Global epidemiology of areca nut usage.  Addict Biol. 2002;7(1):77-83.PubMedGoogle ScholarCrossref
3.
IARC Working Group on the Evaluation of Carcinogenic Risks to Humans.  Betel-quid and areca-nut chewing and some areca-nut derived nitrosamines.  IARC Monogr Eval Carcinog Risks Hum. 2004;85:1-334.PubMedGoogle Scholar
4.
Lee  CH, Ko  AM, Warnakulasuriya  S,  et al.  Intercountry prevalences and practices of betel-quid use in south, southeast and eastern Asia regions and associated oral preneoplastic disorders: an international collaborative study by Asian Betel-quid Consortium of south and east Asia.  Int J Cancer. 2011;129(7):1741-1751.PubMedGoogle ScholarCrossref
5.
Lee  CH, Ko  AM, Yen  CF,  et al.  Betel-quid dependence and oral potentially malignant disorders in six Asian countries.  Br J Psychiatry. 2012;201(5):383-391.PubMedGoogle ScholarCrossref
6.
Lee  CH, Ko  AM, Warnakulasuriya  S,  et al.  Population burden of betel quid abuse and its relation to oral premalignant disorders in South, Southeast, and East Asia: an Asian Betel-quid Consortium Study.  Am J Public Health. 2012;102(3):e17-e24.PubMedGoogle ScholarCrossref
7.
Lee  CH, Chiang  SL, Ko  AM,  et al.  Betel-quid dependence domains and syndrome associated with betel-quid ingredients among chewers: an Asian multi-country evidence.  Addiction. 2014;109(7):1194-1204.PubMedGoogle ScholarCrossref
8.
Lord  GA, Lim  CK, Warnakulasuriya  S, Peters  TJ.  Chemical and analytical aspects of areca nut.  Addict Biol. 2002;7(1):99-102.PubMedGoogle ScholarCrossref
9.
Chu  NS.  Effects of Betel chewing on the central and autonomic nervous systems.  J Biomed Sci. 2001;8(3):229-236.PubMedGoogle ScholarCrossref
10.
Papke  RL, Horenstein  NA, Stokes  C.  Nicotinic activity of arecoline, the psychoactive element of ‘betel nuts’, suggests a basis for habitual use and anti-inflammatory activity.  PLoS One. 2015;10(10):e0140907.PubMedGoogle ScholarCrossref
11.
Patidar  KA, Parwani  R, Wanjari  SP, Patidar  AP.  Various terminologies associated with areca nut and tobacco chewing: a review.  J Oral Maxillofac Pathol. 2015;19(1):69-76.PubMedGoogle ScholarCrossref
12.
Bhat  SJ, Blank  MD, Balster  RL, Nichter  M, Nichter  M.  Areca nut dependence among chewers in a South Indian community who do not also use tobacco.  Addiction. 2010;105(7):1303-1310.PubMedGoogle ScholarCrossref
13.
Benegal  V, Rajkumar  RP, Muralidharan  K.  Does areca nut use lead to dependence?  Drug Alcohol Depend. 2008;97(1-2):114-121.PubMedGoogle ScholarCrossref
14.
Gowing  LR, Ali  RL, Allsop  S,  et al.  Global statistics on addictive behaviours: 2014 status report.  Addiction. 2015;110(6):904-919.PubMedGoogle ScholarCrossref
15.
Mirza  SS, Shafique  K, Vart  P, Arain  MI.  Areca nut chewing and dependency syndrome: is the dependence comparable to smoking? a cross sectional study.  Subst Abuse Treat Prev Policy. 2011;6:23.PubMedGoogle ScholarCrossref
16.
Osborne  PG, Ko  YC, Wu  MT, Lee  CH.  Intoxication and substance use disorder to Areca catechu nut containing betel quid: a review of epidemiological evidence, pharmacological basis and social factors influencing quitting strategies.  Drug Alcohol Depend. 2017;179:187-197.PubMedGoogle ScholarCrossref
17.
American Psychiatric Association.  Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Association; 2013.
18.
Hasin  DS, O’Brien  CP, Auriacombe  M,  et al.  DSM-5 criteria for substance use disorders: recommendations and rationale.  Am J Psychiatry. 2013;170(8):834-851.PubMedGoogle ScholarCrossref
19.
Lee  CH, Ko  YC, Huang  HL,  et al.  The precancer risk of betel quid chewing, tobacco use and alcohol consumption in oral leukoplakia and oral submucous fibrosis in southern Taiwan.  Br J Cancer. 2003;88(3):366-372.PubMedGoogle ScholarCrossref
20.
Ko  YC, Huang  YL, Lee  CH, Chen  MJ, Lin  LM, Tsai  CC.  Betel quid chewing, cigarette smoking and alcohol consumption related to oral cancer in Taiwan.  J Oral Pathol Med. 1995;24(10):450-453.PubMedGoogle ScholarCrossref
21.
Lee  CH, Lee  KW, Fang  FM,  et al.  The neoplastic impact of tobacco-free betel-quid on the histological type and the anatomical site of aerodigestive tract cancers.  Int J Cancer. 2012;131(5):E733-E743.PubMedGoogle ScholarCrossref
22.
Gupta  B, Johnson  NW.  Systematic review and meta-analysis of association of smokeless tobacco and of betel quid without tobacco with incidence of oral cancer in South Asia and the Pacific.  PLoS One. 2014;9(11):e113385.PubMedGoogle ScholarCrossref
23.
Guha  N, Warnakulasuriya  S, Vlaanderen  J, Straif  K.  Betel quid chewing and the risk of oral and oropharyngeal cancers: a meta-analysis with implications for cancer control.  Int J Cancer. 2014;135(6):1433-1443.PubMedGoogle ScholarCrossref
24.
World Health Organization.  Schedules for Clinical Assessment in Neuropsychiatry (SCAN). Geneva, Switzerland: WHO; 1994.
25.
First  MB, Spitzer  RL, Gibbon  M, Williams  JBW.  Structured Clinical Interview for DSM-IV Axis I Disorders. Washington, DC: American Psychiatric Press; 2005.
26.
Kramer  IR, Pindborg  JJ, Bezroukov  V, Infirri  JS; World Health Organization.  Guide to epidemiology and diagnosis of oral mucosal diseases and conditions.  Community Dent Oral Epidemiol. 1980;8(1):1-26.PubMedGoogle ScholarCrossref
27.
Hu  L, Bentler  P.  Fit indices in covariance structure analysis: sensitivity to underparameterized model misspecifications.  Psychol Methods. 1998;3(4):424-453. doi:10.1037/1082-989X.3.4.424Google ScholarCrossref
28.
Bentler  PM.  Comparative fit indexes in structural models.  Psychol Bull. 1990;107(2):238-246.PubMedGoogle ScholarCrossref
29.
Cohen  J.  Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Earlbaum Associates; 1988.
30.
Lee  CH, Wu  DC, Lee  JM,  et al.  Anatomical subsite discrepancy in relation to the impact of the consumption of alcohol, tobacco and betel quid on esophageal cancer.  Int J Cancer. 2007;120(8):1755-1762.PubMedGoogle ScholarCrossref
31.
Grant  BF, Saha  TD, Ruan  WJ,  et al.  Epidemiology of DSM-5 drug use disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions-III.  JAMA Psychiatry. 2016;73(1):39-47.PubMedGoogle ScholarCrossref
32.
Grant  BF, Goldstein  RB, Saha  TD,  et al.  Epidemiology of DSM-5 alcohol use disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions III.  JAMA Psychiatry. 2015;72(8):757-766.PubMedGoogle ScholarCrossref
33.
Chou  SP, Goldstein  RB, Smith  SM,  et al.  The epidemiology of DSM-5 nicotine use disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions-III.  J Clin Psychiatry. 2016;77(10):1404-1412.PubMedGoogle ScholarCrossref
34.
Zhang  X, Reichart  PA.  A review of betel quid chewing, oral cancer and precancer in mainland China.  Oral Oncol. 2007;43(5):424-430.PubMedGoogle ScholarCrossref
35.
Jordan  FM, Gray  RD, Greenhill  SJ, Mace  R.  Matrilocal residence is ancestral in Austronesian societies.  Proc Biol Sci. 2009;276(1664):1957-1964.PubMedGoogle ScholarCrossref
36.
Lin  CC, Tamí-Maury  I, Ma  WF,  et al.  Social and cultural context of betel quid consumption in Taiwan and implications for prevention and cessation interventions.  Subst Use Misuse. 2017;52(5):646-655.PubMedGoogle ScholarCrossref
37.
Strickland  SS.  Anthropological perspectives on use of the areca nut.  Addict Biol. 2002;7(1):85-97.PubMedGoogle ScholarCrossref
38.
Amarasinghe  HK, Johnson  NW, Lalloo  R, Kumaraarachchi  M, Warnakulasuriya  S.  Derivation and validation of a risk-factor model for detection of oral potentially malignant disorders in populations with high prevalence.  Br J Cancer. 2010;103(3):303-309.PubMedGoogle ScholarCrossref
39.
Mehrtash  H, Duncan  K, Parascandola  M,  et al.  Defining a global research and policy agenda for betel quid and areca nut.  Lancet Oncol. 2017;18(12):e767-e775.PubMedGoogle ScholarCrossref
Original Investigation
March 2018

Association of DSM-5 Betel-Quid Use Disorder With Oral Potentially Malignant Disorder in 6 Betel-Quid Endemic Asian Populations

Author Affiliations
  • 1Department of Public Health, College of Health Science, Kaohsiung Medical University, Kaohsiung, Taiwan
  • 2Research Center for Environmental Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
  • 3Department of Medical Research, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan
  • 4Key Laboratory of Vertebrate Evolution and Human Origins, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing, China
  • 5Department of Population Health Sciences, Medical College of Georgia, Augusta University, Augusta
  • 6Department of Psychiatry, China Medical University Hospital, Taichung, Taiwan
  • 7Department of Oral Medicine, King’s College London, World Health Organization Collaborating Centre for Oral Cancer and Precancer, London, United Kingdom
  • 8Department of Biomedicine, University of Bergen, Bergen, Norway
  • 9Faculty of Dentistry, MAHSA University, Kuala Lumpur, Malaysia
  • 10Oral Cancer Research and Coordinating Centre, Faculty of Dentistry, University of Malaya, Kuala Lumpur, Malaysia
  • 11Environment-Omics-Disease Research Center, China Medical University Hospital, China Medical University, Taichung, Taiwan
JAMA Psychiatry. 2018;75(3):261-269. doi:10.1001/jamapsychiatry.2017.4307
Key Points

Question  Can DSM-5–defined betel-quid use disorder determine the risk of oral potentially malignant disorder in Asian populations with wide use of betel-quid?

Findings  In the Asian Betel-quid Consortium study of 8922 participants from 6 populations, betel-quid use disorder met DSM-5 criteria for a substance use disorder, had a high prevalence among users of betel-quid, and was correlated with risk of oral potentially malignant disorder, especially if users of betel-quid demonstrated symptoms of tolerance and used larger amounts or had a longer history of betel-quid use.

Meaning  To reduce the risk of oral potentially malignant disorder, any betel-quid use warrants intervention, and because the prevalence of betel-quid use disorder among users of betel-quid reaches as high as 86%, effective treatment modules addressing dependency on betel-quid should be developed and evaluated.

Abstract

Importance  Betel-quid (BQ) is the fourth most popular psychoactive agent worldwide. An emerging trend across Asia is the addictive consumption of BQ, which is associated with oral cancer and other health consequences.

Objective  To investigate the validity and pattern of DSM-5–defined BQ use disorder (BUD) and its association with oral potentially malignant disorder (OPMD) among Asian populations.

Design, Setting, and Participants  In-person interviews were conducted from January 1, 2009, to February 28, 2010, among a random sample of 8922 noninstitutionalized adults from the Asian Betel-quid Consortium study, an Asian representative survey of 6 BQ-endemic populations. Statistical analysis was performed from January 1, 2015, to December 31, 2016.

Main Outcomes and Measures  Participants were evaluated for BUD using DSM-5 criteria for substance use disorder and for OPMD using a clinical oral examination. Current users of BQ with 0 to 1 symptoms were classified as having no BUD, those with 2 to 3 symptoms as having mild BUD, those with 4 to 5 symptoms as having moderate BUD, and those with 6 or more symptoms as having severe BUD.

Results  Among the 8922 participants (4564 women and 4358 men; mean [SD] age, 44.2 [0.2] years), DSM-5 symptoms showed sufficient unidimensionality to act as a valid measure for BUD. The 12-month prevalence of DSM-5–defined BUD in the 6 study populations was 18.0% (mild BUD, 3.2%; moderate BUD, 4.3%; and severe BUD, 10.5%). The 12-month proportion of DSM-5–defined BUD among current users of BQ was 86.0% (mild BUD, 15.5%; moderate BUD, 20.6%; and severe BUD, 50.0%). Sex, age, low educational level, smoking, and drinking were significantly associated with BUD. Among individuals who used BQ, family use, high frequency of use, and amount of BQ used were significantly linked to moderate to severe BUD. Compared with individuals who did not use BQ, those who used BQ and had no BUD showed a 22.0-fold (95% CI, 4.3-112.4) risk of OPMD (P < .001), whereas those with mild BUD showed a 9.6-fold (95% CI, 1.8-56.8) risk (P = .01), those with moderate BUD showed a 35.5-fold (95% CI, 4.3-292.3) risk (P = .001), and those with severe BUD showed a 27.5-fold (95% CI, 1.6-461.4) risk of OPMD (P = .02). Individuals with moderate to severe BUD who used BQ and had the symptom of tolerance had a 153.4-fold (95% CI, 33.4-703.6) higher risk of OPMD than those who did not use BQ, and those with moderate to severe BUD who used BQ and had a larger amount or longer history of BQ use had an 88.9-fold (95% CI, 16.6-476.5) higher risk of OPMD than those who did not use BQ.

Conclusions and Relevance  This international study gathered data about BQ users across 6 Asian populations, and it demonstrates that DSM-5 symptoms could fulfill a BUD construct. Most current Asian users of BQ already have BUD, which is correlated with risk of OPMD. Among individuals with moderate to severe BUD who used BQ, tolerance and a larger amount or longer history of BQ use are the key symptoms that correlated with enhanced risk of OPMD. These findings play an important role in providing a new indication of an additional psychiatric management plan for users of BQ who have BUD.

Introduction

Betel-quid (BQ) is the fourth most popular self-administered psychoactive agent worldwide, after caffeine, alcohol, and nicotine.1 More than 600 million people use BQ within the Indo-Asia-Pacific biogeographic region, and its use is spreading into Asian migrant communities in Western countries.2 The prevalence of BQ use among adults is particularly high in India, Pakistan, and Sri Lanka (8.4%-40.0%); Nepal, Malaysia, and Indonesia (10.3%-47.8%); mainland China and Taiwan (2.3%-29.0%); Palau and the Solomon Islands (72.0%-83.0%); and among Bangladeshi migrants in the United Kingdom (30.0%-90.0%).3,4 In these places, BQ is easily available in ready-made packaging at low prices ($0.05-$0.30 per BQ),5 and chewing BQ in public is often socially accepted.4

Betel-quid is a masticatory mixture consisting of a fresh, unripe, or dried Areca catechu nut usually wrapped (with or without tobacco) in a betel leaf from the Piper betel vine, smeared with aqueous lime, and packed with flavoring ingredients.3,4,6,7 Arecoline is the principal active agent in the areca nut,3 with a chemical structure analogous to that of nicotine.8 Its biological profile is that of a nonselective agonist of the muscarinic acetylcholine receptors, acting on the α4 and β2, as well as the α6 and β3, subunits of nicotinic acetylcholine receptors, which are the 2 groups of receptors most closely associated with the addictive properties of nicotine.9,10 A high quantity of BQ use can induce cocaine-like physiological states, such as anxiety, dilated pupils, tachycardia, and elevated blood pressure.11 Tolerance and withdrawal have been observed in long-term BQ users.5,7,12,13

Use of BQ is not formally identified as an addictive behavior according to global substance evaluations.14 Nonetheless, studies have begun to use the DSM-IV to measure the dependent use of BQ.5,7,13,15,16 In the DSM-5, abuse and dependence as defined in the DSM-IV were merged into a single disorder with a severity metric based on the symptom count.17 Changes to the criteria for substance use disorder (SUD) in the DSM-5 include the removal of the legal problems criterion, the addition of craving, and a diagnostic minimum threshold of at least 2 symptoms.18 However, no epidemiologic data on BQ use were updated using the DSM-5.

Prolonged BQ use is linked with oral potentially malignant disorder (OPMD).4,19 During BQ chewing, the areca nut–derived nitrosamines and reactive oxygen species produced in the oral cavity can induce genetic damage to exposed oral keratinocytes.3 Persistent BQ exposures predispose oral cells to preneoplastic lesions, leading to full malignant neoplasms.20-23 Frequency of use is a central determinant of this outcome; however, a subsequent controlling addiction, manifesting as impaired control, social impairment, risky use, and pharmacologic symptoms, underlying its increased use should also be considered. Our understanding of the psychiatric dimensions associated with addictive consumption of BQ needs to be improved.

In response to the scope of BQ use and the potential health problems in Asia-Pacific countries, the Asian Betel-quid Consortium conducted a joint study to evaluate the effects of BQ use disorder (BUD) on oral health and offered strategies to activate outreach measures for the prevention of oral diseases.4-7 We define individuals with BUD as users of BQ who met all the DSM-5 SUD diagnostic criteria. We sought to elucidate the following 3 research issues: the validity of DSM-5 symptoms for measuring BUD use and the patterns of BUD use among current chewers; the country-dependent parameters that determine BUD; and the association between BUD and OPMD.

Methods
Participants

Six cross-sectional studies were concurrently conducted across East Asia (Taiwan and mainland China), Southeast Asia (Malaysia and Indonesia), and South Asia (Nepal and Sri Lanka) between January 1, 2009, and February 28, 2010, under the Asian Betel-quid Consortium study. Research details are described in a previous article.4 eTable 1 in the Supplement shows the sociodemographic factors. The number of recruited participants from each study region ranged from 1002 to 2356, indicating a high rate of response (68%-100%). An identical study protocol was administered to all study populations and approved by the Ethics Review Committee of Kaohsiung Medical University (Taiwan), Central South University (mainland China), the University of Malaya (Malaysia), Airlangga University (Indonesia), Kathmandu University (Nepal), and the University of Peradeniya (Sri Lanka). Written informed consent was collected from all participants.

Diagnostic Instrument

A survey questionnaire was designed using proper materials from World Health Organization (WHO) surveys and national prevalence studies. The collected data consisted of sociodemographic factors; disease history; age of initial BQ consumption; quantity of daily use; frequency of use; years of consumption; types of BQ, alcohol, and cigarettes used; family history of substance use; and years since cessation for those who no longer used BQ.

Our interview questions were adapted from the Structured Clinical Interview for DSM-IV, Text Revision Axis I Disorders for SUDs and Schedules for Clinical Assessment in Neuropsychiatry (SCAN).24,25 All included symptoms were updated to fully match the DSM-5 SUD diagnostic criteria. We merged the DSM-IV–derived BQ symptoms for abuse and dependence, removed the legal problem criterion, and used the SCAN-derived craving symptom for DSM-5–defined BUD.18 Questions were initially written in English and translated into the primary language or dialect of each study population. All questions were back-translated into English so that their content and semantic equivalence could be verified by bilingual specialists. The standardized questionnaire was used in all study areas, and our interviews were conducted in local languages.

BQ Use Disorder

Following WHO guidelines,3 a lifetime BQ user was defined as an individual who had used at least 1 quid of any type of BQ product per day for a minimum of 6 months. Among lifetime users, current users were defined as those who had used BQ within the preceding 12 months before the interview, and past users were those who had not used BQ for at least 12 months before the interview.

We used 11 DSM-5 symptoms to assess BUD for current users according to the data obtained. These symptoms included the following: large amount or longer history of BQ use (BQ is used in larger amounts or over a longer period than intended); unsuccessful cutdown (unsuccessful efforts to reduce or control BQ use); time spent chewing (spending a large amount of time chewing BQ); craving (having a strong desire or sense of compulsion to use BQ); neglected major roles (recurrent use of BQ results in a failure to fulfill major role obligations at work or home); social or interpersonal problems (continual BQ use despite having persistent or recurrent social or interpersonal problems caused or exacerbated by the effects of its use); given up activities (a reduction in important social, occupational, or recreational activities owing to BQ use); hazardous use (recurrent substance use in situations in which it is physically hazardous); continued use despite knowing problems (continual BQ chewing despite an awareness of the physical or psychological problems caused by chewing); tolerance; and withdrawal. The symptoms were grouped into 4 pathologic behavioral categories named impaired control (large amount or longer history of BQ use, unsuccessful cutdown, time spent chewing, and craving), social impairment (neglected major roles, social or interpersonal problems, and given up activities), risky use (hazardous use and continued use despite knowing problems), and pharmacologic symptoms (tolerance and withdrawal).

A positive diagnosis of BUD required the presence of at least 2 of the 11 symptoms within 12 months before interview. We followed the DSM-5 criteria for SUD; current users of BQ with 0 to 1 symptoms were classified as having no BUD, those with 2 to 3 symptoms as having mild BUD, those with 4 to 5 symptoms as having moderate BUD, and those with 6 or more symptoms as having severe BUD.17

Oral Potentially Malignant Disorder

All dental and medical professionals completed a standardized OPMD training course for diagnosing oral submucous fibrosis, oral leukoplakia, and oral lichen planus. The characteristics and site of each oral disorder were carefully examined using portable plane dental mirrors for soft-tissue retraction and dental lights for illumination, according to WHO clinical criteria.26

Statistical Analysis

Statistical analysis was performed from January 1, 2015, to December 31, 2016. Survey data modules of Stata, version 15 (StataCorp), were used to adjust for multistage sampling design and complex sampling weights. A 4-step procedure was used to analyze data. First, we performed a confirmatory factor analysis to test for a unidimensional BUD construct. Two goodness of fit criteria were used to determine unidimensionality (the root-mean-square error of approximation <0.05 and the comparative fit index >0.95).27,28 After unidimensionality was established, the item parameters were generated from an Item Response Theory 2-parameter logistic model in Mplus, version 7.11 (Muthén & Muthén). Convergent validity between the antecedent, concurrent, and subsequent correlates and the unidimensional BUD factor scores were conducted through correlation analyses. Cohen conventions for interpreting the effect size of the correlations were used to determine the strength of the associations between each correlate and BUD factor scores.29 Second, we estimated the prevalence and proportion of BUD in each study region based on weighted data. Third, we constructed multinomial logistic regression models to evaluate the associations of sociodemographic factors with no BUD, mild BUD, moderate BUD, and severe BUD compared with nonusers of BQ.30 Adjusted odds ratios (AORs) were used to assess the association of chewing characteristics with pathologic behaviors and BUD diagnoses for current users of BQ. Fourth, 4-stage modeling and global tests of joint significance via stepwise addition of each variable block were used to study the association of OPMD with demographic factors, characteristics of BQ use, DSM-5 symptoms, and BUD diagnosis. Because low OPMD prevalences were observed in Malaysia and Nepal, we performed sensitivity analyses to evaluate the associations between BUD and OPMD, before and after excluding the Malaysian and Nepalese data.

Results
Measurement Validity of DSM-5 Symptoms for BUD

An excellent fit was calculated for a unidimensional construct to measure the BUD factor scores (root-mean-square error of approximation, 0.026; comparative fit index, 0.995; Table 1). The factor loadings for DSM-5 symptoms ranged from 0.701 to 0.985. Item parameters showed that hazardous use provided the most information (item discrimination, 5.671), whereas social or interpersonal problems offered the least information (item discrimination, 0.983) among the 11 symptoms. The corresponding item difficulty parameters ranged from 1.039 to 2.898, which captured the latent trait level of BUD at which a user has a 50% probability of exhibiting the symptom. Regarding convergent validity, the antecedent correlates of educational level, family use of BQ, and friend use of BQ were significantly associated with BUD factor scores (eTable 2 in the Supplement). Betel quid–associated concurrent correlates and substance-associated subsequent correlates had a strong correlation with BUD factor scores (correlation coefficients, 0.64-0.92).

Prevalence and Severity of BUD

The prevalence of current chewers across the 6 populations was 6.7% to 39.5%, with the highest prevalence observed in Nepal (Table 2). In East Asia (Taiwan and mainland China), the 12-month prevalence of current users of BQ with BUD was 4.7% to 8.1%, in Southeast Asia (Malaysia and Indonesia), the 12-month prevalence of current users of BQ with BUD was 14.8% to 29.4%, and in South Asia (Nepal and Sri Lanka), the 12-month prevalence of current BQ users with BUD was 8.4% to 39.2% (18.0% overall). In East Asia (Taiwan and mainland China), the proportion of BUD in current users of BQ was 61.1% to 70.8%, in Southeast Asia (Malaysia and Indonesia), the proportion of BUD in current users of BQ was 75.8% to 98.5%, and in South Asia (Nepal and Sri Lanka), the proportion of BUD in current users of BQ was 55.8% to 99.3% (86.0% overall).

Demographic Factors Associated With BUD

In mainland China, men and younger people were more likely to have any type of BUD (eTable 3 in the Supplement). By contrast, in Malaysia, women and older people had a higher prevalence of BUD. A higher educational level correlated with a lower risk of mild to severe BUD in Taiwan, Malaysia, and Sri Lanka. Cigarette smokers in mainland China were more likely to have all types of BUD (AOR, 3.5-5.7), whereas Malaysian and Indonesian smokers were less likely to have mild to severe BUD (AOR, 0.02-0.30). Alcohol drinkers in Taiwan and Malaysia had a 2.9- to 32.1-fold risk of mild to severe BUD, whereas alcohol drinkers in Indonesia had a 0.3- to 0.4-fold risk of moderate to severe BUD.

Use Characteristics Associated With BUD

Table 3 shows that, in the combined data of users of BQ, family use was associated with a 2.3-fold higher likelihood of impaired control and a 0.6-fold lower likelihood of social impairment than did no family use of BQ. Having friends who used BQ was associated with a 2.8-fold higher likelihood of social impairment than did not having friends who used BQ. Overall, family use was associated with a 2.0- to 3.0-fold risk of developing moderate to severe BUD compared with no BUD, and having friends who used BQ was associated with a 3.6-fold risk of developing mild BUD compared with no BUD. In addition, high frequency of use was associated with impaired control (AOR, 1.2 [95% CI, 1.1-1.3]), social impairment (AOR, 1.1 [95% CI, 1.0-1.2]), risky use (AOR, 1.4 [95% CI, 1.2-1.6]), pharmacologic symptoms (AOR, 1.2 [95% CI, 1.0-1.4]), and moderate (AOR, 1.3 [95% CI, 1.1-1.5]) to severe (AOR, 1.6 [95% CI, 1.4-2.0]) BUD. The amount of BQ used was also associated with impaired control (AOR, 1.1 [95% CI, 1.1-1.2]), risky use (AOR, 1.1 [95% CI, 1.0-1.1]), pharmacologic symptoms (AOR, 1.1 [95% CI, 1.0-1.1]), and mild (AOR, 1.1 [95% CI, 1.0-1.1]), moderate (AOR, 1.1 [95% CI, 1.0-1.2]), and severe (AOR, 1.2 [95% CI, 1.1-1.2]) BUD.

OPMD and BUD

The overall prevalence of OPMD was higher in users of BQ with moderate to severe BUD (26.8%-27.2%) than in those with no BUD or mild BUD (4.3%-5.8%) and nonusers of BQ (4.2%) (eTable 4 in the Supplement). The symptoms of tolerance, larger amount or longer history of BQ use, and continued use despite knowing problems showed the highest prevalence of OPMD (40.8%, 31.6%, and 29.3%, respectively) (eTable 5 in the Supplement). These 3 symptoms also exhibited 4.3-, 2.9-, and 2.0-fold higher likelihoods of OPMD compared with users of BQ without these symptoms, after adjustment for demographic and risk factors, use characteristics, and other symptoms (Table 4; model 3). Higher risk of OPMD was associated with a higher number of pharmacologic symptom behaviors (AOR, 3.4 [95% CI, 1.7-6.6] for 1 increase in symptom numbers; P < .001 for linear trend). We found that having no BUD was associated with a 22.0-fold (95% CI, 4.3-112.4) higher risk of OPMD compared with nonusers of BQ, having mild BUD was associated with a 9.6-fold (95% CI, 1.8-56.8) higher risk, having moderate BUD was associated with a 35.5-fold (95% CI, 4.3-292.3) higher risk, and having severe BUD was associated with a 27.5-fold (95% CI, 1.6-461.4) higher risk (Table 4; model 4). No interaction effects on OPMD were observed between BUD and the consumption of alcohol and cigarettes. Sensitivity analyses, before and after excluding the Malaysian and Nepalese data, confirmed similar results.

Combined and Conditional Effects of BUD Symptoms and Diagnosis

Table 5 shows that, compared with nonusers of BQ, users of BQ with moderate to severe BUD and the symptom of larger amount or longer history of BQ use had an 88.9-fold (95% CI, 16.6-476.5) higher risk of OPMD, and users of BQ with moderate to severe BUD and the symptom of tolerance had a 153.4-fold (95% CI, 33.4-703.6) higher risk of OPMD. The risks of these 2 symptoms were significantly higher (larger amount or longer history of BQ use: AOR ratio, 2.3 [95% CI, 1.1-4.6]; tolerance: AOR ratio, 4.1 [95% CI, 1.9-9.1]) than in users of BQ with BUD of equal severity but without the corresponding symptoms. Users of BQ with moderate to severe BUD without the symptom of neglected major roles had a 39.7-fold (95% CI, 7.8-201.7) higher risk of OPMD. This risk was higher (AOR ratio, 3.3 [95% CI, 1.6-6.9]) than in users of BQ with BUD of equal severity but with this symptom.

Discussion

The 11 DSM-5 symptoms of BUD could be modeled as a unidimensional BUD construct. We applied these measures and uncovered a high prevalence of DSM-5–defined BUD across the 6 BQ-endemic Asian populations investigated. The 12-month prevalence of BUD was 18.0%, and 86.0% of the users of BQ had some form of BUD. The prevalence of BUD in this study exceeds that reported for DSM-5–defined drug use disorder (3.9%)31 and is comparable to the results of national surveys in the United States for the prevalence of DSM-5–defined alcohol (13.9%) and nicotine (20.0%) use disorders.32,33

We identified several country-specific factors associated with BUD. First, in mainland China, male sex was associated with BUD, and in Malaysia, female sex was associated with BUD. Age was negatively associated with BUD in mainland China and positively associated with BUD in Malaysia and Indonesia. These findings accurately capture the ethnodemographic circumstances in those regions; BQ use is an emerging trend in mainland China and is popular among young working-class men,34 whereas in Malaysia and Indonesia, BQ use is associated with the social customs of Austronesian language–speaking matrilocal societies, and thus has a longer history of use and is more closely associated with older female groups.35 Second, as seen in other SUDs,31,33 a lower educational level was associated with mild to severe BUD in Taiwan, Malaysia, and Sri Lanka. Therefore, preventive and interventional strategies should consider the socioeconomic characteristics of the groups using BQ. Third, nationwide reports have revealed that DSM-5–defined alcohol, nicotine, and drug use disorders are interrelated.31-33 Similarly, our analysis found that cigarette-smoking users of BQ were associated with no BUD to mild BUD in Taiwan and all BUD types in mainland China, and alcohol-drinking users of BQ were associated with specific BUD types in all study regions except for Indonesia. A negative association was found between cigarette smoking and mild to severe BUD in Malaysian and Indonesian users of BQ. For these 2 countries, BQ users tended not to be cigarette smokers (BQ users vs nonusers: AOR, 0.05; 95% CI, 0.03-0.08). A possible explanation is a lower purchasing power for cigarettes because the price of cigarettes is higher than that of BQ and BQ is often used by people with relatively lower income in these 2 populations.

Our study showed that a high frequency of use of BQ and using high amounts of BQ correlated with all 4 pathologic behaviors (except between social impairment and amount) and were associated with moderate to severe BUD, thereby confirming the link between dose-enhanced use and substance addiction. The initiation of BQ use is often a result of use among peers or family members.36 In this study, BQ users with friends who chewed exhibited a higher likelihood of social impairment and developing mild BUD. By contrast, BQ users with family members who used BQ had a lower likelihood of social impairment (AOR, 0.6) but a higher likelihood of impaired control (AOR, 2.3) and significant levels of moderate to severe BUD, possibly because BQ use is accepted in certain family and social situations, whereas alcohol consumption and cigarette smoking may be deemed objectionable.37 Based on these findings, family-based interventions should be incorporated into BUD prevention and treatment strategies.

Use of BQ poses the greatest risk of developing OPMD among the principal risk factors.19,38 Based on DSM-IV criteria, 1 previous study found a 2.5- to 51.5-fold risk of OPMD among dependent users of BQ.5 Our study used DSM-5 criteria to identify 9.6- to 22.0-fold risks of OPMD among users of BQ with no BUD to mild BUD and 27.5- to 35.5-fold risks of OPMD among users of BQ with moderate to severe BUD from 6 Asian populations. Frequency and amount of BQ use correlated with BUD; however, our findings revealed that the association between BUD and OPMD was more substantial after adjustment for use characteristics, symptoms, and other covariates. This finding indicates that addictive consumption of BQ is linked to riskier biological outcomes.

A closer investigation into symptom-specific BUD revealed that a 2.0- to 4.3-fold risk of OPMD was associated with continued use despite knowing problems, having a larger amount or longer history of BQ use, and tolerance (Table 3; model 3). These symptoms potentially underlie the development of OPMD among users of BQ. Users of BQ with moderate to severe BUD and the symptom of tolerance had a conditionally higher risk of OPMD (AOR ratio, 4.1 [95% CI, 1.9-9.1]), as did those with a larger amount or longer history of BQ use (AOR ratio, 2.3 [95% CI, 1.1-4.6]), thereby emphasizing the role of pharmacologic symptoms and impaired control behaviors in BUD-associated risk of OPMD. The symptom of neglected major role of social impairment was associated with a lower risk of OPMD because sharing BQ in family and workplace environments can improve interpersonal relationships.36

Strengths and Limitations

The strength of this consortium study was that our investigations were conducted under a single framework using an identical protocol, measuring tools, and diagnostic instruments. A major limitation of this study was its cross-sectional design, which precluded any causal interpretations. Our findings represent only a snapshot of the BUD situation in the study populations. Caution must be exercised when generalizing the findings of this study to other areas because use of BQ might differ even within a single country. We believe that our research framework and methods could be applied to areas or countries where BQ use is prevalent. Further issues regarding intercountry variations in BUD and OPMD are explained in more detail in the eAppendix in the Supplement. Our data are of significance to global research and policy agendas for the BQ and areca nut.39

Conclusions

Symptoms, as defined by the DSM-5, have a unidimensional construct that underlies BUD. Current BQ users in Asian populations exhibit a high prevalence of DSM-5–defined BUD. Tolerance and larger amount or longer history of BQ use were the pathologic symptoms that most correlated with the enhanced risk of OPMD among users of BQ with moderate to severe BUD. We hypothesized that BUD is an intermediary step during the progression from BQ use toward developing OPMD. Appropriate psychiatric treatment should be considered when assisting patients in controlling BQ use to minimize the risk of developing OPMD.

Back to top
Article Information

Accepted for Publication: November 28, 2017.

Corresponding Author: Ying-Chin Ko, MD, PhD, Environment-Omics-Disease Research Center, China Medical University and Hospital. No. 2 Yude Road, Taichung 40447, Taiwan (ycko0406@gmail.com).

Published Online: February 7, 2018. doi:10.1001/jamapsychiatry.2017.4307

Author Contributions: Drs Lee and A.M.-S. Ko contributed equally to this study are considered co–first authors. Dr Y.-C. Ko had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: A.M.-S. Ko, Warnakulasuriya, Lee, Y.-C. Ko.

Acquisition, analysis, or interpretation of data: A.M.-S. Ko, Hung, Yang, Ibrahim, Zain, Lee, Y.-C. Ko.

Drafting of the manuscript: A.M.-S. Ko, Hung, Lee, Y.-C. Ko.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: A.M.-S. Ko, Hung, Yang, Lee, Y.-C. Ko.

Obtained funding: A.M.-S. Ko, Hung, Y.-C. Ko.

Administrative, technical, or material support: A.M.-S. Ko, Hung, Ibrahim, Zain, Y.-C. Ko.

Supervision: A.M.-S. Ko, Y.-C. Ko.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was funded by grant KMU-EM-99-1-1 from the Center of Excellence for Environment Medicine, Kaohsiung Medical University; “Aim for the Top Universities Grant” (grants KMU-TP104A13 and KMU-TP105A12) from the Research Center for Environmental Medicine, Kaohsiung Medical University; and grants MOHW106-TDU-B-212-144003 and MOHW106-TDU-B-212-122016 from the Taiwan Ministry of Health and Welfare Health and Welfare surcharge of tobacco products.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The views and opinions expressed in this article are those of the authors and should not be construed to represent the views of any of the sponsoring organizations.

Additional Contributions: Jennifer Ko, MS, assisted in organizing the various centers’ principal investigators. The authors would also like to express our appreciation to the study staff members, including Yi-Jun Gao, PhD and Sunarjo, DDS, for their diligent work and excellent endeavors in this international cooperative study. They were not compensated for their contributions.

References
1.
Boucher  BJ, Mannan  N.  Metabolic effects of the consumption of Areca catechu Addict Biol. 2002;7(1):103-110.PubMedGoogle ScholarCrossref
2.
Gupta  PC, Warnakulasuriya  S.  Global epidemiology of areca nut usage.  Addict Biol. 2002;7(1):77-83.PubMedGoogle ScholarCrossref
3.
IARC Working Group on the Evaluation of Carcinogenic Risks to Humans.  Betel-quid and areca-nut chewing and some areca-nut derived nitrosamines.  IARC Monogr Eval Carcinog Risks Hum. 2004;85:1-334.PubMedGoogle Scholar
4.
Lee  CH, Ko  AM, Warnakulasuriya  S,  et al.  Intercountry prevalences and practices of betel-quid use in south, southeast and eastern Asia regions and associated oral preneoplastic disorders: an international collaborative study by Asian Betel-quid Consortium of south and east Asia.  Int J Cancer. 2011;129(7):1741-1751.PubMedGoogle ScholarCrossref
5.
Lee  CH, Ko  AM, Yen  CF,  et al.  Betel-quid dependence and oral potentially malignant disorders in six Asian countries.  Br J Psychiatry. 2012;201(5):383-391.PubMedGoogle ScholarCrossref
6.
Lee  CH, Ko  AM, Warnakulasuriya  S,  et al.  Population burden of betel quid abuse and its relation to oral premalignant disorders in South, Southeast, and East Asia: an Asian Betel-quid Consortium Study.  Am J Public Health. 2012;102(3):e17-e24.PubMedGoogle ScholarCrossref
7.
Lee  CH, Chiang  SL, Ko  AM,  et al.  Betel-quid dependence domains and syndrome associated with betel-quid ingredients among chewers: an Asian multi-country evidence.  Addiction. 2014;109(7):1194-1204.PubMedGoogle ScholarCrossref
8.
Lord  GA, Lim  CK, Warnakulasuriya  S, Peters  TJ.  Chemical and analytical aspects of areca nut.  Addict Biol. 2002;7(1):99-102.PubMedGoogle ScholarCrossref
9.
Chu  NS.  Effects of Betel chewing on the central and autonomic nervous systems.  J Biomed Sci. 2001;8(3):229-236.PubMedGoogle ScholarCrossref
10.
Papke  RL, Horenstein  NA, Stokes  C.  Nicotinic activity of arecoline, the psychoactive element of ‘betel nuts’, suggests a basis for habitual use and anti-inflammatory activity.  PLoS One. 2015;10(10):e0140907.PubMedGoogle ScholarCrossref
11.
Patidar  KA, Parwani  R, Wanjari  SP, Patidar  AP.  Various terminologies associated with areca nut and tobacco chewing: a review.  J Oral Maxillofac Pathol. 2015;19(1):69-76.PubMedGoogle ScholarCrossref
12.
Bhat  SJ, Blank  MD, Balster  RL, Nichter  M, Nichter  M.  Areca nut dependence among chewers in a South Indian community who do not also use tobacco.  Addiction. 2010;105(7):1303-1310.PubMedGoogle ScholarCrossref
13.
Benegal  V, Rajkumar  RP, Muralidharan  K.  Does areca nut use lead to dependence?  Drug Alcohol Depend. 2008;97(1-2):114-121.PubMedGoogle ScholarCrossref
14.
Gowing  LR, Ali  RL, Allsop  S,  et al.  Global statistics on addictive behaviours: 2014 status report.  Addiction. 2015;110(6):904-919.PubMedGoogle ScholarCrossref
15.
Mirza  SS, Shafique  K, Vart  P, Arain  MI.  Areca nut chewing and dependency syndrome: is the dependence comparable to smoking? a cross sectional study.  Subst Abuse Treat Prev Policy. 2011;6:23.PubMedGoogle ScholarCrossref
16.
Osborne  PG, Ko  YC, Wu  MT, Lee  CH.  Intoxication and substance use disorder to Areca catechu nut containing betel quid: a review of epidemiological evidence, pharmacological basis and social factors influencing quitting strategies.  Drug Alcohol Depend. 2017;179:187-197.PubMedGoogle ScholarCrossref
17.
American Psychiatric Association.  Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Association; 2013.
18.
Hasin  DS, O’Brien  CP, Auriacombe  M,  et al.  DSM-5 criteria for substance use disorders: recommendations and rationale.  Am J Psychiatry. 2013;170(8):834-851.PubMedGoogle ScholarCrossref
19.
Lee  CH, Ko  YC, Huang  HL,  et al.  The precancer risk of betel quid chewing, tobacco use and alcohol consumption in oral leukoplakia and oral submucous fibrosis in southern Taiwan.  Br J Cancer. 2003;88(3):366-372.PubMedGoogle ScholarCrossref
20.
Ko  YC, Huang  YL, Lee  CH, Chen  MJ, Lin  LM, Tsai  CC.  Betel quid chewing, cigarette smoking and alcohol consumption related to oral cancer in Taiwan.  J Oral Pathol Med. 1995;24(10):450-453.PubMedGoogle ScholarCrossref
21.
Lee  CH, Lee  KW, Fang  FM,  et al.  The neoplastic impact of tobacco-free betel-quid on the histological type and the anatomical site of aerodigestive tract cancers.  Int J Cancer. 2012;131(5):E733-E743.PubMedGoogle ScholarCrossref
22.
Gupta  B, Johnson  NW.  Systematic review and meta-analysis of association of smokeless tobacco and of betel quid without tobacco with incidence of oral cancer in South Asia and the Pacific.  PLoS One. 2014;9(11):e113385.PubMedGoogle ScholarCrossref
23.
Guha  N, Warnakulasuriya  S, Vlaanderen  J, Straif  K.  Betel quid chewing and the risk of oral and oropharyngeal cancers: a meta-analysis with implications for cancer control.  Int J Cancer. 2014;135(6):1433-1443.PubMedGoogle ScholarCrossref
24.
World Health Organization.  Schedules for Clinical Assessment in Neuropsychiatry (SCAN). Geneva, Switzerland: WHO; 1994.
25.
First  MB, Spitzer  RL, Gibbon  M, Williams  JBW.  Structured Clinical Interview for DSM-IV Axis I Disorders. Washington, DC: American Psychiatric Press; 2005.
26.
Kramer  IR, Pindborg  JJ, Bezroukov  V, Infirri  JS; World Health Organization.  Guide to epidemiology and diagnosis of oral mucosal diseases and conditions.  Community Dent Oral Epidemiol. 1980;8(1):1-26.PubMedGoogle ScholarCrossref
27.
Hu  L, Bentler  P.  Fit indices in covariance structure analysis: sensitivity to underparameterized model misspecifications.  Psychol Methods. 1998;3(4):424-453. doi:10.1037/1082-989X.3.4.424Google ScholarCrossref
28.
Bentler  PM.  Comparative fit indexes in structural models.  Psychol Bull. 1990;107(2):238-246.PubMedGoogle ScholarCrossref
29.
Cohen  J.  Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Earlbaum Associates; 1988.
30.
Lee  CH, Wu  DC, Lee  JM,  et al.  Anatomical subsite discrepancy in relation to the impact of the consumption of alcohol, tobacco and betel quid on esophageal cancer.  Int J Cancer. 2007;120(8):1755-1762.PubMedGoogle ScholarCrossref
31.
Grant  BF, Saha  TD, Ruan  WJ,  et al.  Epidemiology of DSM-5 drug use disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions-III.  JAMA Psychiatry. 2016;73(1):39-47.PubMedGoogle ScholarCrossref
32.
Grant  BF, Goldstein  RB, Saha  TD,  et al.  Epidemiology of DSM-5 alcohol use disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions III.  JAMA Psychiatry. 2015;72(8):757-766.PubMedGoogle ScholarCrossref
33.
Chou  SP, Goldstein  RB, Smith  SM,  et al.  The epidemiology of DSM-5 nicotine use disorder: results from the National Epidemiologic Survey on Alcohol and Related Conditions-III.  J Clin Psychiatry. 2016;77(10):1404-1412.PubMedGoogle ScholarCrossref
34.
Zhang  X, Reichart  PA.  A review of betel quid chewing, oral cancer and precancer in mainland China.  Oral Oncol. 2007;43(5):424-430.PubMedGoogle ScholarCrossref
35.
Jordan  FM, Gray  RD, Greenhill  SJ, Mace  R.  Matrilocal residence is ancestral in Austronesian societies.  Proc Biol Sci. 2009;276(1664):1957-1964.PubMedGoogle ScholarCrossref
36.
Lin  CC, Tamí-Maury  I, Ma  WF,  et al.  Social and cultural context of betel quid consumption in Taiwan and implications for prevention and cessation interventions.  Subst Use Misuse. 2017;52(5):646-655.PubMedGoogle ScholarCrossref
37.
Strickland  SS.  Anthropological perspectives on use of the areca nut.  Addict Biol. 2002;7(1):85-97.PubMedGoogle ScholarCrossref
38.
Amarasinghe  HK, Johnson  NW, Lalloo  R, Kumaraarachchi  M, Warnakulasuriya  S.  Derivation and validation of a risk-factor model for detection of oral potentially malignant disorders in populations with high prevalence.  Br J Cancer. 2010;103(3):303-309.PubMedGoogle ScholarCrossref
39.
Mehrtash  H, Duncan  K, Parascandola  M,  et al.  Defining a global research and policy agenda for betel quid and areca nut.  Lancet Oncol. 2017;18(12):e767-e775.PubMedGoogle ScholarCrossref
×