568 SAMT VOL 74 3 DES 1988

Betel-nut chewing and submucous fibrosis

in Durban
H. A. SEEDAT, C. W. VAN WYK

Summary Muslims in their survey. The manner in which the households

were sampled is not described. They found that 30,7% of
A stratified random sample among South African Indians women practised the chewing habit while only 5,5% of men
living in Durban revealed the occurrence of betel-nut chewing were chewers. The habit was more common in the elderly -
and the resultant submucous fibrosis (SF) to be relatively 71,9% of women over 60' years and 10,3% of men in the same
high. Women chewers predominated in a ratio of. 13:1. The age group.
habit increased with age al1d 30,6% of women over '65 years
Shear er al. 9 undertook a study in the JohannesburglPretoria
practised the custom. Thirty-eight per cent of chewers
revealed signs of impending and established SF; women area on 1000 consecutive unselected Indians. The sample was
predominated 70:1, and the majority of sufferers in this not chosen randomly. The overall prevalence rate of SF,
instance belonged to the age group 45 - 54 years (12,9%). which was only encountered in women, was 0,5%. Dockrat and
Forty-six per cent of those with signs associated with SF had Shear lo looked at an unselected sample in Durban of I 200
fibrous bands in the mouth and were regarded as having Indian subjects over the age of 15 years, 800 from a door-to-
established SF. It was calculated that 5% of the total Indian door survey and 400 from a hospital outpatient department.
population in South Africa could be chewers and that 2,3% The prevalence of the disease was 0,6%.
may develop SF. Randeria",12 observed 28 cases of SF in a Durban hospital
S AIr Med J 1988; 74: 568-5~1.
between 1969 and 1972. Van Wyk er al. 13 investigated oral
disease among Indians living in Cape Town and found the
occurrence of SF to be 1,25%. However, their study sample
'Oral submucous fibrosis [SF] is an insidious chronic disease was small (N = 300).
affecting any part of the oral cavity and sometimes the pharynx. A study of the prevalence of betel-nut chewing and the
Although occasionally preceded by and/or associated with presence of impending SF and established SF among South
vesicle formation, it is always associated with a juxta-epithelial African Indians residing in Durban was undertaken with a
inflammatory reaction followed by a a fibro-elastic change of view to initiating preventive and intervention programmes.
the lamina propria with epithelial atrophy, leading to stiffness
of the oral mucosa and causing trismus and inability to eat.' 1
This stiffness and trismus is due to a fibro-elastic change in Subjects and methods
the lamina propria of the oral mucosa. 1 Early signs and
symptoms heralding the onset of the disease are a burning Population
sensation in the mouth often experienced whcm eating spicy
The Indians in South Africa owe their presence to the
food; '-5 blanching of the palate, faucial pillars, buccal mucosa
system of indenture between the British raj in India and the
and lips;6 oral vesicles and ulceration;7 excessive salivation;'
British colonies, which started in 1837 and ended in 1917.
defective gustatory sensation; and dryness of the mouth.' The
They constitute a heterogeneous society composed of many
disease is progressive and once there are palpable fibrous
distinct ethnic groups and many of their attitudes and patterns
bands in the mouth it is established and will not regress.
of social life are traditional. 14 The population is weighted
There is ample evidence that the habit of betel-nut chewing is
towards younger people and can be classified as a relatively
a prime cause. 1
young population (Fig. 1).
The habit of betel-nut chewing and the resultant SF is
In Durban about 60% of the Indian population is unskilled
primarily confined to Indians and people of south-east Asian
and semi-skilled. Almost 23% are in white-collar jobs, 10% are
stock. It has been reported elsewhere in the world where
skilled workers in the trades and professional people comprise
Indians reside, including South Africa. The RSA has an
about 5%.15 The majoritY are of the Hindu faith (75,26%),
Indian community of almost 1 million, the biggest concentra-
followed by Muslims (13,05%) and Christians (11,69%). The
tion being in the city of Durban.
Although several studies have reported the presence of the
disease in South Africa, figures reflecting the real occurrence
of the habit and the disease are still lacking. Schonland and
Bradshaw8 carried out a prevalence study of the chewing habit
in Durban among 500 domestic establishments, comprising
659 families. Since the majority of people belonged to the
Hindu faith, adjustments were made to increase the number of

Department of Maxillofacial Surgery, King Edward VDI

Hospital, Durban
H. A. SEEDAT, M.B. B.CH., B.D.S., PH.D.
Oral and Dental Research Institute, University of Stellen-
bosch, Parowvallei, CP
C. W . VAN WYK, B.CH.D., F.D.S., R.CS., PH.D. 18 11 o
%

Reprint requests [0: Professor C. w. van Wyk, Oral and Dental Research Institute, Private
Bag X I, Tygerberg, 7505 RSA. Fig. 1. The age and sex distribution of the South African Indian
Accepted 5 Joo 1988. population in 1980. .
 SAMJ VOL 74 3 DEC 1988 569

overwhelming maJonty are English speaking. According to home language (vernacular), religion and indulgence in the
Durban City Council statistics, there were 364 752 Indians in betel-ehewing habit. The intra-oral examination consisted of a
the metropolitan area in 1978; this figure had increased to visual inspection and palpation and was carried out under
almost 400 000 in 1980, the date that this survey was planned. natural light near a window or doorway. Artificial light was
From the status of know edge regarding the prevalence of used when there was insufficient natural light. Only one
betel-nut chewing and SF in South Africa, it was calculated author (H.A.S.) undertook the examination and interviewing.
that a sample size of 2400 subjects would be sufficient. Indicators of impending 'early' SF were a history Of betel-
Personal contact and enquiry into the betel habit by one of the nut chewing with two or more of the following symptoms and
authors (H.A.S.) revealed that it is almost exclusively practised signs: (I) a history of a burning sensation on eating spicy foods;
by adults and very infrequently by children. It was therefore this would have to be a new symptom for people used to
decided that the sample should include subjects of 10 years eating spicy foods; (il) dry mouth; (iil) complaint or exhibition
and older only, and that it would be stratified into the age of vesicles and ulcers; (iv) localised blanching of the palate,
groups 10 - 14, 15 - 24, 25 - 34, 35 - 44, 45 - 54, 55 - 64 and 65+ faucial regions, cheeks and lips; (v) smooth silk)' mobile feeling
years, and include equal numbers of each sex in each age of the mucosa lost and replaced by a linen- or leather-like
group. sensation; this feature was included because of its common
From maps available in the city of Durban, 38 Indian occurrence in patients referred to hospital with SF. A diagnosis
suburbs were identified in metropolitan Durban and it was of established SF was made if the chewer also had palpable
planned that the sampling points would be proportionally fibrous bands in the mouth.
distributed according to the size of the residential areas.
However, owing to redevelopment in some suburbs and because
others were too small, some areas were disregarded and others Results
consolidated, resulting in 29 well-delineated areas. This reduced
the sample size to 2058 and accordingly 147 subjects were The survey started during 1981 and was completed in 1983.
needed for each age-sex stratum. One hundred and forty- Since equal numbers per age/sex strata were selected it was
seven clusters of 14 individuals were to be examined at each not possible to correlate the sample characteristics with the
sampling point. Special maps were used to determine the population characteristics. However, when those parameters
sampling points. They showed the respective density of the which the sampling method did not influence, such as place of
population in the various areas and were of the scale 1:6000 birth, home language and religion, were compared an acceptable
and 1:15000, which showed all the streets and plot numbers. correlation was found. Ninety-nine per cent of the subjects
A grid was placed over a specific area of the map and two
numbers from a set of random numbers were then selected.
The first number was used as a co-ordinate on the X axis and
the other on the Y axis. The sampling point was taken where
TABLE I. AGE AND SEX DISTRIBUTION OF ALL CHEWERSIN
the projections met. This procedure was repeated until all
SURVEY
sampling points had been selected.
Provision was made before the sampling for a number of Age (yrs) Total (%)* Females (%) Males (%);
eventualities such as the selection of houses on the same side 10-14 2 ( 0,7) 1 ( 0,7) 1 (0,7)
of the street, the opposite side of the street, around corners 15 -24 6 ( 2,0) 6 ( 4,1) o
and in apartment buildings. These rules were followed until 25-34 17 ( 5,8) 17 (11,6) o
the required 14 subjects were interviewed. 35 - 44 32 (10,9) 32 (21,8) o
45-54 41 (13,9) 37 (25,0) 4 (2,7)
55-64 37 (12,6) 35 (23,8) 2 (1,4)
65+ ~(17,3) ~(30,6) ~(4,1)
House interviews Total 186 173 13
These interviews consisted of a screening questionnaire and • % of total, 294 per age group.
an intra-oral examination if required. A questionnaire was t % of females ~ 147 per age group.
completed for each selected subject detailing the age, sex, : %of males 5
marital status, occupation, birthplace (India or South Africa),

TABLE 11. SURVEY CHEWERS WITHOUT FEATURES OF SF AND CHEWERS WITH FEATURES OF IMPENDING
AND ESTABLISHED SF
Chewers with impending and
Chewers without SF established SF
Age (yrs) Total (%)* Females (%It Males (%); Total (%)* Females (%)t Males (%)t
10 -14 1 ( 0,3) o 1 (0,7) 1 (0,3) 1 ( 0,7) o
15-24 2 ( 0,7) 2 ( 1,4) o 4 (1,5) 4 ( 2,7) o
25-34 4 ( 1,5) 4 ( 2,7) o 13 (4,4) 13 ( 8,8) o
35-44 19 ( 6,5) 19 (12,9) o 13 (4,4) 13 ( 8,8) o
45-54 21 ( 7,1) 18 (12,2) 3 (2,0) 20 (6,8) 19(12,9) 1 (0,7)
55-64 28 ( 9,5) 26 (17,7) 2 (1,4) 9 (3,1) 9 ( 6,1) o
65+ 40 (13,6) 34 (23,1) 6 (4,1) 11 (3,7) 11 ( 7,5) o
Total m 103 12 71 70 1
• % of fofal. 294 per age group.
t % of temales ~ 147 per age group.
; % of males ~
Without SF ratio F:M 8,6:1.
With SF ratio F:M 70:1.
 570 SAMT VOL 74 3 DES 1988

were born in South Mrica, 72,5% belonged to the Hindu faith, The estimated crude prevalence rates of this smdy differ
17,8% were Muslim and 9,7% were Christian. Very few subjects markedly from the findings of Schonland and Bradshaw,8 who
were not able to converse in English. found that 30,7% of females and 5,5% of males in Durban
Of the 2058 subjects examined, 186 (9,04%) indulged in the chewed betel nut. The above smdy was not a random survey,
chewing habit and 71 (3,4%) of the sample had signs and making comparison difficult. Should one accept that the fmd-
symptoms of impending (early) and established SF. The symp- ings reflect the situation at that stage, then it means a drastic
toms and signs were as follows: burning and pain 9, history of reduction in the practice has taken place. Some reduction is
\'esicles and ulcers 9, linen- or leather-like mucosae 71, blanch- probable as there is a growing awareness of the inherent
ing 63, and fibrous bands 33. danger of betel-nut chewing.
Females far outnumbered males in all age groups (Table I). On the other hand, the prevalence rates for people with
The habit was age-related. However, when the age distribution signs indicating the onset of the disease or having the disease
was compared between the chewers without SF and those with is substantially higher than the results of Shear et al. 9 (0,5%)
feamres of impending and established SF a significant dif- and Dockrat and Shear 10 (0,6%) obtained in Pretoria-Johannes-
ference was found (chi-square test = 23,572; df = 5; P < burg and Durban. The rates for people with fibrous bands in
0,001). The majority of chewers without signs of SF were in this smdy were similar to the findings of Van Wyk et al.
the age group 65 years and older while the majority of those (1,25%).13
with signs of the disease fell in the age group 45 - 54 years A similar discrepancy is noted in results recorded in India
(Table 11). (Table IV). The differences may be partly related to the
There were no significant differences between the two groups method of sampling but it is more likely that the diagnostic
with regard to marriage, occupation, birthplace, home language criteria we used resulted in the diagnosis of a wider spectrum
or religion. of SF cases. As mentioned earlier, we included subjects who,
we believed, had early signs of SF as well as those with fully
established disease. The occurrence of established cases is
Discussion and conclusions higher than the majority of Indian smdies but corresponds to
the Trivandrum study of Zachariah et al. 18 (Table IV).
This survey was stnitified according to age and gender, which
allows for the calculation of related prevalences in the popula-
tion. The survey was planned according to the required criteria
for such a type of survey, thus there was every reason to TABLE IV. PREVALENCE OF SUBMUCOUS FIBROSIS IN INDIA
believe that the fmdings reflected the true position in Durban.
The Indian population of Durban in 1980 constimted 46% of Prevalence
the total Indian population of South Mrica, which is over- Source Location rate (%)
whelmingly urban. According to Meer 1; the age, religion, Pindborg et al.'· Lucknow 0,51
language and social distribution (including the customs) are Bombay 0,50
similar for Indians throughout the country. Therefore it can Pindborg et alY Bangalore 0,18
be argued that the findings in Durban can be extrapolated to Zachariah et al.'8 Trivandrum 1,22
the rest of the country. Pindborg et al.'9 Andhra Pradesh (Srikakulam) 0,04
By using the 1980 census figures for this group of South Bihar (Darbhanga) 0,07
Mricans (Fig. 1), it· was possible to make rough estimates Gujarat (Bhavnagar) 0,16
regarding the prevalence of chewers and people showing signs Kerala (Ernakulam) 0,36
of established and impending SF. It is estimated that 5% of Wahi etal. 20 Uttar Pradesh (Mainpuri) 0,59
the population (38699) are chewers and that 2,3% (18884) Mehta et al. 21 Maharashtra (Poona) 0,03
could develop the disease (Table Ill). If only established SF Pindborg et al. 22 Parakadavu 0,14
(the presence of palpable fibrous bands in the mouth) was Gupta et al.23 Kerala (Ernakulam) 0,65
considered then the calculated crude prevalence of the disease Gujarat (Bhavnagar) 0,06
in the total population comes to 1,2% (9844) and 1,6% in
people older than 10 years (Table Ill).

TABLE Ill. PROJECTED OCCURRENCE OF CHEWERS AND· OF CHEWERS WITH FEATURES OF IMPENDING AND ESTABLISHED SF IN
INDIANS ACCORDING TO THE 1980 POPULATION CENSUS
Projected chewers with features of impending
Projected ch ewers and established SF
Survey General population Survey General population
Age (yrs) cases Females (%) Males (%) Total (%) cases Females (%) . Males (%) Total (%)
10-14 2 0,7 0,7 0,7 1 0,7 0,3
15-24 6 4,0 2,0 4 3,0 1,5
25-34 17 12,0 6,0 13 9,0 4,0
35-44 32 22,0 11,0 13 9,0 4,0
45-54 41 25,0 3,0 14,0 20 13,0 0,7 7,0
55-64 37 24,0 1,4 13,0 9 6,0 3,0
65+ 51 31,0 4,0 17,0 11 7,5 4,0
10+ 186 12,0 0,5 6,4 71 6,0 0,06 3,0
All ages 9,0 0,4 5,0 4,4 0,04 2,3
Projected occurrence of chewers with fibrous bands (SF)
10+ 33 3,2 1,6
All ages 2,4 .1,2

The overwhelming preponderance of female chewers reflects
the panern of betel-nut chewing among South African Indians.
It also reflects the social attitude of this population. Since the
habit is not accepted in Western-type society one fmds that
th~ men, who form an active part of this society, refrain and
prefer smoking. The women, on the other hand, are more
home-bound and can participate. Also, smoking by women is
frowned upon in Indian society while there is a certain accept-
ance of the betel-nut habit. The above fmdings are in agreement
with others. 1,4,24
According to Schonland and Bradshaw8 the percentage of
chewers increases with age, which corresponds with our fmd-
ings. The resulting disease, however, appears to affect younger
people. Paymaster3 believes that younger persons are usually
affected and Pindborg and Sirsat! reported that the maximum
prevalence occurs in the age range 20-40 years. These proposi-
tions are supported by our work where the mean age for
chewers without SF was 56 years while it was 43 years for
those depicting early and late features of the disease. Seventy-
five per cent of those with late features belonged to the age
group 25 - 54 years. The reason is not clear but it is possible
that the pattern of the practice, such as frequency of chewing,
or individual susceptibility for the disease may induce early
onset.
From this study it is 'concluded that the noxious habit of
betel-nut chewing is still very prevalent among South African
Indians and will result in a number of people developing
serious health problems, including permanent trismus and oral
cancer. Intervention programmes are strongly advocated.
This research was carried out in partial fulfilment for the degree
of Ph.D. at the University of Stellenbosch by Dr H. A. Seedat.
The investigation was supported by the South African Medical
Research Council.
REFERENCES
1. Pindborg H, Sirsat SM. Oral submucous fibrosis. Oral Surg 1966; 22:
764-779.
2. Joshi SG. Submucous fibrosis of the palate and pillars. Indian] Owlaryngol
1953; 4: 110-113.
SAMJ VOL 74 3 DEC 1988 571
3. Paymaster JC. Cancer of the buccal mucosa: a clinical study of 650 cases in
Indian patients. Canar 1956; 9: 431-435.
4. De Sa JV. Submucous fibrosis of the palate and cheek. Ann Owl Rhinol
Laryngo/1957; 66: 1143-1159.
5. Mehta FS, Pindborg H, Hamner JE et al. Reporl on Investigations of Oral
Cancer and Precancerow Condilions in Indian Rural PopuIarions. Copenhagen:
Munksgaard, 1971: 1966-1969.
6. Pindborg H, Chawla TN, Srivastava AN, Gupta D, Mehrotra ML. Oinical
aspects of oral submucous fibrosis. ACla Odonwl Scand 1964; 22: 679-691.
7. Su IP. Idiopathic scleroderma of the mouth: report of three cases. Arch
OlolaryngoI1954; 59: 330-332.
8. Schonland MN, Bradshaw E. Upper alimentary tl'ag cancer in Natal
Indians with special reference to the betel-chewing habit. Br] Cancer 1969;
23: 670-682.
9. Shear M, Lemmer J, Dockrat IS. Oral submucous fibrosis in South Mrican
Indians: an epidemiological study. S Afr J Med Sci 1967; 32: 41-46.
10. Dockrat IS, Shear M. Oral submucous fibrosis in Natal. In: Cahn L, ed.
Proceedings of lhe Fourlh Internalional Academy of Oral Palhology. New
York: Gordon & Breach, 1969: 57-63.
I!. Randeria JD. Orogastric cancer among betel chewers in Natal. In: Shapiro
HA, ed. Prpceedings of lhe Firsl Nalional Cancer Congress of Soulh Africa.
Johannesburg: National Cancer Association of South Mrica. 1972: 69-77.
12. Randeira JD. Study of oral lesions with special reference to young nut
chewers. In: Severi L, ed. Tumours of Early Life in Man and Animals.
Perugia: Perugia Quadrennial International Conference on Cancer, 1977:
557-567.
13. Van Wyk CW, Stax J, Farman AG. The prevalence of oral mucosal lesions
among a random sample of Asians resident in Cape Town.] Denl Assoc S
Afr 1977; 32: 589-592.
14. Mistry SD. Ethnic groups of Indians in South Mrica. S Afr Med] 1965; 39:
691-694.
15. Meer F. Porlrail of Indian Soulh Africans. Durban: Avon House, 1969: 86.
16. Pindborg H, Kalapessi HK, Kale SA, Singh B, Talyerkhan BN. Frequency
of oralleukoplakias and related conditions among 10000 Bombayites.] AI/-
India Denl Assoc 1965; 37: 228-229.
17. Pindborg H, Bhan M, Devanath KR, Narayana HR, Rarnachandra S.
Frequency of oral white lesions among 10000 individuals in Bangalore,
South India: A preliminary report.IndianJ Med Sci 1966; 2(), 349-352.
18. Zachariah T, Mathew B, Varma NAR, Iqbal AM, Pindborg H. Frequency
of oral mucosal lesions among 5000 individuals in Trivandrum, South India.
] Indian Denl Assoc 1966; 38: 290-294.
19. Pindborg H, Mehta FS, Gupta PC, Daftary DK. Prevalence of oral
submucous fibrosis among 50915 Indian villagers. Br] Cancer 1968; 22:
646-654.
20. Wahi PN, Mital VP, Lahiri B, Luthra UK, Seth RK, Arora GD. Epidemio-
logical study of precancerous lesions of the oral cavity: a preliminary report.
Indian] Med Res 1970; 58: 1361-1391.
21. Mehta FS, Gupta PC, Daftary DK, Pindborg H, Choksi SK. An epidemio-
logic study of oral cancer and precancerous conditions among 101761
villagers in Maharashtra, India. 1nl] Cancer 1972; 10: 134-141.
22. Pindborg H, Bhonsle RB, Murti PR, Gupta PC, Daftary DK, Mehta FA.
Incidence and early forms of oral submucous fibrosis. Oral Surg 1980; 22:
40-44.
23. Gupta PC, Mehta FS, Daftary DK el al. Incident rates of oral cancer and
natural history of oral precancerous lesions in a 10-year follow-up study of
Indian villagers. CommunilY Denl Oral Epidemiol198O; 8: 287-233.
24. Wahi PN, Kapur VL, Luthra UK, Srivastava MC. Submucous fibrosis of
the oral cavity: two studies on epidemiology. Bull WHO 1966; 35: 793-799.