An intervention study of oral cancer and precancer in rural Indian populations: a preliminary report*

Bulletin of the World Health Organiztion, 60 (3): 441 - 446 (1982) An intervention study of oral cancer and precancer in rural Indian populations: a preliminary report* FALI S. MEHTA,1 M. B. GUPTA, 3 j. J. PINDBORG,4 R. B. BHONSLE,5 P. N. JALNAWALLA,5 & P. N. SINOR5 A large-scale epidemiologicalstudy oforal cancer andprecancer was initiated in 1966 in several regions of India. Phase 1 of the study consisted of a cross-section survey to deter- mine theprevalence ofcancer andprecancerous lesions, whilephase 2 was a 1J-yearfollow- up survey to determine the incidence and natural history oforalprecancer. Following these preliminary investigations, an intervention study was started to try to persuade subjects to give up tobacco, and to measure the subsequent changes in the incidence and regression rate oforalprecancer. In each ofthe three study areas, 12 000 adult tobacco users were selected, examined, and interviewed in a house-to-house survey. Information on the ill effects of tobacco use was given through individual discussions with a social scientist, film shows, and posters. The population was reviewed after one year and the proportions ofsubjects who had discontinued tobacco use werefound to be 2% in Ernakulam, 1% in Bhavnagar, and 5% in Srikakulam. Additionally, between 1%o and 16% had reduced their tobacco use. In Bhavnagar and Ernakulam the regression rate of leukoplakia was significantly higher among those who had stopped or reduced their tobacco consumption. In Srikakulam there was a highly significant correlation between reduction in tobacco consumption and regression ofpalatal lesions. A large-scale epidemiological survey of oral cancer and precancer was started in 1966 in several areas of India. The preliminary survey was divided into 2 phases: (a) Phase 1 of the project was a cross-sectional field survey undertaken in 1966 -67 covering five districts in four states of India, each chosen for the particular tobacco smoking and chewing habits of the inhabitants. The objective was to determine the prevalence of oral precancerous lesions and their association with tobacco usage. The results of this baseline survey, covering 50 915 rural individuals, were published in 1969 (1). (b) Phase 2 of the project was a 10-year follow-up of 60/o of the original study population. Two of the original districts were not included in this phase because very few precancerous lesions were detected there. A comprehensive report of this follow-up, * Requests for reprints should be addressed to Professor J. J. Pindborg, 4 Universitetsparken, DK-2100 Copenhagen, Denmark. X Head, Basic Dental Research Unit, Tata Institute of Funda- mental Research, Bombay, India. 2 Consultant Behavioural Scientist, Basic Dental Research Unit, Tata Institute of Fundamental Research, Bombay, India. 3 Epidemiologist, Tata Institute of Fundamental Research, Bombay, India. 4 Professor of Oral Pathology, Department of Oral Pathology, Royal Dental College, and Head, Department of Oral Medicine and Oral Surgery, Rigshospitalet, Copenhagen, Denmark. 5 Dentist, Basic Dental Research Unit, Tata Institute of Funda- mental Research, Bombay, India. giving the incidence of oral cancer and precancer and an analysis of the natural history of these lesions, has been published elsewhere (2). The results obtained in phases 1 and 2 indicated that: (a) oral cancer and precancerous lesions occurred almost solely among those who smoked or chewed tobacco, and (b) oral cancer was almost always preceded by some type of precancerous lesion. On the basis of these results, an intervention trial was initiated to try to persuade the target population to give up their tobacco habits, and to investigate any effect this might have on the incidence and regression rate of the precancerous lesions. Many anti-smoking campaigns and clinics have tried to persuade people to give up smoking through education, counselling, and medication, but the impact of these programmes on the disease process has not often been assessed. There have, however, been several studies of the effects of a naturally occurring intervention. Doll & Pike (3) compared mortality among British phys- icians with that in the general population, on the premise that a higher proportion of the physicians had discontinued smoking. In several other studies, data for ex-smokers have been analysed separately. The intervention in these studies, however, occurred without planning or control. In the present study, a carefully designed compre- 4194 -441- F. S. MEHTA ET AL. hensive programme of intervention is aimed at a specific target population. Every individual in this population is examined annually to measure any change in the incidence and the regression pattern of the disease. Although there is no separate control population, the baseline values have been well estab- lished in the early phases of the study. The prelimi- nary findings of the survey are presented in this report. MATERIALS AND METHODS Study population The study population was selected from inhabi- tants of Ernakulam district in the state of Kerala, Bhavnagar district in the state of Gujarat, and Srika- kulam district in the state of Andhra Pradesh (Fig. 1). This population was similar to that studied in phases 1 and 2, but was not composed of the same indi- viduals. A large number of subjects were screened and about 12 000 tobacco users aged 15 years and over were selected in each district. In Bhavnagar district, females were not included as they rarely used tobacco. The tobacco habits in each district have been described earlier (1). I N D A Fig. 1. Map of India showing the three areas where the intervention study was carried out. Survey methods The personnel involved in the survey, the procedure of oral examination, and the criteria for diagnosis were similar to those used in phase 1 (1). The record- ing of tobacco use and the oral examination were carried out separately and independently. As this was an intervention phase, the study team also included two social scientists, who were trained in conducting in-depth interviews in a considerate, sympathetic, warm, and positive manner. Intervention methods A programme of intervention was devised after appropriate pilot surveys. In-depth interviews of the subjects were conducted to investigate the reasons for using tobacco, the implications of the habit in social, economic, and health terms, and the possible motivations for giving up the habit. Pilot surveys were conducted to assess the communication media available to the target population, that could be used to convey the intervention messages. The programme was designed to use different communication media in such a way as not to overwhelm the target popu- lation at any one time. Each item of communication was pretested on the target population and, if necessary, modified to make it easily under- standable. Baseline survey The first stage of intervention was aimed at creating an awareness of the ill effects of tobacco. Each sub- ject was seen by a dental surgeon, who carried out a full clinical examination of the mouth and gave his professional advice to the individual. A social scien- tist then interviewed the subject and explained the purpose of the examination and the implications of the advice. He encouraged the individual to ask ques- tions, and explained the relationship between tobacco use and oral cancer using an illustrated booklet. A version of the film Tobacco habits and oral cancer was produced for each region. This film out- lined the relationship between tobacco use, precancer- ous lesions, and oral cancer with specific emphasis on the tobacco habits of the particular area. The films were pretested for their impact on the target popula- tion, and were shown using a portable super-8-mm projector. The results indicated that the film was capable of producing the desired awareness among the audience if it was presented within a specific format, i.e., an introduction to the film, the first showing of the film, a group discussion, and a second showing. Based on the results of the preliminary surveys, it was decided that the film shows were to be limited to 20-25 individuals and were to be held indoors. The discussion was intended to emphasize 442 ORAL CANCER AND PRECANCER IN INDIA Table 1. Changes in tobacco use at the first follow-up in Ernakulam district, one year after intervention Level of use Bidia Cigarette Chewing Mixed use Total Increased 3.3 2.6 0.9 1.7 2.1 Unchanged 88.3 83.6 89.9 87.2 88.7 Reduced 6.9 7.8 6.5 10.7 7.3 Stopped 1.5 6.0 2.7 0.4 1.9 Total 100.0 100.0 100.0 100.0 100.0 Total no. of subjects 5046 268 4428 1646 11 388 ' Bidi is a cheap Indian cigarette (6). the salient points of the film, using the queries and Again, an illustrated booklet was used in the indi- comments of the audience. vidual interviews, the film was shown in the same After the oral examinations, posters containing format, and posters were displayed at prominent simple, pretested, handwritten messages were dis- places in the villages. played at prominent places in each village. The objec- tive of these posters was to maintain an awareness of the relationship between tobacco use and oral RESULTS cancer. Firstfollow-up The prevalence of leukoplakia was found to be2.9% in Ernakulam and 5% in Bhavnagar district. In One year after the intervention, the villagers were the first baseline survey (1), the prevalence of leuko- re-examined clinically and interviewed about their plakia among tobacco users was 2.7% in Ernakulam tobacco habits. The specific tobacco habits of each and 4.6% in Bhavnagar. individual were discussed, and if the subject was Tables 1, 2, and 3 show the changes in tobacco convinced about the ill effects of tobacco, he was habits established at the first follow-up, one year after advised on ways of giving up the habit. Otherwise, the the initial examination. In Ernakulam district, 1.9% ill effects of using tobacco were explained again of subjects discontinued and 7.3%o reduced their and the importance of giving up tobacco was tobacco habits; the highest percentage of discontinu- emphasized. ation was among cigarette smokers (6%). In Bhav- Table 2. Changes in tobacco use at the first follow-up in Bhavnagar district, one year after intervention Bidi, pipe, Level of use Bidi Clay pipe and others Chewing Mixed use Total (%) (%) (%) (%) (%) (%) Increased 0.2 - - 0.7 0.6 0.2 Unchanged 96.3 97.9 97.8 93.6 99.4 96.8 Reduced 1.2 0.7 0.6 0.5 - 0.9 Stopped 2.3 1.4 1.6 5.2 - 2.1 Total 100.0 100.0 100.0 100.0 100.0 100.0 Total no. of subjects 5329 3269 885 748 170 10 401 443 F. S. MEHTA ETAL. Table 3. Changes in tobacco use at the first follow-up in Srikakulam district, one year after intervention Conventional Reverse Level of use smoking smoking' Chewing Mixed use Total ( (%) % % (%) Increased 18.0 32.1 21.4 16.1 26.1 Unchanged 48.0 54.3 41.9 58.9 52.7 Reduced 28.0 9.2 26.7 21.7 16.3 Stopped 6.0 4.4 10.0 3.3 4.9 Total 100.0 100.0 100.0 100.0 100.0 Total no. of subjects 2852 6348 401 1319 10 920 a Smoking of cigars with the lighted end inside the mouth. nagar district, 2.1% gave up tobacco completely and Table 4. Regression of leukoplakia according to change 0.9% reduced their usage. In Srikakulam district, in tobacco use at the first follow-up examination, one 4.9% stopped and 16.3% reduced their tobacco year after intervention habits. In Bhavnagar and Srikakulam, the highest No. of No. of Regression percentage of subjects giving up the habit was among Level of use leuko- regressed rate tobacco chewers (5.2% and 10%, respectively). plakia leukoplakia (%) Table 4 shows the relationship between the changes in tobacco use and the regression of leukoplakia in Ernakulam Ernakulam and Bhavnagar. Although very few Unchanged patients with leukoplakia gave up their tobacco or increased 263 3 1.1 habits, the difference in the regression rates was stat- Reduced or stopped 76 4 5.3a istically significant. The regression rate of leuko- plakia among those discontinuing or reducing their Total 339 7 2.3 level of tobacco use was more than four times higher than among those who continued their habits. Bhavnagar Table 5 shows the relationship between changes in Unchanged . . . or increased520 67 12.9tobacco use and palatal changes among subjects in or increased Srikakulam. The regression rate was similar for those Reduced or stopped 9 5 55.6 who either did not change their habits or increased Total 529 72 13.6 their level of tobacco use, while rates among those who reduced or gave up tobacco, were approximately x2 = 4.96, P < 0.05. 3 and 7 times higher, respectively (P < 0.01). b x2 = 13.7, P < 0.01. Table 5. Palatal changes observed in subjects in Srikakulam, one year after intervention Increase in size Partial Complete Regression rate Level of use of lesion Persistence regression regression Total (%) Increased 172 1290 89 51 1602 3.2 Unchanged 225 2529 150 85 2989 2.8 Reduced 29 464 25 58 576 10.1 Stopped 11 104 15 34 164 20.7e Total 437 4387 279 228 5331 4.3 a X2 = 290.6, P < 0.01. "44 ORAL CANCER AND PRECANCER IN INDIA DISCUSSION It has often been suggested that, since tobacco is a primary etiological agent for leukoplakia, cessation of its use may cause the leukoplakia to regress. The present study has provided a direct epidemiological confirmation of this hypothesis. In a study in Hungary, 34%o of leukoplakias were reported as cured during a mean observation period of 10 years following surgical treatment, conservative treatment, or elimination of etiological agents, such as mechan- ical factors, alcohol, Candida albicans, and smoking (4). However, the effect of giving up smoking was not assessed separately. Several investigators (5 - 8) have reported spontaneous regression of leukoplakia but have not correlated the changes with tobacco habits. An increase in the regression rates of the palatal lesions following discontinuation of tobacco use has already been demonstrated in Srikakulam (9). The present study appears to confirm these results on a larger scale. As palatal changes were assessed in terms of separate components, it was not always easy to distinguish between persistent or partially regressed lesions. However, in spite of these difficulties, the association between reduction in tobacco use and regression of lesions was clear-cut and significant. The percentage of subjects giving up or substan- tially reducing their tobacco use ranged from 3% in Ernakulam to 21 o in Srikakulam. These figures are quite low but it should be noted that the target population in the present study was not motivated. The individuals had no reason to give up tobacco and in fact, a high proportion of subjects thought that tobacco possessed a range of medicinal properties and had beneficial effects. Thus the primary task was to create an awareness among the target population of the ill effects of tobacco usage in general, and of the relationship between tobacco and oral cancer in particular. The results achieved can therefore be considered as an index of the level of awareness thus created. Although it is generally thought that awareness alone is insuf- ficient to modify a behaviour pattern, in the present study it led to a partial or complete discontinuation of tobacco use by many individuals. There have been two reports recently on the effects of intervention on smoking habits in a selected group of individuals, one from Stanford, USA (10) and the other from North Karelia, Finland (11). The objec- tive in both projects was to persuade the population to give up smoking in order to reduce the risk of cardiovascular disease. In both projects the target population was motivated and total community support was available. A variety of persuasive techniques and communication media were used, including television. In the Stanford project, 17%/ of the smokers discontinued their smoking habits in a 2- year period, while in North Karelia the proportion of smokers was reduced from 54% to 43% of the population over a period of 5 years. In the present study the population was not motivated and there was no involvement at community level. However, the results obtained a year after the start of the project appear to be comparable with the two reported studies. The present study is still continuing and the inter- vention methods are revised and updated every year. It is planned in the future to use newspaper articles, radio broadcasts (slogans, talks, and plays), posters, cinema slides, and folk art and theatre forms to convey information on the effects of tobacco. It is felt that it will soon be possible to determine whether there is a significant reduction in the inci- dence of leukoplakia following discontinuation of tobacco use. ACKNOWLEDGEMENTS This project was supported by funds from the National Institutes of Health, US Public Health Service, under PL 480 Research Agreement No. 01-022-01. RItSUME ETUDE AVEC INTERVENTION SUR LES LtSIONS CANCtREUSES ET PRtCANCR.EUSES DE LA BOUCHE DANS DES POPULATIONS RURALES: RAPPORT PRELIMINAIRE Une etude epidemiologique de grande envergure sur les l6sions cancereuses et precancereuses de la bouche a e entreprise en 1966 dans differentes r6gions de l'Inde parmi 50 000 villageois. La phase 1 de cette etude a e une enquete transversale destin6e a determiner les taux de prevalence, alors que la phase 2, une enquete de surveillance ulterieure 445 446 F. S. MEHTA ET AL. d'une dur6e de 10 ans, a permis de determiner l'incidence et 1'histoire naturelle des l6sions pr6canc6reuses de la bouche. A la suite de ces investigations pr6liminaires, une 6tude avec intervention a ete entreprise afin de persuader les sujets adonn6s au tabac d'y renoncer, et de mesurer les modifica- tions de l'incidence qui ont suivi et le taux de r6gression des l6sions pr6canc6reuses de la bouche. Dans chacune des trois zones, 12 000 adultes ayant des habitudes de tabagisme ont ete choisis et l'on a recherche chez eux des l6sions precanc6- reuses de la bouche, notamment leucoplasie, 6rythroplasie, k6ratose palatine chez les sujets fumant des cigares a 1'envers, lichen plan, et fibrose sous-muqueuse buccale. La population a ete examin6e au cours d'une enquete de porte en porte et les sujets ont 6t6 interrog6s sur leurs habitudes en mati&e de tabac. L'intervention au cours de la premiere annee a e realisee au moyen d'entretiens individuels avec un sp6cialiste des sciences sociales, de presentations de films et d'affiches. Apres une periode d'un an, la proportion de la population qui avait interrompu ses habitudes en ce qui concerne le tabac representait 2% en Ernakulam, 1% en Bhavnagar et 5% en Srikakulam. En outre, la proportion de ceux qui avaient r6duit leur usage du tabac allait de 1% A 16%. En Bhavnagar, de meme qu'en Ernakulam, le taux de regression de la leucoplasie etait notablement plus eleve parmi ceux qui avaient arrete ou r6duit leur tabagisme. En Srikakulam, l'association entre la r6duction de l'usage du tabac et la regression des lesions palatines etait hautement significative. REFERENCES 1. MEHTA, F. S. ET AL. Epidemiologic and histologic study of oral cancer and leukoplakia among 50,915 villagers in India. Cancer, 24: 832- 849 (1969). 2. GUPTA, P. C. ET AL. Incidence rates of oral cancer and natural history of oral precancerous lesions in a 10-year follow-up study of Indian villagers. Community dentis- try and oral epidemiology, 8: 287 - 333 (1980). 3. DOLL, R. & PIKE, C. Trends in mortality among British doctors in relation to their smoking habits. Journal of the Royal College of Physicians, 2: 216- 222 (1972). 4. BAN6czy, J. Follow-up studies in oral leukoplakia. Journal of maxillo-facial surgery, 5: 69- 75 (1977). 5. PINDBORG, J. J. ET AL. Studies in oral leukoplakia. A preliminary report on the period prevalence of malignant transformation in leukoplakia based on a follow-up study of 248 patients. Journal of the Amer- ican Dental Association, 76: 767 - 771 (1968). 6. MEHTA, F. S. ET AL. Oral leukoplakia in relation to tobacco habits: a ten-year follow-up of Bombay police- men. Oral surgery, oral medicine and oral pathology, 34: 426-433 (1972). 7. MEHTA, F. S. & PINDBORG, J. J. Spontaneous regres- sion of oral leukoplakias among Indian villagers in a 5-year follow-up study. Community dentistry and oral epidemiology, 2: 80- 84 (1974). 8. SILVERMAN, S. JR ET AL. Malignant transformation and natural history of oral leukoplakia in 57,518 industrial workers of Gujarat, India. Cancer, 38: 1790- 1795 (1976). 9. MEHTA, F. S. ET AL. Reverse smoking in Andhra Pra- desh, India: Variability of clinical and histologic appearances of palatal changes. International journal of oral surgery, 6: 75 - 83 (1977). 10. MACCOBY, N. ETAL. Reducing the risk of cardiovascular disease. Journal of community health, 3: 100- 114 (1977). 11. TUOMILEHTO, J. ET AL. A community anti-smoking pro- gramme: Interim evaluation of the North Karelia Pro- ject. International journal of health education, 21 (suppl.) (1978).