March 1998, Volume 48, Issue 3

Original Article

Prevalence and Pattern of Smoking in Pakistan

Syed Ejaz Alam  ( PMRC Research Centre, Jinnah Postgraduate Medical Centre, Karachi. )


Objective: To determine the prevalence and pattern of smoking amongst Pakistanis. Sampling Design: A two stage stratified sample design was adopted for the survey, Primary Sampling Units (PSU’s) and Secondary Sampling Units (SSU’s).
Setting: Urban and Rural, Pakistan 1990- 94.
Subjects: Stratified systematic sample of 9441 males and females aged 15 years and above.
Main outcome measures: Smokers were defined as, subjects who were currently smoking and who had smoked _100 cigarettes/beedis or chillum/huqqa in their life time.
Results: In Pakistan 21.6% (36% males and 9% females) of 9441 subjects were smokers. In urban areas it was 20.7% and in rural 22.0%, males were predominant in both urban and rural areas. Proportion of smokers who used cigarette/beedi were significantly higher in ‘males (60%) while chillumfhuqqa were more in females (62%). Prevalence increased with age upto 64 years, after which it declined but in urban females it continued to rise with age. Among both males and females; illiterate, married individuals with poor general health were more likely to smoke. These factors were 2 to 3 times more in males and 2 to 5 times more in females who were more likely to be smokers than those who were literate, single individuals with good general health. Conclusion: Smoking was more prevalent in illiterate, married persons and those with poor general health (JPMA 48:64, 1998).


The Tobacco industry in Pakistan is expanding at a rate of 5% per year1 and Pakistani cigarettes have amongst the highest levels of tar and nicotine in the world2. In addition to cigarettes, tobacco is smoked in unique local ways which include “beedi” (tobacco rolled in dry leaves), “Huqqa” comprises of a clay pot (chillum) containing burning coal over a layer of tobacco fixed to a water containing pot with two pipes, one for smoking and the other for filtering the smoke through water. Cigarette smoking or tobacco use has been very strongly associated with a wide range of diseases including cancers of mouth, throat, larynx. lungs, bladder, emphysema and coronary heart disease. This study reports the prevalence and patterns of smoking as recorded ina survey carried out on a stratified systematic sample of adults (15 years and above) during 1990-94.

Subjects and Methods

During a National Health Survey of Pakistan a structured questionnaire was used to obtain data on smoking, in relation to age. sex, literacy, marital status and general health in urban and rural poptilation of Pakistan. The sampling frame of National Health Survey of Pakistan (NHSP) consisted of all urban and rural areas of the four provinces (Punjab, Sindh, NWFP and Ralochistan) as defined in 1981 population census3 using atwo stage (Primary Sampling Units and Secondary Sampling Units) stratified sample design. The primary sampling units (PSU ‘s) in the urban areas were enumeration blocks and in rural areas they were mouzas/dehs/villages. Eighty PSU’s were drawn, of these 32 were urban and 48 from rural areas. Sample of 30 households from each PSU’s were selected using systematic sampling technique by random start. The calculated sample size consisted of 2400 house-holds, However, when the systemic sample was drawn, there were preliminary 2359 households in 80 PSU’s. The overall household non-response was 2.1%4. The population covered in 2395 sampled households was 19747. Out of them 18322 individuals couldbe examined. Out of 18322 subjects 8881 (48,5%) were infants and children (under 15 years) and 9441 (51.5%) were adults (15 years and above). Smoking information was taken from both sexes in urban and rural population of Pakistan. Respondents who were currently smoking and who had smoked _100 cigarettes orbeedis or Chillum/Haqqa in their life time were defined as smokers. The data was entered and analyzed on computer package “Epi. Info 5.1”, the differences in smoking prevalence between groups were compared with X2-test.


Smoking Prevalence and Associations
Overall, 2 1.6% of 9941 subjects were smokers. In the study population as a whole and in urban and rural groups, proportion of males were significantly higher than females (P<0.001)(Table 1).

Cigarette/beedi smoking was significantly more in males and chillum/huqqa in females (P<0.00 1) (Figure 1).

Males smoking cigarette/beedi were sinificantly more in urban and female smokers of chilluinfhuqqa in rural population (p
Prevalence of smoking increased with the age upto 64 years amongst males, after which it declined, while inurbanfernales it continued to increase with age, while it declined in rural females (Figure 2).

The average age of onset for cigarette smokers was 18 years in males and 24 years in females, while for chillum/huqqa smokers it was 20 years in males and 29 years in females. Male cigarette users smoked on an average times per day. The prevalence of smoking was univariately associated with literacy, marital status and general health condition (p<0.00l). Smoking was more prevalent in illiterate, married with poor general health in both sexes. The presence of these factors were 2 to 3 (odd ratio) times more in males and 2 to 5 (odd ratio) times more in females who were smokers than those who were literate, single and had good general health (Table III).


This, population based study reports the prevalence of smoking in Pakistani adult population aged 13 years and above which was 2 1.6% during 1990-94. These rates were higher in males (36%) than females (9%). Among cigarette smokers average age at onset of smoking was 18 years in males. 24 years in females and in chillum/huqqa smokers the average age at onset was 20 years in males and 29 years in females. In Delhi (Urban India) 13558 subjects aged 25-64 years were studied. The ovenli prevalence of smoking was 24.5%, with males being 45% and females 7%5, In Pakistan, smoking habits have been reported ma few studies comprising of different population groups6-8. The frequency of smoking in male medical students was 21.3% in Karachi7 while and 22.4% male college students were smoker in Peshawar8. . The average age at the start of smoking was 17 yeats amongst students in Karachi7 which is almost similar to the present study. The most vulnerable ages for cigarette consumption arc 10 through 18 years, when most users start smoking and become addicted to tobacco9. It was observed that in both sexes literate, single individuals with good general health were significantly less likely to be smokers than those who were illiterate, married and had poor genemi health.


1. Pakistan Health Education Survey 1991-92. Islamabad, Ministry of Health, Government of Pakistan, 1993.
2. Asghar, M. and Jan. Z.A. Monitoring of harmful constitutenta of cigarette and tobacco in Pakistan J. Pak. Med. Assoc., I 989;39:66-8.
3. Malik, M.D. ‘The sampling methodology for the National Health Survey of Pakistan’ Pak. J. Med. Res., I 992;3 1:289-90.
4  National Health Survey ofPakiatan (Preliminary Report) Islamabad, Pakistan Medical Research Council (PMRC), 1996, p.10.
5. Narayn. V.K.M,, Chadha, S.L,, Hanson, R.L. et al. ‘Prevalence and patterns of smoking in Delhi: Cross sectional study’ Br. Med. J., 1996;312:1 576-1579
6. Mahmood, Z.. ‘Smoking and chewing habits ofpeople ofKarachi - 1981’. J. Pak, Med Assoc., 1982.32:34-37.
7. Ahmed, EN. and Jafarcy, NA. ‘Smoking habits amongst medical students of Sindh Medical College, J.Pak.Med.Assoe., 1983:33:39-44.
8. Ahmad, Z., UlIah, H., Stddiqui, M.K. et al. Blood parameters asd smoking pattern in Peshawar Colleges. Pak. J. Med. Res., 1995;34:190-193.
9. Sadruddin. A. and Agha. F Teenager’s smoking . A great Public Health Problem, Renewing the Pool of Smokers. J. Pak. Med. Assoc., 1996;46:284-286.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: