Daily tobacco use and problem drinking among urban adults in South Africa: a longitudinal study

Daily tobacco use and problem drinking among urban adults in South Africa: a longitudinal study

Karl Peltzer^1,2,&, Li-Wei Chao^1,3,4,5, Shandir Ramlagan¹, Helena Szrek^3,4

 

¹HIV/AIDS/STIs and TB Research Programme, Human Sciences Research Council, Pretoria, South Africa, ²Department of Research and Innovation, University of Limpopo, Turfloop, South Africa, ³University of Porto, Porto Business School, Porto, Portugal, ⁴University of Porto, Center for Economics and Finance (cefUP), Porto, Portugal, ⁵University of Pennsylvania, Population Studies Center and Leonard Davis Institute of Health Economics, Philadelphia, USA

 

 

^&Corresponding author
Karl Peltzer, HIV/AIDS/STIs and TB Research Programme, Human Sciences Research Council, Pretoria, South Africa

 

 

 

 

In the "South African National Health and Nutrition Examination Survey", conducted in 2012, of adult South Africans, 9.6% were engaged in concurrent current tobacco use and problem drinking, 20.3% in problem drinking only and 18.2% in current tobacco use only [1]. Various studies have shown that the concurrent use of alcohol and tobacco is more detrimental to health than each drug on its own [2-4]. There is a lack of studies in Africa investigating problem drinking and daily tobacco use over time in a longitudinal study. Evaluating tobacco use and drinking change is of "importance to epidemiologic studies because it is often the persistence of lifestyle behaviours, such as smoking and drinking, that influences health" [5]. As reviewed by Phaswana-Mafuya et al [1], risk factors for concurrent alcohol and tobacco use have been identified as sociodemographic variables (lower education, male gender, younger age, lower socioeconomic status) and psychosocial (drug use, poor mental health). Urban populations may be at greater risk of daily tobacco use and problem drinking in South Africa [1, 6-8]. Therefore, the aim of this study was to explore the prevalence of daily tobacco use and problem drinking and its determinants among urban dwellers in a longitudinal study in South Africa.

 

 

Sample and procedure: the sample included second wave (N = 2213, in 2012) and third wave (N = 2213, in 2014) of the "South African Panel Study of Small Business and Health, a longitudinal survey in African townships in South Africa that collects data on health, psychology, and entrepreneurship from owners of small businesses and from randomly selected respondents who do not own businesses" [9]. A two stage stratified probability sampling design was used to select the full sample from twenty-eight African dominated enumeration areas in the Tshwane Municipality [9]. The interviews were conducted in the preferred language of the respondent, "using Google Phones with Android 1.5 operating system and interview software Open Data Kit (ODK) Collect version 1.1 [9]. Study participants signed informed consents and the protocol was approved by participating institution's ethics review boards.

 

Measures: Tobacco use was assessed with four questions: 1) Do you currently smoke any tobacco products, such as cigarettes, cigars or pipes? If they answered yes, they were asked 2) Do you currently smoke tobacco products daily? 3) Do you currently use any smokeless tobacco, such as snuff, chewing tobacco, betel? If they answered yes, they were asked 4) Do you currently use smokeless tobacco products daily? [10]. Daily tobacco use included the current daily use of smoking tobacco products and/or current daily use of smokeless tobacco products. Hazardous or harmful alcohol use (or problem drinking) defined with a cut-off score of four for men and three for women on the Alcohol Use Disorder Identification Test (AUDIT)-C [11]. The Cronbach alpha for the AUDIT-C was 0.81 for this sample.

 

Concurrent daily tobacco use and problem drinking: based on the above measures on daily tobacco use and problem drinking, concurrent was defined as using both tobacco daily and problem drinking, not at exactly the same time but within a brief time period. The general health status was measured using SF-12, an instrument with 12 items that reflect eight sub-domains, such as physical functioning (2 items, Cronbach α: 0.81) and mental health (2 items, α: 0.82). For each participant, the SF-12 scoring algorithm generates a Physical health Component Summary (PCS-12) score and a Mental health Component Summary (MCS-12) score, with lower scores indicating higher activity limitations [12]. Psychological distress was assessed with the 10-item "Kessler Psychological Distress scale" (K-10) [13]. (Cronbach alpha 0.78). The participant's feelings of stress is measured by the Perceived Stress Scale (PSS-4) [14]. (Cronbach alpha 0.63). Sociodemographic factors that included gender, age, marital status, education, income and born in current city or not, were also assessed. Body mass index (BMI) was calculated as "weight in kg divided by height in metre squared" [15], using the participant's weight and height that were measured by trained research nurses. A list of ever-diagnosed chronic conditions was also included as control variables and these included migraine headache, lower back pain, hypertension, stomach ulcer, arthritis, heart attack or angina, diabetes, emphysema/bronchitis, asthma, cancer, epilepsy and stroke.

 

Data analysis: data analysis was conducted using STATA software version 15.0 (Stata Corporation, College Station, Texas, USA). Frequencies, means, and standard deviations were calculated to describe the sample. Changes in daily tobacco use, problem drinking, or concurrrent daily tobacco use and problem drinking were calculated and the significance of the difference over time was tested using the Wilcoxon matched pairs signed-ranks test, using a significance level at 10%. The association between sociodemographic factors, health variables and daily tobacco use only, problem drinking only and concurrent daily tobacco use and problem drinking was modelled using logistic regression with generalised estimating equations (GEE) to account for repeated measures. This analysis provides estimates of the average effects over the two surveys accounting for within-person correlation in the two surveys, reporting Odds Ratios and 95% Confidence Intervals. Time dependent variables measured at each survey included daily tobacco use, problem drinking, concurrent daily tobacco use and problem drinking, physical health (PCS), mental health (MCS), psychological distress and perceived stress. Sociodemographic variables (age, gender, income, education, marital status) and body weight status (available only at one time) were also added in the GEE model. Potential multi-collinearity between variables was assessed with variance inflation factors, none of which exceeded the value of 2.0. P < 0.05 was considered significant.

 

 

Sample characteristics and prevalence rates of tobacco and alcohol use: the sample included 2213 participants at Time 1 (mean age 44.7 years, SD = 15.08; range 20-95) and 2213 participants at Time 2 (mean age 46.7 years, SD = 15.08; range 22-97). Almost two thirds (62.5%) were female, 35% were 50 years and older, 36.8% had Grade 12 or more education, 54.1% were single, divorced or widowed and 56.4% had been born in the current city they were living in. More than half (56.3%) were overweight or obese and 48% had one or more chronic condition (Table 1).

 

Tobacco and alcohol use and other health variables over time: daily tobacco use was at time 1 23.99% and at time 2 23.41%, a non-significant decrease of 0.58% (P = 0.323). Problem drinking was at Time 1 19.61% and at time 2 21.1%, an increase of 1.5%, which is significant at the 10% level (z = 1.83 by Wilcoxon signed-ranks test, p = 0.067). Concurrent daily tobacco use and problem drinking has increased from Time 1 9.46% to 10.3% at time 2 (Table 2), an increase of 0.84%, also significant at the 10% level (z = 0.175 by Wilcoxon signed-ranks test, p = 0.080).

 

Associations with daily tobacco use and problem drinking: in longitudinal regression analyses, men's odds of daily tobacco use was 2.99 times that of females, 2.71 for problem drinking, and 6.56 for concurrent daily tobacco use and problem drinking. Daily tobacco use increased with age, while problem drinking decreased with age. Daily tobacco users were less educated, more likely to be currently unmarried and were more likely to have been born in the current city. Overweight or obese urban dwellers were less likely to engage in daily tobacco use, but higher perceived stress was associated with daily tobacco use. Problem drinkers were also more likely to have been born in the current city, but not associated with education level. Better mental health (MCS), psychological distress, physical health (PCS) and the presence of chronic conditions had no influence on daily tobacco use, problem drinking and concurrent daily tobacco use and problem drinking (Table 3).

 

 

The study found at baseline high rates of daily tobacco use only (24.0%), problem drinking only (19.6%) and concurrent daily tobacco use and problem drinking (9.5%) in this urban population in South Africa. These results seem similar to a previous national population-based survey in South Africa, with 18.2% current tobacco users only, 20.3% problem drinkers only and 9.6% concurrent current tobacco users and problem drinkers [1]. In this longitudinal study over one year, compared to Time 1 problem drinking only significantly increased to Time 2, while concurrent daily tobacco use and problem drinking also increased but not significantly. The increase in problem drinking may be attributed to the identified risk factor of having higher income and thus better purchasing power that may have led to increased alcohol consumption. On the other hand, problem drinking only decreased with age, which is consistent with another study in South Africa [1] and a longitudinal study in USA [16].

 

Several risk factors (sociodemographic and health factors but not psychological distress, mental and physical problems as well as chronic conditions) were jointly associated with concurrent daily tobacco use and problem drinking, as well as problem drinking or daily tobacco use. Similar to previous studies [1, 17], we found that male gender was strongly associated with daily tobacco use, problem drinking and concurrent daily tobacco use and problem drinking. While older people were more likely than younger people to use tobacco daily, the pattern was reversed for problem drinking, within younger people being more likely to pursue problem drinking. The prevalence of concurrent daily tobacco use and problem drinking was about 6 times higher among male than female respondents. This finding conforms to previous studies in South Africa [1]. In agreement with other studies [18], a lower education level was associated with daily tobacco use.

 

However, unlike in other studies [3, 4, 17, 18], there was no association between psychological distress, poor physical quality of life and problem drinking and daily tobacco use. We found that problem drinking was highly associated with daily tobacco use and vice versa, similar to other studies [1, 19-22]. The identification of several sociodemographic and health behaviour factors may help in better targeting intervention programmes for this urban dweller population. This study had several limitations. Information in this study was self-reported; hence, subject to bias. Further, the follow-up period in this longitudinal study was only one year and future studies should have follow-up assessments over a longer period.

 

 

We found high prevalences of daily tobacco use, problem drinking and concurrent daily tobacco use and problem drinking among urban dwellers in South Africa. Further research is required for determining appropriate interventions in this study population.

 

 

The authors declare no competing interests.

 

 

Karl Peltzer, Li-Wei Chao, Shandir Ramlagan and Helena Szrek designed and conducted the study and analysis, drafted and revised the paper. All authors read and agreed to the final manuscript.

 

 

This project was supported by the U.S. National Institutes of Health R01- HD-051468 (Chao P.I.), by the European Regional Development Fund through COMPETE 2020 - Competitiveness and Internationalisation Operational Programme (POCI), and by Portuguese public funds through FCT (Fundação para a Ciência e a Tecnologia) in the framework of the projects POCI-01-0145-FEDER- 006890, POCI-01-0145-FEDER-016894, POCI-01-0145-FEDER-016894 (PTDC/IIM-ECO/6808/2014), IF/00994/2013, IF/01721/2012 and CONT_DOUT/14/FEC-UP/459/9716/3/2008. We would like to thank excellent research assistance provided by Ana Rita Farias and Joana Gomes da Costa.

 

 

Table 1: baseline sample characteristics

Table 2: change in tobacco and alcohol use over 12 months (N = 2213)

Table 3: logistic regression model with generalized estimating equations (GEE) for the association between sociodemographic and health variables on daily tobacco use, problem drinking and dual tobacco and alcohol use

 

 

1. Phaswana-Mafuya N, Peltzer K, Pengpid S. Concurrent tobacco use and risky drinking in South Africa: results from the South African National Health and Nutrition Examination Survey. J Psychol Afr. 2018; 28(4): 316-322. Google Scholar


2. Diehl A, Scherbaum N. Nicotine dependence as comorbidity of alcohol dependence-epidemiology, etiology and therapy. Fortschr Neurol Psychiatr. 2008; 76(1): 14-20. PubMed | Google Scholar


3. Subramaniam M, Abdin E, Vaingankar J, Phua AM, Tee J, Chong SA. Prevalence and correlates of alcohol use disorders in the Singapore Mental Health Survey. Addiction. 2012; 107(8): 1443-1452. PubMed | Google Scholar


4. Farrell M, Howes S, Bebbington P et al. Nicotine, alcohol and drug dependence and psychiatric comorbidity-results of a national household survey. Int Rev Psychiatry. 2003; 15(1-2): 50-56. PubMed | Google Scholar


5. Mulder M, Ranchor AV, Sanderman R et al. The stability of lifestyle behavior. Int J Epidemiol. 1998; 27(2): 199-207. PubMed | Google Scholar


6. Williams CT, Grier SA, Marks AS. "Coming to town": the impact of urbanicity, cigarette advertising and network norms on the smoking attitudes of black women in Cape Town, South Africa. J Urban Health. 2008; 85(4): 472-85. PubMed | Google Scholar


7. Lau YK, Tam J, Fleischer NL, Meza R. Neighbourhood deprivation, smoking and race in South Africa: a cross-sectional analysis. Prev Med Rep. 2018; 11: 202-208. PubMed | Google Scholar


8. Oyebode O, Pape UJ, Laverty AA, Lee JT, Bhan N, Millett C. Rural, urban and migrant differences in non-communicable disease risk-factors in middle income countries: a cross-sectional study of WHO-SAGE data. PLoS One. 2015; 10(4): e0122747. PubMed | Google Scholar


9. Chao LW, Szrek H, Peltzer K et al. A Comparison of EPI Sampling, Probability Sampling and Compact Segment Sampling Methods for Micro and Small Enterprises. J Dev Econ. 2012; 98(1): 94-107. PubMed | Google Scholar


10. World Health Organization (WHO). Guidelines for controlling and monitoring the tobacco epidemic. Geneva, Switzerland: WHO. 1998. Google Scholar


11. Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA. The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking, Ambulatory Care Quality Improvement Project (ACQUIP), Alcohol Use Disorders Identification Test. Arch Intern Med. 1998; 158(16): 1789-95. PubMed | Google Scholar


12. Ware JE, Kosinski M, Dewey JE, Gandek B. How to Score and Interpret Single-Item Health Status Measures: a manual for users of the SF-8 Health Survey. Lincoln RI: qualityMetric incorporated. 2001.


13. Kessler RC, Barker PR, Colpe LJ et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry. 2003; 60(2): 184e-189e. PubMed | Google Scholar


14. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health and Soc Behav. 1983; 24(4): 385-396. PubMed | Google Scholar


15. World Health Organization (WHO). Obesity: preventing and managing the global epidemic-report of a WHO consultation. Geneva, Switzerland: WHO. 2000. Google Scholar


16. Eigenbrodt ML, Mosley TH Jr, Hutchinson RG, Watson RL, Chambless LE, Szklo M. Alcohol consumption with age: a cross-sectional and longitudinal study of the AtherosclerosisRisk in Communities (ARIC) study, 1987-1995. Am J Epidemiol. 2001; 153(11): 1102-11. PubMed | Google Scholar


17. Lee JY, Ko YJ, Park SM. Factors associated with current smoking and heavy alcohol consumption among women of reproductive age: the Fourth Korean National Health and Nutrition Examination Survey 2007-2009. Public Health. 2013; 127(5): 473-81. PubMed | Google Scholar


18. Horn K, Gao X, Williams J, Helmkamp J, Furbee M, Manley W. Conjoint smoking and drinking: a case for dual-substance intervention among young emergency department patients. Acad Emerg Med. 2000; 7(10): 1126-1134. PubMed | Google Scholar


19. Manwell LB, Ignaczak M, Czabala JC. Prevalence of tobacco and alcohol use disorders in Polish primary care settings. Eur J Public Health. 2002; 12(2): 139-44. PubMed | Google Scholar


20. Maatoug J, Harrabi I, Hmad S, Belkacem M, Al'absi M, Lando H, Ghannem H. Clustering of risk factors with smoking habits among adults, Sousse, Tunisia. Prev Chronic Dis. 2013; 10: E211. PubMed | Google Scholar


21. Rosal MC, Ockene JK, Hurley TG, Reiff S. Prevalence and co-occurrence of health risk behaviors among high-risk drinkers in a primary care population. Prev Med. 2000; 31(2 Pt 1): 140-7. PubMed | Google Scholar


22. Bonevski B, Regan T, Paul C, Baker AL, Bisquera A. Associations between alcohol, smoking, socioeconomic status and comorbidities: evidence from the 45 and Up Study. Drug Alcohol Rev. 2014; 33(2): 169-76. PubMed | Google Scholar