Page 164 - HIVMED_v21_i1.indb

Page 2 of 8 Original Research

smoking prevalence in PLWHA is approximately twofold to upright position. Weight to the nearest gram was measured
threefold higher than in the general population in developed using Analog and Digital (A&D) Medical PersonalScale (Model
countries and ranges from 40% to 74%. 6,7 UC-321, Japan) with the participant in light clothes and without
shoes. Biochemical analyses included serum cotinine levels and
The greater adverse effects of tobacco smoking in PLWHA lipid profiles, which were acquired at an ISO 15189-accredited
are highly relevant to SA. The country has the greatest burden pathology laboratory (Path Care, Reference Laboratory, Cape
of HIV worldwide with approximately 7.97 million PLWHA Town, SA), as previously described in detail. 15
in 2019. With almost 20% of 15–49-year-old South African
8
adults being HIV positive, determining the burden of tobacco Definitions
smoking in the HIV infected can inform strategies for tobacco
cessation in this high-risk population. This is particularly Participants were classified as either ‘current smoker’, ‘former
pertinent in the era of widespread dissemination of smoker’ or ‘never a smoker’, considering all forms of smoked
9
antiretroviral therapy (ART) in SA and increased longevity tobacco, including cigarettes, cigars or pipes. Current smokers
of the HIV-infected population. People living with HIV and included participants who smoked daily or occasionally.
AIDS are now at an increased risk of dying from cardiovascular Former smokers refer to participants who indicated that they
and other non-communicable diseases (NCDs), including had quit smoking at the time of the interview, regardless of
tobacco-related conditions, rather than from AIDS. 10,11,12 the duration since quitting. ‘Smokeless tobacco users’ referred
to the use of chewing tobacco, snuff or betel leaf and the areca
This study, therefore, aims to determine the prevalence of nut at the time of the survey. Exposure to second-hand smoke
smoking and associated factors including HIV-specific was determined from ‘household smoke’. Cotinine, a major
factors in PLWHA receiving ART at public healthcare facilities metabolite of nicotine, is commonly used as a biomarker to
in the Western Cape province of SA. identify exposure to tobacco. 16,17 Serum cotinine levels were
used to define the different smoking categories as follows: ‘no
Methods tobacco exposure’: cotinine <10 mg/mL, ‘environmental smoke
Population and sampling exposure or light smoking’: cotinine levels of 10 ng/mL –
100 ng/mL and ‘moderate to heavy smoking’: cotinine
This cross-sectional study was conducted in a sample of >100 mg/mL, in line with the 2012 South African National
≥ 18-year-old HIV-infected adults who were randomly Health and Nutrition Examination Survey (SANHANES). 18
selected from a list of patients attending the clinic on the
study day. Participants were recruited between March 2014 Alcohol use was defined as drinking at least one standard
and February 2015 from healthcare facilities in the Western alcoholic drink per day. A standard alcoholic drink consists
Cape that provided ART to at least 325 HIV-infected patients of a can (340 mL) of beer, one glass (125 mL) wine or ‘one-
per month. This was to ensure the recruitment of an adequate shot’ (25 mL) of spirits. Body mass index (BMI) was calculated
number of participants within a reasonable period. Of the 17 as weight in kilograms divided by height in metres squared
2
healthcare facilities selected, 10 were in Cape Town and (kg/m ), and overweight and obesity was defined as BMI ≥
seven were in the surrounding rural municipalities. Excluded 25 kg/m . Cut-points for HIV-related variables were set at
2 19
participants were those who were pregnant, breastfeeding, median values, that is, ≥ 396 cell/mm for CD4 counts and of
3
+
bedridden, undergoing cancer treatment, on corticosteroid ≥ 5 years duration of HIV diagnosis.
treatment, or unwilling or unable to provide consent. The
detailed methods have been described previously. 13 Statistical analysis
The Statistical Package for Social Sciences (IBM SPSS Inc,
Data collection
Chicago, IL, USA) V.25.0 software was used for the data
Trained clinicians, nurses and fieldworkers collected data analyses. Continuous variables are presented as means
via standardised international questionnaires, clinical (± standard deviation [SD]) or medians (25th – 75th
measurements and biochemical analyses. Data were captured percentiles) and categorical variables are presented as counts
on personal digital assistants (PDAs), using electronic case and percentages. Analysis of variance (ANOVA), χ tests and
2
report forms with built-in checks for quality control. The non-parametric equivalents were used as appropriate for
interviews and physical examinations were conducted on the group comparisons. Logistic regression models adjusted for
recruitment day, and following an overnight fast, participants age and gender were used to determine associations with
returned the next day to have their blood samples taken. current smoking. A p-value < 0.05 defined statistically
significant results.
Participants provided their socio-demographic history,
including tobacco use, which was adapted from the World Approval to conduct the study
Health Organization’s (WHO) STEPwise approach to
14
Surveillance (STEPS) tool. Information on the duration of Permission to conduct the survey was obtained from the
diagnosed HIV infection, cluster of differentiation 4 (CD4) Health Research Office of the Western Cape Department of
counts and ART regimens was extracted from clinical records. Health and the relevant healthcare facilities. The study was
Height to the nearest millimetre was measured using a Leicester approved by the South African Medical Research Council
Height Scale (Seca, UK) with the participant barefoot and in the Ethics Committee and conducted in accordance with the

http://www.sajhivmed.org.za 156 Open Access