Epidemiology of human immunodeficiency virus-1 and hepatitis B virus co-infection and risk factors for acquiring these infections in the Fako division of Southwest Cameroon

Sample

The study was made up of a cross-sectional convenience sample of 761 participants
over the age of 18 who attended a free screening in one of five hospitals in the Fako
division of Southwest Cameroon. Age under 18 was the only criteria excluding individuals
from participation. Each hospital was located in one of the five health districts
of the Fako division: Muea (Muea Health centre), Tiko (Tiko Central clinic), Limbe
(Limbe Regional Hospital), Muyuka (Muyuka District Hospital) and Buea (Buea Regional
Hospital). The study was conducted from June 13th to July 6th, 2011. During this period,
data was collected for 5 days from each of the five hospitals. Between 150 and 250
participants took part in the study in each hospital.

Recruitment

Participants were recruited through many means including television, radio, community
informers, and informative posters and pamphlets. We advertised free testing for HIV
and HBV on the local television channel in the area and over the radio 3 days prior
to and during the entire study. Recruitment posters were mounted throughout the towns
and around the health centres where the screenings took place. Information about the
study was also spread by word over a megaphone by town criers who dispensed important
information to the community. Furthermore, announcements were made and flyers handed
out at busy marketplaces.

Screening

Upon arrival at the screening at one of the designated health centres, subjects took
part in pre-test counselling with an experienced HIV counsellor. Each of the counsellors
was recruited from the hospital where the screening was taking place and worked at
the hospital as an HIV counsellor. The counsellor educated the participant on transmission
risk factors and means to avoid contraction of HIV and HBV. The counsellor then administered
a 30-question survey, detailed below, and the patient signed the IRB-approved consent
form. After completing these steps, the participant had 2 mL of blood drawn by a qualified
laboratory technician. The technician then ran the blood samples using rapid HIV and
HBV test kits and results were available for the patient about 60 minutes later. The
laboratory technicians were recruited from the hospital where the screening took place.
Upon receiving results, patients underwent post-test counselling, which was dependent
on the test outcome. HIV-positive patients were directed towards HIV treatment centres
in the area to receive appropriate follow-up care. Those who were HBV-positive were
told to consult a physician for proper care and further viral load testing. Those
who were HIV and HBV negative were given further instructions on means to protect
themselves from HIV and HBV infection in the future.

Consent

This study was approved by the Cameroon National Ethics Committee (CNEC) and administrative
authorisations were obtained from the Southwest Regional Delegation for Public Health
and the health institutions where participants were enrolled. Participants were required
to sign a consent form outlining the study procedures and detailing the risks involved
in the study. For participants who were unable to read or understand the consent,
the counsellor verbally explained the information.

Measures

Questionnaire

The counsellors administered a questionnaire to each participant that included questions
on demographics, past HIV and HBV testing practices, sexual practices and drug use.
Questions on demographics included age, sex, marital status, pregnancy status, religion,
education, career, and income. Questions about HIV and HBV testing included the number
of times the participant had been tested for each disease, the reason for being tested,
the results of the test, and if positive, age of diagnosis and whether the patient
was receiving ART if HIV positive. Questions pertaining to sexual practice included
age of first intercourse, number of sexual partners in the last year and lifetime,
history of sexually transmitted diseases, frequency of condom use and whether the
participant had sexual intercourse with a known HIV-positive person. Finally, the
questionnaire contained additional questions on known risk factors for HIV and HBV
acquisition such as injection drug use and blood transfusions.

Disease measurement

Three test kits were used in this study. The Determine HIV-1/2 test kit, which tests
for the presence of HIV-antibodies in serum was used to screen all participants for
HIV. Those who tested positive on the Determine test kits were confirmed using the
HEXAGON HIV 3rd Generation Immuno-chromatographic Rapid Test for the Detection of
Antibodies to Human Immune Deficiency Viruses 1 and 2. Hepatitis B virus was diagnosed
by screening for the hepatitis B surface antigen (Acon Laboratory) using the DiaSpot
HBsAg: One Step Hepatitis B Surface Antigen Test Strip. Each of these tests kits had
sensitivity 99 % and a specificity 97 %.

Socio-demographic variables

The following socio-demographic variables were included in the analysis: age, sex,
education level, religion, marital status, income level, and blood transfusion status.
The age variable was divided into older and younger age, depending on the median age
of 32. Sex was coded as either male or female. Education level was a binary variable
compromised of low education (completion of no school or primary school) and high
education (completion of secondary school or post-secondary education). Four religion
variables were examined: Catholic and not Catholic, Protestant and not Protestant,
Muslim and not Muslim, and other and not other. Marital status was divided into a
binary variable made up as ever married (married, separated/divorced, and widowed)
and never married (single). Income, which was divided into four levels in the questionnaire
(10,000 CFA, 10–50,000 CFA, 50–100,000 CFA and 100,000 CFA), was a binary variable
in the model divided into high (50,000 CFA/month) and low (?50,000 CFA/month), based
on the median income level of 10,000–50,000 CFA per month. Students and housewives
were excluded from income level analysis as they had incomes of zero, which was not
necessarily representative of their socioeconomic statuses. Blood transfusion status
was divided into simply “yes” to indicate ever having a transfusion and “no” to mean
never having a transfusion.

Sexual behaviour variables

Sexual risk factors examined in the analysis included condom use, lifetime number
of sexual partners, and age at first sexual intercourse. Condom use was a binary variable
divided into low (never, rarely, and sometimes) and high (often, usually). Lifetime
number of sexual partners was a binary variable divided into 0–9 and 10+. Age at first
sex was a binary variable divided into less than 18 years old and greater than 18 years
old, based on the median value of age at first sexual intercourse, 18.

Statistical analysis

The data was analysed using SAS 9.2 looking for the prevalence of HIV, HBV, HIV/HBV
co-infection and a potential association between HIV and HBV. Additionally, we looked
at whether high-risk sexual behaviour variables were associated with each disease.
To conduct final model building, chi-squared tests and logistic regression were used
to investigate associations.

Model building

Chi-squared analysis was used to analyse the crude relationship between each of the
above variables in the questionnaire and HIV status and HBV status. Multivariate logistic
regression models were built based on the results of the Chi-squared test and univariate
logistic regression; only significant variables (p-value 0.05) were included in the final model.