CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 5, June 2013
AFRICA
181
(biochemistry) investigations. Data presented in this publication
are related only to hypertension.
A list of the districts in the city was used for sampling.
Each district was divided into squares and each square was
subdivided into concessions (a group of households). A list of
all concessions was obtained from the regional statistics office.
This list was used as a sampling frame for the random selection
of squares.
In each square, concessions to be visited were randomly
selected and inside the concession, a household was also
randomly selected. In each household, all the persons matching
the selection criteria were invited to participate in the study. One
hundred and twenty households were randomly selected, giving a
total of 1 424 participants; 32 sets of data were not been analysed
because of missing biological and/or clinical data.
Eligible criteria were age
≥
15 years and being a resident
of Saint Louis. Formal written consent was obtained.
Non-consenting patients and pregnant women were not included.
Participants were involved in the survey for one day. Those
with abnormal physical or laboratory findings were counselled
and referred to the regional hospital as defined by the National
Health reference system. Interviews, body measurements and
laboratory tests were performed by nurses and clinical officers.
The survey questionnaire consisted of socio-demographic
(age, gender, education, marital status), lifestyle variables (fruit
and legume consumption, exposure to tobacco and alcohol, and
physical activity), and medical and health history.
Physical body measurements included blood pressure
(BP), height, weight, and waist circumference. Blood pressure
measurements were taken using an electronic digital blood
pressure machine (OMRON
®
M6). Three BP measurements were
performed on both arms, in a seated position, legs uncrossed, after
a five- to 10-minute rest. The highest BP value was recorded.
Waist circumference was measured in centimeters using a tape
measure, and the measurement was made at the mid-axillary line,
midway between the last rib and the superior iliac crest. Height
was measured with the participant standing upright against a wall
on which a height mark was made. Weight measurements were
taken on a pre-calibrated weighing scale (Seca 750). Participants
were weighed dressed in light clothing and barefoot.
Blood samples were analysed in a single laboratory using
an automate Reflotron- Plus
®
. Cholesterol [total, high-density
lipoprotein (HDL) and low-density lipoprotein (LDL)],
triglyceride, fasting blood glucose, uric acid and creatinine levels
were analysed.
Hypertension was defined as a systolic BP
≥
140 mmHg or
a diastolic BP
≥
90 mmHg, or a documented medical history of
antihypertensive treatment.
7
Obesity was defined as body mass
index (BMI)
≥
30.0 kg/m
2
, and overweight by a BMI
>
25 but
<
30 kg/m
2
.
Diabetes mellitus was defined as two fasting blood glucose
levels
>
1.26 g/l and/or a documented medical history of diabetes
or diabetes treatment. The threshold for normal values were
<
2 g/l for total cholesterol,
<
1.6 g/l for LDL cholesterol,
>
0.4
g/l for HDL cholesterol, and
<
1.5 g/l for fasting triglycerides.
Physical inactivity was defined as the absence of daily
physical activity or the presence of physical activity lasting
less at 150 minutes per week. Abdominal obesity was defined
according to NCEP, with a waist circumference greater than 102
cm in men and 88 cm in women.
Ethics committee approval to undertake the survey was in
accordance with national and local regulations. Written, signed
consent was obtained for each of the patients included. The study
was conducted in accordance with the Helsinki II Declaration.
Statistical analysis
Data recorded in the standard questionnaire were double checked
by external monitor and double-entered using Epi Data software.
Entered data were cleaned and analysed by an experienced
biostatistician using Epi info version 3.5.1 software.
Binary variables were described by their proportion and
continuous variables by means and standard deviation (SD).
Pearson and Yates (when appropriate) chi-square test were used
for the comparison of qualitative variables and Student’s
t-
test
for the comparison of quantitative variables between groups.
A logistic regression model was built with variables associated
with hypertension. Age and gender were forced into the final
model. The results were statistically significant if
p
<
0.05.
Results
We recruited 1 424 participants (983 female, 69%). Mean age
was 43.4 years (SD: 17.8), (range 15–96 years); 70.8% were
<
55
years and 87.5% were
<
65 years. Fig. 1 shows the distribution
of the population by age. Table 1 shows the characteristics of the
enrolled population and Table 2 shows the prevalence of various
cardiovascular risk factors.
Six hundred and fifty-five participants had HTN, giving a
prevalence of 46.0% (95% confidence interval: 43.4–48.6%).
Among these 655 cases, 327 (50%) were previously undiagnosed.
HTN was more frequent in females [47.9% (44.8–51.1%)] than
in males [41.7% (37.1–46.5%),
p
=
0.015, OR
=
1.29 (1.02–
1.62)]. The mean age was significantly higher in the hypertensive
participants (53.6, SD: 15.8 years) than in non-hypertensive
participants (34.7 years, SD: 14.5,
p
<
0.001). The prevalence of
HTN increased with age (
p
=
0.001) (Fig. 2). Mean duration of
HTN was 6.9 years (range: 1 month – 60 years).
Among HTN participants, mean systolic BP was 136 mmHg
and mean diastolic BP 88 mmHg. Grade 1 HTN was more
frequent (48%) than grade II (25%) and grade III (27%). HTN
tended to be more frequent in participants who had primary school
level education (42.1%) than in those who had higher levels of
education (28.4%,
p
=
0.18). Table 3 shows the distribution of
hypertension according to socio-professional category. There
Fig. 1. Distribution of study population by age (
n
=
1 424).
15–24
25–34 35–44 45–54 55–64 65–74
≥
75
Years
18.8%
17.3% 16.4%
18.3%
16.7%
7.4%
5.1%
