52 / 62
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 31, No 4, July/August 2020
214
AFRICA
participant had been sitting for five minutes. SBP and DBP were
measured three times at one- to two-minute intervals with the
average of the last two readings taken as the mean clinic reading.
Participants were randomly assigned in a 1:1:1 fashion (to
receive community health worker support, home blood pressure
monitoring or usual care) by simple ballot conducted by health
records officers in each of the two health centres.
Study personnel were instructed to treat participants with SBP
≥
140 mmHg and
<
159 mmHg and/or DBP
≥
90 mmHg and
<
100 mmHg who had not previously been on blood pressure-
lowering therapy with amlodipine 5 mg. All other patients were
recommended to be treated with fixed-dose combinations of
amlodipine 5 mg/ramipril 5 mg or amlodipine 5 mg/losartan
50 mg for those intolerant of ramipril, according to local practice.
Community healthworker support consisted of four structural
education sessions and eight home visits over four weeks for
tailored counselling related to health behaviours, medication
adherence and clinic follow up. Home blood pressure monitoring
included training and provision of an automated home blood
pressure-monitoring device for daily monitoring.
At baseline, we collected data on demographics, medical
and social history, anthropometry and laboratory studies. Four
weeks after randomisation, participants returned to the clinical
site for an evaluation of change in SBP and hypertension control
(co-primary outcomes). Secondary outcomes included self-
reported blood pressure-lowering medication adherence and side
effects.
Statistical analysis
We reported baseline data using means (standard deviation)
or medians (interquartile range) as appropriate for continuous
variables, and proportions for categorical variables. We used
analysis of variance (ANOVA) and Pearson’s chi-squared test
to compare baseline continuous and categorical data. We
calculated the mean change in SBP from baseline to follow up
and compared these results across groups using analysis of
covariance (ANCOVA), adjusting for baseline SBP. We defined
statistical significance as a two-sided
p
<
0.05 and used SAS v9.4
(Cary, North Carolina) for analyses.
Results
Table 1 summarises the baseline data and co-primary and
secondary results. Among the 60 participants recruited, mean
(SD) age was 41 (11), 46 (8) and 42 (7) years in the community
health worker-supported, home blood pressure-monitoring,
and usual-care groups, respectively (
p
=
0.18). Overall, 35% of
participants were male with a higher proportion in the home
blood pressure-monitoring group (65%) compared with other
groups.
Most (75%) participants had been diagnosed with
hypertension for five to 10 years. Baseline mean (SD) SBP
were 159 (11), 151 (13) and 155 (12) mmHg in the community
health worker-supported, home blood pressure-monitoring and
usual-care groups, respectively (
p
=
0.12). Baseline mean (SD)
DBP were 99 (11), 91 (8) and 98 (8) mmHg in the same groups,
respectively (
p
=
0.86).
At the four-week follow up, the mean SBP differences were
–31 (12), –27 (14) and –21 (8) mmHg in the community health
worker-supported, home blood pressure-monitoring and usual-
care groups, respectively (
p
=
0.02). There were no differences
in DBP at the four-week follow up. Only one adverse event
(dizziness) occurred in one participant in the home blood
pressure-monitoring group and no adverse events occurred in
the other groups. Self-reported
use of
two BP-lowering drugs
at the four-week follow up was
higher in the community health
worker-supported (80%) and home blood pressure-monitoring
(70%) groups compared with the usual-care group (65%), but
these differences were not statistically significant (
p
=
0.12).
Discussion
Our study demonstrates that community health worker support
and home blood pressure monitoring are feasible and may
be effective in primary care settings in Nigeria. However,
Table 1. Baseline characteristics and outomes of
participants by intervention group
Parameters
Community
health work-
er support
(
n
=
20)
Home blood
pressure
monitoring
(
n
=
20)
Usual care
(
n
=
20)
p
-value
a
Baseline characteristics
Age, mean (SD), years
42 (11)
46 (8)
42 (7)
0.18
Male,
n
(%)
5 (25)
13 (65)
3 (15)
<
0.01
Height, mean (SD), cm
161 (12)
168 (12)
163 (10)
0.17
Weight, mean (SD), kg
73 (13)
77 (11)
69 (9)
0.06
Duration of hypertension,
n
(%)
0.86
<
5 years
14 (70)
15 (75)
16 (80)
5–10 years
5 (25)
4 (20)
4 (20)
>
10 years
1 (5)
1 (5)
0 (0)
Occupation,
n
(%)
0.04
Caterer
1 (5)
0 (0)
0 (0)
Clergy
0 (0)
1 (5)
0 (0)
Driver
0 (0)
0 (0)
1 (5)
Farming
1 (5)
0 (0)
1 (5)
Housewife
1 (5)
0 (0)
2 (10)
Lecturer
0 (0)
1 (5)
0 (0)
Public servant
1 (5)
11 (55)
3 (15)
Retired
1 (5)
0 (0)
0 (0)
Trader
15 (75)
7 (35)
13 (65)
Baseline medication,
n
(%)
0.29
None
5 (25)
5 (25)
11 (55)
CCB
7 (35)
7 (35)
6 (30)
ACEI
6 (30)
6 (30)
1 (5)
BB
2 (10)
2 (10)
2 (10)
Baseline SBP, mean (SD), mmHg
159 (11)
151 (13)
155 (12)
0.12
Baseline DBP, mean (SD), mmHg
99 (11)
97 (8)
98 (8)
0.86
Follow-up characteristics
Follow-up medication,
n
(%)
0.12
CCB
4 (20)
6 (30)
7 (35)
CCB and ACEI
6 (30)
3 (15)
0 (0)
CCB and ARB
10 (50)
11 (55)
13 (65)
4-week SBP, mean (SD), mmHg
128 (5)
125 (5)
133 (11)
<
0.01
4-week DBP, mean (SD), mmHg
81 (8)
80 (5)
82 (6)
0.51
Decrease in SBP, mean (SD), mmHg
b
31 (12)
27 (14)
21 (8)
0.02
Decrease in DBP, mean (SD), mmHg
b
18 (9)
17 (9)
16 (6)
0.88
Adverse events,
n
(%)
0 (0)
1 (5)
0 (0)
0.36
Non-adherence to study medication,
n
(%)
0 (0)
1 (5)
1 (5)
0.60
CCB, calcium channel blocker; ACEI, angiotensin converting enzyme inhibitor; BB,
beta-blocked; ARB, angiotensin II receptor blocker; DBP, diastolic blood pressure;
SBP, systolic blood pressure; SD standard deviation;
a
ANOVA or chi-squared test;
b
Decrease
=
baseline – 4-week follow up.