13 / 78
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 28, No 6, November/December 2017
AFRICA
351
The exact prevalence rate of medication non-adherence among
cardiac patients in Sudan is not known since there are limited
published studies. A study was conducted at Elshaab Hospital in
Khartoum to determine the adherence to secondary-prevention
medication among 210 patients and it was found to be 66%.
15
Another study was performed among 76 patients with heart failure
admitted to the Sudan Heart Institute in Khartoum, which indicated
that 75% of the respondents were adherent to their medications.
16
These previous studies have used limited study populations
or a small number of patients in single clinical settings and may
not enable meaningful conclusions to be drawn regarding levels
of adherence to cardiovascular medications. This highlights the
need to expand this area of research to include patients attending
multicentre, out-patient cardiovascular clinics and to improve the
quality of such research. Therefore, this study was conducted to
evaluate prevalence, predictors and barriers of non-adherence to
medications among cardiac patients attending the three largest
national referral cardiac centres located in Khartoum State.
Methods
This was a descriptive, quantitative and cross-sectional study
designed to describe the adherence of patients with cardiovascular
diseases to their medications.
Sudan is one of largest countries in Africa (total area of
1 861 484 km
2
) with an estimated population of 36 million people
as of July 2015 (CIA fact book, 2016). It is a federal nation
consisting of 18 states. Khartoum State, the capital of Sudan,
covers an area of 28 165 km
2
and contains almost 20% of the
population, 84% of whom live in urban areas.
This study was conducted between September 2014 and
March 2015 in Khartoum State, Sudan. The study population
consisted of out-patients attending the cardiac clinics in Ahmed
Gasim Cardiac Surgery and Renal Transplantation Centre,
Elshaab Teaching Hospital and Sudan Heart Institute, because
they represent the three largest national referral cardiac centres
located in Khartoum State.
The study was conducted in accordance with the Declaration
of Helsinki and national and institutional standards. Ethical
approval for this study was obtained from the Directorate
of Research, Ministry of Health, Khartoum State. Inclusion
criteria for a patient to enter the study were patients aged 18
years or older diagnosed with cardiovascular disease or its major
risk factor, hypertension, who started using a cardiovascular
medication for a duration of three months or more. Patients who
had psychiatric disorders or cognitive impairment were excluded.
The sample size was determined using PS power and sample
size calculator V.3.05.
17
A sample of 260 patients would be
necessary to determine a 20% difference in proportion between
two groups; for example, male versus female with 90% power
and at 5% significance level. Assuming a response rate of 60%, a
sample size of 433 patients was approached to be included in the
study. The total number of patients selected from each hospital
was proportional to the out-patient population attending the
hospital per year. The patients at each hospital were randomly
selected, using systematic random sampling from the patients’
registration lists.
The content validity of the study questionnaire was
established by a research group at Kuwait University. The
questionnaire was translated into Arabic and subjected to a
process of forward and backward translation. The accuracy and
meaning of the translated versions both forward and backward
were checked, and recommended amendments where necessary
were discussed before being finalised. It was pre-tested for
content, design, readability and comprehension on 16 patients
with cardiovascular diseases, and modifications were made as
necessary so that the questionnaire was simple to understand
and answer, yet gave accurate data.
The final version of the pre-tested questionnaire was
composed of four sections, and it contained both open-ended
and closed questions. The first section included items to provide
information about the sociodemographic characteristics of
the respondents (age, gender, marital status, educational level,
residence and monthly income). Section two consisted of
questions to provide information about the clinical variables
of the study population (type and duration of cardiovascular
disease, and type and duration of medications used by the
patient).
The third section evaluated adherence to medications using
the validated eight-item Morisky medication adherence scale
(MMAS-8).
18
Each item measures a specific medication-taking
behaviour; response categories are yes/no for each item with a
dichotomous response and a five-point Likert response for the
last item (never/rarely, once in a while, sometimes, usually and all
the time). The negative response for each item was coded as one,
except for the item asking if the patient took the medications
yesterday (where a positive response was coded as one). The
total score was calculated by summing the values from all
eight question items. Optimal adherence was defined as having
a MMAS-8 score of greater than six out of a total of eight,
according to the methodology used in previous literature.
19,20
Section four included questions to explore the reasons for not
taking the medications regularly.
Data were collected via structured face-to-face interviews of
the respondents in the waiting rooms of the cardiac clinics using
the pre-tested questionnaire. The interview lasted approximately
15–20 minutes. The selected patients were contacted and given
an explanation about the purposes of the research. They were
assured of confidentiality and gave verbal consent to participate
in the study. Data about clinical variables were checked with the
attending physicians from the patients’ medical records.
Statistical analysis
Data were entered into the Statistical Package for Social Sciences
[IBM SPSS Statistics for Windows, version 23 (IBM Corp,
Armonk, NY, USA)] and descriptive analysis was conducted.
The results were reported as percentage (95% confidence
interval) and mean (standard deviation). Univariate logistic
regression was performed to determine the relationship of each
independent variable with adherence to cardiac medications. All
variables with
p
≤
0.25 in the univariate analysis were included
in the multiple logistic regression analysis to determine the
factors that were independently associated with non-adherence
to cardiac medications. The excluded variables were gender (
p
=
0.38), marital status (
p
=
0.83), residence (
p
=
0.36), hospitals (
p
=
0.57) and duration of medication use (
p
=
0.45). Only the results
of multivariate logistic analysis are reported showing odds ratio
(OR) and 95% confidence interval (CI). Statistical significance
was accepted at
p
<
0.05.