Cardiovascular Journal of Africa: Vol 29 No 6 (November/December 2018)

CARDIOVASCULAR JOURNAL OF AFRICA • Volume 29, No 6, November/December 2018

AFRICA

345

Methods

This cross-sectional study was conducted in a large community

health centre (CHC) south of Johannesburg. At the time of the

study, from March to May 2012, the CHC provided ambulatory

curative, preventative and rehabilitation services, and served

as a referral centre for five smaller clinics. In the out-patient

department (OPD), there were three doctors, three PHC nurse

clinicians and five staff nurses.

In this facility, patients with hypertension are booked by

appointment for their monthly follow-up visits. A day before the

appointment, an administrative clerk retrieves patients’ medical

records from the archives for all patients booked. On the day of

the clinic visit, patients collect their files and proceed to sit in the

waiting hall, from where a nurse invites a group of 10 patients

for measurement of vital signs at the nurses’ station. Thereafter,

the patients are directed to a doctor or PHC nurse clinician,

depending on the complexity of the presenting problem.

In 2011, an estimated 1 974 patients with hypertension attended

this CHC. Assuming an annual increase of 10%, the estimated

patients with hypertension for the year 2012 was rounded off

to be 2 100. To determine the sample size, the researchers used

the Raosoft sample size calculator.

The required sample size

was calculated to be 328, based on a 5% margin of error, 95%

confidence level and a response distribution of 50%.

Patients were recruited into the study by the first author

and two staff nurses who were trained as research assistants.

During the sampling, a research assistant approached patients

as they presented for measurement of vital signs at the nurses’

station. Every third patient with hypertension was sampled

for recruitment until the sample size was attained. If the third

patient refused, the next willing patient (fourth or fifth) who met

the inclusion criteria was approached and recruited.

To be selected, a patient had to consent to participate in

the study, be 18 years or older and have attended the CHC for

at least two months of hypertension treatment. Patients who

participated in the pilot study, emergencies, those with mental

impairment and those who presented after hours were excluded.

Patients willing to participate in the study were taken to an

adjacent room where the researcher and the second research

assistant obtained informed written consent.

The following CV risk factors were considered in this study,

informed by relevance to PHC and financial costs: advancing

age, gender, obesity, left ventricular hypertrophy (LVH) on

electrocardiograph (ECG), tobacco and alcohol use, diabetes,

physical inactivity, hypercholesterolaemia, family history of

hypercholesterolaemia and past family history of fatal CV events.

After obtaining informed consent, the first author and

second research assistant (when interpretation was necessary)

administered the questionnaire, performed anthropometric and

clinical measurements, and performed ECGs in an adjacent room.

A structured researcher-administered questionnaire was

developed

de novo

in English, based on the literature. Each

questionnaire had a serial number identical to the corresponding

participant’s file number, in case of need for retrieval. Personal

identifying information was not collected. The questionnaire

collected information on:

•

Sociodemography: age, gender, marital status, educational

attainment, employment status and ethnic group, tobacco use

(cigarette smoking, exposure to second-hand smoke and snuff

use) and alcohol consumption. Current smoking was defined

as self-reported active smoking within the last year before

examination.

Current alcohol use was defined as a patient

who regularly consumed alcohol (regardless of the type) in

the past 12 months.

•

Clinical co-morbidity: self-reports of diabetes mellitus, family

history of hypercholesterolaemia and family history of fatal

CV events (defined as death of a close family member due to

heart attack or stroke before the age of 55 years in men or 65

years in women).

•

Engagement in physical activity: level of involvement and

duration of time spent on physical activity, whether at the

workplace, during leisure time or as household chores.

Engagement in physical activity was defined as reporting

moderate-intensity activities in a usual week for ≥ 30 minutes

per day, ≥ five days per week; or vigorous-intensity activities

in a usual week, ≥ 20 minutes per day, ≥ three days per week or

both.

The following examples of physical activities were used

to categorise participants. Moderate-intensity physical activ-

ity: brisk walking, dancing, gardening, housework, domestic

chores, walking domestic animals, active involvement in

games and sport with children. Vigorous-intensity physi-

cal activity: running, walking up hill, fast cycling, aerobics,

competitive sport and games such as soccer, hockey basket-

ball, fast swimming and heavy shovelling or digging.

Anthropometric

measurements

included

abdominal

circumference, weight, height and body mass index (BMI). These

were done using equipment and procedures as recommended

in the South African hypertension guideline of 2011.

Each

instrument was calibrated before use and research assistants

were trained to minimise inter-rater differences. Measurements

were done using:

•

Weight and height: Seca

scale, manufactured by Seca

CE in Germany (model: 767-1321004, serial number:

1767002071575). Heights and weights were measured in

metres and kilograms to two decimal points. The research

assistant used the technique recommended by de Onis

et al

for weight and height measurements. Thereafter, the research-

er computed the BMI in kg/m

. Obesity was considered if

BMI was > 30 kg/m².

•

Waist circumference: flexible, non-elastic metric tape. The

research assistant used a flexible non-elastic tape measure

154 cm long and 15 mm wide, with 1-cm graduation intervals

for the waist circumference measurement. The measure-

ment was recorded at the end of a normal expiration at the

midpoint between the lower margin of the last palpable rib

and the top of the iliac crest on the mid-axillary line, with the

tape parallel to the floor at the level at which the measurement

was done.

Each measurement was repeated twice and if the

measurements came within 1 cm of one another, the average

was calculated. If there was > 1 cm difference, the two meas-

urements were repeated. Central obesity was present if waist

circumference was ≥ 102 cm in men and ≥ 88 cm in women.

In the Asian population, central obesity was considered in

men if the abdominal circumference was ≥ 90 cm and in

women if the waist circumference was ≥ 80 cm.

Clinical measurements were done including:

•

BP: automated Dinamap

, General Electric Medical Systems

(model: DPC321N-EN, item number: 2019194-001). After

resting for five minutes, the BP was measured according to

the method described in the JNC VII and South African