CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 9/10, October/November 2013
AFRICA
371
transition’ is now commonly used to refer to changes in the diet
that occur with urbanisation.
Recent data also suggest however that the nutrition transition
is occurring within rural areas, possibly explaining the increasing
prevalence of obesity in less developed settings.
40-42
Another
major contributing factor to the high prevalence of obesity in
urban versus rural communities in SA and SSA is the increase in
physical inactivity and the adoption of a more sedentary lifestyle
with urbanisation.
30,31,43
Maternal and early life factors
Nutrient deprivation and the timing thereof during the intra-
uterine period leads to foetal programming, resulting in genetic
and epigenetic adaptations.
44,45
These biological adaptations
predispose an individual to obesity when exposed later in life
to an environment abundant with energy-dense and/or high-fat
foods, as is currently experienced in middle-income countries
such as SA.
46
The prevalence of low birth weight (
<
2.5 kg), very often
the consequence of nutrient deprivation in utero, is 15% in SA,
which is marginally higher than the overall prevalence of 13%
in SSA.
47
The COHORTS initiative, a study of birth cohorts in
five low- or middle-income countries including SA, has shown
that size at birth is linked to major features of the metabolic
syndrome in adulthood, including obesity.
48,49
However the
relationship between pre-natal exposure and obesity in later life
has been shown to fit a U-shaped curve. More specifically, low
birth weight has been associated with increased levels of adult
abdominal adiposity, while high birth weight was associated with
overall adult adiposity.
50,51
High birth weight has been shown to be a result of excessive
maternal body weight or excessive weight gain during
pregnancy.
52,53
This is of concern in SA, given the high prevalence
of obesity in SA adolescents and adult women. It is compounded
further by healthcare inequalities, associated perceptions of the
healthcare system, and the periodic lack of adequate resources
that have led to late or poor attendance rates at antenatal clinics.
54,55
Under-nutrition during the first six months of life increases
the risk of stunting. Global statistics indicate that in SSA, the
prevalence of stunting under the age of five years is 39%, with
stunting rates ranging from 27% in Ghana to 55% in Niger,
and SA reporting a stunting prevalence of 24%.
47
In transitional
societies of SSA, stunting and adolescent obesity may co-exist in
the same geographic population.
56
A cross-sectional growth survey conducted in rural SA
children and adolescents aged one to 20 years showed that
an estimated one in five children aged one to four years was
stunted. Concurrently, the prevalence of combined overweight/
obesity was 20–25% among girls in late adolescence.
56
Steyn
et
al
. showed that stunting in children under the age of nine years
resulted in a 1.8-fold increased risk of obesity.
57
Moreover, other
evidence suggests that individuals who were stunted as children
were more likely to be overweight as adults.
58
Furthermore,
excessive weight gain during childhood was associated with
adult body composition.
59
Physical activity
Physical activity may be defined as any bodily movement
produced by skeletal muscle that requires energy expenditure.
60
Prior to the early 2000s, the evidence base for physical activity
and health in SSA was limited, fragmented and localised,
with few nationally representative samples. Self-report physical
activity questionnaires were not standardised, often not validated
in the populations in which they were being applied, and the
focus was primarily on energy balance and seasonal agriculture-
related physical activity and under-nutrition.
Recent WHO Stepwise surveillance initiatives, using
a common instrument called the Global Physical Activity
Questionnaire (GPAQ), have yielded a growing body of evidence
on the global trends in physical activity and inactivity.
61
The
physical activity recommendations for health in adults are
defined as engaging in at least 150 minutes of moderate-intensity
activity per week, or 75 minutes of vigorous-intensity activity
per week, or an equivalent combination of moderate- and
vigorous-intensity activity.
62
Physical inactivity has been defined
as ‘doing no or very little at work, at home, for transport or
during discretionary time’.
63
In the African region, estimates of the prevalence of inactivity
are widely varying, ranging from as low as 3.8 and 1.5% in
women and men in the Comoros, to 15 and 9% in Ghanaian
women and men, and 48 and 45% in SA women and men,
respectively.
61,63,64
The highest reported prevalences of inactivity
in this region are similar in magnitude to those seen in North
America, and higher than those reported in South America,
Western Pacific or Asia.
10
It appears that the inactivity gradient and obesity seem to be
related to development within the region and within the country.
An ecological evaluation of inactivity in women and men in 13
SSA countries demonstrated a significant correlation between
gross national income (per capita) and prevalence of self-
reported inactivity.
65
Sobngwi
et al
.
66
studied over 1 600 Cameroonian adults living
in either rural or urban settings and found that lowered the odds
for overweight and obesity in a dose-dependent manner, and that
the odds for overweight and obesity, as well as impaired glucose
tolerance, were significantly increased with increased lifetime
exposure to an urban environment (percentage of life in a city).
Conversely, in SA, results from the THUSA study showed
that among a group of black adult women, physical inactivity
was a stronger correlate of obesity than socio-economic status
and dietary factors.
31
As physical activity has been identified
as playing a key role in influencing health outcomes, even in
communities undergoing epidemiological transition, trends in
physical activity behaviour have implications for public health
and the emerging burden of NCDs in the region.
Armstrong and Bull
67
highlighted that in developing countries,
‘occupational-, domestic- and transport-related activities may
contribute more to overall energy expenditure than leisure-time
or recreational activity’, and therefore a multi-domain approach
to the measurement of physical activity is essential. A recent
study including data from 22 African countries showed a higher
proportion of adult men compared to women (84 vs 76%) meeting
the global physical activity recommendations.
64
Although levels
of physical activity varied greatly across these countries and
population sub-groups, the study found that leisure time activity
(5%) was consistently low, irrespective of gender, whereas work
activity (moderate and vigorous combined) contributed the most
(49%) to total physical activity time, followed by transport-
related activity (46%).
64
