Abstract:
South Africa is experiencing an increasing burden of noncommunicable
diseases (NCDs). There is evidence of co-morbidity
of several NCDs at small geographical areas in the country.
However, the extent to which this applies to joint spatial autocorrections
of NCDs is not known. The objective of this study was to
derive and quantify multivariate spatial autocorrections for NCD-related
mortality in South Africa. The study used mortality
attributable to cerebrovascular, ischaemic heart failure and hypertension
captured by the country’s Department of Home Affairs for
the years 2001, 2007 and 2011. Both univariate and pairwise spatial
clustering measures were derived using observed, empirical
Bayes smoothed and age-adjusted standardised mortality rates.
Cerebrovascular and ischaemic heart co-clustering was significant
for the years 2001 and 2011. Cerebrovascular and hypertension
co-clustering was significant for the years 2007 and 2011, while
hypertension and ischaemic heart co-clustering was significant for
the year 2011. Co-clusters of cerebrovascular-ischaemic heart disease
are the most profound and located in the south-western part
of the country. It was successfully demonstrated that bivariate spatial
autocorrelations can be derived for spatially dependent mortality
rates as exemplified by mortality rates attributed to three cardiovascular
conditions. The identified co-clusters of spatially
dependent health outcomes may be targeted for an integrated
intervention and monitoring programme.
