As the temperatures soar and weather patterns continue to be affected by global warming and climate dynamics, more people – particularly in developing countries – are likely to die from natural deaths such as respiratory disease, cardiovascular disease and stroke deaths.
The logic behind the study
Although many temperature studies have been conducted in disciplines such as climatology, it is only in the last decade that they have received greater attention in public health and epidemiology fields. Epidemiology is the study of the distribution of diseases and the determinants of the diseases.
Epidemiological studies are crucial to understanding the impact that climate change could have on human health. They help guide how adaption strategies could be developed. Very few of these studies have been conducted in Africa. There is a need for them in developing countries because people have different vulnerabilities compared with those living in North America and Western Europe.
For example, building characteristics are different and peoples' activities differ. The heat affects people living in corrugated iron structures (shacks) in informal settlements differently. South African census data shows there are 1.2 million households with 3.3 million people living in informal corrugated structures which are not on the same property as formal brick houses.
In addition, people living with chronic illnesses such as diabetes, HIV or TB, or those suffering from diarrhoea, are more vulnerable to heat effects. There are more than six million people living with HIV in South Africa – the highest burden of any country.
How we did it
The study applied complicated statistical models. Death data from Statistics South Africa and data from WeatherSA were used to see if apparent temperature, which is a combination of temperature and humidity, increased the number of natural deaths in three of South Africa’s major cities: Cape Town, Durban and Johannesburg. Data from 2006 to 2010 were used.
The study looked at the effect of the heat when the apparent temperature went above a threshold in each of the cities. The city-specific threshold was determined in the statistical models. This is 15℃ in Cape Town, 20℃ in Durban and and 13℃ in Johannesburg.
The world is divided in various climatic zones. Cape Town, Durban and Johannesburg are classified in three different zones:
Cape Town has a Mediterranean climate;
Durban has a humid subtropical climate, which borders a tropical wet and dry climate (Aw); and
Johannesburg has a subtropical highland climate (Cwb).
In Cape Town we analysed the natural deaths when the apparent temperature was between 15℃ and 27℃. We found there was a 1% increase in natural deaths across all ages each time the apparent temperature increased by one degree above the threshold of 15℃. For those over the age of 60, there was a 2% increase in natural deaths.
In Durban, where natural deaths that happened between an apparent temperature of 20℃ and 31℃ were considered, there was a 1% increase in natural deaths across all ages for each one degree increase in apparent temperature above the threshold of 20℃. For those over 60, there was a 1.4% increase while children up to four-years-old saw a 1.9% increase in natural deaths for each one degree increase in apparent temperature above the threshold.
In Johannesburg, where there were 94 900 natural deaths in total during 2006 to 2010, there was a 0.5% increase in natural deaths for all ages and 1.2% increase in natural deaths for people over the age of 60 for each one degree increase in apparent temperature above the threshold of 13℃. The apparent temperature varied from 13℃ to 24℃.
Analysing the data shows that as it gets hotter, more people are dying from natural deaths at lower apparent temperature thresholds compared to other global cities with similar climate. This means that South Africa is more vulnerable to heat effects than other developed countries.
Our next step is to expand our study to include four other South African cities located in different climatic zones. And our plan is to look at specific causes of death, such as respiratory and cardiovascular disease, to see if there is a link to heat.