Causes of Death in South Africa 1997- 2001: briefing by Stats SA

Meeting report

SOCIAL SERVICES SELECT COMMITTEE
15 April 2003
CAUSES OF DEATH IN SOUTH AFRICA 1997- 2001: BRIEFING BY STATS SA

Chair: Ms L Jacobus (ANC)

Documents handed out:
Powerpoint Presentation on Causes of Death in South Africa 1997-2001
Causes of Death in South Africa 1997- 2001 (Advance Release of Recorded Causes of Death) -
Statistics South Africa website

SUMMARY
Statistics South Africa briefed the Committee on the leading causes of death in South Africa
between 1997 and 2001. They emphasised that although "unspecified unnatural causes" was still the first leading cause of death, it steadily declined in this period and was offset by a steep rise in mortality due to Tuberculosis, HIV, influenza and pneumonia. The statistics were described in terms of age, sex and population group. Matters discussed included: the high prevalence of HIV related deaths in the female population, the accuracy of the data collected as HIV/AIDS was not a notifiable disease, poverty and malnutrition and the need for further research into the socio-economic variables affecting mortality patterns.

MINUTES:
Ms Jacobus welcomed the delegation from Statistics South Africa (StatsSA). She said that members of the Departments of Health and Home Affairs were present and should feel free to ask any questions or give any input.

Mr P Lehohla (Statistician General, StatsSA) and Mr H Philips (Executive Manager: Demographic Analysis, StatsSA) briefed the Committee. The briefing followed the Powerpoint presentation. The following notes are added for clarification:

Mr Lehohla gave a brief description of the mandate and framework in which StatsSA works (see slides 2-3).

He gave an example of the kind of information that StatsSA researches and collates. He said that the RDP (Reconstruction and Development Programme) had specified that all people should be within a five-kilometre radius of the necessary social services. StatsSA had mapped the actual distribution of social services, which showed many discrepancies in the level of service provision and infrastructure between different geographical areas and populations in South Africa.

Mr Philips gave a brief introduction to the Report on the underlying causes of death in South Africa for the period 1997 to 2001: its history, aim and the methodology undertaken:
- The Report was produced in collaboration with the Departments of Home Affairs and Health and the Medical Research Council.
- 'Underlying cause of death' referred to the cause that started a chain of events that lead to death. For example if a person had diabetes, and due to this, suffered from renal failure and then died of a stroke, the underlying cause of death would be diabetes.
- The quality of information gathered from the death certificates was dependent on the accuracy with which the medical doctors completed the death certificates.
- "Unspecified unnatural causes" referred to violent deaths, for example: car crashes, suicide, murder, drowning etc.
- They were grouping many causes of death together, and that further research needed to be carried out to disentangle the variables so as to address specific problems such as violence against women.
- "Ill-defined causes" refer to cases where the cause of death is not clear, for example sudden infant death.
- Unspecified unnatural causes, ill-defined causes, Tuberculosis (TB), HIV and lastly, influenza and pneumonia (one class / group) are the five leading causes of death in South Africa and account for 41% of all deaths in the country.
- Some of the TB and influenza deaths, may in fact be caused by HIV (and were therefore opportunistic diseases), but were noted as TB or Influenza on the death certificates by the doctor. Therefore the prevalence of deaths caused by HIV may in fact be much higher than stated.
- The pattern of causes of death has changed between 1997 and 2001. By 2001, HIV and TB were the leading causes of death.
- Differences in causes of death between males and females: HIV is the leading cause of death for women and women are far more likely to die of HIV/AIDS than men. Men are more likely to die of TB. He noted that the five leading causes for men where different to those for women. For example, one of the five leading causes of death for women was cerebrovascular diseases (brain diseases), which is not the case for men.
- "Unnatural unspecified" was still the leading cause of death amongst men, but one should be more worried about the fact that there had been a 95% increase in deaths due to HIV for males between 1997 and 2001.
- In terms of age, the leading cause of death for people below the age of fifteen years, was intestinal infections, but that there had been an increase in the number of deaths due to HIV and influenza and pneumonia. Malnutrition was the seventh leading cause of death for those below the age of fifteen.
- For females, HIV was the leading cause of death in the age groups: 15-29 years (22.5% of deaths) and 30-39 years. TB was the second leading cause of death for these age groups amongst women. Another cause for concern was that diabetes was the fifth leading cause of death for women.
- In terms of population / 'race' group. Coloured and African males die mainly from unspecified unnatural causes and TB. White and Indian males die mainly from heart diseases. The leading cause of death for Coloured women was cerebrovascular diseases, followed by unspecified unnatural causes. African women are ten times more likely to die from HIV than Indian or White women.

Mr Philips said that one of StatsSA's biggest challenges in continuing these studies was enhancing interdepartmental collaboration and infrastructure for registration. Mr Lehohla gave some examples of the need for interdepartmental collaboration in answering questions related to registration of births, deaths and others with Home Affairs.

Mr Philips concluded by saying that there have been changes in the underlying morbidity patterns over time with differentials due to population group, age and sex. Although there has been a decline in unspecified unnatural causes of death, this has been offset by an increase in HIV, TB and influenza as a cause of death, which will have policy implications.

Discussion
Ms Jacobus asked why there was a separation between HIV and TB, influenza etc, when most instances of TB, influenza where opportunistic diseases caused by HIV. She commented that there was no mention of meningitis, which surprised her, as it was also a disease people often died of after contracting HIV.

Ms Jacobus asked if there had been any analysis of the economic status of people dying of HIV, TB etc. Lastly Ms Jacobus asked why more women than men died of HIV.

Mr J Thagale (UCDP) noted that some insurance companies do not pay out insurance to people who die of AIDS. He had heard that some people negotiated with their doctor to state TB rather than HIV/AIDS as the cause of death. To what extent are the statistics accurate given this influence?

Ms N Kondlo (ANC) asked if the reason more women were dying from AIDS was that women were a majority in South Africa or that the main form of transmission of HIV was via sexual intercourse. Was it true that those who died of opportunists diseases such as TB and influenza were not all HIV positive?

Mr Philips replied that the studies were based on the death certificates they received from the Department of Home Affairs. If the attending physician or nurse stated X, then that is what was recorded for the study. It was a long process from the death of a person to the registration of it by the doctor to having it recorded by Home Affairs.
Due to the relationship between TB and HIV, the prevalence of HIV could be higher than recorded. They acknowledged this but HIV was not a notifiable disease, in other words the medical doctor was under no obligation to state that HIV had started the chain of medical events that led to death.

Regarding conducting research into economic status, Mr Philips said that they were looking into this, as well as looking at other socio-economic variables relating to mortality in general.

Mr Philips said that he was not fully equipped to say why more women die of HIV/AIDS than men, but that it had something to do with female biology making women more prone to contracting HIV than men. He agreed that more research needed to be done to disentangle the reasons for the difference.

Mr Philips said that not all of those who die of TB or influenza have HIV.

Mr Lehohla said that if a doctor became sympathetic to a person's call for the doctor not to state HIV as the cause of death, it leads to absurdities in the data.

Mr H Sogoni (UDM) said that from his experience and observation, reporting of deaths in the rural areas was often low and if a person died at home and no claims needed to be made, the death was not reported to Home Affairs. Did StatsSA had any indication of the rate of reporting of deaths in the rural areas?

A committee member asked if in the future they would get statistics on the number of deaths caused by cancer.

Ms B Sono (DA) asked if the sample for the study had been taken from the Eastern Cape or nationally. If the sample had been taken from the whole country it was problematic in light of Mr Sogoni's question, as the level of infrastructure and pace of development was not the same in all provinces.

Mr D Kgware (ANC) referred to the causes of death for different population groups, and asked if differences in lifestyle had something to do with the higher prevalence of HIV related deaths in the African population and lower prevalence in the White population.

Ms E Gouws (DA) asked if they could say from their analysis what role poverty and malnutrition played in the prevalence of HIV deaths.

Mr Lehohla replied that they were aware that death notification was not complete nation wide and that female deaths were even less reported due to the fact that the issue of succession and property plays less of a role than with men.

Mr Lehohla noted that the slides showing social services for the Eastern Cape, was intended to illustrate a point, but that the sample was national. StatsSA would soon be mapping the distribution of the causes of death for the whole country.

Responding to Ms Gouws's question on poverty, Mr Lehohla said that their data was limited to that shown on the death certificate i.e. age, sex and population group and what the person died of. The next Demography and Health Survey will attempt to look at HIV prevalence and socio-economic variables and the risks different people are exposed to.

Mr Philips said that details on deaths caused by cancer could be found on page 63 of the report: Causes of Death in South Africa 1997- 2000. Page 32 and 33 showed an example of what the death certificate looked like and the information they could glean from it.

Mr Philips said that poverty was linked to the causes of death and that malnutrition was the seventh leading cause of death for children below fifteen years of age. He added that more research needed to be done in this area.

A Department of Home Affairs official pointed out that the death certificate shown on page 33 of the report was not the normal one, but one used when citizens died in a foreign country.

Mr Philips said that lifestyle did play a role in the different patterns of death between different populations, for example diabetes was partially linked to eating habits. He added that they should start to merge their data sets to answer these kinds of questions.

Ms Kondlo said that the study was carried out using data from 1997. Now that it was 2003 she wondered whether programmes on education and prevention of HIV had changed the statistics positively or whether the country was getting deeper and deeper into the HIV/AIDS problem.

Mr Lehohla said that it was difficult to answer this question and he did not want to draw premature conclusions. The kind of comprehensive research being undertaken by StatsSA was just starting (and thus fraught with issues) and unlike other countries such as India for example, they did not yet have years of research with which to assess patterns.

Ms Jacobus said that it would be very useful to conduct further research into the particular forms that "unspecified unnatural causes" took in order to address problems such as violence against women and children.

She noted that there had been whispers in the Department of Health that HIV should be made a notifiable disease and that members of civil society had objected as they saw it as an infringement of human rights. The government would soon be placed in a difficult position as these statistics had to be gained in order to assess the size of the problem. She asked when the next report was due to be released.

Mr Lehohla responded that a report on statistics from 2001-2002 would be released towards the end of the year.

Ms Jacobus thanked the StatsSA delegation and the meeting was adjourned.