Letter from South Africa
http://www.redflagsweekly.com/allen/2005_feb27.html
HIV/AIDS facts from the epicentre of an assumed pandemic
By Red Flags Columnist, Anita Allen
Statistics dominated the headlines on the Great Aids Debate in the
past week due to the release of the Mortality and Causes of Deaths
in South Africa 1997 - 2002 by Statistics SA (1)
Key findings of the report based on information from 2,88 million death
certificates classified according to the World Health Organisation's
International Classification of Diseases, revision 10 (ICD-10) were:
The number of reported deaths from all causes (natural and unnatural)
increased 57% from 318 287 in 1997 to 499 268 in 2002, which was
partly attributed to population growth of 10% and improved death
notification from an estimated 70% in 1997 to 90% in 2002.
Tuberculosis, influenza and pneumonia, and cerebrovascular diseases
emerged as the leading causes of death. The proportion of deaths in the
age group 20-49 increased, with the proportion of female deaths in this
category rising relative to male deaths.. The question is: does the increase
in deaths represent a real rise proportional to the total population?
After applying every computational check I could think up, the
answer is yes.
Compensating for the under-reporting and the rise in population I came
up with rough estimated mortality rates expressed as a percentage of the
population: 1.0% (1997), 1.09% (1998), 1.06% (1999), 1.08% (2000),
1.16% (2001), 1.19% (2002). This is a respectable death rate compared
to developed nations of the world and does not indicate a pandemic of
any kind. Nevertheless it does indicate a real increase.
Much of the increase is attributable to the rise in reported deaths due to
tuberculosis and influenza/pneumonia (the latter combined in one category
in ICD-10). In the period TB deaths rose 58% from 22021 in 1997 to
59951 in 2002, and influenza/pneumonia 44% from 11503 to 37637.
Expressed as a proportion of all causes of death TB rose from 6.9% in
1997 to 12% in 2002 and influenza/pneumonia rose from 3,6% to 7.5%.
The remaining leading cause of death cerebrovascular diseases decreased
slightly from 5.3% in 1997 and 5% in 2002.
For me the most surprising result in the Statssa report is the number of
deaths attributed to HIV and AIDS as reported by doctors, who remain
the only people legally entitled to sign death certificates. This was 6234
(1997), 7266 (1998), 9925 (1999), 10420 (2000), 9212 (2001) and
10425 (2002). Given the publicity and scope of the estimated AIDS
pandemic a total of 53,482 for the period is an anomaly. It is further
underlined by the fact that expressed as a proportion of all natural
deaths, HIV/AIDS remained static at about 2% for the period.
According to a study by the Medical Research Council (MRC) in 2004,
the numbers of deaths attributable to HIV/AIDS over the period 1997
to 2002 was estimated to be 847,079.(2)
The estimated HIV infection in the South African population according to
the two biggest surveys is an agreed 10 - 11%.(3, 4) This translates to
about 4 million people being infected in 1997. Given an average five years
from infection to death about 50% (2 million) of these should have died by
2002. Even if one takes an average 10 years from infection to death then
25% (1 million) should have died in the period 1997 -2002.
One of the most popularly quoted statistics of the HIV/AIDS fraternity is
that 600 - 1000 people daily (218400 - 364000 a year) die in the epidemic.
Total deaths due to natural causes were reported by Statssa only for 1997,
1999 and 2001, namely 264285, 328671 and 401812, which can only mean
a vast over estimate by the HIV/AIDS industry.
The MRC in a study published in December 2003 estimated that deaths in
the 0-4 year old age group due to HIV/AIDS in 2000 as 106070 (5).
Whereas the only comparable figure in the Statssa report is deaths from
all causes (natural and unnatural) for the 0-4 year old age group in 2000
to be 40768.
Another sign that all does not add up for the HIV/AIDS crusaders is that
according to their paradigm one would expect deaths in the provinces with
the highest number of people estimated to be HIV infected, namely
KwaZulu-Natal and Gauteng, to have risen disproportionately to other
provinces. This is not the case. Both provinces have remained static in
terms of the percentage of HIV/AIDS deaths reported and in the years
1997, 1999 and 2001 HIV/AIDS figured in the top ten causes of death
only in 1999.
There are three possible explanations for the low numbers of reported HIV
and AIDS deaths in the Statssa report. Firstly, doctors could be extremely
reluctant to record HIV as a cause of death, even when they know it is a
cause. Misclassification has been the view of the majority of HIV-causes-AIDS
crusaders here, as expressed in a major report in the Journal of the International
AIDS Society this month (6) Their view is that due to stigma doctors are saving
families from public embarrassment by purposeful misclassification without
reference to HIV. I cannot support this view because death certificates are
confidential documents. A copy is given to the family and the only other
parties which see a certificate are undertakers and the officials of the
Department of Home Affairs. I also do not think that doctors would open
themselves to being struck off the SA Health Professional Council roll by
wrongly, and illegally, filling out death certificates.
A second - more likely - explanation in my view, is that given the raging
debate over HIV and AIDS in South Africa since the high-profile stance
taken by the president of the country, Thabo Mbeki, doctors are being
honest about the uncertainty surrounding this assumed new germ and
disease by definition. In other words, doctors on a large scale are
sceptics.
A third explanation is that due again to the publicity surrounding President
Mbeki's stance on HIV and AIDS and his refusal to be drawn in anyway
since April 2002 into commenting on, apologising for or retracting his views,
people are refusing to take HIV tests. Therefore the doctors would have no
option but to enter the cause of death as other than HIV-related. This would
also account for the modest take-up of antiretrovirals in the public health
system (PHS) since their roll-out in April 2004, which now stands at an
estimated 23,000.
The media generally has given the Statssa report superficial attention. Most
have attributed the increase in deaths to HIV/AIDS by the simple expedient
of combining all diseases listed under the Bangui definition of AIDS as
HIV-related. (6, 7, 8). Not surprising, is that coverage has been limited to
all the usual dial-a-quotes uttering dire warnings of even greater catastrophe
if antiretrovirals are not more widely distributed. (9, 10).
In searching for an explanation for the disproportionate rise in deaths among
young adults, especially females in the 20-49 year old age group, it is probable
that part of the increase could be because this group has the most children
registered for government child support grants. This means the group would
have the most complete records of any other group. The disproportionate
rise in female deaths could be as a result of their attendance at ante-natal
clinics, which would again imply a more complete record. This interpretation
is supported by the fact that an estimated 98% of black pregnant women
attend ante-natal clinics and deaths among all population groups with the
exception of blacks remains virtually unchanged.
Most of the increase in deaths in the 20 to 49 year old group is attributable
to the rise in TB and influenza/pneumonia. Expressed as a proportion of all
deaths across all age groups, TB has risen from 11.9% in 1997 to 19% in
2002 and influenza/pneumonia from 4.1% in 1997 to 9.3%. A possible
explanation for this overall rise could be that word is out on the African
grapevine that these diseases equate to a diagnosis of HIV/AIDS and people
with these symptoms are simply not checking in to clinics. If this is the case
then linking especially TB with HIV would be the single most tragic
consequence of the HIV/AIDS era.
Since antiretrovirals were rolled out in the PHS only in April 2004, there
are as yet no statistics on the number of deaths. As of February 2005 an
estimated 23,000 people are on ARVs in the PHS with a further about
45,000 registered through medical schemes. To gauge the possible extent
of deaths among people on antiretrovirals, a recent article noted that in
one hospital in KwaZulu-Natal of the 300 people on ARV treatment
since May 2004, 20 have died in the seven months to end January 2005.(10)
This is an astonishing admission for drugs that HIV/AIDS crusaders invariably
tag on the adjective "lifesaving" or "life-prolonging".
If ARVS were being rolled out according to the procedures for monitoring
and reporting laid out by the Department of Health, the facts would be forth
coming. Unfortunately, this is not yet being realised, as the Minster of Health
Dr Manto Tshabalala-Msimang was forced to admit during parliamentary
media briefings this week when she was unable to give any figures on ARVs
because "patient information is not up to scratch".(9)
"We are working in the dark and you just give the medicine and hope God
will be on your side" she said.(11)
References.
http://www.statssa.gov.za/news_archive/18Feb2005_1.asp
Dorrington, RE et al. 2004. The Demographic Impact of HIV/AIDS in South Africa.
Annual ante-natal clinic surveys of HIV prevalence. Department of Health 1990 - 2003.
Nelson Mandela Foundation/HSRC Study of HIV/AIDS: Household Survey 2002;
Policy Brief No 3. December 2003. The Burden of Disease Research Unit, Medical Research Council, in
conjunction with the United Nations Children's Fund (UNICEF).
Groenewald, P et al. Identifying deaths from AIDS in South Africa. AIDS. 19(2):193-201. 28 January
2005.
AIDS underlies sharp rise in South Africa's death rate - AllAfrica Global Media. 20 February 2005
AIDS-linked death data stir political storm in South Africa - The New York Times. 19 February 2005.
Call for more AIDS drugs as deaths rocket. Business Day. 21 February 2005.
Deaths soar as AIDS drug rollout falters. Sunday Independent. 20 February 2005.
Sutherland, RK. A New Dawn for Antiretrovirals. British Medial Journal 330: 371. 12 February 2005.
TB, HIV and AIDS, diabetes, hypertension, cancer remain challenges. Health-e News Service. 21
February 2005.
