Rapid responses are electronic comments to the editor. They enable our users
to debate issues raised in articles published on bmj.com. A rapid response
is first posted online. If you need the URL (web address) of an individual
response, simply click on the response headline and copy the URL from the
browser window. A proportion of responses will, after editing, be published
online and in the print journal as letters, which are indexed in PubMed.
Rapid responses are not indexed in PubMed and they are not journal articles.
The BMJ reserves the right to remove responses which are being
wilfully misrepresented as published articles or when it is brought to our
attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not
including references and author details. We will no longer post responses
that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I read Pat Sidley's uncritical summary of the 175-page report THE
IMPACT OF HIV/AIDS ON THE HEALTH SECTOR: NATIONAL SURVEY OF
HEALTH PERSONNEL, AMBULATORY AND HOSPITALISED PATIENTS AND
HEALTH FACILITIES, 2002, funded in part by the US Centers for Disease
Control. A closer look at the report reveals a very different picture than
The most important and telling revelation appears on page xiii of the
Executive Summary under the sub-heading Increased patient load: "The
study results showed that overall there has not been an increase in the
mean number of admissions to the medical wards of all patients (AIDS
and non-AIDS) reported between 1995 and 2000." As if to drive home
the point that there is no explosive growth in health problems in South
African hospitals, in the same paragraph the authors go on to say,
"Interestingly, during this period, the total bed occupancy rates have
remained about the same."
In fact, the mean hospital bed occupancy rates overall for 1995,
1997,
and 2000 were 78.1%, 77.2% and 78.4% respectively. Furthermore,
"When all hospitals are considered, there has been an overall decline in
utilisation of paediatric beds" dropping from 88.3% in 1995 to 79.6% in
2000.
Magically, however, the authors transformed their own simple, clear,
readily documented results into "a very large increase in the mean
number of HIV/AIDS-related admissions between 1995 and 2000" even
though total hospital admissions remained virtually flat over that period.
Faced with the dilemma of how to explain a growing AIDS epidemic
when there was no net change in total morbidity and mortality, the
authors were forced to speculate "that non-AIDS patients have been
'crowded out' of the health care system to give way to HIV/AIDS
patients." That argument implies that the hospitals were filled to
capacity. However, their own report says the hospital bed occupancy rate
was consistently below 80% during the five-year period studied.
The zero-sum game the authors are playing (i.e., constant overall
rates
of death and disease yet dramatic increase in HIV/AIDS) requires that
there must be compensatory reductions in other diseases and causes of
death. For example, as more and more TB deaths are reclassified as AIDS
deaths, there is a corresponding reduction in the number of TB deaths. It
is the insidious Bangui definition of what constitutes AIDS in Africa that
catalyzes this despicable practice.
In Appendix 2 under the heading WHO surveillance (Bangui/WHO/
clinical) on page 145 we find the specifics of the Bangui definition of
African AIDS:
"For surveillance purposes, an adult is said to have AIDS when two
major
and one minor sign/s are present.
"Major signs in adults
• Weight loss more ?10% of body weight
• Chronic diarrhoea >1 month
• Prolonged fever >1 (intermittent or constant)
"Minor signs in adults
• Persistent cough >1 month
• Generalised itching skin rash
• Recurrent herpes zoster
• Oro-pharengeal candidiasis
• Chronic progressive and disseminated herpes simplex infection
• Generalised lymphadenopathy
"For surveillance purposes, a child is said to have AIDS when two
major
and at least two minor signs are present.
"Major signs in children
• Weight loss or abnormally slow growth
• Chronic diarrhoea >1 month
• Prolonged fever >1 month
"Minor signs
• Generalised lymphadenopathy
• Oro-pharyngeal candidiasis
• Repeated common infections (otitis, pharyngitis, etc.)
• Persistent cough
• Generalised dermatitis
• Confirmed maternal LAV/HTVL-111 infection (PAHO, June 2000)".
Table 79 on page 146 of the report shows that based on the Bangui
definition the WHO's cumulative number of AIDS cases from 1980
through 1997 for all of Africa was 620,000. For comparison, the
cumulative total of AIDS cases in the USA through 1997 was 641,087.
This is a far cry from the tens of millions of African AIDS cases in
newspaper headlines. The report cites a member of Mbeki's Presidential
AIDS Advisory Panel, Christian Fiala, as the source of the information in
Table 79. Fiala, C (2000) AIDS in Africa - the way forward, 6/7 May. http://perso.wanadoo.fr/sidasante/critique/fiala.htm.
The authors took the unprecedented and much welcomed step of
comparing the prevalence of AIDS based on the Bangui definition to the
prevalence of what is loosely and imprecisely described as "HIV/AIDS"
based on the HIV antibody test. Such an analysis requires defining a
gold-standard for the prevalence of AIDS and, hence (in the minds of the
authors), the prevalence of HIV. The authors used the Bangui symptoms
as the gold-standard for AIDS in South Africa. Since the authors did not
describe how they were able to separate non-HIV-caused Bangui
symptoms from HIV-caused Bangui symptoms the uncertainty in the
Bangui gold-standard to accurately capture true cases of AIDS ranged
from 0-100%. In line with mainstream dogma, the authors believe that
antibodies to HIV mean that a person is actively infected with infectious
HIV. Furthermore, the belief goes, the presence of antibodies against HIV
meant that a person's Bangui symptoms were, in some as yet
undemonstrated and wholly mysterious way, ultimately caused by HIV.
Armed with this belief and the HIV antibody test the authors hoped to
get a handle on just how accurately (between 0 and 100%) the Bangui
symptoms were able to diagnose AIDS in South Africans. However, this
approach presents its own basket of problems.
The HIV antibody test is documented to suffer from over 60 sources of
false-positive results. Examples pertinent to South Africa are: naturally-
occurring antibodies, tuberculosis, herpes simplex I /II, upper
respiratory tract infection (cold or flu), recent viral infection or
exposure
to viral vaccines, pregnancy in multiparous women, malaria, alcoholic
hepatitis/liver disease, naturally acquired hepatitis B antibodies or
hepatitis B vaccination, and healthy individuals as a result of poorly-
understood cross-reactions. Although the authors acknowledged in
Appendix 4 the existence of false positive HIV antibody results, they did
not take into consideration the contribution of false-positive HIV
antibody results in their analysis or conclusions. Instead, Appendix 5
shows that the authors relied upon the test manufacture's claim that,
"The kit protocol indicates sensitivity = 100% and specificity of 99.9% to
99.2%".
Given the arbitrary nature of a gold-standard (Bangui definition)
that
redefines a list of general symptoms of long-established African
morbidity as AIDS, plus the large range of unknown errors, plus the
unconsidered sources of false-positive HIV antibody results, on top of
the circular practice of attributing HIV as the cause of Bangui symptoms,
it is not surprising that the authors reported a very poor correlation
between Bangui AIDS and HIV-antibody-determined "HIV/AIDS". The
positive predictive value for adults and children combined ranged from
16-29%. This means that using Bangui AIDS as the gold-standard, the
HIV antibody test had a false positive rate of 71-84%.
Notwithstanding all the uncertainty, confounding and massive hand-
waving, the authors decided that the problematic HIV antibody results
were a more accurate measure of AIDS in South Africa than the recorded
Bangui diseases and deaths.
The other things described in the report said to be influenced by
"HIV/
AIDS" (i.e., Absenteeism among health workers, Low staff morale,
Increased patient load, Affected sub-systems of the health care system,
Progression of the impact of HIV/AIDS over time, AIDS mortality in
health care workers) are immediately rendered mute by the disclaimer
that begins the section titled HIV/AIDS and Conduct of Professional
Duties:
"HIV/AIDS is not a notifiable disease. It is therefore not necessary
for
patients to make their HIV status known to health workers, or for health
workers to notify their patients of their HIV status. In fact, due to the
stigma attached to the disease, many patients visiting clinics or lying in
hospitals, as well as many health workers, are unaware of their HIV
status. If they know that they are HIV positive, they usually do not want
to make this known to other people.
"During the Gauteng survey it was established that professional
health
workers have access to the medical files of patients in the public health
sector. However, it was difficult to determine a patient's HIV status from
these files. (During the Gauteng survey, fieldworkers (professional
nurses) could obtain the HIV status of only a small number of patients
visiting public clinics and hospitals from their medical files.) In the
other
eight provinces the situation might be different, although professional
health workers generally reported that they were not officially informed
about the HIV status of patients.
"Nurses in private hospitals did not have access to patients' medical
files. Normally the medical practitioner responsible for the patient keeps
these files. Even medical practitioners may be unaware of a patient's
status. Unless informed by a medical practitioner or the patient, most
professional health workers did not know the status of their patients.
Many nurses in public and private health facilities were, however, of the
opinion that they could determine a patient's status by looking at
presenting [Bangui?] symptoms.
"Non-professional health workers were normally not familiar with the
status of patients unless informed by the patient or other staff
members."
Given such profound ignorance and confusion from the outset, there is
little basis on which to measure how "HIV/AIDS" affects the conduct of
professional duties. Undaunted by the dismal prospects of obtaining any
accurate or meaningful results, the authors forged ahead, saying:
"It is against this background that health workers were asked if
patients
suffering from HIV/AIDS affect how they perform their duties as health
care givers. Table 36 shows that 43.9 per cent of health workers felt that
the prevalence of HIV/AIDS impacted on their work. Nearly half (48.9 per
cent) of professional and 30.8 per cent of non-professional health
workers reported that they had experienced this effect."
Such opinions are not the basis of a scientific study, but they make
great
headlines.
Pat Sidley says that, "This report, although commissioned by the
health
department, was kept under wraps by the department until it was leaked
in the past month." Given its obvious and numerous flaws, the real
wonder is why a sympathizer would want to bring attention to the report
and subject it to critical review.
Critical analysis of the report on The Impact of HIV/AIDS on the Health Sector of South Africa
Dear Editor,
I read Pat Sidley's uncritical summary of the 175-page report THE
IMPACT OF HIV/AIDS ON THE HEALTH SECTOR: NATIONAL SURVEY OF
HEALTH PERSONNEL, AMBULATORY AND HOSPITALISED PATIENTS AND
HEALTH FACILITIES, 2002, funded in part by the US Centers for Disease
Control. A closer look at the report reveals a very different picture than
that portrayed by Sidley. (The report can be downloaded at http://
www.hsrcpublishers.co.za/index.html?health_sector.html~content)
The most important and telling revelation appears on page xiii of the
Executive Summary under the sub-heading Increased patient load: "The
study results showed that overall there has not been an increase in the
mean number of admissions to the medical wards of all patients (AIDS
and non-AIDS) reported between 1995 and 2000." As if to drive home
the point that there is no explosive growth in health problems in South
African hospitals, in the same paragraph the authors go on to say,
"Interestingly, during this period, the total bed occupancy rates have
remained about the same."
In fact, the mean hospital bed occupancy rates overall for 1995,
1997,
and 2000 were 78.1%, 77.2% and 78.4% respectively. Furthermore,
"When all hospitals are considered, there has been an overall decline in
utilisation of paediatric beds" dropping from 88.3% in 1995 to 79.6% in
2000.
Magically, however, the authors transformed their own simple, clear,
readily documented results into "a very large increase in the mean
number of HIV/AIDS-related admissions between 1995 and 2000" even
though total hospital admissions remained virtually flat over that period.
Faced with the dilemma of how to explain a growing AIDS epidemic
when there was no net change in total morbidity and mortality, the
authors were forced to speculate "that non-AIDS patients have been
'crowded out' of the health care system to give way to HIV/AIDS
patients." That argument implies that the hospitals were filled to
capacity. However, their own report says the hospital bed occupancy rate
was consistently below 80% during the five-year period studied.
The zero-sum game the authors are playing (i.e., constant overall
rates
of death and disease yet dramatic increase in HIV/AIDS) requires that
there must be compensatory reductions in other diseases and causes of
death. For example, as more and more TB deaths are reclassified as AIDS
deaths, there is a corresponding reduction in the number of TB deaths. It
is the insidious Bangui definition of what constitutes AIDS in Africa that
catalyzes this despicable practice.
In Appendix 2 under the heading WHO surveillance (Bangui/WHO/
clinical) on page 145 we find the specifics of the Bangui definition of
African AIDS:
"For surveillance purposes, an adult is said to have AIDS when two
major
and one minor sign/s are present.
"Major signs in adults
• Weight loss more ?10% of body weight
• Chronic diarrhoea >1 month
• Prolonged fever >1 (intermittent or constant)
"Minor signs in adults
• Persistent cough >1 month
• Generalised itching skin rash
• Recurrent herpes zoster
• Oro-pharengeal candidiasis
• Chronic progressive and disseminated herpes simplex infection
• Generalised lymphadenopathy
"For surveillance purposes, a child is said to have AIDS when two
major
and at least two minor signs are present.
"Major signs in children
• Weight loss or abnormally slow growth
• Chronic diarrhoea >1 month
• Prolonged fever >1 month
"Minor signs
• Generalised lymphadenopathy
• Oro-pharyngeal candidiasis
• Repeated common infections (otitis, pharyngitis, etc.)
• Persistent cough
• Generalised dermatitis
• Confirmed maternal LAV/HTVL-111 infection (PAHO, June 2000)".
Table 79 on page 146 of the report shows that based on the Bangui
definition the WHO's cumulative number of AIDS cases from 1980
through 1997 for all of Africa was 620,000. For comparison, the
cumulative total of AIDS cases in the USA through 1997 was 641,087.
This is a far cry from the tens of millions of African AIDS cases in
newspaper headlines. The report cites a member of Mbeki's Presidential
AIDS Advisory Panel, Christian Fiala, as the source of the information in
Table 79. Fiala, C (2000) AIDS in Africa - the way forward, 6/7 May.
http://perso.wanadoo.fr/sidasante/critique/fiala.htm.
The authors took the unprecedented and much welcomed step of
comparing the prevalence of AIDS based on the Bangui definition to the
prevalence of what is loosely and imprecisely described as "HIV/AIDS"
based on the HIV antibody test. Such an analysis requires defining a
gold-standard for the prevalence of AIDS and, hence (in the minds of the
authors), the prevalence of HIV. The authors used the Bangui symptoms
as the gold-standard for AIDS in South Africa. Since the authors did not
describe how they were able to separate non-HIV-caused Bangui
symptoms from HIV-caused Bangui symptoms the uncertainty in the
Bangui gold-standard to accurately capture true cases of AIDS ranged
from 0-100%. In line with mainstream dogma, the authors believe that
antibodies to HIV mean that a person is actively infected with infectious
HIV. Furthermore, the belief goes, the presence of antibodies against HIV
meant that a person's Bangui symptoms were, in some as yet
undemonstrated and wholly mysterious way, ultimately caused by HIV.
Armed with this belief and the HIV antibody test the authors hoped to
get a handle on just how accurately (between 0 and 100%) the Bangui
symptoms were able to diagnose AIDS in South Africans. However, this
approach presents its own basket of problems.
The HIV antibody test is documented to suffer from over 60 sources of
false-positive results. Examples pertinent to South Africa are: naturally-
occurring antibodies, tuberculosis, herpes simplex I /II, upper
respiratory tract infection (cold or flu), recent viral infection or
exposure
to viral vaccines, pregnancy in multiparous women, malaria, alcoholic
hepatitis/liver disease, naturally acquired hepatitis B antibodies or
hepatitis B vaccination, and healthy individuals as a result of poorly-
understood cross-reactions. Although the authors acknowledged in
Appendix 4 the existence of false positive HIV antibody results, they did
not take into consideration the contribution of false-positive HIV
antibody results in their analysis or conclusions. Instead, Appendix 5
shows that the authors relied upon the test manufacture's claim that,
"The kit protocol indicates sensitivity = 100% and specificity of 99.9% to
99.2%".
Given the arbitrary nature of a gold-standard (Bangui definition)
that
redefines a list of general symptoms of long-established African
morbidity as AIDS, plus the large range of unknown errors, plus the
unconsidered sources of false-positive HIV antibody results, on top of
the circular practice of attributing HIV as the cause of Bangui symptoms,
it is not surprising that the authors reported a very poor correlation
between Bangui AIDS and HIV-antibody-determined "HIV/AIDS". The
positive predictive value for adults and children combined ranged from
16-29%. This means that using Bangui AIDS as the gold-standard, the
HIV antibody test had a false positive rate of 71-84%.
Notwithstanding all the uncertainty, confounding and massive hand-
waving, the authors decided that the problematic HIV antibody results
were a more accurate measure of AIDS in South Africa than the recorded
Bangui diseases and deaths.
The other things described in the report said to be influenced by
"HIV/
AIDS" (i.e., Absenteeism among health workers, Low staff morale,
Increased patient load, Affected sub-systems of the health care system,
Progression of the impact of HIV/AIDS over time, AIDS mortality in
health care workers) are immediately rendered mute by the disclaimer
that begins the section titled HIV/AIDS and Conduct of Professional
Duties:
"HIV/AIDS is not a notifiable disease. It is therefore not necessary
for
patients to make their HIV status known to health workers, or for health
workers to notify their patients of their HIV status. In fact, due to the
stigma attached to the disease, many patients visiting clinics or lying in
hospitals, as well as many health workers, are unaware of their HIV
status. If they know that they are HIV positive, they usually do not want
to make this known to other people.
"During the Gauteng survey it was established that professional
health
workers have access to the medical files of patients in the public health
sector. However, it was difficult to determine a patient's HIV status from
these files. (During the Gauteng survey, fieldworkers (professional
nurses) could obtain the HIV status of only a small number of patients
visiting public clinics and hospitals from their medical files.) In the
other
eight provinces the situation might be different, although professional
health workers generally reported that they were not officially informed
about the HIV status of patients.
"Nurses in private hospitals did not have access to patients' medical
files. Normally the medical practitioner responsible for the patient keeps
these files. Even medical practitioners may be unaware of a patient's
status. Unless informed by a medical practitioner or the patient, most
professional health workers did not know the status of their patients.
Many nurses in public and private health facilities were, however, of the
opinion that they could determine a patient's status by looking at
presenting [Bangui?] symptoms.
"Non-professional health workers were normally not familiar with the
status of patients unless informed by the patient or other staff
members."
Given such profound ignorance and confusion from the outset, there is
little basis on which to measure how "HIV/AIDS" affects the conduct of
professional duties. Undaunted by the dismal prospects of obtaining any
accurate or meaningful results, the authors forged ahead, saying:
"It is against this background that health workers were asked if
patients
suffering from HIV/AIDS affect how they perform their duties as health
care givers. Table 36 shows that 43.9 per cent of health workers felt that
the prevalence of HIV/AIDS impacted on their work. Nearly half (48.9 per
cent) of professional and 30.8 per cent of non-professional health
workers reported that they had experienced this effect."
Such opinions are not the basis of a scientific study, but they make
great
headlines.
Pat Sidley says that, "This report, although commissioned by the
health
department, was kept under wraps by the department until it was leaked
in the past month." Given its obvious and numerous flaws, the real
wonder is why a sympathizer would want to bring attention to the report
and subject it to critical review.
David Rasnick, PhD
Member of the Presidential AIDS Advisory Panel
Competing interests:
None declared
Competing interests: No competing interests