Spread of AIDS in Africa driven by poor medical practice, report saysBMJ 2003; 326 doi: https://doi.org/10.1136/bmj.326.7387.466/a (Published 01 March 2003) Cite this as: BMJ 2003;326:466
All rapid responses
I find it difficult to understand why there should be so much furore
about the assertion by David Gisselquist and John Potterat in their
article that 'poor medical practice also drives the spread of AIDS in
Africa'. I cannot also understand why their point should adversely affect
whatever public health initiatives there are currently in place in the
fight against the epidemic in Africa.
The spread of HIV / AIDS in Africa is multifactorial as in other parts of
the world. In the western world homosexual sex ( and heterosexual sex to a
lesser extent), infected blood transfusion and needle stick injury were
accepted as the routes of transmission. All these are also, potentially,
risk factors in Africa to a lesser or greater degree because of different
social, economic and cultural circumstances. But in Africa other risk
factors have been identified including a large element of heterosexual sex
especially where this takes the form of peno-anal rather than peno-vaginal
penetration ( Konotey-Ahulu 1989).
In addition, anyone who knows anything about the current state of health
clinics and hospitals in Africa will know that disposable needles and
syringes are used and re-used again and again on different patients until
they are actually blunt. If this is not a sure way to spread a ravaging
HIV/AIDS epidemic, I wonder what is.
Then as I pointed out in an editorial in BMJ West Africa edition ( April,
2001) another (but till now less talked about), source of spread is the
uncontrolled and unregulated 'surgery' carried out by both well meaning
and quack native doctors, healers and birth attendants. Unfortunately, the
shortage of well trained doctors and other health professionals and the
lack of resources for those on the ground to work efficiently and safely,
means that these unregulated, untrained healers actually attend to more
I think that the position should be that whilst the message of safe sex
should continue unabated, all those involved in fighting the epidemic,
including the W.H.O, should highlight the deliterious effect and
consequences of bad medical and health practices such as re-using
disposable equipment. The W.H.O., I.M.F., the world bank and other
agencies could assist these countries financially to replenish stocks of
needles and equipment. With less corruption and better management African
countries can also help themselves by setting up revolving accounts to
ensure that their meagre resources are better applied to replenish old
The public health message of safe, less risky sex, must go hand in hand
with better, safer medical practices if the epidermic of HIV/AIDS is to be
Joseph Ana is managing editor of BMJ West Africa edition which carried an editorial in its April 2001 issue making this same point: that unsafe sex is not the only means of spread of HIV / AIDS in Africa.
Competing interests: No competing interests
I was interested to read your report on articles suggesting that the spread of AIDS in Africa may not be driven by unsafe sex but by unsafe medical practice. Whilst working in South Africa a year or so ago, I noticed several examples of unsafe medical practice and decided to conduct a simple investigation to determine whether there was evidence that doctors might be using the same needle and/or syringe or other instruments on two or more successive patients.
There is some evidence that this has occurred particularly in private practice 1,2,3. I decided to investigate the hypothesis that patients with a history of possible blood contamination have a higher chance of testing HIV positive by using a series of questions administered by an HIV/AIDS counsellor during the course of normal clinical work.
All patients being interviewed by the HIV/AIDS counsellor in Taung hospital in the North West Province for pre-test counselling, were asked to participate in the study after normal counselling had taken place. Each patient was asked whether, over the last five years (since the first democratic general election), they had had injections from a private doctor, had dental treatment from a private dentist, had injections or scarification from a traditional healer, or had ever had scarification by a traditional healer. This information was added to a questionnaire together with the patient’s hospital number, their age and gender. When the HIV result became available, this was added to the questionnaire. Ethical approval for the study was obtained from the Medunsa Research and Ethics Committee. The results were analysed by the Statistical Package for Social scientists (SPSS, Version 9) using the chi-squared test for univariate analysis and logistic regression for multivariate analysis.
One hundred and four consecutive adult patients were studied. There were 62 females and 42 males and their mean age was 34.5 years SD 15.9 (males 38.2(18.7) years and females 32.0 (13.2) years). Positive HIV tests were found in 48% of the total group (53% of females and 41% of males). The results of a univariate analysis are shown in Table 1. The application of a stepwise logistic regression model allowing for age, gender and the four univariate factors yielded ‘injection by doctor’ as the only significant factor (risk ratio 2.8, 95% confidence interval 1.2 to 6.3).
Recall of events by patients cannot be considered to be an accurate method of recall, but by giving patients a well recognised event such as the first general election in 1994, and asking them to recall events since that date, may well reduce the chance of forgetting events such as injections given by doctors, visits to dentists and traditional healers. Of these three types of intervention, only injections given by private doctors were significantly associated with the patient being HIV positive, thereby strengthening the possibility that private doctors in South Africa are contributing to the current HIV/AIDS epidemic. A patient is almost three times more likely to have HIV if they have been injected by such a doctor. Although this suggestion has been made before, only one previous study2 has shown such an association, and that was in children in Zaire.
Only patients who were attending Taung hospital and who were considered ill enough to merit an HIV test, were included in this study. However, the HIV positive group may have been more likely to have sought out injections over the last few years, compared with the HIV negative group as suggested by Hrdy 1. . Findings such as these must always be interpreted with caution, because of possible confounding factors such as a history of sexually transmitted disease, whether the male patients have been circumcised and so on. These findings may help to explain the finding reported in Fassin and Schneider’s article4 where they describe an extraordinarily high prevalence of HIV in women with a single lifetime sexual partner.
1.Hrdy DB Cultural practices contributing to the transmission of human immunodeficiency virus in Africa. Rev Infect Dis 1987; 9 No 6; 1109-1119
2. Mann JM, Francis H, Davachi F, Baudoux P, Quinn TC, Nzilambi N, Bosenge N, Colebunders RL, Kabote N, Piot P, Asila PK, Curran JW. Human Immunodeficiency Virus seroprevalence in pediatric patients 2-14 years of age at Mama Yemo Hospital, Kinshasa, Zaire. Pediatrics 1986; 78 (4): 673-76.
3. Reid SJ and Giddy J. Rural health and human rights – summary of a submission to the Truth and Reconciliation Commission health sector hearings 17 June 1997. S.Afr.med. J. 1998; 88 (8); 980-981.
4. Fassin D, Schneider H. The politics of AIDS in South Africa: beyond the controversies. BMJ 2003: 326: 495-497.
Michael Whitfield (retired General Practitioner)
Table 1 Association of HIV status with reported injections from private doctors or traditional healers and visits to dentists over the previous five years. HIV positive (%)HIV negative (%)Significance Injection from private doctor Yes 36 (58) 26 (42) P <0.02 No 14 (33) 28 (67) Visited private dentist Yes 17 (57) 13 (43) Not significant No 33 (45) 41 (55) Ever had scarification Yes 19 (49) 20 (51) Not significant No 31 (48) 34 (52) Injection or scarification Yes 16 (53) 14 (47) Not significant No 34 (46) 40 (54) Total Sample 50 (48) 54 (52)
Competing interests: Table 1 Association of HIV status with reported injectionsfrom private doctors or traditional healers and visits todentists over the previous five years. HIV positive (%)HIV negative (%)SignificanceInjection from private doctor Yes 36 (58) 26 (42) P <0.02 No 14 (33) 28 (67) Visited private dentist Yes 17 (57) 13 (43) Not significant No 33 (45) 41 (55) Ever had scarification Yes 19 (49) 20 (51) Not significant No 31 (48) 34 (52) Injection or scarification Yes 16 (53) 14 (47) Not significant No 34 (46) 40 (54) Total Sample 50 (48) 54 (52)
Is it not odd that one of the main difficulties facing most sub-
Saharan people is a lack of access to any medical care at all? And in
South Africa at least, where medical care is available throughout the
country at basic level, the syringes and needles are all disposable (and
presumably disposed of after use).
Human Immunodeficiency Virus infection is undoubtedly a social
problem more than a medical one, for with the best will in the world and
with all the drugs that science has so far provided, nobody has yet been
cured of it or its consequent syndrome - or have I missed something?
Until the men of Africa change their beliefs that women are there for
their pleasure and at their command, and women become empowered to say no,
the problem will continue. And condoms are only part of the answer, as
their use is often unacceptable to the men (though not usually to women).
(For those who insist on a political slant, this might be one consequence
of the bad old days of apartheid and attempts to limit the size of black
families, or it might not be.)
Even though there may be some cases of transmission through unsafe
medical practices, the future for HIV transmission control HAS to lie with
education and behaviour modification.
Established and ran an HIV clinic in South Africa in 1990-91.
Competing interests: No competing interests
Why HIV/AIDS is so Prevalent in Africa
It is my hypothesis that increased testosterone increases HIV
infection rates and AIDS. This will explain why AIDS is so high in Africa.
Blacks produce more testosterone than whites, and the infection rate of
blacks far exceeds the rate in whites. (Blacks males produce significantly
more testosterone than while males, (J Natl Cancer Inst 1986; 76: 45), and
black females produce more testosterone than white females, (J Clin
Endocin Metab 1996; 81: 1108). In the following quotations regarding
establishment of a virus, equine arteritis virus (EAV) in horses, it is
demonstrated that testosterone is directly involved in infection and
maintenance of the EAV. It has been determined that: "The findings confirm
that persistent EAV infection is unlikely to occur in geldings and support
the results of previous studies, which demonstrated that testosterone
plays an essential role in the establishment and maintenance of the
carrier state." (J Comp Pathol 1994; 111: 383), first quotation below).
Also: "These findings confirm that the virus can replicate in the
reproductive tract of a significant proportion of colts for a variable
period of time after clinical recovery in the absence of circulating
concentrations of testosterone equivalent to those found in sexually
mature stallions." ( J Comp Pathol 1993; 109: 29), second quotation
below). I suggest the same influence of testosterone is occurring in the
infection rate of HIV, and resultant AIDS, in high testosterone people,
that is, blacks.
McCollum WH, et al., "Resistance of Castrated Male Horses to
Attempted Establishment of the Carrier State with Equine Arteritis Virus,"
(J Comp Pathol 1994; 111: 383)
"Twelve geldings all became infected when inoculated intranasally
with the KY-84 strain of equine arteritis virus (EAV), a strain previously
shown to be capable of establishing the carrier state in the stallion.
With the exception of one animal that showed no effects other than
pyrexia, all of the geldings developed clinical signs characteristic of
equine viral arteritis (EAV). The geldings were febrile for varying
periods within the range of 2-10 days after inoculation. Viraemia occurred
from day 2 onwards, for periods varying from 9 to at least 19 days. Nasal
shedding of virus began 2-4 days after inoculation and persisted for
periods ranging from 7-14 days. All geldings "seroconverted" to EAV by day
11, with serum neutralization titres ranging from 8 to 64. The titres
ranged from 8 to 32 after 4 weeks. Low concentrations of EAV were detected
in the kidney and blood of one gelding killed 30 days after inoculation
and in the blood of another killed after 57 days. Virus was not isolated
from any tissue or fluid collected from the remaining 10 geldings, all of
which were killed between days 30 and 148. The findings confirm that
persistent EAV infection is unlikely to occur in geldings and support the
results of previous studies, which demonstrated that testosterone plays an
essential role in the establishment and maintenance of the carrier state."
Holyoak GR, et al., "Relationship between Onset of Puberty and
Establishment of Persistent Infection with Equine Arteritis Virus in the
Experimentally Infected Colt," (J Comp Pathol 1993; 109: 29)
"The relationship between stage of reproductive tract maturity and
susceptibility to the experimental establishment of persistent infection
with equine arteritis virus (EAV) was investigated in 21 prepubertal and
15 peripubertal colts. Five of six peripubertal colts inoculated
intranasally remained infected in the reproductive tract from post-
challenge day 28 to 93 and two of six from post-challenge day 120 to 180.
No virus was detected in five of these animals killed on post-challenge
day 210. Each of two peripubertal colts remained infected in the
reproductive tract at post-challenge day 60 and one of nine was found to
be persistently infected with EAV 15 months after challenge. These
findings confirm that the virus can replicate in the reproductive tract of
a significant proportion of colts for a variable period of time after
clinical recovery in the absence of circulating concentrations of
testosterone equivalent to those found in sexually mature stallions. Long-
term persistent infection with EAV does not appear to occur in colts
exposed to the virus before the onset of peripubertal development. We
suggest that colts should be vaccinated at approximately 6 months of age,
before peripubertal development but after the disappearance of maternally
There are a number of investigations that support my contention that
testosterone adversely affects the immune system vis-à-vis the HIV. One
contains this generality: "...sexually mature male vertebrates are often
more susceptible to infection and carry higher parasite burdens in the
field." (Int J Parasitol 1996; 26: 1009). Another investigation determined
"CONCLUSIONS: Castration before soft-tissue trauma and hemorrhagic
shock maintains normal immune function in male mice, but sham-castrated
male mice show significant immunodepression. The maintenance of immune
function by androgen deficiency does not seem to be related to changes in
the release of corticosterone. We conclude that male sex steroids are
involved in the immunodepression observed after trauma-hemorrhage. Thus,
the use of testosterone-blocking agents following trauma-hemorrhage should
prevent the depression of immune functions and decrease the susceptibility
to sepsis under those conditions." (Arch Surg 1996; 131: 1186).
For sake of brevity, I have not gone in to detail regarding my
explanation of how I think testosterone reduces the immune response. The
next citation supports the negative effect of testosterone on the immune
system, regarding malaria, and continues to say that the testosterone
effect is not due to the classical explanation. My explanation does not
rely on the "classical AR response."
"Our data suggest that testosterone suppresses the development of
protective immunity against P. chabaudi malaria, and that this
immunosuppressive effect of testosterone is not primarily mediated by the
classical AR response." (J Endocrinol 1992; 135: 407).
Another investigation examined a specific phase of the immune
response and found the same anti-immune function of testosterone. "These
results suggest that the male sex hormone, testosterone, but not the
female sex hormone has a role in the down-regulation of the systemic
eosinophil responses of C57BL/6 mice to infection with B. pahangi."
(Immunopharmacol 1992; 23: 75). You may know that tuberculosis,
Mycobacterium tuberculosis, is more prevalent in blacks than whites, in a
socioeconomically controlled study (New Eng J Med 1990; 322: 422). I
explained this, then in 1990, as an example that testosterone adversely
affects the immune system in humans, especially in blacks. In a study of
mice exposed to Mycobacterium marinum, testosterone was found to increase
susceptibility in males and females. This study carefully controlled for
the presence of testosterone. "Although this ordering corresponded to the
susceptibilities of both male and female mice to the organisms, much
greater strain dependency was seen in males than in females. Castration
caused an increase in the host resistance of males, but this effect was
substantially reversed by continuous testosterone treatment. Testosterone
also increased the susceptibility of female mice to this infection. These
findings imply that the male sex hormone is involved in the lowered anti-
M. marinum resistance of males." (Infect Immun 1991; 59: 4089).
Now, the data in the paragraph above, and citations regarding
testosterone and the EAV in horses, all support negative effects of
testosterone on differing types of infection in different mammals. Bearing
in mind that I cited solid evidence that black males produce significantly
more testosterone than white males and black females produce more
testosterone than white females, it is my hypothesis that increased
testosterone in blacks is why the HIV infects blacks more readily in the
U.S.A. and much more so in Africa. For the very same reason, it also
explains why blacks exhibit more tuberculosis, and other infections, than
Competing interests: No competing interests
The following observation has a curious interpretation.
"The authors point to the exceptionally rapid spread of the disease in
countries with widespread health services, such as South Africa and
Zimbabwe. In Zimbabwe in the 1990s, HIV infections increased by 12% a year
whereas other sexually transmitted disease were declining by 25%"
It is suggested that this shows the use of dirty needles may be the more
likely cause of AIDS.
The observation should have stated that the DETECTION of HIV and other
sexually transmitted diseases (STD)have changed.
This is a more logical interpretation because of the wider availability of
testing for HIV and more effective cheap treatment of other STD.
Competing interests: No competing interests