Circumcision and HIV/AIDS Prevention

(Male circumcision clinic in Namibia)

Introduction

That medical circumcision is a vital weapon in the war on HIV is now consensus in the medical scientific community. From the first suggestions in 1986, to multiple epidemiological studies through the 1990s and early 2000s, and culminating in a series of three randomized controlled trials published in 2005 and 2007 (Auvert, et al., 2005; Gray et al., 2007; Bailey et al., 2007), a wealth of compelling evidence has convinced every medical body dealing with the epidemic. Circumcision reduces the risk of female to male HIV transmission by up to 70 % (Lei et al., 2015). Safe medical circumcision is now being rolled out in 14 high priority African countries where the epidemic is at its worst. (See: http://www.who.int/hiv/topics/malecircumcision/fact_sheet/en/).

It is also being considered for countries outside Africa, such as in the Caribbean, where HIV prevalence can reach 3 %. Debate continues about whether it is worthwhile in low-HIV settings but, in principle, even there it should be protective. The effect is a biological one, related to the presence or absence of a foreskin (a major route of entry for the virus), and viruses do not care which continent they are on.

Initially targeting adults, the circumcision program is now being extended to children and infants, much to the consternation of intactivists who continue to rail against the evidence, churning out pseudoscientific articles, spreading misinformation to Africans through social media, websites and comments threads, and doing all they can to undermine this life-saving initiative. The irresponsibility, and immorality, of this is disgusting. It is already well documented how similar behaviour by anti-vaccinationists (many of whom are also intactivists) has led to outbreaks of preventable disease. Circumcision is no different.

The scope for disaster is terrifying. When former South African president Thabo Mbeki fell for HIV/AIDS denial pseudoscience on the internet he halted his country’s implementation of another effective anti-HIV strategy: antiretroviral therapy (ART). The result was 330,000 dead (Chigwedere et al., 2008). As circumcision is the third most successful intervention, after ART and condoms (WHO, 2016), undermining it could be catastrophic.

In this section we bring you rebuttals of common intactivist arguments relating to circumcision and HIV. Many relate to Africa, others are more general. But the Internet is global, and intactivist arguments can reach everywhere. Intactivists would rejoice if they dissuaded Africans from participating in the WHO-sponsored circumcision program. Africans would die in droves if they did. We at circfacts.org believe that African lives matter.

Table: Summary of the principal published attacks on the African RCTs, or the circumcision program following them, and the debunkings of these.

Intactivist articleDebunked by
Green, et al., (2010)Banerjee et al., (2011); Wawer et al., (2011)
Van Howe & Storms (2011)Morris et al. (2011)
Boyle & Hill (2011)Wamai et al. (2012)
Ncayiyana (2011)Kessinger & Millard (2012); Venter et al. (2012)
Earp (2012)Geffen (2012)
McQuoid-Mason (2013)Millard & Goldstuck (2013)
Adler & Svoboda (2014)Moreton (2014)
Darby (2014)Morris et al. (2015)
Van Howe (2015)Morris et al. (2016)
Van Howe (2017)Morris et al. (2017)

References
Adler, P., Svoboda, J.S. (2014) Open letter to the Bill & Melinda Gates Foundation. Submitted by Harry Crouch of the National Coalition for Men on 16 July 2014. On-line: http://ncfm.org/wp-content/uploads/2014/07/141617-Cover-letter-AND-the-open-letter-to-the-Gates-Foundation.pdf

Auvert, B., Taljaard, D., Lagarde, E., Sobngwi-Tambekou, J., Sitta, R., Puren, A. (2005) Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial. PLoS Medicine, 2(11), 1112-22. On-line: https://journals.plos.org/plosmedicine/article/comments?id=10.1371/journal.pmed.0020298

Bailey, R.C., Moses, M., Parker, C.B., Agot, K., Maclean, I., Krieger, J.N., Williams, C.F.M., Campbell, R.T., Ndinya-Achola, J.O. (2007) Male circumcision prevention for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet, 369(9562), 643-56. On-line: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60313-4/fulltext

Banerjee, J., Klausner, J.D., Halperin, D.T., Wamai, R., Schoen, E.J., Moses, S., Morris, B.J., Bailis, S.A., Venter, F., Martinson, N., Coates, T.J., Gray, G., Bowa, K. (2011) Circumcision denialism unfounded and unscientific. Am. J. Prev. Med., 40(3), e11-e12. On-line: https://ghdonline.org/uploads/Circumcision_Denialism_Unfounded_and_Unscientific_AJPM.pdf

Boyle, G.J., Hill, G. (2011) Sub-Saharan African randomised clinical trials into male circumcision and HIV transmission: Methodological, ethical and legal concerns. J. Law Med., 19(2), 316-34. On-line: http://artemide.bioeng.washington.edu/2011-12_JLM-Boyle-Hill.pdf

Chigwedere, P. Seage, G.R., Gruskin, S., Lee, T.H., Essex, M. (2008) Estimating the lost benefits of antiretroviral drug use in South Africa, J. Acquir. Immune Defic. Syndr., 49(4), 410-5. On-line: https://journals.lww.com/jaids/fulltext/2008/12010/estimating_the_lost_benefits_of_antiretroviral.10.aspx

Darby, R. (2014) Syphilis 1855 and HIV-AIDS 2007: Historical reflections on the tendency to blame human anatomy for the action of micro-organisms. Global Public Health, 10(5-6), 573-88. Abstract: https://www.tandfonline.com/doi/abs/10.1080/17441692.2014.957231?journalCode=rgph20

Earp, B.D. (2012) When bad science kills, or how to spread AIDS. Originally on the Oxford University “Practical Ethics” blog, it has since been reproduced here: http://www.creativitypost.com/activism/when_bad_science_kills_or_how_to_spread_aids

Geffen, N. (2012) Getting circumcision science right in the media. Posted on the Quackdown blog: http://www.quackdown.info/article/getting-circumcision-science-right-media/

Gray, R.H., Kigozi, G., Serwadda, D., Makumbi, F., Watya, S., Nalugoda, F., Kiwanuka, N., Moulton, L.H., Chaudhary, M.A., Chen, M.Z., Sewankambo, N.K., Wabwire-Mangen, F., Bacon, M.C, Williams, C.F.M., Opendi, P., Reynolds, S.J., Laeyendecker, O., Quinn, T.C., Wawer, M.J. (2007) Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet, 369(9562), 657-66. On-line: https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60313-4/fulltext

Green, L.W., Travis, J.W., McAllister, R.G., Peterson, K.W., Vardanyan, A.N., Craig, A. (2010) Male circumcision and HIV prevention insufficient evidence and neglected external validity. Am. J. Prev. Med., 39(5), 479–82. On-line: http://www.thewholenetwork.org/uploads/4/2/0/2/4202929/ajpmgreenetal2010_pub.pdf

Kessinger, M., Millard, P.S. (2012) Voluntary male medical circumcision. S. Afr. Med. J., 102(3), 123-4.  (A brief reply by Ncayiyana followed on p. 125-6). On-line: http://www.samj.org.za/index.php/samj/article/view/5491

Lei, J.H., Liu, L.R., Wei, Q., Yan, S.B., Yang, L., Song, T.R., Yuan, H.C., Xiao, L.V., Han, P. (2015) Circumcision status and risk of HIV acquisition during heterosexual intercourse for both males and females: A meta-analysis. PLoS ONE, 10(5): e0125436. doi:10.1371/journal. pone.0125436. On-line: https://journals.plos.org/plosone/article/authors?id=10.1371/journal.pone.0125436

McQuoid-Mason, D.J. (2013) Is the mass circumcision drive in KwaZulu-Natal involving neonates and children less than 16 years of age legal? What should doctors do? S. Afr. Med. J. 103(5), 283-4. On-line: https://pdfs.semanticscholar.org/2a45/d5b233af66869fae36f074b27240da1d71d4.pdf?_ga=2.187139243.2042035954.1609675434-1941952992.1609259545

Millard, P.S., Goldstuck, N. (2013) Legality of circumcision in South Africa. S. Afr., Med. J., 103(7), 436. On-line: http://www.samj.org.za/index.php/samj/article/view/7115/5183

Moreton, S. (2014) An annotated version of the National Coalition for Men’s letter to the Bill and Melinda Gates Foundation. On-line: http://www.quackdown.info/article/debunking-anti-circumcision-pseudoscience/ and https://thecaseagainstintactivism.wordpress.com/2014/08/07/ncfm-refutation/

Morris, B.J., Waskett, J.H., Gray, R.H., Halperin, D.T., Wamai, R., Auvert, B., Klausner, J.D. (2011) Exposé of misleading claims that male circumcision will increase HIV infections in Africa. J. Public Health Africa, 2:e28, 117-122. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5345501/

Morris, B.J., Wamai, R.G., Krieger, J.N., Banerjee, J., Klausner, J.D. (2015) Male circumcision to prevent syphilis in 1855 and HIV in 1986 is supported by the accumulated scientific evidence to 2015: Response to Darby. Global Public Health, 12(10), 1315-1333. Abstract: https://www.tandfonline.com/doi/abs/10.1080/17441692.2015.1104371?journalCode=rgph20

Morris, B.J., Barboza, G., Wamai, R.G., Krieger, J.N. (2016) Circumcision is a primary preventive against HIV infection: Critique of a contrary meta-regression analysis by Van Howe. Global Public Health, 13(12), 1889-1899. Abstract: https://www.tandfonline.com/doi/abs/10.1080/17441692.2016.1164737?journalCode=rgph20

Morris, B.J., Barboza, G., Wamai, R.G., Krieger, J.N. (2017) Expertise and ideology in statistical evaluation of circumcision for protection against HIV infection. World Journal of AIDS, 7(3), 179-203.  On-line: https://www.scirp.org/journal/PaperInformation.aspx?PaperID=78405

Ncayiyana, D.J. (2011) The illusive promise of circumcision to prevent female-to-male HIV infection – not the way to go for South Africa. S. Afr. Med. J., 101(11), 775-6. On-line: http://www.samj.org.za/index.php/samj/article/view/5384/3655

Van Howe, R.S., Storms, M.R. (2011) How the circumcision solution in Africa will increase HIV infections. J. Publ. Health Africa, 2:e4, 11-15. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5345479/

Van Howe, R.S. (2015) Circumcision as a primary HIV preventive: Extrapolating from the available data. Global Public Health, 10(5-6), 607-25. Abstract: https://www.tandfonline.com/doi/abs/10.1080/17441692.2015.1016446?journalCode=rgph20

Van Howe, R.S. (2017) Expertise or ideology? A response to Morris et al. 2016, ‘Circumcision is a primary preventive against HIV infection: Critique of a contrary metaregression analysis by Van Howe’. Global Public Health, 2017, 13(12), 1900-1918. First page: https://www.tandfonline.com/doi/abs/10.1080/17441692.2016.1272939?journalCode=rgph20

Venter, F., Rees, H., Pillay, Y., Simelela, N., Mbengashe, T., Geffen, N., Conradie, F., Shisana, O., Rech, D., Serenata, C., Taljaard, D., Gray, G. (2012) The medical proof doesn’t get much better than VMMC. S. Afr. Med. J., 102(3), 124-5.  (A brief reply by Ncayiyana followed on p. 125-6). On-line: http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S0256-95742012000300012

Wamai, R.G., Morris, B.J., Waskett, J.H., Green, E.C., Banerjee, J., Bailey, R.C., Klausner, J.D., Sokal, D.C. & Hankins, C.A. (2012) Criticisms of African trials fail to withstand scrutiny: Male circumcision does prevent HIV infection. J. Law Med., 20(1): 93-123. On-line: https://ghdonline.org/uploads/JLM2012_Male_circ_does_prevent_HIV_infection.pdf

Wawer, M.J., Gray, R.H., Serwadda, D., Kigozi, G., Nalugoda, F., Quinn, T.C. (2011) Male circumcision as a component of human immunodeficiency virus prevention. Am. J. Prev. Med. 40(3), e7-e8. On-line (behind paywall): http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S0256-95742012000300012

WHO (2016) Voluntary medical male circumcision for HIV prevention in 14 priority countries in east and southern Africa. Progress Brief, June. On-Line: www.who.int/iris/bitstream/10665/246174/1/WHO-HIV-2016.14-eng.pdf

Evidence vs. Ideology: Intactivism and the HIV Epidemic

This article, by a contributor who wishes to remain anonymous, summarizes the early history of the discovery of the HIV/foreskin connection, and the reaction of intactivists to this hugely important, and life-saving, finding. Whilst the fact that circumcision is highly protective against female to male transmission of HIV is now consensus among all reputable medical and scientific bodies, intactivists are ideologically opposed to circumcision and cannot tolerate the idea that the procedure might actually have significant benefits. The article here describes the distortions, misrepresentations and denialism the intactivists engaged in to try and discredit the growing body of evidence against them. It covers the period from the initial indications in 1986 that circumcision may protect against HIV up to just before publication of the randomized controlled trials (in 2005 & 2007) that settled the matter – at least for everyone not resident in the crankosphere. Like creationists, intactivists are still stuck in denial and continue to rail against the evidence to this day, but that’s pseudo-scientists for you.

Evidence vs. Ideology: Intactivism and the HIV Epidemic

Although the earliest suggestions that circumcision might protect against HIV were printed in 1986, (Alcena, 1986; Fink, 1986) epidemiological evidence would not emerge for another two years. By the end of 1988, two studies were published finding evidence of increased risk of HIV in uncircumcised males (Greenblatt, 1988; Simonsen, 1988). The second, by Simonsen et al., (Simonsen, 1988) received some attention from the press.

These events evidently made an impact upon the anti-circumcision lobby. In 1989, Snyder made a presentation at NOCIRC’s symposium in 1989, in which he described the “recent speculation” that circumcision might protect against HIV as “most cruel” (Snyder, 1989a). In a second presentation only days later to the California Medical Association, (Snyder, 1989b) he referred to Simonsen’s study, asked why American men had not been protected by the fact that many were circumcised, and argued that circumcision had “failed to prevent” HIV and other conditions. This suggests that he had failed or refused to understand the concept of reduction, as opposed to elimination, of risk. By the end of that year, two more studies followed showing further evidence of a protective effect (Bongaarts, 1989; Cameron, 1989).

By 1991, a NOCIRC newsletter complained, “[the United States] circumcision rate increased eight-tenths of a percent from 1988 to 1989 following suspect reports that babies left intact are at increased risk for urinary tract infection and claims that men with foreskins are more likely to acquire AIDS” (NOCIRC, 1991). The anti-circumcision lobby may have felt threatened: the 1989 review of circumcision by the American Academy of Pediatrics had, to quote the American Medical Association, “reversed a long-standing opinion that medical indications for routine circumcision were lacking”(AMA, 1999).

The volume of published work concerning circumcision and HIV continued to increase until 1995, with 28 PubMed entries for that year alone – a cumulative total of 137 since 1986. At a NOCIRC symposium in 1996, Van Howe outlined the “problem” of the persistence of circumcision in the United States. Reasons, he believed, included “censorship in american [sic] medical journals” and “exploitation of the mainstream medical journals by pro-circumcision zealots.” He outlined his strategies, which included “aggressively countering the various circumcision “myths”” (Van Howe, 1996a). Such “myths” apparently included HIV: “[one] common scare tactic seen in the literature today posits that, if a child is not circumcised at birth, he is at risk for … HIV” (Van Howe, 1996b).

A year earlier, in a letter to the American Academy of Pediatrics, Van Howe claimed that several new studies “have shown that circumcised males are at higher risk for developing … HIV infections” (Van Howe, 1995). Investigation of his chosen sources is revealing. Of the three studies that Van Howe cited, two found no significant difference – the 95% confidence interval was 0.1-1.1 in one (Guimaraes, 1991) and 0.3-2.6 in another (De Perre, 1987). The third was a partner study (Chao, 1994). Van Howe himself later commented that most studies in this class “failed to show the foreskin to be a significant factor” (Van Howe, 1996).

Selectivity and especially misrepresentation of sources, as we shall see, would soon become a staple strategy. For example, in January 1998, Van Howe claimed that “large random cross-sectional studies which have the least degree of bias have found that circumcised men are more likely to develop or transmit HIV infection” (Van Howe, 1998). As Bailis observed, however, in all but one case, Van Howe’s references found, after adjustment for confounding factors, “no statistically significant association between HIV and circumcision” (Bailis, 1998).

Yet another tactic adopted was more disturbing. Not content with rejecting the evidence of a protective effect, anti-circumcision lobbyists argued that the opposite was true: circumcision, they argued, increased the risk of HIV! In 1998, Van Howe, along with Paul Fleiss and Frederick Hodges (both of whom have also presented at NOCIRC symposia; see NOCIRC, 2002), published a paper purporting to review the immunological functions of the prepuce (Fleiss, 1998). Remarkably, they claimed that “the most common reason given is that it [circumcision] inhibits the transmission of STDs, even though rigorously controlled studies have consistently shown that circumcised males are at greater risk for all major STDs than males whose penises are intact” (Fleiss, 1998). (Five of the six references cited in support of this claim report finding no significant differences.) In their view, the association between HIV and lack of circumcision was a new claim of “circumcision advocates”, and describe epidemiological data as an “attempt to verify this theory” of “an American circumciser … who first hypothesized that circumcision prevents HIV infection” (Fleiss, 1998).

Curiously, these authors claimed that “[t]he inner prepuce contains apocrine glands, which secrete … lysozyme” (Fleiss, 1998). In making such a statement, they overlooked two histological studies that found no evidence of glandular tissue in the preputial mucosa, (Taylor, 1996; Parkash, 1973). Remarkably, they cited one of these elsewhere in the same article (Taylor, 1996). By 1999, an anti-circumcision webpage attributed to George Hill declared that preputially-secreted lysozyme protected the owner against HIV (Hill, 1999). Disturbingly, this myth was repeated as though it were factual, and without further investigation, in a 2003 Cochrane Review (Siegfried, 2003). If nothing else, this makes clear the need for rigorous checking of bibliographies as a vital part of the peer-review process.

In 1999, Van Howe published a paper, purporting to be a meta-analysis, in which he concluded “When the raw data are combined, a man with a circumcised penis is at greater risk of acquiring and transmitting HIV than a man with a non-circumcised penis” (Van Howe, 1999). Articles were soon published which described its statistical flaws and – more importantly – the results of correct analysis of his sources (Moses, 1999; O’Farrell, 2000). These were subsequently cited in some 50 or so papers, including further meta-analyses (e.g., Weiss, 2000). Indeed, as the first published meta-analysis investigating the protective effect of circumcision against HIV, Van Howe’s 1999 paper may have accelerated research into this association.

Interestingly, the paper was cited in a comprehensive 387 reference review of techniques in meta-analysis (Barker, 2005), which noted that “Although this [Simpson] paradox is usually illustrated with hypothetical examples, in one metaanalysis that used simple data pooling to test the protective effect of circumcision against AIDS infection in sub-Saharan African men, conclusions were reached that were diametrically opposed to those of a subsequent metaanalysis in which appropriate techniques were used.”

It should be noted that, although there is no clear evidence of deliberate deception on Van Howe’s part, at least one author has cited these events as an example of Van Howe’s limited credibility (Schoen, 2007), noting Van Howe’s “long history of opposition to circumcision”. Interestingly, Van Howe authored another paper in the same year, in which he reviewed the evidence concerning STIs other than HIV (Van Howe, 1999a). As with the HIV meta-analysis, his reported results were at odds with those of other reviews, (Weiss, 2006; Moses, 1999) and indeed the source material. Perhaps the term “hostile review” should be introduced for papers such as these, which seek not to summarize but to contradict their source material.

References
Alcena, V. (1986) AIDS in Third World countries. N. Y. State J. Med., 86(8), 446.

AMA (1999) Report 10 of the Council on Scientific Affairs (I-99). American Medical Association. Retrieved 08-04-2007 from: http://www.ama-assn.org/ama/pub/category/13585.html [Link now broken, but here is an archived copy:
http://web.archive.org/web/20070206053938/http://www.ama-assn.org/ama/pub/category/13585.html]

Bailis, S.A. (1998) Circumcision and infectious diseases. Pediatr. Infect. Dis. J., 17(8), 762-3.

Barker, F.G., Carter, B.S. (2005) Synthesizing medical evidence: systematic reviews and metaanalyses. Neurosurg. Focus, 19(4), E5.

Bongaarts, J., Reining, P., Way, P., Conant, F. (1989) The relationship between male circumcision and HIV infection in African populations. AIDS, 3(6), 373-377.

Cameron, D.W., Simonsen, J.N., D’Costa, L.J., Ronald, A.R., Maitha, G.M., Gakinya, M.N., et al. (1989) Female to male transmission of human immunodeficiency virus type 1: risk factors for seroconversion in men. Lancet, 2(8660), 403-407.

Chao, A., Bulterys, M., Musanganire, F., Habimana, P., Nawrocki, P., Taylor, E., et al. (1994) Risk factors associated with prevalent HIV-1 infection among pregnant women in Rwanda. National University of Rwanda-Johns Hopkins University AIDS Research Team. Int. J. Epidemiol., 23(2), 371-380.

Dagher, R., Selzer, M.L., Lapides, J. (1973) Carcinoma of the Penis and the Anti-Circumcision Crusade. J. Urol., 110, 79-80.

Fink, A.J. (1986) A possible explanation for heterosexual male infection with AIDS. N. Engl. J. Med., 315(18), 1167.

Fleiss, P., Hodges, F., Van Howe, R.S. (1998) Immunological functions of the human prepuce. Sex. Trans. Inf., 74(5), 364-7.

Foley, J.M. (1966) The unkindest cut of all. Fact Magazine, July-August, 3(4), 2-9.

Greenblatt, R.M., Lukehart, S.A., Plummer, F.A., Quinn, T.C., Critchlow, C.W., Ashley, R.L., et al. (1988) Genital ulceration as a risk factor for human immunodeficiency virus infection. AIDS, 2(1), 47-50.

Guimaraes, M., Castilho, E., Sereno, A., Cavalcante, S., Lima, L.A., Gomes, V., et al. (1991) Heterosexual transmission of HIV-1: a multicenter study in Rio de Janeiro, Brazil. VII International Conference on AIDS. Florence, June 1991 [abstract MC3098]

Hill, G. (1999) Summary of evidence that the foreskin and lysozyme may
protect against HIV infection. Retrieved 08-04-2007 from:
http://www.foreskin.org/lysozyme.htm

Morgan, W.K.(1965) The rape of the phallus. JAMA, 193, 223.

Moses, S., Nagelkerke, N.J., Blanchard, J. (1999) Analysis of the scientific literature on male circumcision and risk for HIV infection. Int. J. STD AIDS, 10(9), 626-8.

NOCIRC/About http://www.nocirc.org/about.php

NOCIRC (1991) Newsletter. National Organization of Circumcision Information and Resource Centers, Fall 1991. Retrieved 28-12-2006 from
http://www.nocirc.org/publish/9-91.pdf.

NOCIRC 2002 http://www.nocirc.org/symposia/seventh/symposium.pdf

O’Farrell, N., Egger, M. (2000) Circumcision in men and the prevention of HIV infection: a ‘meta-analysis’ revisited. Int. J. STD AIDS, 11(3), 137-42.

Okino, B.M., Yamamoto, L.G. (2004) Survey of Internet web sites on circumcision. Clin. Pediatr. (Phila.). 43(7), 667-9.

Parkash, S., Jeyakumar, K., Subramanya, K., Chaudhuri, S. (1973) Human sub-preputial collection: its nature and formation. J. Urol., 110, 211-212.

Schoen, E.J. (2007) Meatal stenosis following neonatal circumcision. Clin. Pediatr. (Phila.). 46(1), 86.

Siegfried, N., Muller, M., Volmink, J., Deeks, J., Egger, M., Low, N., Weiss, H., Walker, S., Williamson, P. (2003) Male circumcision for prevention of heterosexual acquisition of HIV in men (Cochrane Review). In: The Cochrane Library, Issue 3, 2003. Oxford: Update Software

Silverman, E.K. (2004) Anthropology and circumcision. Annual Review of Anthropology, 33, 419-445.

Simonsen, J.N., Cameron, D.W., Gakinya, M.N., Ndinya-Achola, J.O., D’Costa, L.J., Karasira, P, et al. (1988) Human immunodeficiency virus infection among men with sexually transmitted diseases. Experience from a center in Africa. N. Engl. J. Med., 319(5), 274-278.

Snyder, J.L. (1989a) The circumcision problem: inertia of the scientific community. Retrieved 08-04-2007 from
http://www.nocirc.org/symposia/first/snyder1b.html.

Snyder, J.L. Testimony against circumcision. Retrieved 08-04-2007
from http://www.nocirc.org/symposia/first/snyder1a.html.

Taylor, J.R., Lockwood, A.P., Taylor, A.J. (1996) The prepuce: specialized mucosa of the penis and its loss to circumcision. Br. J. Urol., 77, 291-295.

Van de Perre, P., Carael, M., Nzaramba, D., Zissis, G., Kayihigi, J., Butzler, J.P. (1987) Risk factors for HIV seropositivity in selected urban-based Rwandese adults. AIDS, 1, 207-211.

Van Howe, R.S. (1995) Letter to American Academy of Pediatrics. Retrieved 08-04-2007 from
http://www.cirp.org/library/statements/letters/1995-11_VanHowe/

Van Howe, R.S.(1996a) Why does circumcision persist in the United States? Retrieved 08-04-2007 from http://www.nocirc.org/symposia/fourth/vanhowe.html.

Van Howe, R.S. (1996b) Why does circumcision persist in the United States? In: Denniston GC, Milos MF, editors. Sexual Mutilations: A Human Tragedy. New York: Plenum Press; 1997. p. 111-119. (note: combine with 1996a)

Van Howe, R.S. (1998) Circumcision and infectious diseases revisited. Pediatr. Infect. Dis. J., 17(1), 1-6.

Van Howe, R.S. (1999) Circumcision and HIV infection: review of the literature and meta-analysis. Int. J. STD AIDS, 10, 8-16.

Van Howe, R.S. (1999a) Does circumcision influence sexually transmitted diseases?: A literature review. BJU Int., 83, Suppl 1, 52-62.

Weiss, H.A., Quigley, M.A., Hayes, R.J. (2000) Male circumcision and risk of HIV infection in sub-Saharan Africa: a systematic review and meta-analysis. AIDS, 14(15), 2361-70.

Weiss, H.A., Thomas, S.L., Munabi, S.K., Hayes, R.J. (2006) Male circumcision and risk of syphilis, chancroid, and genital herpes: a systematic review and meta-analysis. Sex Transm. Infect., 82(2), 101-9.

WHO/UNAIDS (2007) Technical Consultation. Male Circumcision and HIV Prevention: Research Implications for Policy and Programming Montreux, 6- 8 March 2007.
http://www.who.int/entity/hiv/mediacentre/MCrecommendations_en.pdf

 


Risk Compensation – an Intactivist Half-Truth

By Stephen Moreton PhD.

April 2020

This article is an updated version of an earlier one dated April 2017. Many more references have been added, and the argument discussed in more detail. But the conclusions remain the same.

The risk compensation argument takes the form:

Circumcised men may believe they are immune to HIV and so abandon safe sex practices thereby putting themselves at risk. Consequently, implementing circumcision as an anti-HIV measure may actually increase HIV.

This is the gist of an argument used by husband and wife intactivists Robert S. Van Howe & Michelle Storms (2011), and uncritically copied by intactivist Brian Earp (Earp, 2012). For their blunders Earp got a well-deserved spanking on the Quackdown blog (http://www.quackdown.info/article/getting-circumcision-science-right-media/?print=true) and Van Howe & Storms were rebutted in detail by Morris et al. (2011).

Undeterred, Earp used it again 7 years later (Earp & Darby, 2019). It crops up commonly in discussions of circumcision in relation to HIV, particularly in Africa where circumcision is heavily promoted to protect against infection during heterosexual intercourse. In this context it is commonly referred to as “risk compensation”. It is also an example of a half-truth (see: http://circfacts.org/sloppy-logic/#slog20 for other examples of this common intactivist ploy).

As with so much relating to circumcision and African HIV, there exists a wealth of research on this topic, allowing us to point to what is true in the argument, and to what has been left out. Firstly, what is true? Yes, risk compensation can happen. Earp & Darby (2019) cite a series of studies “(Riess et al., 2010; Grund and Henninck, 2012; Feldblum et al. 2015; Jung et al., 2016; Kibira et al., 2017; but see Shi et al., 2017)” and go on to say that such behaviour may increase a man’s risk of infection “over and above whatever advantage he may initially have had (in this respect) in being circumcised.”

So, do the references they cite support the claim? Not very well – they are summarised below. Risk compensation is hereafter abbreviated to “RC”, male circumcision to “MC”.

Riess et al., 2010: In Kenya, 5 out of 30 participants reported riskier behaviour post-MC, the rest either the same (17) or less (8). In only one case did this behaviour take the form of lower condom use, and even that was temporary and only with his wife. In the other 4 cases the men increased their number of sexual partners, but maintained consistent condom use, thereby offsetting the risk. In other words, RC was slight, and only in a minority of men. In addition, 6 men reported fewer coital injuries post-MC and attributed this to their MC (a clear health benefit). Several also said that condom use was easier. The authors conclude “MC does not necessarily lead to risk compensation” and that “Counselling placed within or related to MC programs appears to be influential in promoting effective behavoir change among participants”. In short, RC can be prevented by education.

Grund & Hennick, 2012: Qualitative study in Swaziland (now called eSwatini), no numbers given. Most men reported engaging in less risky behaviour post-MC and attributed this to the counselling they received when they were circumcised. It was also reported by some men that condoms were easier to use and less prone to slippage or breakage after MC. “Very few participants cited engaged in occasional ‘reckless’ sexual behavior following circumcision” and this was temporary as they tried out their circumcised penis, “during a brief period of sexual experimentation shortly after circumcision. … These men were often those who reported receiving more limited counselling about circumcision. This underscores the importance of providing HIV counselling with circumcision, as it promotes protective behaviors and reduces sexual risk-taking and risk compensation.” In other words, RC was slight, temporary and could be countered by education.

Feldblum et al., 2015: In Kenya, 76.9 % used condoms more after MC, and 94.4 % found that condom use was easier. 28.1 % reported an increased number of partners post-MC, but they were also using condoms more. Only 3.4 % used condoms less. It is not clear that this increase in sexual partners was related to RC. As almost half of the participants were youths aged 20 or less it may simply represent these lads maturing and having more opportunities for sex. Again, any RC effect in this study is slight.

Jung et al., 2016: This is marked “PRELIMINARY DRAFT. PLEASE DO NOT CITE OR CIRCULATE WITHOUT AUTHORS’ PERMISSION.” It is not a peer-reviewed paper. Self-report by Malawian males found no difference in RC except inconsistent condom use. But HSV-2 testing indicated high rates in circumcised males, despite MC being modestly protective against HSV-2. No difference was reported for HIV testing.

Kibira et al., 2017: Uganda. Most men maintained safe sex behaviour, but some had a false belief that they should have one-off sex with another partner without a condom before returning to their usual partner. 2 men had sex before healing was complete. So, again, RC is only seen in a minority, and was temporary.

Shi et al., 2017: Surveyed data from 10 African countries and found no evidence of RC.

So, four of the five papers E&D cite in support of RC found only minorities of men exhibit RC, while majorities did not. Even then it was often only temporary and could be countered by education. The 5th study is not a peer-reviewed paper and its own authors say it is a draft not to be cited (which begs the question why they posted it on the internet). There are many other studies finding no RC. E&D ignore most of these, giving a brief nod only to one (Shi et al., 2017) this is despite Earp having been made aware of others in 2012 on the Quackdown blog. These were:

http://i-base.info/htb-south/1787 – a commentary on the Orange Farm results that observed no RC, and

Mattson et al., (2008) who found that circumcision did not result in increased RC in a large (n = 1319) study in Kenya.

Other studies that found no RC are:

Agot et al., 2007: No RC in Kenya.

Ansari et al., 2017: No RC in Indonesia.

Ayiga & Letamo, 2011: No significant difference in condom use between circumcised and non-circumcised men in Botswana.

Chikutsa et al., 2013: No RC in Zimbabwe.

Godlonton et al., 2016: No RC in Malawi.

Govender et al., 2018: No RC in S. Africa.

Harper, 2016: No RC in Dominican Republic.

Johnson et al., 2013: No difference in condom use between circumcised & non-circumcised men in S. Africa.

Kabwama et al., 2018: Meta-analysis of cohort & RCT studies. No difference in condom use between circumcised and non-circumcised men.

Lau et al., 2015: MC correlates with safe sexual behaviour in 11 African countries.

L’Engle et al., 2014: No RC in Kenya.

Maffioli 2017: No RC in Lesotho.

Mukudu et al., 2019: No RC in Soweto.

Ortblad et al., 2019: Large cohort study. No RC in S. Africa.

Westercamp et al., 2014: No RC in Kenya.

Westercamp et al., 2017: No RC in Kenya.

That’s 18 published, peer-reviewed studies that found no RC, of which at least 13 were likely available to E&D when they wrote their 2019 paper. Yet they cite only one.

Studies which do find evidence of RC:

Balekang, G.B., Dintwa, K.F. 2016: In Botswana, riskier behaviour was seen in circumcised men, but was also linked to lower education.

Chatsika et al., 2020: In Malawi most participants knew that MC was only partly effective, but 38.2 % still engaged in risky behaviour. But there is no non-circumcised control group. 12.7 % believed that MC gave total protection against HIV.

Crosby et al., 2016: Circumcised black US gay men less likely to use condoms during insertive anal sex, but the difference is barely significant (p = 0.04).

Greevy et al., 2018: A minority of women in KwaZulu-Natal lacked accurate knowledge about MC, but this could be improved with education. Despite this there was no actual evidence of RC by these women.

Ilyasu et al., 2013. Risky sexual behaviour common in Nigerian men, but there was no control of non-circumcised ones to compare with. Therefore, this study cannot be used as evidence for RC. There was a lack of knowledge about MC and health benefits.

Kalichman et al., 2018: Some RC in S. African women, and lack of accurate knowledge. Need for education.

Kibira et al., 2014: RC seen in circumcised Ugandan men, but they also had lower HIV, so they were still better off despite engaging in riskier behaviour.

Shi et al., 2020: RC associated with traditional circumcision, but not medical circumcision, in Africa. Likely due to lack of education.

Wang et al., 2016: 26.9 % of Chinese men intending to be circumcised expressed intention to commit RC. This was associated with inaccurate knowledge. Education needed.

Wilson et al., 2014: Some RC in Kenyans, but only in those who did not believe that MC was effective against HIV. (An issue that education would address.)

Zungu et al., 2016: No RC with respect to sexual debut, risk perception and condom use, but some with respect to number of partners and alcohol use, in S. Africa.

That’s 16 studies of which one can be discounted due to lack of a non-circumcised control, one that has not been peer-reviewed and even the authors say it should not be cited, and one (Greevy et al., 2018) that only indicated potential for RC, not actual RC. That leaves 13 usable studies confirming (a slight) RC effect, 12 of them available when Earp & Darby wrote their 2019 paper, in which they cite five. So, they cite just 1 of 12 contrary papers, and 5 of 12 (weakly) supportive ones. That is bias.

To summarise, just over half of the studies find no evidence for RC. When evidence is found, the effect is usually slight, often temporary, and in all cases likely to be amenable to education. In the Ugandan study where risk compensation was observed (Kibira et al., 2014) the circumcised men were still less likely to become infected with HIV than their non-circumcised peers. There is a hint of this in the famous RCTs too. In the Kenyan and S. African trials there was a very slight tendency for the circumcised males to have more sexual partners but still they were better off than their uncut peers (see http://circfacts.org/medical-benefits/hivaids/#hiv12). Circumcision is so effective that even if some men do take more risks, they are still better off without their foreskins! In short, we know that MC does not increase a man’s risk “over and above whatever advantage he may initially have had (in this respect) in being circumcised.”

Even by the time Van Howe, Storms, and Earp, wrote their earlier articles there were studies which they ignored, showing that when men were properly counselled about circumcision, so that they understood it provided partial protection only and they should continue to be sexually responsible (wear condoms etc.), then they did so. In short, with proper education risk compensation does not happen, so it is really an argument to ensure that men are educated, rather than an argument against circumcision.

Despite all this, Earp and others persist in using this argument (e.g. in Earp & Darby, (2019), in dated articles still on intactivist websites, in some of the submissions to the CDC in 2014, and in comments threads). It is high time they gave it a rest.

References

Agot, K.E., Kiarie, J.N., Nguyen, H.Q., Odhiambo, J.O., Onyango, T.M., Weiss, N.S. (2007) Male circumcision in Siaya and Bondo Districts, Kenya. J. Acquir. Immun. Defic. Syndrome, 44, 66-70. On-line: https://journals.lww.com/jaids/Fulltext/2007/01010/Male_Circumcision_in_Siaya_and_Bondo_Districts,.10.aspx

Ansari, M.R., Lazuardi, E., Wignall, F.S., Karma, C., Sumule, S.A., Tarmizi, S.N., Magnani, R. (2017) Voluntary male medical circumcision to prevent HIV in Tanah Papua, Indonesia: Field trial to assess acceptability and feasibility. Current HIV Res., 15(5), 361-71. On-line abstract: https://www.ncbi.nlm.nih.gov/pubmed/28990535

Ayiga, N., Letamo, G. (2011) Impact of male circumcision on risk compensation through the impediment of condom use in Botswana. African Health Sciences, 11(4), 550-9. On-line: http://www.bioline.org.br/request?hs11111

Balekang, G.B. & Dintwa, K.F. (2016) A comparison of risky sexual behaviours between circumcised and uncircumcised men aged 30-44 years in Botswana. African Health Sciences, 16(1), 105-115. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4915434/

Chatsika, J.Z., Kumitawa, A., Samuel, V., Azizi, S.C., Jumbe, V.C. (2020) Voluntary male medical circumcision and sexual practices among sexually active circumcised men in Mzuzu, Malawi: a cross-sectional study. BMC Public Health, 20:211. On-line: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-020-8309-5

Chikutsa, A., Ncube, A.C., Mutsau, S. (2013) Male circumcision and risky sexual behaviour in Zimbabwe: Evidence from the 2010-11 Zimbabwe demographic and health survey. USAID DHS working papers, No. 102. On-line: https://www.dhsprogram.com/publications/publication-WP102-Working-Papers.cfm

Crosby, R.A., Graham, C.A., Mena, L., Yarber, Y.L., Sanders, S.A., Milhausen, R.R., Geter, A. (2016) Circumcision status is not associated with condom use and prevalence of sexually transmitted infections among young black MSM. AIDS Behav., 20, 2538-42. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5800511/

Earp, B. (2012) When bad science kills, or how to spread AIDS. Article posted on Oxford University’s Practical Ethics blog. http://blog.practicalethics.ox.ac.uk/2012/05/when-bad-science-kills-or-how-to-spread-aids/ (Accessed 12 April 2020).

Earp, B., Darby, R. (2019) “Circumcision, autonomy and public health”, Public Health Ethics, 12(1), 64-81. Abstract: https://academic.oup.com/phe/article-abstract/12/1/64/4764011

Feldblum, P.J., Oketch, J., Ochieng, R., Hart, C., Kiyuka, G., Lai, J.J., Veena, V. (2015) Longer-term follow-up of Kenyan men circumcised using the ShangRing device. PLoS ONE 10(9), e0137510. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4569077/

Godlonton, S., Munthali, A., Thornton, R. (2016) Responding to risk: Circumcision, information, and HIV prevention. Review of Economics and Statistics, 98(2), 333-49. Abstract: https://www.mitpressjournals.org/doi/10.1162/REST_a_00516

Govender, K., George, G., Beckett, S., Montague, C., Frohlich, J. (2018) Risk compensation following medical male circumcision: Results from a 1-year prospective cohort study of young school-going men in KwaZulu-Natal, South Africa. Int. J. Behav. Med., 25(1), 123-130. Abstract: https://link.springer.com/article/10.1007/s12529-017-9673-0

Greevy, C., King, R., Haffejee, F. (2018) Male circumcision for HIV prevention: female risk compensatory behaviour in South Africa. AIDS Care, 30(9), 1083-9. Abstract: https://www.tandfonline.com/doi/full/10.1080/09540121.2018.1476658

Grund, J. M., and Hennink, M. M. (2012). A qualitative study of sexual behavior change and risk compensation following adult male circumcision in urban Swaziland. AIDS Care, 24, 245–251. Abstract: https://www.tandfonline.com/doi/abs/10.1080/09540121.2011.596516

Harper, K.N. (2016) No changes in sexual behaviour after voluntary medical male circumcision in the Dominican Republic. AIDS, 30(9), N13. On-line: https://journals.lww.com/aidsonline/fulltext/2016/06010/No_changes_in_sexual_behaviour_after_voluntary.2.aspx

Ilyasu, Z., Abubakar, I.S., Sani, I.H., Jibo, A.M., Karaye, I.M., Salihu, H.M., Aliyu, M.H. (2013) Male circumcision and HIV risk behavior among university students in Northern Nigeria. American J. Men’s Health, 7(2), 94-101. On-line: https://journals.sagepub.com/doi/full/10.1177/1557988312456698

Johnson, S., Kincaid, D.L., Figueroa, M.E, Delate, R., Mahlasela, L., and Magni, S. (2013). The Third National HIV Communication Survey, 2012. Pretoria: JHHESA. On-line: https://www.ccisa.org.za/sites/default/files/hiv_survey.pdf

Jung, J., Kim, B., Kim, H. B., and Cristian, P.-E. (2016). Long-term effects of male circumcision on risky sexual behaviors and STD infections: Evidence from Malawian schools. KDI School Working Paper Series. Working Paper 16-08, pp. 1–43. Available from https://ssrn.com/abstract=2833153 [accessed 10 April 2020. Earp & Darby name the first author as Jung, in the link it is Kim].

Kabwama, S.N., Ssewanyana, D., Berg-Beckhoff, G. (2018) The association between male circumcision and condom use behavior – a meta-analysis. Mater. Sociomed., 30(1), 62-6. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5857052/

Kalichman, S., Mathews, C., Kalichman, M., Eaton, L.A., Nkoko, K. (2018) Male circumcision for HIV prevention: Awareness, risk compensation, and risk perceptions among South African women. Global Public Health, 13(11), 1682-90. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6396304/

Kibira, S.P.S., Nansubuga, E., Tumwesigye, N.M., Atuyambe, L.M., Makumbi, F. (2014) Differences in risky sexual behaviors and HIV prevalence of circumcised and uncircumcised men in Uganda: evidence from a 2011 cross-sectional national survey. Reproductive Health, 11(25), 1-8. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3994333/

Kibira, S. P. S., Atuyambe, L. M., Sandøy, I. F., Makumbi, F. E., and Daniel, M. (2017). “Now that you are circumcised, you cannot have first sex with your wife”: Post circumcision sexual behaviours and beliefs among men in Wakiso District, Uganda. Journal of the International AIDS Society, 20, 21498. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5515054/

Lau, F.K., Jayakumar, S., Sgaier, S.K. (2015) Understanding the socio-economic and sexual behavioural correlates of male circumcision across eleven voluntary medical male circumcision priority countries in southeastern Africa. BMC Public Health, 15(813) 1-11. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4546248/

L’Engle, K., Lanham, M., Loolpapit, M., Oguma, I. (2014) Understanding partial protection and HIV risk and behavior following voluntary medical male circumcision rollout in Kenya. Health Educ. Res., 29(1), 122-130. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3894669/

Maffioli, E.M. (2017) Is traditional male circumcision effective as an HIV prevention strategy? Evidence from Lesotho. PLoS ONE, 12(5): e0177076. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5428932/

Mattson, C.L., Campbell, R.T., Bailey, R.C., Agot, K., Ndinya-Achola, J.O., Moses, S. (2008) Risk compensation is not associated with male circumcision in Kisumu, Kenya: A multi-faceted assessment of men enrolled in a randomized controlled trial. PLoS One, 3(6), 1-9. On-line: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0002443

Morris, B.J., Waskett, J.H., Gray, R.H., Halperin, D.T., Wamai, R., Auvert, B., Klausner, J.D. (2011) Exposé of misleading claims that male circumcision will increase HIV infections in Africa. J. Public Health Africa, 2(e28), 117-122. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5345501/

Mukudu, H., Dietrich, J., Otwombe, K., Manentsa, M., Hlongwane, K., Haas-Kogan, M., Sartorius, B., Martinson, N. (2019) Voluntary medical male circumcision (VMMC) for prevention of heterosexual transmission of HIV and risk compensation in adult males in Soweto: Findings from a programmatic setting. PLoS ONE, 14(3): e0213571. On-line: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0213571

Ortblad, K.F., Harling, G., Chimbindi, N., Tanser, F., Salomon, J.A., Bärnighausen, T. (2019) Does incident circumcision lead to risk compensation? Evidence from a population cohort in KwaZulu-Natal, South Africa. J. Acquir. Immun. Defic. Syndr., 80(3), 269-75. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6375765/

Riess, T.H., Achieng, M.M., Otieno, S., Ndinya-Achola, J.O., Bailey, R.C. (2010) “When I was circumcised I was taught certain things”: risk compensation and protective sexual behaviour among circumcised men in Kisumu, Kenya. PLoS One, 5(8), 1-9. On-line: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0012366

Shi, C-F., Li, M., Dushoff, J. (2017) Evidence that promotion of male circumcision did not lead to sexual risk compensation in prioritized Sub-Saharan countries. PLoS ONE, 12(4), e0175928. On-line: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0175928

Shi, C., Li, M., Dushoff, J. (2020) Traditional male circumcision is associated with sexual risk behaviors in sub-Saharan countries prioritized for male circumcision. AIDS and Behavior, 24(3), 951-9. On-line: https://www.ncbi.nlm.nih.gov/pubmed/30955178

Van Howe, R.S., Storms, M. (2011) How the circumcision solution in Africa will increase HIV infections. J. Public Health Africa, 2(e4), 11-15. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5345479/

Wang, Z., Feng, T., Lau, J.T.F. (2016) Prevalence and associated risk factors of behavioral intention for risk compensation following voluntary medical male circumcision among male sexually transmitted disease patients in China. AIDS Care, 28(1), 1332-7. Abstract: https://www.ncbi.nlm.nih.gov/pubmed/27120407

Westercamp, N., Agot, K., Jaoko, W., Bailey, R.C. (2014) Risk compensation following male circumcision: Results from a two-year prospective cohort study of recently circumcised and uncircumcised men in Nyanza province, Kenya. AIDS Behav., 18(9), 1764-75. On-line abstract: https://www.ncbi.nlm.nih.gov/pubmed/25047688

Westercamp, N., Jaoko, W., Mehta, S., Abuor, P., Siambe, P., Bailey, R.C. (2017) Changes in male circumcision prevalence and risk compensation in the Kisumu, Kenya population 2008-2013. J. Acquir. Immun. Defic. Syndr., 72(2), e30-e37. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5233580/

Wilson, N.L., Xiong, W., Mattson, C.L. (2014) Is sex like driving? HIV prevention and risk compensation. J. Dev. Econ., 106, 78-91. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4794434/

Zungu, N.P., Simbayi, L.C., Mabaso, M., Evans, M., Zuma, K., Ncitakalo, N., Sifunda, S. (2016) HIV risk perception and behavior among medically and traditionally circumcised males in South Africa. BMC Public Health, 16:357. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4845373/


The USA/Europe comparison: another intactivist fail

By Stephen Moreton, PhD

March 2017

In a discussion of circumcision and HIV it will not be long before an intactivist objects with something along the lines of:

“The USA has the highest rate of circumcision and the highest rate of HIV of any developed nation. Contrast with Europe, which has low rates of both. Therefore circumcision is not protective against HIV.”

This argument is common in comments threads and intactivist propaganda. It is an intactivist favourite, so merits detailed examination.

Before we proceed any further, we must do what should be done with all intactivist claims. We must fact-check it. We need to know the circumcision and HIV rates across the developed world. Such data is readily available. The Human Development Index (HDI) ranks countries by how developed they are. It may be viewed here: https://en.wikipedia.org/wiki/Human_Development_Index  Restricting ourselves to only those listed as “highly developed” we find 49 countries listed for 2014. These are in the table below, ranked by HDI (most developed first). The most up to date list of circumcision rates by country is by Morris et al., (2016), and is used in the table below. Finally, CIA data on HIV prevalence for the age group 15 to 49, in the years 2011 & 2014 is freely available here:
https://en.wikipedia.org/wiki/List_of_countries_by_HIV/AIDS_adult_prevalence_rate .

Country% Circumcised% HIV
Norway3.00.15
Australia26.60.17
Switzerland5.90.35
Denmark5.30.16
Netherlands5.70.2
Germany10.90.15
Ireland0.90.28
USA71.20.6
Canada31.90.3
New Zealand33.00.1
Singapore14.90.1
Hong Kong280.1
Liechtenstein4.80.1
Sweden5.10.18
UK20.70.33
Iceland0.10.3
South Korea77.00.1
Israel91.70.2
Macau0.1No data
Luxembourg2.40.3
Japan90.1
Belgium22.6*0.3
France140.4
Austria5.80.4
Finland0.80.1
Thailand23.41.13
Slovenia8.50.08
Spain6.60.42
Italy2.60.28
Czech Republic0.10.05
Greece4.70.2
Estonia0.251.30
Brunei51.90.1
Cyprus22.70.06
Qatar77.50.1
Andorra1.1No data
Slovakia0.150.02
Poland0.10.07
Lithuania0.20.1
Malta0.30.1
Saudi Arabia97.10.1
Argentina2.90.47
UAE760.2
Chile0.20.29
Portugal0.60.7
Hungary0.80.1
Bahrain81.20.3
Latvia0.40.7
Croatia1.30.1
Kuwait86.40.1
Montenegro18.5No data

* actually 15% according to Bronselaer (2013).

HIV data for Andorra, Macau and Montenegro is lacking.

Straight away we can spot a few nations that do not fit the intactivists’ narrative. Highly developed nations with HIV rates higher than the USA’s 0.6 % are (HIV rate in brackets): Thailand (1.13), Estonia (1.30), Portugal (0.7) and Latvia (0.7). Likewise, South Korea, Israel, Saudi Arabia, UAE, Bahrain and Kuwait all have circumcision rates higher than in the USA, but lower HIV prevalence.

As the most common form of the argument compares USA with Europe, we shall look at that in more detail. Whilst a few European countries do have higher HIV prevalence than the USA, most are noticeably lower. By tediously adding up the total numbers infected, and the total populations of each country, one can arrive at an average prevalence of 0.3 % HIV for Europe, whether “Europe” is defined as everything between Russia and the Atlantic, or just the European Union. This is half the USA’s 0.6 %. And circumcision is, indeed, much less common in Europe than in America. So, superficially, it seems as if the intactivists may have a point.

But they don’t. In using this argument intactivists are ignoring one of the most common, and well-known, pitfalls of epidemiology: confounding factors. Circumcision is just one of a variety of factors that influence the transmission of HIV (and other STIs). If those other factors are strong enough they may hide the influence of the factor of interest, in this case circumcision. This is called confounding, and is a fundamental weakness of country to country comparisons. There may be some other factor, or factors, influencing the result. Or, to put it simply, correlation is not causation.

Many factors influence HIV transmission: condom use, availability of anti-retroviral therapy (ART), promiscuity, genetic susceptibility, unhygienic tribal circumcision rituals, migration patterns, use of prostitutes, intravenous drug use, homosexual sex and, doubtless, others, all affect HIV transmission and, therefore, prevalence. Some (e.g. tribal circumcision) are of little or no relevance here, but the others do matter to varying degrees.

So comparing one country with another is very risky. What appears to be a correlation with the factor of interest may be due to something entirely unrelated. This is why randomized controlled trials (RCTs) are so beloved of epidemiologists. Their design specifically eliminates confounding factors, allowing the researchers to focus on the factor under study, confident any effect they observe is not due to interference from something unrelated. RCTs are often referred to as the “gold standard” of epidemiology for this reason, and their results taken as authoritative.

The connection between foreskins and HIV was originally detected on the basis of regional comparisons. There was a tendency, in Africa, for HIV to be more common in regions where most men were not circumcised. But, as intactivists pointed out (and still do) there are a few places where the association is reversed. Or (as they were keen to point out) the observed association could be due to confounding factors. Maybe men in one region were circumcised, but were also prudish in their sexual habits. Men in another region were not circumcised, and also promiscuous. Or perhaps circumcised men in one country lived in isolated rural communities, with little contact with the outside world, whilst non-circumcised ones belonged to urban communities, or travelled far in search of work, so had more exposure to the virus. And so on.

To eliminate these possibilities, three large RCTs were conducted independently in three different African countries. Each one showed conclusively that circumcision does indeed protect against female to male transmission of HIV, and the observed trends across countries really were due to this effect, and not to confounding factors.

So, given that circumcision does protect against HIV, why do we fail to see this in the USA/Europe comparison? In light of the above discussion the answer is obvious: confounding factors. So what are those factors?

To begin with, we now know that HIV has been present in the USA since about 1971, possibly even earlier (Worobey et al., 2016), although not first detected there until the 1980s. It arrived in the USA via Haiti, and was already established in the gay community there by the time it began turning up in Europe. So the epidemic had a head start in the USA.

This, however, is not the main confounder. Comparing data for unwanted pregnancy rates reveals a startling disparity between the USA and the rest of the developed world. The bar chart on page 4 of UNICEF (2001) is a “must see”. The USA has almost twice as many teenage pregnancies as the next developed nation (the UK) and about four times the European Union average. When it comes to unwanted pregnancies, the USA is the capital of the developed world. One can repeat the exercise for abortion rates, and find the same. The same report noted that 40 % of sexually active US teenagers do not use contraception.

In short, there are huge differences in sexual behaviour between America and the rest of the developed world. Americans are more promiscuous, are sexually active at younger ages, and are less likely to use condoms, than many other countries. These are all hugely important confounding factors, likely to have major effects on HIV transmission. And they are completely ignored every time intactivists deploy the USA/Europe comparison.

Their argument is hypocritical too. Intactivists advocate sexual restraint, and condoms, as effective ways of controlling HIV (which they are, if used) seemingly unaware that these are likely reasons why Europe is so far ahead of the USA with respect to HIV (and other STIs). They also forget that they were quick to point to confounding factors as an explanation for the observed foreskin/HIV association prior to the RCTs, but ignore them as an explanation for the USA/Europe disparity. Do they think that confounding factors only apply to arguments that favour circumcision?

So we can dismiss the intactivists’ USA/Europe comparison as baseless, ignorant and hypocritical, but it doesn’t end there. It is superficial too. A more detailed look at the data reveals a story the intactivists would not like told.

HIV has been in the USA since about 1971, yet in all that time it has remained largely confined to high risk groups (mostly men who have sex with men, also some who share needles (CDC, 2016)). In none of these cases does circumcision make a difference. It works for female to male heterosexual transmission, and when HIV is transmitted that way, it is disproportionately so in black and Hispanic people. Now this could be down to confounding factors (socio-economic status may affect sexual behaviour, for instance) but they also happen to be the groups in which circumcision is least common (CDC, 2010)! Could it be circumcision that has kept the infection from crossing over into the heterosexual population in the USA?

It is difficult to gather good epidemiological evidence in settings where the prevalence of the infection under study is low. It requires large sample sizes, and/or long study durations. This makes Africa a good setting for HIV studies because it is so common there. But, in countries with mere fractions of a percent prevalence, it is hard to tease out the effect of one factor (e.g. circumcision) from the others. Consequently, observational studies in low-HIV settings are less likely to find an association, and there are only a handful of such studies. Some find no link, and are eagerly cited by intactivists, but some do.

A study of heterosexual African-Americans in Baltimore found that HIV was more common in those not circumcised (Warner et al., 2009), and a recent comparison between Israel, France and Netherlands (Chemtob et al., 2015) found a higher HIV incidence in the latter two, non-circumcising, cultures. Again, this might be confounding, but the comparison is better than the USA/Europe one, as sexual behaviour between these countries is more alike. Data on specifically female to male transmission in developed countries is very scarce but, given that what is known of the mechanisms of transmission indicates the foreskin is involved, there is no reason in principle why circumcision should not afford protection regardless of country.

In short, intactivists are showing their ignorance, and double standards, whenever they make the USA/Europe comparison. Looked at in more detail, the data do not support their cause.

References:
Bronselaer, G. (2013) Reply, BJU Int., 111, E270-1.

CDC (2010) Trends in circumcision among newborns. On-line: https://www.cdc.gov/nchs/data/hestat/circumcisions/circumcisions.htm

CDC (2016) HIV in the United States: At a glance. On-line: https://www.cdc.gov/hiv/statistics/overview/ataglance.html

Chemtob, D., Op de Coul, E., van Sighem, A., Mor, Z., Cazein, F., Semaille, C. (2015) Impact of male circumcision among heterosexual HIV cases: comparisons between three low HIV prevalence countries. Israel J. Health Policy Research, 4(36), 8 pages.

Morris, B.J., Wamai, R.G., Henebeng, E.B., Tobian, A.A.R., Klausner, J.D., Banerjee, J., Hankins, C.A. (2016) Estimation of country-specific and global prevalence of male circumcision. Population Health Metrics, 14(4), 13 pages.

UNICEF (2001) A league table of teenage births in rich nations. On-line: http://www.unicef-irc.org/publications/pdf/repcard3e.pdf

Warner, L., Ghanem, K.G., Newman, D.R., Macaluso, M., Sullivan, P.S., Erbelding, E.J., (2009) Male circumcision and risk of HIV infection among heterosexual African American men attending Baltimore sexually transmitted disease clinics. J. Infect. Dis., 199(1), 59-65.

Worobey, M., Watts, T.D., McKay, R.A., Suchard, M.A., Granade, T., Teuwen, D.E., Koblin, B.A., Heneine, W., Lemey, P., Jaffe, H.W. (2016) 1970s and ‘Patient 0’ HIV-1 genomes illuminate early HIV/AIDS history in North America. Nature, 539(3), 98-101.

In 10 out of 18 countries circumcised men were more likely to have HIV than non-circumcised men

Summary answer:

Data for most of the countries do not achieve statistical significance. Even for those that do, there are grave doubts about the reliability of the data as many men report being circumcised when they are not, or are only partially so, having undergone various ritual, tribal cutting practices. Those same practices also spread HIV, making it likely that some men caught their HIV in this way, further complicating the data. Confounding from other sources, such as migration, and high infidelity, divorce and remarriage rates also explain some of the results. This is a very weak and problematic data set which does not trump better quality data from dozens of epidemiological studies, several reviews, and meta-analyses, indicating that the association between lack of circumcision, and HIV infection, is real.

Detailed answer:

At first glance this would appear to undermine the assertion that circumcision protects against HIV. After all if, in more than half of the countries examined, the correlation is the reverse it contradicts the assertion. So it is no surprise to find this argument heavily touted by intactivists. Often a quote, from the USAID DHS report number 22 (from which it is derived, see: Mishra et al., 2009), is used:

“in 8 of 18 countries with data, HIV prevalence is lower among circumcised men, while in the remaining 10 countries it is higher”

A Google search will find this quote everywhere. Repeat offenders are Mark Lyndon, an intactivist from Manchester, England, and Hugh Young, who runs the circumstitions website. It is also used by prominent, and widely published, intactivist Robert S Van Howe (Van Howe & Storms, 2011; Van Howe, 2015).

Of course the argument is spurious, and on multiple grounds. Aside from the old adage that correlation is not causation, it has already been used elsewhere on this site as an example of quote-mining (http://circfacts.org/sloppy-logic/#slog28).

Reading the original quickly throws up issues with the sample sizes. At the top of page 14 we find, “some of the estimates presented in this report are based on small numbers of cases in survey samples, and should be interpreted with caution”. And a footnote on p. 110 reads, “Note: HIV prevalence estimates for ‘not circumcised’ men for Guinea and Niger are based on small numbers of cases”. Looking at the actual sample sizes we find that the non-circumcised samples for Guinea, Niger and Senegal are just 18, 14 & 56 men respectively. Given the relatively low rates of HIV in those countries (0.9, 0.8 & 0.5 respectively) one does not have to be a statistician to realise that it is quite likely there would not be a single HIV positive man in such small samples (and, indeed, there isn’t). So how can one possibly tell, for those countries, if there is a correlation? On that basis alone one can therefore discount 3 of those 10 countries. And in no case do the authors attempt any statistical analysis to determine if the differences they report are significant. So let’s do that!

Happily, one does not need a degree in statistics to do this, just a calculator, and this link: https://graphpad.com/quickcalcs/contingency1/  Take the data in table 9.4 of the USAID report, work out the actual numbers who are HIV positive and negative for each group (circumcised or not) and enter them into the table in the link. One can do either a Chi-squared test, or a Fisher’s exact one. For our purposes it makes little difference, the results are similar. The ones given in the table below are Fisher’s, except for India, for which the program indicated Chi-squared should be used owing to the large data sets. The countries in which circumcised men seemed to be more likely to have HIV are here called the “contrary countries”, the rest “conforming countries”. The p values are in brackets after the country. Any p value more than 0.05 is not considered significant, and is marked below by an asterisk.

Contrary countries                      Conforming countries

Cameroon (0.005)                       Burkina Faso (0.12*)

Guinea (1.0*)                                Cambodia (1.0*)

Haiti (0.03)                                   Côte d’Ivoire (0.64*)

Lesotho (<0.0001)                      Ethiopia (0.40*)

Malawi (0.02)                              Ghana (0.74*)

Niger (1.0*)                                   India (0.03)

Rwanda (0.04)                            Kenya (<0.0001)

Senegal (1.0*)                              Uganda (0.002)

Tanzania (0.25*)

Zimbabwe (0.12*)

So, for only eight countries were the data sets, and differences between them, large enough to achieve statistical significance. This makes nonsense of the claim about 10 out of 18 countries having higher HIV prevalence in circumcised men. It should be 5 out of 8. And for one of those (Rwanda) the significance is borderline.

This issue with the small sample sizes (though without the more detailed analysis above) has been pointed out in comments threads ad nauseam but intactivists take little notice (as is typical). The problems are only beginning, however. The above analysis assumes that the data are accurate, but it is not clear how many of those men who claim to be circumcised actually are. The study was based on a questionnaire. Nobody got the men to drop their pants to check if they really were circumcised. This matters, a lot. As Van Howe delights in (correctly) pointing out, self-report of circumcision status can be unreliable (Van Howe, 2013; and references therein). Many men report being circumcised when, in fact, they are not.

The problem is related to the nature of many circumcisions in Africa. For many African men circumcision is part of a coming of age tribal ritual. It is performed by a shaman, with no regard to hygiene or pain control (in fact being able to bear the pain is part of the ritual). But not all that goes on actually counts as circumcision in the medical sense. Various foreskin cutting practices that fall short of a circumcision may be carried out instead. Sometimes the cutting is dropped altogether, but the other elements preserved, in a ritual that may have included circumcision in the past.

Consequently, many men who have gone through such ceremonies consider themselves circumcised, and identify themselves as so, but they are not. In Lesotho, for example, when men who identified themselves as circumcised were asked to drop their pants for confirmation, only half actually were fully circumcised. 27 % had undergone some sort of foreskin cutting, but which fell short of circumcision, and 23 % had undergone no cutting at all (Thomas et al., 2011). Lesotho, it should be noted, is one of those contrary countries.

The problem with ritual tribal circumcisions is not restricted to whether the men actually are circumcised in the first place. Ritual circumcision in Africa is one of the few instances where a non-sexual mode of transmission of HIV has been shown to be significant. Many of the victims (which is an appropriate term regarding circumcision in this case) are youths and young men who are already sexually active and, consequently, HIV infected. As the shaman works his way along the line-up, going slice, slice, slice, with his razor blade, spear tip, or whatever sharp implement is used, and one just one fellow in the line-up has HIV, the implications for those that follow are obvious. HIV gets on the blade, is transferred to the next victim, and so on down the queue. In fact one can find freshly circumcised virgins in these countries who are HIV positive, having contracted it from their tribal circumcision.

This issue was identified in 2007 by Brewer et al (2007) in Tanzania, Lesotho and part of Kenya. The first two are “contrary countries”. It has since been highlighted in Cameroon, another “contrary country” (Ndiwane, 2008). So, aside from whether these men actually are properly circumcised, or have just had some other foreskin cutting, there is a high likelihood they many got their HIV from their unhygienic tribal circumcision. It is safe, hygienic medical circumcision that is being touted as protective. Not dangerous, unhygienic tribal cutting practices. Consequently, data from any countries in which such practices prevail must be treated as suspect. The men may not even be circumcised in the first place and, even if they are, they may have caught their HIV from some dangerous tribal ritual, not by sex.

So, we can now discount Lesotho, Tanzania and Cameroon. The Zimbabwe data too must be treated as suspect. In a study conducted in 2010 – 2011, it was found that 58 % of circumcised Zimbabwean men had been circumcised by a traditional provider, or family friend, 31.2 % of them at age 15 or more, rather than by medical professionals (Chikutsa et al., 2013). The difference reported in the USAID report did not achieve statistical significance anyway.

In Haiti there is some indication that foreskin cutting practices that fall short of full circumcision are practiced (George et al., 2012). Haiti also has the issue of a rather modest sample size that is, statistically, only moderately significant. Using the data in the report, the difference amounts to just five individuals. Enough to be skewed by just a few households (these were household surveys).

So now we are left with Malawi and Rwanda as the sole remaining “contrary countries” out of the original ten. In fact the observation that Malawi appeared to contradict the protective effect of circumcision was used to great effect in that country to stall the implementation there of circumcision (Poulin & Muula, 2011), doubtless with many lives lost as a result. However, we now have some explanations.

Tribal, and partial, circumcision is practised by the Lomwe and some other ethnic groups in Malawi (Bengo et al., 2010; Rennie et al., 2015), but others are likely fully circumcised, being Muslim. Thus, although Mutombo et al. (2015) did not consider tribal circumcision to be a significant driver of the epidemic in Malawi, we know that many Malawian men participating in the USAID survey were not fully circumcised. Mutombo et al. did note a much higher HIV rate in those who had been through a tribal circumcision and, as part of the ceremony, had had unprotected sex shortly after (“ritual sex”). So having unprotected sex whilst the penis is still healing from a tribal circumcision puts men at much higher risk of catching HIV. Quelle surprise.

Most men in Southern Malawi are circumcised, in contrast to the Central and Northern regions, where most men are not; but the south is also the poorest part of the country, and the men there frequently travel to the cities in search of work. Whilst there they are not celibate, and some end up infected. Meanwhile, back home, their wives get lonely and stray, thereby exposing themselves to the virus too. They also have a high divorce and remarriage rate. Those who stay at home, and are in faithful relationships, are less likely to have HIV. Those who travel far in search of work, or who repeatedly divorce and remarry, are more likely to be infected. So the anomaly there is due to patterns of migration and multiple marriages (Poulin & Muula, 2011). In short, that bane of epidemiology – confounding factors (see also: http://circfacts.org/medical-benefits/hivaids/#hiv4), and compounded by the familiar problem of whether those men who report being circumcised actually are.

So now only Rwanda remains. The difference reported is barely significant (in fact it amounts to just 7 individuals) and, with only one doctor for every 16,700 people in 2010
(https://www.cia.gov/library/publications/the-world-factbook/fields/2226.html), we can be quite confident that the few men who claimed to be circumcised are unlikely to have had it done by a doctor. And that’s assuming they are accurate in claiming to be circumcised in the first place. So the same doubts apply as with the other “contrary countries”.

These problems have been pointed out to intactivists repeatedly, whether in comments threads or in published rebuttals (e.g. Morris et al., 2016). But still they keep peddling the claim. Lyndon has, at least, taken on board the fact that unhygienic tribal circumcisions are to blame for some of the anomalous data, but still peddles the argument because:

“the whole rationale for the RCTs into female-to-male transmission was a purported correlation between high rates of male circumcision and low rates of HIV” (comment on 24 July 2017 at http://www.patheos.com/blogs/friendlyatheist/2017/07/13/there-are-good-reasons-for-male-circumcision-debunked).

But his argument for using the USAID data is nonsense. The correlation is there. One cannot trump dozens of epidemiological studies, several reviews, and meta-analyses showing the correlation, with one contrary, but weak, data set that includes sample sizes too low to be of use, unreliable self-reports about being circumcised, and the clear presence of major confounding. To continue to use this argument, knowing there are so many problems, is just dishonest. No reputable body gives credence to the argument. Only intactivists do.

References
Bengo, J.M., Chalulu, K., Chinkhumba, K., Kazembe, L., Maleta, K.M., Masiye, F., Mathanga, D. (2010) Situation analysis of male circumcision in Malawi. Report by the College of Medicine, 136 pages.

Brewer, D. D., Potterat, J. J., Roberts, J. M. Jr., Brody, S. (2007). Male and female circumcision associated with prevalent HIV infection in virgins and adolescents in Kenya, Lesotho, and Tanzania. Annals of Epidemiology, 17, 217–226.

Chikutsa, A., Ncube, A.C., Mutsua, S. (2013) Male circumcision and risky sexual behaviour in Zimbabwe: Evidence from the 2010 – 2011 Zimbabwe Demographic and Health Survey. DHS Comparative Reports No. 102, USAID.

George, C., Brennen, D., Roberts, R., Rabie, A., Read, S. (2012) Getting to zero: Male circumcision for HIV prevention in the Bahamas. West Indian Med. J., 61(9), 921-3.

Morris, B.J., Barboza, G., Wamai, R.G. & Krieger, J.N. (2016). Circumcision is a primary preventive against HIV infection: Critique of a contrary metaregression analysis by Van Howe, Global Public Health, ePub ahead of print.

Mishra, V., Amy M., Rathavuth H., Yuan G., Robey, B. (2009). Levels and Spread of HIV Seroprevalence and Associated Factors: Evidence from National Household Surveys. DHS Comparative Reports No. 22. USAID.

Mutombo, N., Maina, B., Jamali, M. (2015) Male circumcision and HIV infection among

sexually active men in Malawi. BMC Public Health, 15(1051), 1-9.

Ndiwane, A. (2008). Laying down the knife may decrease risk of HIV transmission: Cultural practices in Cameroon with implications for public health and policy. Journal of Cultural Diversity, 15, 76–80.

Poulin, P., Muula, A.S. (2011) An inquiry into the uneven distribution of women’s HIV infection in rural Malawi. Demographic Research, 25, article 28, 869-902.

Rennie, S., Perry, B., Corneli, A., Chilungo, A., Umar, E. (2015) Perceptions of voluntary medical male circumcision among circumcising and non-circumcising communities in Malawi, Global Public Health, 10(0), 679-91.

Thomas, A.G., Tran, B.R., Cranston, M., Brown, M.C., Kumar, R., Tlelai, M. (2011) Voluntary medical male circumcision: A cross-sectional study comparing circumcision self-report and physical examination findings in Lesotho. PLoS One, 6(11), e27561.

Van Howe, R. S. (2013). Sexually transmitted infections and male circumcision: A systematic review and meta-analysis. ISRN Urology, 2013, article 109846, 1–42.

Van Howe, R. S. (2015). Circumcision as a primary HIV preventive: Extrapolating from the available data. Global Public Health, 10, 607–625.

Van Howe, R. S., & Storms, M. R. (2011). How the circumcision solution in Africa will increase HIV infections. Journal of Public Health in Africa, 2(1), 11-5.

Lead-time bias argument

“It doesn’t take a rocket scientist to know that if you have part of your penis cut off you are probably not going to have sex in the near future, thereby reducing the probability of catching the virus over a set time; the study never addresses this issue.”

The above quote is from “The Science of Circumcision”, an error-riddled YouTube video by self-styled British “skeptic” Myles Power. It is one of the “methodological flaws” intactivists claim compromised the African randomised controlled trials on circumcision and HIV. In published articles it is referred to as “lead-time bias”, and is used by Boyle & Hill (2011), and mentioned by Darby & Van Howe (2011), and Van Howe & Storms (2011). In parallel to the Ugandan trial on HIV, one was also conducted on HSV, HPV and syphilis, so Storms (2009) also used the lead-time bias argument to attack its finding that circumcision protected against HSV. As did the widely-published Robert S. Van Howe, who took his usual haughty tone (Van Howe, 2013):

“The potential for lead-time bias in any cohort study or clinical trial is taught and emphasized in the most basic classes on research design. How was this potential source of bias missed by the highly regarded researchers at Johns Hopkins, the reviewers who approved funding for these studies at the National Institutes of Health, and the editors and peer-reviewers at highly regarded medicals [sic] journals such as The Lancet and The New England Journal of Medicine? To compensate for the deficiencies of these individuals, a post hoc adjustment of six weeks lead time was made. Six weeks were chosen to be on the conservative side.”

Van Howe applied a “correction” on this basis to a series of results for various STDs that did not suit his desired conclusion that circumcision is ineffective against them.

The argument appears also on intactivist websites (e.g. Doctors Opposing Circumcision), in the popular intactivist YouTube video “The Elephant in the Hospital Room” (which is even worse than Myles Power’s one), and in comments threads.

At first glance it seems reasonable. Freshly circumcised men are not going to use their tender organs right away, so this would seem to introduce a bias into the results in favour of an apparent protective effect. The problem is that, contrary to the claims of intactivists, it was taken into account and found not to be important.

The S. African trial took it into account directly by incorporating a 6-week healing period into their calculations. They found it played only “a minor role in explaining the effect of the intervention” (Auvert et al., 2005). In response to the use of the lead-time bias argument against the HSV findings, the authors of the original study calculated the effect it would have and found that the protection afforded by circumcision was still significant (p = 0.02). They also mentioned that 85 % of subjects healed in as little as four weeks, so six is erring on the high side (Tobian et al., 2009), not Van Howe’s “conservative side”.

The Kenyan and Ugandan trials used a different approach. Instead, they measured HIV at baseline, 6, 12 and 24 months. Now, if the effect was due to abstinence during healing it should all be concentrated in the first six month period. It wasn’t. In fact it was sustained throughout the study period, if anything growing stronger over time (Bailey et al., 2007; Gray et al., 2007). So whatever was causing the effect, it was not abstinence during healing. The S. African trial, it should be noted, found the same, their subjects being monitored at 0, 3, 12 and 21 months. So that trial accounted for it in two different ways.

Of course this has been pointed out to intactivists (Wamai et al., 2012; Morris et al., 2014) but they take no notice and carry on peddling the argument as if it still stood. This is normal behaviour for pseudoscientists.

References
Auvert, B., Taljaard, D., Lagarde, E., Sobngwi-Tambekou, J., Sitta, R., Puren, A. (2005) Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial. PLoS Medicine, 2(11), 1112-22.

Bailey, R.C., Moses, M., Parker, C.B., Agot, K., Maclean, I., Krieger, J.N., Williams, C.F.M., Campbell, R.T., Ndinya-Achola, J.O. (2007) Male circumcision prevention for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet, 369(9562), 643-56.

Boyle, G.J., Hill, G. (2011) Sub-Saharan African randomised clinical trials into male circumcision and HIV transmission: Methodological, ethical and legal concerns. J. Law Med., 19, 316-34.

Darby, R., Van Howe, R.S. (2011) Not a surgical vaccine: there is no case for boosting infant male circumcision to combat heterosexual transmission of HIV in Australia. Australian and New Zealand Journal of Public Health, 35(5), 459-65.

Gray, R.H., Kigozi, G., Serwadda, D., Makumbi, F., Watya, S., Nalugoda, F., Kiwanuka, N., Moulton, L.H., Chaudhary, M.A., Chen, M.Z., Sewankambo, N.K., Wabwire-Mangen, F., Bacon, M.C, Williams, C.F.M., Opendi, P., Reynolds, S.J., Laeyendecker, O., Quinn, T.C., Wawer, M.J. (2007) Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet, 369(9562), 657-66.

Morris, B.J., Hankins, C.A., Tobian, A.A.R., Krieger, J.N., Klausner, J.D. (2014) Does Male Circumcision protect against sexually transmitted infections? Arguments and meta-analyses to the contrary fail to withstand scrutiny. ISRN Urology, 2014, Article ID 684706, 23 pages.

Storms, M.R. (2009) Male circumcision for the prevention of HSV-2 and HPV infections (Letter to the editor). New England J. Med., 361(3), 307.

Tobian, A.A.R., Serwadda, D., Gray, R.H., (2009) Male circumcision for the prevention of HSV-2 and HPV infections (Reply). New England J. Med., 361(3), 307-8.

Van Howe, R.S., Storms, M.R. (2011) How the circumcision solution in Africa will increase HIV infections. J. Public Health in Africa, 2(e4), 11-5.

Wamai, R.G., Morris, B.J., Waskett, J.H., Green, E.C., Banerjee, J., Bailey, R.C., Klausner, J.D., Sokal, D.C., Hankins, C.A. (2012) Criticisms of African trials fail to withstand scrutiny: Male circumcision does prevent HIV infection. J. Law Med. 20, 93-123.

 

In Malawi HIV doubled after the circumcision campaign began

This claim began to circulate after it appeared in an on-line newspaper, “Malawi24”, in 2015:

https://malawi24.com/2015/08/04/circumcision-disaster-malawi-hiv-infection-rate-doubles/

The article asserts that:

“Statistics cited by the BBC in a report marketing Medecins Sans Frontieres (MSF), one of the world’s renowned NGOs working on public health, show that HIV infection rate in Malawi has doubled in recent years despite a range of interventions put in place to tackle the spread of the virus that have included relentless campaign on condomisation and circumcision.

According to the report, HIV rates have doubled in Malawi moving from 10% to 20% in 1 year. Strangely, this has been the same period that Malawians have been manipulatively forced to go through circumcision in masses with the promise that it reduces the contraction of HIV.”

The anti-circumcision tone “manipulatively forced”, and lack of details about where to find this BBC report so it can be checked, should set alarm bells ringing. And one only has to read a little further to find the author conceding that “some Malawians trashed the results claiming they were sensationalised and exaggerated”.

But the best part is at the end:

“This article has been updated. An earlier version credited the statistics cited by the BBC to MSF. However, MSF have denied that they gave the statistics to the BBC. The organisation says the claims the BBC made that 1 in every 5 people are HIV positive in Malawi were “grossly unfounded and sensationalised” with the aim of ‘grabbing headlines’.  Malawi24 apologises for any inconveniences caused.”

In short, the figures are bogus, and the claim was, in effect, retracted by the same on-line paper that originally made it! Yet it keeps turning up in comments threads, opinion pieces, and (alarmingly) is being peddled to Africans by the disturbing intactivist website vmmcproject.org.

Do intactivists not bother to read their own sources?

Why do they think that an unreliable, un-sourced, retracted, on-line newspaper report trumps hard data carefully collected by medical and scientific professionals?

Some facts (real ones, not intactivist ones), from more credible sources:

The circumcision program in Malawi was formally launched in 2012, although uptake was initially rather low, at 11.5 % after two years, being hampered by religious and cultural beliefs, as well as a lack of resources to fully implement it. But over the period 2011 to 2013, i.e. straddling the initiation of the circumcision program, new HIV infections declined from 55,000 in 2011 to 34,000 in 2013 (National AIDS Commission, 2014).

As of 2016, the HIV prevalence in Malawi was 10.6 % for the age group 15 to 64 years (MPHIA 2016).

References
MPHIA (2016) Malawi Population-based HIV Impact Assessment, December 2016. On-line: http://phia.icap.columbia.edu/wp-content/uploads/2016/09/MALAWI-Factsheet.FIN_.pdf

National AIDS Commission (2014) National HIV Prevention Strategy, 2015-2020. National AIDS Commission, Malawi. On-line: http://hivstar.lshtm.ac.uk/files/2016/05/Malawi-National-HIV-Prevention-Strategy-2015-2020.pdf

 

Circumcision increases the risk of women contracting HIV

A claim sometimes made on intactivist websites, articles, and in comments threads, is that women are actually at more risk of contracting HIV if their partners are circumcised, e.g., here: http://www.intactamerica.org/dangerousmistake, here: http://www.circumstitions.com/HIV.html#chao and here: http://www.organiclifestylemagazine.com/10-circumcision-myths-lets-get-the-facts-straight (caution, almost every assertion in these sources is bogus or misleading, but here we focus only on the alleged HIV risk to women).

As is usual for intactivist claims this one has very little evidence in support. In fact, it rests on just two papers. The earlier paper is by Chao et al. (1994) who studied women in Rwanda. They found that several factors were associated with a higher risk of HIV infection in Rwandan women, and one of them was having circumcised male partners. Intactivists have seized upon this, but without looking for alternative explanations.

Aside from the old adage that correlation is not causation, what we may be seeing is confounding. The authors themselves suggest that the circumcised males “may constitute a population with high risk sexual behaviour. It is also likely that some men were circumcised as a result of treatment for venereal disease.” They also found that women of higher socio-economic status were more likely to be infected. It has since been found that in 11 African countries, including Rwanda, circumcision is more common in those of higher socio-economic status (Lau et al., 2015) offering another possible confounder. Those in higher socio-economic classes tend to have more sexual partners, including sex workers, thereby putting themselves at greater risk.

Further uncertainties arise from the nature of the “circumcisions” themselves. The authors “did not collect data on the reason for or age at circumcision”. The study was conducted long before the medical circumcision program began. As explained here, http://circfacts.org/medical-benefits/hivaids/#hiv5, prior to the medical circumcision program, circumcision in Africa was often an unhygienic tribal, ritual affair, carried out by a shaman with a razor blade, or other sharp implement. These circumcisions actually spread HIV, so such men are more likely to be HIV positive, and hence a danger to their female partners. Most circumcisions in Rwanda nowadays are medical, but as late as 2015 it was found that around 20 % were not: Lau et al., (2015). Data for the time of Chao’s study is lacking but, because it predates the medical circumcision program and doctors are in short supply in Rwanda, the proportion of non-medical circumcisions was likely higher. There is also the issue of whether the “circumcised” men in the study actually were circumcised. Many men, who have been through rituals that may involve some sort of cutting, describe themselves as circumcised even though they are not. Although in the Rwandan study it was the women who were asked if their male partners were circumcised, adding further scope for error. All this is covered in detail in the link above. In short, it is uncertain if the men actually were circumcised and, even if they were, they may not all have been done by a doctor in a hygienic setting.

The other study was an attempt to directly assess the risk, or benefit, to women of having a circumcised male partner, with respect to male-to-female HIV transmission (Wawer et al., 2009). In a randomized controlled trial, 922 non-circumcised, HIV positive Ugandan men were randomly split into two groups. One group was circumcised (the intervention group), the other not (the control group). Their HIV negative female partners were then monitored for signs of infection.

Slightly more women in the intervention group contracted HIV during the course of the study, implying that having a circumcised partner put them at greater risk. What intactivists don’t tell their audience is that this effect was confined to those women whose partners resumed sexual intercourse before their circumcision wound had completely healed, and even then the statistical significance of the association was marginal (p = 0.038). So, if a man starts using his penis when the circumcision wound has not yet healed, the weeping of HIV positive serous fluid, even blood, increases the risk his female partner will contract HIV. Quelle surprise! For those women whose partners waited the recommended 6 weeks to heal before resuming sex, there was no additional risk, a fact omitted by intactivists.

The misleading way this study is portrayed by intactivists does not end there, however. The study was terminated early due to futility. The researchers had been unable to recruit enough volunteers to achieve statistical significance, even if they had carried it through to completion, a circumstance that appeals to the intactivists’ conspiracy theorising mind-set. English intactivist Mark Lyndon (whose cut-and-paste posts under the name “ml66uk” infest comments threads across the Internet, and who often cites this study) argues that it was stopped early because the researchers, “were worried that if it continued, it may show a statistically significant increase later” (post, dated 15 July, 2012 in comments here: http://www.quackdown.info/article/getting-circumcision-science-right-media/). In other words, he thinks it was an attempt to prevent findings being made that might contradict the official view that circumcision is effective against HIV.

In reality, “Recruitment into the trial was stopped at interim analysis on the recommendation of the data safety monitoring board. This decision was on the basis of the futility of continuing to recruit because fewer serodiscordant couples had been enrolled concurrently than anticipated, decreasing the power to detect a protective effect.” It would require the enrolment of around 10,000 serodiscordant couples, which “is likely to be logistically unfeasible” (Weiss et al., 2009).

So a randomized controlled trial would be impractical, but that is not to say the subject cannot be investigated. By 2009 a total of 19 epidemiological studies had been conducted on male circumcision and HIV risk to women, most finding no adverse effect, making the intactivists’ selection of the only two that did an example of cherry-picking (http://circfacts.org/sloppy-logic/#slog7). Systematic review and meta-analysis of these studies found no evidence for any increased risk of HIV transmission to women if their male partners were circumcised (Weiss et al., 2009).

No one claims a direct benefit from circumcision with regards to male to female HIV transmission. The direct benefit is for the reverse route, but that is not to say that there is no benefit. Circumcision interrupts the cycle of transmission, female to male, then to other females and then to other males, and so on. By (partially) breaking the female to male part of the cycle, it reduces the overall number of infected males in a population. A female having sex with a new male partner is therefore at a reduced risk – there is less chance that her new man will be HIV positive.

And there is evidence that this is actually happening following the implementation of the WHO circumcision program. In 2014 it was reported that results from a large (n = 2452) cross-sectional study of South African women found that those who only had sex with circumcised male partners were at much lower risk of contracting HIV (Jean et al., 2014).

In another large (n = 1356) cohort study, also of South African women, those with a circumcised partner were at less risk of HIV infection (p = 0.011). The authors concluded, “The protective effect of partner circumcision in our study is likely an indirect effect related to lower HIV prevalence amongst circumcised partners who had unknown HIV status, as amongst males with known HIV status, uncircumcised males had considerably higher HIV prevalence” (Fatti et al., 2017).

So, although the effect is indirect, circumcision also protects women from HIV.

References
Chao, A., Bulterys, M., Musanganire, F., Habimana, P., Nawrocki, P., Taylor, E., Dushimimana, A., Saah, A. (1994) Risk factors associated with prevalent HIV-1 infection among pregnant women in Rwanda. National University of Rwanda-Johns Hopkins University AIDS Research Team. Int. J. Epidemiol., 23(2), 371-80. On-line abstract: https://www.ncbi.nlm.nih.gov/pubmed/8082965

Fatti, G., Shaikh, N., Jackson, D., Goga, A., Nachega, J.B., Eley, B., Grimwood, A. (2017) Low HIV incidence in pregnant and postpartum women receiving a community based combination HIV prevention intervention in a high HIV incidence setting in South Africa. PLoS One, 12(7), e0181691. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5531482/

Jean, K., Lissouba, P., Taljaard, D., Taljaard, R., Singh, B., Bouscaillou, J., Peytavin, G., Sitta, R., Mahiane, S.G., Lewis, D., Puren, A., Auvert, B. (2014) HIV incidence among women is associated with their partners’ circumcision status in the township of Orange Farm (South Africa) where the male circumcision roll-out is ongoing (ANRS-12126). 20th International AIDS Conference, Melbourne, Australia. Abstract FRAE0105LB. On-line: http://pag.aids2014.org/Abstracts.aspx?AID=11010

Lau, F.K., Jayakumar, S., Sgaier, S.K. (2015) Understanding the socio-economic and sexual behavioural correlates of male circumcision across eleven voluntary medical male circumcision priority countries in southeastern Africa. BMC Public Health, 15(813). On-line: https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-015-2135-1

Wawer, M.J., Makumbi, F., Kigozi, G., Serwadda, D., Watya, S., Nalugoda, F., Buwembo, D., Ssempijja, V., Kiwanuka, N., Moulton, L.H.,Sewankambo, N.K., Reynolds, S.J., Quinn, T.C., Opendi, P., Iga, B., Ridzon, R., Laeyendecker, O., Gray, R.H. (2009) Circumcision in HIV-infected men and its effect on HIV transmission to female partners in Rakai, Uganda: a randomised controlled trial. Lancet, 374(9685), 229-37. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2905212/

Weiss, H.A., Hankins, C.A., Dickson, K. (2009) Male circumcision and risk of HIV infection in women: a systematic review and meta-analysis. Lancet Infect. Dis., 9(11), 669-77. On-line: http://www.thelancet.com/journals/laninf/article/PIIS1473-3099(09)70235-X/fulltext

Foreskins contain Langerhans cells which protect against HIV.

This intactivist claim is an oversimplification, and an attempt to sow confusion. The argument asserts that foreskins contain cells called Langerhans cells, which is true. These produce a protein, langerin, which binds to HIV and enables it to be destroyed by the body’s natural defences, which is at least partially true. The study usually cited by intactivists is the one by De Witte et al. (2007) demonstrating this effect, at least in vitro. The authors cautioned that “strategies to combat infection must enhance, preserve or, at the very least, not interfere with Langerin expression and function.” Music to the ears of intactivists as removing the foreskin will obviously remove those wonderful Langerhans cells and their precious langerin.

But things are rarely that simple in molecular biology. A beneficial effect of langerin is only observed at low HIV loads. At high concentrations of HIV langerin is simply overwhelmed. De Witte et al. admitted this, saying, “We observed transmission at a very high viral load, strongly suggesting that Langerin is saturated at high HIV-1 concentrations.” In fact Langerhans cells themselves have been implicated in the mechanisms of infection, being targeted by HIV. This is one of several likely foreskin-related mechanisms of infection (Morris & Wamai, 2012; Jayathunge et al., 2014).

That HIV is readily transmitted from female to male, and that circumcision is so effective at preventing this, is proof that in practice any putative benefit of langerin is minor, and eclipsed by the strong protective effect of removing the foreskin altogether.

References
De Witte, L., Nabatov, A., Pion M., Fluitsma, D., de Jong, M.A.W.P., de Gruijl, T., Piguet, V., van Kooyk, Y., Geijtenbeek, T.B.H. (2007) Langerin is a natural barrier to HIV-1 transmission by Langerhans cells. Nature Medicine, 13(3): 367–371. https://www.ncbi.nlm.nih.gov/pubmed/17334373 (abstract only).

Jayathunge, P.H.M., McBride, W.J.H., MacLaren, D., Kaldor, J., Vallely, A., Turville, S. (2014) Male circumcision and HIV transmission; what do we know? Open AIDS J., 8, 31-44. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4192839/

Morris, B.J., Wamai, R.G. (2012) Biological basis for the protective effect conferred by male circumcision against HIV infection. Int. J. STD AIDS, 23(3), 153-9. https://www.ncbi.nlm.nih.gov/pubmed/22581866 (abstract only).

 

The absolute risk reduction is a mere 1.31 %

The African trials indicated that circumcision reduces the risk of a man contracting HIV through heterosexual intercourse by around 60 %. Subsequent meta-analyses indicated that the actual figure is about 70 % (Lei et al., 2015; Sharma et al., 2017). That is excellent. In fact it compares favourably with influenza vaccine (Osterholm et al., 2012).

This is not what intactivists want to hear, so they will look for any way they can to downplay this impressive statistic, so they come up with their own instead: 1.31 %! Before going into details as to why this is misleading, even dishonest, it is necessary to explain the difference between relative and absolute risk reduction.

The 60 or 70 % figures are for relative risk reduction. They mean that a circumcised man is 60 or 70 % less likely to contract HIV during heterosexual intercourse, than his non-circumcised peers. This is the usual way efficacies are stated, and they allow workers to estimate absolute risk reduction for whatever population they are interested in. Here, the absolute value is the difference between the percentage of men in the control (non-circumcised) group who became HIV positive, and the corresponding percentage in the intervention (circumcised) group.

The 1.31 % figure is an estimate for the absolute risk reduction in the African trials, and was deployed by intactivists Boyle and Hill (2011) and continues to pop up in intactivist articles, and on comments threads. The intactivists’ chief academic Robert S. Van Howe, who likes to bamboozle his readers with statistics, also uses this figure (Van Howe, 2015). After all, 1.31 % sounds a lot less impressive than 70, or even 60 %, so intactivists use it and claim it is “not statistically significant”, even though the original papers did the relevant statistical analyses and found their results to be highly significant.

The correct figure is actually 1.8 % (Siegfried et al., 2009), as was pointed out in a rebuttal to Boyle & Hill (Wamai et al., 2012) and which Van Howe, in keeping with normal intactivist behavior, ignored. But, for the sake of argument, we will use the intactivists’ figure of 1.31 % and see where it leads.

Firstly it must be noted that the figure is for the two years of the trials. If the protective effect is maintained (and follow-up data indicates that it is) then after 20 years that 1.31 % becomes 13.1 % which, in a population of a million sexually active men, translates into 131,000 new HIV infections averted. Of course, this is a simplification as factors may change over that 20 years, e.g. sexual behaviours, prevalence and number of new entrants to the cohort, but as an approximation it suffices to make the point. Sub-Saharan Africa is home to about half a billion males, most of them under 30 years old. If even just half of them are sexually active (and bear in mind the children will soon grow up and become active), then a 1.31 % absolute risk reduction means millions of new infections averted, tens of millions if one extrapolates to the end of the century. Not so small now, is it?

The absolute risk reduction is, in part, a function of how prevalent the disease is in a population. Thus, if the disease is exceedingly rare, then even a relative risk reduction of 100 % will translate into a very low absolute one, or even zero if that disease is now extinct (as with smallpox). Vaccines for uncommon diseases, such as whooping cough, may have very high relative risk reductions, but very low absolute ones. But we still vaccinate.

As mentioned earlier, the relative risk reduction (70 %) is similar to that for influenza vaccine: figures of 59 %, 69 % and 83 % are cited by Osterholm et al., (2012), for various vaccine types and age groups. Using data from the same source Mike Adams, a notorious anti-vaxxer, derived an absolute risk reduction of 1.5 % and triumphantly declared that the relative risk reduction for influenza vaccine of 60 % was “a total lie” (Adams, 2011). Using language strikingly similar to that used by intactivists, he complained about “the vaccine-pushing CDC”, branded his opponents (i.e. all of medical science) the “quack science community” and accused them of lying with statistics! He went on to complain about the dangers of vaccines (cf. the dangers of circumcision), throw in some conspiracy theorising (big pharma) and listed alleged “problems” with the study, whilst all the time laying it on thick about how small this 1.5 % protection is.

Sound familiar? It ought to. One could almost take a standard intactivist tract deploying the absolute vs. relative risk argument, and just replace the word “circumcision” with the word “vaccine”. The numbers are even strikingly similar! Adams’ argument was swiftly, and comprehensively, demolished on a skeptic blog (Gorski, 2011) but, before long, it was copied by another anti-vaxxer (Huff, 2013), ignoring the previous exposé, just as intactivists routinely ignore their own critics. And, again, it was swiftly shot down on a skeptic website (Karlsson, 2013). Adams, Huff, and “Natural News” are familiar to skeptics who debunk anti-medical pseudoscience and quackery (https://rationalwiki.org/wiki/NaturalNews).

If the argument is pseudoscience when applied to vaccines, why should it be any less so when applied to circumcision? Next time intactivists use this ploy, do the sums above, and then remind them of the crank company they have.

References
Adams, M. (2011) Shock vaccine study reveals influenza vaccines only prevent the flu in 1.5 out of 100 adults (not 60% as you’ve been told). Natural News, 27th October. On-line: http://www.naturalnews.com/033998_influenza_vaccines_effectiveness.html

Boyle, G.J., Hill, G. (2011) Sub-Saharan African randomized clinical trials into male circumcision and HIV transmission: Methodological, ethical and legal concerns. J. Law Med., 19(2), 316-34.On-line abstract: https://www.ncbi.nlm.nih.gov/pubmed/22320006

Gorski, D. (2011) Mike Adams vs. the flu vaccine. Respectful Insolence blog. On-line: https://respectfulinsolence.com/2011/10/27/mike-adams-vs-the-flu-vaccine/

Huff, E.A. (2013) Why you should avoid taking the seasonal flu vaccine. Natural News, 21st September. On-line: https://www.naturalnews.com/042159_flu_shot_vaccine_myths_influenza.html

Karlsson, E. (2013) Anti-vaccine misinformation about the seasonal influenza vaccine. On-line article: https://debunkingdenialism.com/2013/09/28/anti-vaccine-misinformation-about-the-seasonal-influenza-vaccine/

Lei, J.H., Liu, L.R., Qiang, W., Yan, S.B, Yang, L., Song, T.R., Yuan, H.C., Lu, X., Han, P. (2015) Circumcision status and risk of HIV acquisition during heterosexual intercourse for both males and females: A meta-analysis. PLoS One, 10(5), e0125436. On-line: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4420461/

Osterholm, M.T., Kelley, N.S., Sommer, A., Belongia, E.A. (2012) Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis. Lancet, 12(1), 36-44. On-line abstract: http://www.thelancet.com/journals/laninf/article/PIIS1473-3099(11)70295-X/abstract

Siegfried, N., Muller, M., Deeks, J.J., Volmink, J. (2009) Male circumcision for prevention of heterosexual acquisition of HIV in men. Cochrane Database Syst Rev, Issue 2, (CD003362 (38 pages)). On-line: http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003362.pub2/abstract;jsessionid=D166B5CFFAF25DAE3C8949C3011EF7A8.f02t02

Sharma, S.C., Rainson, N., Khan, S., Shabbir, M., Dasgupta, P., Ahmed, K. (2017) Male circumcision for the prevention of HIV acquisition: A meta-analysis. BJU Int., ePub ahead of print. On-line abstract: https://www.ncbi.nlm.nih.gov/pubmed/29232046

Van Howe, R.S. (2015) Circumcision as a primary HIV preventive: Extrapolating from the available data. Global Public Health, 10(5-6), 607-25. On-line abstract: https://www.ncbi.nlm.nih.gov/pubmed/25760456

Wamai, R.G., Morris, B.J., Waskett, J.H., Green, E.C., Banerjee, J., Bailey, R.C., Klausner, J.D., Sokal, D.C., Hankins, C.A. (2012) Criticisms of African trials fail to withstand scrutiny: Circumcision does prevent HIV infection. J. Law Med., 20(1), 93-123. https://www.ncbi.nlm.nih.gov/pubmed/23156651

 

The ELISA test for HIV used in the African trials has a 3 month “window period”. This affected the results.

This canard was used by Dr Ryan McAllister (PhD physics) in his misleading video “Child circumcision: an elephant in the hospital”. His video is probably the most popular intactivist video on YouTube, and is cited endlessly by gullible intactivists who have fallen for its many errors. Maybe if they tried fact-checking its claims they would not talk so glowingly of it.

McAllister says: “In addition, they used an antibody test, which you know, if you go to an STD clinic and they use an antibody test, they tell you there is a three-month window. You’ve got to wait until three months after your last exposure if you really want to know if you’ve got HIV. But again they didn’t start the clock after three months, and most of the benefit or half the benefit occurred during that first three months. So those infections occurred prior to the start of the trial, prior to the randomization and intervention.” (McAllister, 2011).

Accompanying this commentary is a slide saying:

  • ELISA test ~ 3 month “window period” seroconversion
    Half the claimed protection appears in this window
  • Corrected results insignificant

The implication is that the results were compromised because most of the infections actually “occurred prior to the start of the trial”.

ELISA stands for Enzyme-Linked ImmunoSorbent Assay. It tests blood samples for antibodies to pathogens the patient has been exposed to. But it takes time for antibodies to form, so if the infection has occurred in the very recent past, there may not be enough antibodies present for the test to detect. In the case of HIV, it can take up to 3 months after exposure for the test to detect the resulting antibodies. This is called the “window period”. Accordingly, an infection that occurred in that period may be missed by the test. In practice, in the great majority of cases the test does detect HIV antibodies long before 3 months has passed, and the method continues to be improved, reducing this period further (Gilbert & Krajden, 2010), so McAllister is exaggerating.

Now for where McAllister has gone completely wrong.

In the Kenyan trial, blood samples were taken at 1, 3, 6, 12, 18 and 24 months for diagnostic testing. But samples were also taken right at the start (baseline) to eliminate any already infected (8 % were, and a further 1 % were indeterminate). Initially they were tested by two rapid tests. If either was positive then confirmation was by ELISA. Those who gave a positive result at 3 months had their 1 month sample tested by PCR (Polymerase Chain Reaction) which has a much shorter window period (as little as a week, but the test is much more expensive). If this was also positive they had their baseline sample tested by PCR. In this way four men were found to have been positive at the start, and so were excluded from the study.

As with the Kenyan trial, participants in the Ugandan trial were screened at the start to eliminate those already HIV positive. Those who seroconverted at later visits had their previous samples tested by PCR. In this way three men were found to have been positive at the start, and so were excluded from the study.

In the South African trial, three different variants of the ELISA test were used on samples collected at start, 3, 12 and 21 months. Only those that were positive for all three tests were treated as positive, the rest negative. Those testing positive in the first month were excluded from the statistical analysis. By the third month the window period would be over.

So in all three trials the window period was accounted for, whether by checking against the better (but costlier) PCR test, or by simply excluding any who seroconverted early. But McAllister’s deception goes beyond simply ignoring what the trials said about the HIV testing. He also asserted that “half the benefit occurred during that first three months”. This is just false. In all three studies the protective effect was maintained throughout the study period, if anything increasing with time, as is clear from the graphs in each of the studies.

Does McAllister not bother to read the studies he criticises? Or does he just make things up? Given the number of things he gets wrong, perhaps it is both.

References
Auvert, B., Taljaard, D., Lagarde, E., Sobngwi-Tambekou, J., Sitta, R., Puren, A. (2005) Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial. PLoS Medicine, 2(11), 1112-22. On-line: http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020298

Bailey, R.C., Moses, M., Parker, C.B., Agot, K., Maclean, I., Krieger, J.N., Williams, C.F.M., Campbell, R.T., Ndinya-Achola, J.O. (2007) Male circumcision prevention for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet, 369(9562), 643-56. On-line: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60312-2/fulltext

Gilbert, M., Krajden, M. (2010) Don’t wait to test for HIV. BC Med. J., 52(6), 308. On-line: http://www.bcmj.org/bc-centre-disease-control/don%E2%80%99t-wait-test-hiv

Gray, R.H., Kigozi, G., Serwadda, D., Makumbi, F., Watya, S., Nalugoda, F., Kiwanuka, N., Moulton, L.H., Chaudhary, M.A., Chen, M.Z., Sewankambo, N.K., Wabwire-Mangen, F., Bacon, M.C, Williams, C.F.M., Opendi, P., Reynolds, S.J., Laeyendecker, O., Quinn, T.C., Wawer, M.J. (2007) Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet, 369(9562), 657-66. On-line: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60313-4/fulltext

McAllister, R. (2011) Child circumcision: An elephant in the hospital. YouTube video: https://www.youtube.com/watch?v=Ceht-3xu84I&t=1743s

 

In the African trials men in the circumcised groups received additional safe sex counselling and condoms

Last updated: March 2023

This is another of the so-called “methodological flaws” in the African circumcision trials, and appears occasionally in intactivist critiques of those trials, thus:

“Why did men in the male circumcision groups receive additional counselling on safe sex practices?” (Boyle & Hill 2011).

This was uncritically copied by Earp (2012): “As Boyle and Hill point out, the men who were circumcised got additional counseling about safe sex practices compared to the control group”.

And in the immensely popular, error-riddled and thoroughly dishonest YouTube video “Child Circumcision: an elephant in the hospital” (McAllister 2011): “Each time the people came to the clinic they gave them condoms and safe sex counselling, but the circumcised group came at least twice more because they had to come to get circumcised and they had to come to get a follow up study” whilst showing a slide that says “More safe sex counselling, condoms to circumcised group”.

And a related claim by Green et al. (2010): “Participants received continuous counseling, free condoms, and monitoring for infection, which was unlikely in real-world campaigns”, although they do not say that the groups received differential counselling, rather that they all got extra counselling. This was swiftly shot down by critics: “The argument that the clinical trials of medical male circumcision lack external validity because of ideal counselling conditions and condom promotion is nonsensical because both study arms were equally exposed to those noncircumcision interventions” (Bannerjee et al. 2011).

And contained in that last quote is the answer to the previous objection – that the circumcised groups received more safe sex counselling than the controls. They didn’t. The claim, like many intactivists make, is false.

One only has to read the original studies to see this.

The Kenyan trial (Bailey et al. 2007)
In the Kenyan trial, when the participants were randomly assigned to their group those assigned to the intervention group were circumcised the same day or very soon after (88 % were done within 3 days). The study authors wrote: “Postcircumcision visits were scheduled for 3, 8, and 30 days to check the wound, record any complications, and ask about sexual activity, level of pain, resumption of normal activities, and satisfaction with the procedure. Participants were counselled to refrain from sexual activity for at least 30 days after the procedure.” Note that there is no suggestion that the men received any safe sex counselling, or condoms, during this period. Advising them to refrain from sex during healing is not safe sex counselling (but has also been incorrectly used by intactivists to attack the trials: http://circfacts.org/medical-benefits/hivaids/#hiv6 ).

A little later the authors continue: “At each study visit—1, 3, 6, 12, 18 and 24 months after randomisation—all participants received HIV counselling and testing, underwent a genital examination to check circumcision status, and were asked questions about sexual activity.” Note “all participants” (emphasis added).

So there were two types of return visit: “Postcircumcision visits” to check on wound healing and ask if the men had engaged in sexual activity, but without mention of safe sex counselling. And “study visits” in which safe sex counselling was given to all participants. In the course of these latter visits, “Individually tailored risk reduction counselling occurred at every visit.”

In an email from the lead author it was confirmed that the subjects did not receive safe sex counselling during the 3 & 8 day visits to check on healing (Prof. Bailey, personnel communication, March 2023).

The Ugandan trial (Gray et al. 2007)
As with the Kenyan trial, most men in the intervention group were circumcised within days of enrolment, 91 % within two weeks. Again, check-ups followed to monitor wound-healing and sexual behaviour: “Postoperative follow-up visits were scheduled at 24–48 hours, 5–9 days, and 4–6 weeks. The first visit was done at the surgical clinic site; subsequent visits occurred in mobile clinics in the communities. … At each postoperative follow-up visit, participants were questioned about symptoms suggestive of complications, and the wound was inspected. Participants were asked about resumption of sexual intercourse, and those who had resumed such activity were asked about condom use”. Again, note there is no mention of them receiving any safe sex counselling, or condoms, at these visits.

The authors continue: “All participants in both groups were followed up at 4–6 weeks, and at 6, 12, and 24 months post-enrolment.” At these visits the men were questioned about their sexual behaviour, and samples taken for tests. In relation to these visits the authors state, “Free condoms were offered to all sexually active participants at all study visits, and were also available through community-based condom depots stocked by the Rakai programme.” So again there were two types of visit, “postoperative” and “study”, but it was at the latter that condoms were available, and were available to both groups. Both groups will also have had equal access to the “condom depots”.

The South African trial (Auvert et al. 2005)
After initial screening, and circumcision of the intervention group (within a week), visits were arranged for 3, 12 and 21 months, so four visits in all (months 1, 3, 12 & 21). “At each of the four visits, each participant was invited to answer a face-to-face questionnaire, to provide a blood sample, and to have a genital examination and an individual counselling session.” Circumcisions were by three general practitioners in their surgeries, and it is not stated if they called the men back for check-ups, or if those practitioners also provided condoms. So, unlike the men in the Kenyan and Ugandan trials, the South African ones in the circumcised group may not even have had follow-up checks to see how well they were healing, or additional access to condoms.

The last nail in the coffin
As a further blow to the intactivists’ argument, in each trial, during the study visits, all participants were asked about their sexual behaviour (number of partners, use of condoms, etc.).  These details were recorded and the data analysed for statistical significance. So we actually know how each group behaved. In the Kenyan trial there was very little difference in sexual behaviour between the two groups, if anything, the circumcised ones had very slightly more sexual partners, and were very slightly less likely to use condoms, but the results were barely significant.

In the Ugandan study there was a slighter greater tendency for the circumcised group to use condoms, but only for the first six months, thereafter condom use was the same for both groups, yet the protective effect persisted, or even grew stronger. So it cannot have been due to a slightly higher use of condoms near the start. The control group were slightly more likely to drink alcohol around the time of sexual intercourse, but other parameters (number of partners and use of prostitutes) were the same for both groups.

There was a slight tendency for the circumcised men to show riskier behaviour in the South African trial, but only the number of partners was statistically significant. Yet still the circumcised group were less likely to contract HIV than the controls. This contradicts the intactivist narrative that these men had more safe sex counselling, and hence behaved more responsibly, they didn’t on both counts.

Conclusion
The claim is a demonstrable falsehood, as with so much of what intactivists say, and it is a trivial matter to read the original studies to see this. In addition sexual behaviour, for both circumcised and control groups, was monitored throughout each trial. There was little difference.

Once again we see the importance of fact-checking every claim intactivists make.

References
Auvert, B., Taljaard, D., Lagarde, E., Sobngwi-Tambekou, J., Sitta, R., Puren, A. (2005) Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 trial. PLoS Medicine, 2(11), 1112-22. On-line: http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020298

Bailey, R.C., Moses, M., Parker, C.B., Agot, K., Maclean, I., Krieger, J.N., Williams, C.F.M., Campbell, R.T., Ndinya-Achola, J.O. (2007) Male circumcision prevention for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet, 369(9562), 643-56. On-line: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60312-2/fulltext

Bannerjee, J., Klausner, J.D., Halperin, D.T., Wamai, R., Scheon, E.J., Moses, S., Morris, B.J., Bailis, S.A., Venter, F., Martinson, N., Coates, T.J., Gray, G., Bowa, K. (2011) Circumcision denialism unfounded and unscientific. Am. J. Prev. Med., 40(3), e11-e12. On-line: http://www.ajpmonline.org/article/S0749-3797(10)00791-9/fulltext

Boyle, G.J., Hill, G. (2011) Sub-Saharan randomised clinical trials into male circumcision and HIV transmission: Methodological, ethical and legal concerns. J. Law Med., 19(2), 316-34. On-line: http://artemide.bioeng.washington.edu/2011-12_JLM-Boyle-Hill.pdf

Earp, B.D. (2012) When bad science kills, or how to spread AIDS. Originally on the Oxford University “Practical Ethics” blog, it has since been reproduced here: http://www.creativitypost.com/activism/when_bad_science_kills_or_how_to_spread_aids

Gray, R.H., Kigozi, G., Serwadda, D., Makumbi, F., Watya, S., Nalugoda, F., Kiwanuka, N., Moulton, L.H., Chaudhary, M.A., Chen, M.Z., Sewankambo, N.K., Wabwire-Mangen, F., Bacon, M.C, Williams, C.F.M., Opendi, P., Reynolds, S.J., Laeyendecker, O., Quinn, T.C., Wawer, M.J. (2007) Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet, 369(9562), 657-66. On-line: http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(07)60313-4/fulltext

Green, L.W., Travis, J.W., McAllister, R.G., Peterson, K.W., Vardanyan, A.N., Craig, A. (2010) Male circumcision and HIV prevention: Insufficient evidence and neglected external validity. Am. J. Prev. Med., 39(5), 479-82. On-line (first page): http://www.ajpmonline.org/article/S0749-3797(10)00439-3/fulltext

McAllister, R. (2011) Child circumcision: An elephant in the hospital. YouTube video: https://www.youtube.com/watch?v=Ceht-3xu84I&t=1743s