Proving a cause-effect relationship between the defendant’s behaviour and the alleged outcome

Edwin J. Bernard
Published: 18 July 2010


  • RT @gnpplus: STATEMENT: GNP+ @HIVJusticeNet condemn dismissal of appeal in Singapore HIV criminalisation case.“HIV prevention is a shared r… 03 Jul 2019
  • Breaking news! @gnpplus and @HIVJusticeNet condemn dismissal of appeal in Singapore #HIVcriminalisation case… 03 Jul 2019
  • Please help us document the process of supporting HIV criminalisation survivors. We are seeking input from legal o… 27 Jun 2019
  • We wholeheartedly agree. A strong @UNAIDS means a strong human rights-based approach to @HIV #HIVisnotacrime… 26 Jun 2019
  • RT @alexmcclelland: The Global Commission on HIV & the Law is concerned about the increased misuse of individuals’ personal health informat… 21 Jun 2019
  • BREAKING: Justice Committee report recommends wide-ranging reforms to HIV criminalisation in Canada. Congratuations… 18 Jun 2019
  • RT @AIDSLAW: We've spent decades advocating for this critical change. We're very happy that it's a key part of the Justice Committee's repo… 17 Jun 2019
  • Published today - 'Using Research in the Fight Against #HIVCriminalisation - A Guide for #Activists' to help advoca… 13 Jun 2019
  • RT @HIVLawCom: A new @HIVJusticeNet report analyses the growing global movement against #HIV criminalisation. Six countries saw precedent… 06 Jun 2019
  • #HIVIsNotaCrime! Now in Spanish, French & Russian -- #MakingMediaWork for #HIVJustice: An introduction to media eng… 06 Jun 2019
  • RT @WOLA_org: New report by @WOLA_org, @IDPCnet, and @Dejusticia shed light on the disproportionate impact of pretrial detention for drug o… 05 Jun 2019
  • Thanks to #RobertCarrFund for recognising a global network focused on ending #HIVcriminalisation was work that need… 29 May 2019
  • BREAKING - JUST RELEASED: New report analyses the successes and challenges of the growing global movement against… 29 May 2019
  • Previewing our new report, Advancing #HIVJustice 3, published on May 29th, at the #ChallengeCrim intersectional dia… 23 May 2019
  • HJN is delighted to be part of this important cross-movement project to #ChallengeCrim globally. 23 May 2019
  • RT @Account4All: "You have to work with grassroots civil society to make access to justice a reality and to ensure justice reaches every pe… 23 May 2019
  • Coming May 29th....Advancing HIV Justice 3: Growing the global movement against HIV criminalisation… 17 May 2019
  • RT @alisapower: If you have NGO managerial skills and a commitment to human rights in the context of HIV - this is an enormously important,… 15 May 2019
  • Could you be one of our new HIV Justice Network Supervisory Board Members? Applications close 31st May 2019. Full… 15 May 2019
  • RT @AIDSLAW: What this means is people living with HIV can face years in jail and be designated "sex offenders" for life if they don't tell… 23 Apr 2019

Difficulty in proving that one individual actually infected another is one of the main reasons why some prosecutors have called for HIV-specific laws, since those laws can be written in ways that lower the evidential bar.1

In jurisdictions without HIV-specific laws, and where HIV exposure is criminalised (usually as a result of non-disclosure of HIV-status prior to unprotected sex), the defendant may argue that they were not, in fact, ‘infectious enough’ to cause ‘harm’.

This relies on evidence regarding:

  • the type and frequency of sexual activity

  • condom use, noting frequency and reliability

  • viral load and whether or not the defendant was on antiretroviral therapy.

Since a variety of factors can make a major difference to the risk of HIV exposure in each individual case, prosecution and defence typically rely on expert witnesses who may present very different points of view regarding the risks of the behaviours in question.1 It is then left up to the jury or judge to decide on the merits of the expert testimony.

As assessing the evidence relating to risk of HIV exposure is covered in detail in the chapter: Risk, the remainder of this section will focus on the issue of proving causality in cases of alleged HIV transmission.

1. It has been observed that generating proof of who infected whom is neither easy nor cheap.For an example from the US context, see Shevory T Notorious HIV: the media spectacle of Nushawn Williams Minneapolis: University of Minnesota Press 2004 (pp 140-141). [Chautaqua County District Attorney James] Subjack explained to this author that there are only two medical laboratories in the United States that do the phylogenetic testing needed to show that a particular victim was infected by a specific defendant, and that the test costs $25,000. Moreover, even if a state or county could afford to pay for the test, questions regarding the validity of the results exist even within the scientific community. The prosecution would need to hire experts to appear in court to vouch for the test results, at further costs of up to $100,000. Even that, as Subjack said, “may not show that the test is reliable. It’s up to the judge..”. Subjack's suggestion for dealing with these evidentiary problems was that the state should attempt to eliminate them.“The statute ought to by itself say, 'You have unprotected sex, know of your HIV-positive status, with an unknowing victim', that's it, crime over'.”

Three elements of proving causality: the fact, timing and direction of the alleged HIV-transmission event

The exact moment that sexual HIV exposure turns into the ‘harm’ of HIV transmission is almost always unknown. This is partly because, as discussed in the chapter: Risk, each sexual act that may lead to HIV exposure carries a very  low individual probability of producing HIV infection. Furthermore, even if symptoms of infection appear in the days or weeks that follow, they are often mild or nonspecific (see Detecting HIV infection, below). Consequently, criminal HIV-transmission cases are often investigated and/or tried many months or years after the alleged event.

Proving criminal HIV transmission requires the use of a combination of scientific evidence about HIV itself (known as 'virological evidence'), medical records and testimony in order to attempt to reconstruct the fact, timing and direction (i.e., who infected whom) of the HIV-transmission event under investigation. It may be difficult to prove any of the three beyond reasonable doubt, particularly if the complainant has never previously taken an HIV test, because there may be plausible alternative explanations for how HIV infection could have occurred.

Is the defendant the source of the complainant’s HIV infection?

Could the complainant have become infected with HIV by someone else, or not via sexual activity at all (e.g., through injecting-drug use or via a past medical procedure)? This would require investigating the medical, sexual and behavioural history of the complainant. Someone who has recently been diagnosed HIV-positive may well be mistaken regarding the source of their infection. Just because they have discovered that their most recent sexual partner is HIV-positive and had not informed them, does not necessarily mean that they acquired their own recently diagnosed infection from that partner or from the sexual transmission route.

Studies from very different settings have shown that people can often be mistaken when they identify a sexual partner as the source of their HIV infection. These studies all used phylogenetic analysis (explained below) to check whether the partner who was thought to be the source of their infection was the person who could have actually infected them.

A study of heterosexual couples in Cuba found that around two-thirds were mistaken when they named one of their recent sexual partners as the source of their infection during routine contact tracing.2 A study of gay men in California who had been very recently infected found that a third were mistaken when they were asked to name the sexual partner they believed to be the source of their infection.3

It is also plausible that someone may lie regarding the source of their infection in order to avoid admitting that they had sex with someone outside their primary relationship. An HIV-prevention study of long-term heterosexual couples in southern and eastern Africa found that 30% of those who became HIV-positive during the two-year study could only have acquired their infection from having sex with someone other than their primary partner. Although there are no data from the study regarding how many admitted to their primary partner that they were having sex outside their relationship, only three-quarters admitted to the researchers that they behaved in this way.4

Does the timing of events support the plausibility of the accusation?

Did the alleged HIV transmission take place before or after the person being investigated was made aware of his or her HIV status? Did the alleged HIV transmission take place before or after the complainant was made aware of the HIV status of the person being investigated? Medical histories can suggest – but not definitely prove – timing, and may help corroborate testimony relating to the defendant’s intent and the complainant’s awareness of the risk of becoming infected.

Can the direction of transmission be proven?

Instead of the defendant transmitting HIV to the complainant, might the complainant actually have been the source of the defendant’s HIV infection? Does evidence exist to prove or disprove either scenario? Again, one source of evidence may be the medical, sexual and behavioural histories of both the person being investigated and the complainant.

Using evidence to establish the fact of transmission by the defendant

The criminal justice system, particularly in high-income countries, is now familiar with the use of scientific evidence following the successful introduction of DNA evidence into court in order to match the genetic fingerprints of individuals accused of crimes with evidence from the crime scene.5

However, a common misconception – shared by police, prosecutors, defence and judges – is that scientific evidence that examines the DNA (or RNA) of the virus in both complainant and defendant (known as phylogenetic analysis) can prove with the same sort of certainty that the defendant infected the complainant. Such virological evidence alone cannot provide enough information to ascertain who might have infected whom or who might have become infected first.6,7,8,9

What phylogenetic analysis reveals and does not reveal about the relationship between two slightly different strains of HIV

Phylogenetic analysis requires the use of complex computational tools to create a hypothetical diagram (known as a phylogenetic tree) to estimate how closely related the samples of HIV taken from the complainant(s) and defendant are likely to be in comparison to other samples.

Although there are a variety of methods by which scientists can increase their confidence that the viral samples are very closely related in comparison to other samples, they could never be completely confident that the defendant infected the complainant based on phylogenetic analysis alone.

However, if the samples are not closely related with a high degree of confidence, this is evidence enough to show that the defendant could not, and did not, infect the complainant. Consequently, there is enough reasonable doubt to allow the prosecution to drop all charges, or for the judge to recommend to the jury that they acquit.

Experts in virology state that the only ‘safe’ use of phylogenetic analysis in criminal HIV transmission cases is to exonerate the accused.8 This important evidential issue resulted in an acquittal in a 2006 case in England,10 subsequent to which guidance was produced regarding the pitfalls and acceptable standards of using phylogenetic analysis in a criminal HIV transmission case;7,6 these have been incorporated into prosecution guidelines for England and Wales.11 Subsequently, many cases have been dropped during investigation12 and several cases have been dismissed during early stages of trials.13 This has also been the case in Sweden, which was one of the first countries in the world to routinely use phylogenetic analysis in criminal cases.14

It is crucial to note that the inverse is not true – virological evidence of a close relationship between the samples does not prove the defendant’s guilt. This is because an expert’s testimony about being confident that the samples are ‘very closely related’ does not definitively answer the question of whether one particular person was the source of another person’s infection. Other ‘closely related viruses’ may be found in people who are part of a wider transmission network (for example, individuals who have current or former sex partners in common, whether they know it or not, and the sex partners of those partners).

This means there are multiple ways to account for why a complainant and defendant may have highly similar strains of HIV. For example, the defendant could have transmitted HIV to an intermediary who was the actual source of the complainant’s infection, or both the defendant and complainant could have acquired very similar strains of HIV from someone else.

Some possible routes of transmission where B alleges that A is the source

Since phylogenetic analysis is costly and time-consuming, however, it is unlikely to be available in all jurisdictions, particularly those in resource-limited countries.15 Given the strong possibility that complainants can be mistaken regarding the source of their infection – especially in cases where the complainant has no previously documented HIV-negative test and a history of unprotected sex with at least one other person – in jurisdictions where phylogenetic analysis is not available, and other evidence is less than compelling, there should be strong scepticism from all members of the criminal justice system, and the media, regarding all claims about the alleged source of the transmission.

Establishing the timing and direction of HIV transmission through medical histories

The medical histories of the accused and the complainant can often help shed light on the timing and direction of the alleged HIV transmission under investigation. However, as with virological evidence, there are limitations as to what can be concluded from this information. This is often because complainants in HIV-transmission cases do not have a history of HIV testing.

Establishing when the accused and the complainant were infected (rather than diagnosed) with HIV is extremely important, but may not be possible depending on HIV testing frequency. An HIV-positive diagnosis may occur at any time following infection, from several weeks to a decade or more later.

Complainants in criminal HIV transmission investigations might not have undergone HIV testing until after ending the relationship with the accused. However, unless medical history suggests no other possible prior HIV risks – sexually or otherwise – it would be wrong to assume that a complainant was HIV-negative prior to his or her relationship with the defendant in the absence of a documented, previous negative HIV antibody test.

In principle, establishing the timing and direction of infection would require the highly invasive process of contacting everyone the complainant had sexual relationships with before testing HIV-positive. If any of those people are HIV-positive as well, phylogenetic analysis can be used to help determine whether or not they, rather than the defendant, might have been the source of the complainant’s infection. (As stated earlier, phylogenetic analysis only allows for, or excludes, this possibility – a strong similarity between viral strains is not definitive proof that one person infected the other.) If any of these partners cannot be traced, each untraced partner should be considered a plausible alternative source of infection.

Similarly, the defence may require access to the complainant’s sexual history, which may include other information supporting the plausibility of an alternative HIV source. If an HIV test was refused (or not offered) at about the same time as an STI was detected, because HIV and STI transmissions can happen together, the STI diagnosis can suggest prior HIV infection of the complainant at about this time as a possibility. However, since HIV is much more difficult to acquire than other STIs,16 caution should be used when interpreting such information.

Detecting HIV infection

NB: Summarised from Transmission and Testing.

Very few people are diagnosed with HIV within the first few weeks after infection, although some people develop symptoms, usually between two and four weeks after infection with HIV. This is known as seroconversion illness. Noticeable symptoms may occur in around 50% of individuals, with the severity of the symptoms varying from a mild flu-like illness at one extreme to an illness severe enough to require hospitalisation at the other. The symptoms are commonly misattributed to other illnesses such as influenza, glandular fever, or tonsillitis. HIV-positive people may believe in retrospect that a period of symptoms that they attributed at the time to something else indicates the approximate date and source of HIV infection, but this would be speculative.

The immune system usually begins to produce antibodies to fight HIV within two to three weeks of infection and the majority of infected people will test positive one month after infection. In some people, antibodies might not be measurable for up to three months. The time between infection and the detection of antibodies is called the ‘window period’. During the window period, recently infected people may already have very high levels of HIV in their blood and sexual fluids even though the most commonly used HIV tests – which detect antibodies or antigens but not the virus itself – would give a negative result. It should be noted that people with HIV are at their most infectious during the window period because their immune systems have not yet been able to effectively control the virus following the initial explosion of viral activity.

Although the standard means of identifying HIV-positive people has previously been through the use of antibody testing alone, fourth-generation HIV tests are able to detect the presence of HIV in the body earlier by looking for antigens as well as antibodies. Fourth-generation HIV tests reduce the window period by around five days compared to third-generation HIV tests, but the window period still ranges, on average, between two weeks and one month following infection. Fourth-generation HIV tests may not be available in all settings, however. Further details of HIV-testing methods and their reliability and window periods can be found in Testing.

Other immunological and virological data

Attempting to work out the timing of HIV infection via the immune system (measured by absolute numbers of key immune cells targeted by HIV, known as CD4 cells) or viral-load levels is extremely difficult, and any information obtained during an investigation should always be interpreted with great caution. During chronic HIV infection there is such a wide individual variability in both viral load and CD4 count that it would be nearly impossible to draw firm conclusions regarding the length of time the virus has been reproducing in a person’s body.17

In order to calculate rates of recent HIV infection in a particular population, public health agencies may use a test known as a ‘detuned assay’. There are several variations of detuned assays, depending on setting, and these can be known as: the BED capture enzyme immunoassay (CEIA); Serological Testing Algorithm Recent HIV Seroconversion (STARHS); or Recent Infection Testing Algorithm (RITA). A detuned assay is a much less sensitive antibody test than is regularly used to detect HIV infection. People recently infected – those within about the last four or six months – may not have produced enough antibodies yet to test positive on the detuned test, even though they tested positive on the regular HIV test. Therefore a negative detuned assay combined with a positive regular HIV test suggests recent infection, usually within the last six months.

Sometimes, these results are passed on to individual patients who may desire to know how long they have been living with HIV. Although detuned assays can be used to make reasonable predictions about the wider epidemic based on the proportion of recent infections, they are not reliable enough to be accurate on an individual level.  This is because studies have found that it wrongly classifies a significant proportion of individual patients.18,19 Consequently, results must be treated with caution and should not be exclusively relied upon as evidence of recent transmission in a legal context.20


  1. Center for HIV Law & Policy Sample expert statement on HIV transmission risk. Center for HIV Law & Policy, 2009
  2. Resik S et al. Limitations to contact tracing and phylogenetic analysis in establishing HIV type 1 transmission networks in Cuba. AIDS Research and Human Retroviruses 23 (3); 347-356, 2007
  3. Smith DM et al. A public health model for the molecular surveillance of HIV transmission in San Diego, California. AIDS 23, 225-232, 2009
  4. Celum C et al. Acyclovir and transmission of HIV-1 from persons infected with HIV-1 and HSV-2. N Engl J Med 362(5):427-39, 2010
  5. Jones T The rise of DNA analysis in crime solving. Guardian, 10 April 2010
  6. Bernard EJ et al. HIV Forensics: the use of phylogenetic analysis as evidence in criminal investigation of HIV transmission. London: NAM/NAT. Available at:, 2007
  7. Bernard EJ et al. HIV forensics: pitfalls and acceptable standards in the use of phylogenetic analysis as evidence in criminal investigations of HIV transmission. HIV Med. 8(6):382-7, 2007
  8. Pillay D et al. HIV phylogenetics: criminal convictions relying solely on this to establish transmission are unsafe. BMJ 335:460-61, 2007
  9. Maroof L The Limits of Science. Counsel, 2009
  10. Carter M Prosecution for reckless HIV transmission in England ends with not guilty verdict. Available at:, 9 August 2006
  11. Crown Prosecution Service Guidelines on Intentional or Reckless Sexual Transmission of Infection. CPS (England and Wales), 2008
  12. Terrence Higgins Trust, personal correspondence with the author. Unpublished, 2010
  13. Bernard EJ Reckless HIV transmission case dismissed due to insufficient evidence. Available at, 11 February 2008
  14. Leitner T and Albert J Reconstruction of HIV-1 transmission chains for forensic purposes. AIDS Rev 2: 241-251, 2000
  15. Willyard C Africa’s HIV transmission laws based on questionable science. Nature Medicine 13, 890, 2007
  16. Pinkerton SD et al. All STDs are not created equal: an analysis of the differential effects of sexual behaviour changes on different STDs. Int J STD AIDS 14:320-328, 2003
  17. Rodriguez B et al. Predictive value of plasma HIV RNA level on rate of CD4 T-cell decline in untreated HIV infection. JAMA 296 (12): 1498-1506, 2006
  18. Linley L and Reed C Applicability of population-based STARHS HIV incidence measure in determining recency of individual infection among patients attending STD Clinics. Eleventh CROI, San Francisco, abstract 854, 2004
  19. WHO WHO Technical Working Group on HIV Incidence Assays: meeting report. Cape Town, South Africa, 16 and 17 July 2009
  20. Phillips M et al. HIV Transmission, the Law and the Work of the Clinical Team. British HIV Association (BHIVA)/British Association of Sexual Health and HIV (BASHH), 2010

Proving a cause-effect relationship between the defendant’s behaviour and the alleged outcome

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.