The Limits of Science

Lara Maroof looks at the misuse of expert virological evidence in HIV prosecutions


Criminal lawyers are familiar with the significant role that DNA evidence can play in a case by adding a sense of scientific certainty. It would be a mistake though to assume that a similar role can be played by expert virological evidence to support a prosecution under s 18 or s 20 of the Offences Against the Person Act 1861 (“OAPA”) on the grounds that the defendant had intentionally or recklessly transmitted HIV to the complainant. In fact it is now clear that this sort of expert evidence is very limited in what it can show. A recent case illustrates the dangers. 


A tragic story

In March 2004 a young woman went for a night out drinking with friends. After drinking to excess, she ended up going home with a young man she met in the pub. She woke up in bed the following morning with him lying naked next to her. She believed they had had sex. A few weeks later, she began to feel ill with a mixture of symptoms—gastroenteritis, a rash, fever. This was described by the HIV physician as a “possible sero conversion illness”. (This is an illness experienced by about two-thirds of patients between two weeks to six months after infection with HIV.) A couple of months later, having met a new boyfriend, they both decided to go for HIV tests. Her test was positive, his negative. She suspected the one-night stand. She made a complaint against the defendant in November. He was arrested a few months later by which time he was serving a prison sentence for an unrelated offence. The police began their investigation by obtaining blood samples from both people. They were eventually sent to a virologist for analysis. Due to difficulties in identifying an appropriate expert and the police’s inexperience with obtaining this sort of evidence, the results only became available in September 2006, some 18 months after the defendant was first identified. It was confirmed that both had HIV. The virologist was asked to provide evidence of how similar the strains were. He claimed that there was a close genetic similarity which supported the allegation of direct transmission.

The defendant was not charged for another two years after the results became available, which itself was a few days before he was due to be released from his previous sentence.  The issue of bail arose. The prosecution submitted this was a very strong case. The defendant had HIV. Expert evidence had been obtained from a virologist. Through “phylogentic analysis” he had found they had “exactly the same strain of HIV.” An overwhelming case? The judge hearing the bail application thought so. He commented that it was a bit of a coincidence that the complainant had slept with the defendant and next thing you know she’s got HIV. In addition, it was said by the prosecution that the complainant was not dealing well with her diagnosis. It was feared that she didn’t have much longer to live. This was effectively a “death sentence”’—not your average s 20 offence. Bail was refused.
 

A complex and misunderstood subject

The judge’s decision on bail had been based on the impression given of the expert evidence and some “common sense” assumptions. It soon became clear that common sense and existing knowledge are inadequate when trying to understand HIV. This is an area which requires specialist knowledge and the guidance of experts for both prosecution and defence. There are many misconceptions and a high level of ignorance. The judge’s assumption that because the complainant had been diagnosed shortly after sex with the defendant, it was very likely he’d given it to her was based on one such misconception. Defence experts were eventually able to establish that she might have been infected up to two years before her diagnosis. This was a best estimate based on a back calculation considering the rate of her viral load decline. If the symptoms suffered in the months preceeding her diagnosis were caused by the onset of HIV, then she could have been infected up to six months before that date. It further turned out that she was not at death’s door. She was receiving anti-retrovirals and was in reasonable health. Fortunately, if there is early access to anti-retrovirals, HIV sufferers can live for another 40 or more years. It was not as straightforward as it first appeared.


The limitations

What of the claim that they had “exactly the same strain of HIV”? This was based on a misunderstanding of the expert evidence of the virologist. He had said they both had the same sub-type of HIV. This was correct but it turned out that there were approximately 8,000 people in the UK who also shared this sub-type. The expert had then gone on to compare the strains to see how similar they were using a process called “phylogenetic analysis”. Phylogenetic analysis is a complex procedure which is used by virologists to analyse the genetic variations between HIV strains. It is able to show which strains are more closely related than others. In criminal cases, the virologist will compare the sample from complainant and suspect with many other control samples to see how close a relationship there is between the two in comparison with the controls. In this case, the virologist was able to say that the two samples were more closely related to each other than they were to the control samples. Once understood, it was quite clear that the expert was not saying they had “exactly the same strain” of HIV or that this evidence proved direct transmission. It simply meant it was one possibility.

In fact it is now absolutely clear—contrary to the assumptions made at the start of the case—that this sort of evidence can never prove that one person infected another. It is essential for all parties involved to understand the limitations. First, it would never be right to suggest that a “unique” strain of HIV had been transmitted from one person to another, as if it were DNA. Even when one person has definitely infected another, for instance mother to infant, the virus strain transmitted will be different to the strain of the infector. This is because HIV constantly mutates. Second, this sort of evidence will only ever be able to support other existing evidence and can never, on its own, prove transmission. What it can show is that the two sample strains (from suspect and victim) are more similar to each other than to the control samples. It is therefore obviously essential to check that appropriate control samples are used in order for the results to have any real significance. The controls should come from individuals who are as similar as possible to the suspect and victim. For instance, if two samples from the UK are compared with controls from individuals from outside the UK, it would be no surprise to find the UK samples are more alike. For the most reliable results, the controls should come from the same geographical area as suspect and victim and be of the same sexual orientation.

However, if appropriate comparison controls are used, phylogenetic evidence can show that two strains are so similar that the individuals are part of the same “transmission network”. This could mean that one had directly infected the other. It could also mean that there is a third, fourth, fifth, etc party involved. In order to determine whether there has been direct transmission, it is necessary to look at the other surrounding evidence.


Informed investigation

In this particular case, it was quite clear that the police and the prosecution had overestimated the significance of the phylogenetic evidence. The result of thinking that the complainant and defendant had “exactly the same type of HIV” led to there being no further investigation of other possible sources of infection. There should have been. In HIV prosecution cases, the prosecution need to identify the relevant period during which infection could have taken place. They need then to identify all possible sources of infection from that period. This includes finding out whether the complainant has been an intravenous drug user and taking a full sexual history. It requires probing questions to be asked about the most personal aspects of a person’s life: the number of sexual partners and frequency of unprotected sex. Efforts will have to be made to trace any partners to discount them as possible sources of infection. The complainant should be made aware of this from an early stage as it is likely that his or her HIV status will be revealed in the process; otherwise the prosecution will never be able to prove their case to the necessary standard. 


Consequences?

The effect of the prosecution overestimating the expert evidence was catastrophic in this case. Instead of the prosecution understanding the key issues and making the necessary enquiries early on, it was not until disclosure after the start of the trial (following repeated requests from the defence) that it was revealed that there were a number of possible other sources of infection. There were numerous sexual partners who were untraceable. All this should have been known before charge. The result was that the defendant spent seven months on remand when there was no prospect of his ever being convicted. On top of that, his HIV status was revealed to any and everyone who happened to be sitting in court. He lived in constant fear that the word would get back to prison that he was HIV positive. 

But what of the complainant? She had waited for years for the case to come to trial during which time she was perhaps told no doubt that there was a strong case against the accused. After all that she was asked to provide a detailed breakdown of all sexual partners including one-night stands. Her lengthy medical records had to be disclosed and gone through with a fine toothcomb. She had to be questioned about all her untraceable sexual partners before a room full of strangers. The complainant would have been spared the whole humiliating and painful ordeal had the police and prosecution had a proper understanding of the limits of the expert evidence and carried out the necessary investigation from the start.


Lessons to be learned

The experience in this particular case demonstrates the need for police and prosecution to be fully aware of the limits of this sort of expert evidence. Phylogenetic analysis cannot provide anywhere near the degree of certainty that DNA evidence can. The experts themselves must emphasise from an early stage the limited extent to which such evidence can be used. If this is not done, it is likely that similar mistakes will be made again, leading to injustice both for accused and the complainant. 

Lara Maroof is a barrister at Charter Chambers

Category: 
Tags: