Great strides have been made in embedding international quality standards in forensic science work that is to be relied upon by investigators and the courts. But these standards alone can’t guarantee that mistakes won’t continue to be made with forensic science at the heart of them. Professor Angela Gallop explains why, and what should be done to minimise the risks.
Mainstream forensic science laboratories are regarded as ‘Calibration and Testing Laboratories’ for accreditation purposes and come under the ISO/IEC 17025 international Standard. Combined with adherence to the Forensic Science Regulator’s Codes of Practice and Conduct, this helps ensure that:
- All items submitted to the laboratory for examination and analysis are handled in such a way as to preserve their continuity (through unbroken chains of custody) and integrity (by avoiding contamination). This provides confidence that test results relate to the items they are supposed to, and that these items have not been contaminated whilst in the Laboratory with anything that could be mistaken for evidence
- All processes and procedures applied to items during testing are fit for their respective purposes, are recorded in a comprehensive quality management manual, and followed by all staff. Together with participation in proficiency tests and inter-laboratory comparisons, this means that it shouldn’t matter which scientist has been involved in which case – their results should all be the same
- All scientific instruments are in good working order, methods to be used on them properly validated, and all staff appropriately trained – as demonstrated by initial proficiency, and then regular competency testing. This means that the results they produce can be relied upon
- All staff understand, and have individually signed up to the Forensic Regulator’s Codes of Practice and Conduct. This means that their honesty and integrity should be beyond reproach
- Critical results and interpretations in each case are checked by another, similarly qualified scientist in a system of peer review. This provides an additional safety net
So where’s the problem?
Problems can arise in interpreting what the results of a particular examination or analysis are likely to mean in the context of a specific case. This is because interpretation can be critically influenced by eg. the amount of background information available to the scientist about the case, the selection of which items they are to examine and for what, the depth and breadth of experience of the scientist doing the work, the amount of time allocated to the case, and the type of report they are commissioned to produce at the end of it.
Amount of background information available to the scientist
Forensic science has become increasingly commoditised in recent years with police customers choosing from a menu of examinations and tests precisely how many and of which type they want to be applied to items they submit to the laboratory. This has been taken to obviate the need for much background information about the crime, the surrounding circumstances, and activities of the people involved. It follows that the scientist conducting the tests may not appreciate the potential for evidence linking a suspect with a crime to have arisen in innocent circumstances. Take the case of a man charged with murdering his wife by strangling her with a pair of her own leggings. Evidence presented by the prosecution included what looked like the husband’s DNA on the leggings. But when, as a result of a review on behalf of the defence, this was put into the context of what the husband said he did and the fact that he and his wife lived together, it lost its force as evidence of his direct involvement in her death, and the case was dismissed at half time.
Selection of which items to examine and for what
Pressure on police budgets has brought commissioning of forensic tests into sharp focus – not because forensics forms a large part of such budgets, but because it represents a large part of a Force’s external spend. This means that it stands out like a sore thumb and, in an attempt to reduce costs, there has been a concerted attempt to reduce the number of items submitted for examination.
On the basis that you can only find potential evidence where you look for it, this can result in distorted interpretations and some evidence being missed altogether. For instance, at the scene of a domestic burglary, multiple footwear marks were observed on the conservatory floor. Gel lifts were taken from the clearest marks and submitted to the laboratory where they were shown to represent two pattern types, neither of which matched the tread pattern of the suspect’s shoes. That would normally have been the end of the matter. But the scientist involved noticed that there were other fainter, more fragmentary footwear marks in photographs of the scene. When these were explored in detail, they were found to correspond with the suspect’s shoes – in pattern, pattern arrangement and unique damage features to the extent that they provided conclusive evidence that these marks were made by the suspect’s shoes.
Depth and breadth of experience
The adverse effects of inexperience are seen most often in newer forensic disciplines, and where traditional forensic science activities have been taken over by police or related personnel. For instance, there are numerous examples of faulty cell site evidence suggesting the owner of a mobile phone was somewhere he or she wasn’t, and of arson cases in which the wrong suspect has been identified, and which wouldn’t have come to light if the defence had not commissioned their own review of the work.
But there can also be problems in the best traditional forensic laboratory. A recent case concerned a man who was accused of injuring another in a fight. There was blood staining on his shoe which, the prosecution’s scientist suggested had resulted from him kicking or stamping on the other man. But our independent review suggested that it simply represented the sort of secondary back spatter you get when drops of blood fall into a pool of liquid blood nearby. In other words, it could have resulted from the defendant simply standing close to the other man while he was bleeding. There was certainly nothing to suggest any kicking or stamping; indeed, there was no complaint about this. But without our intervention, the initial opinion of the first scientist would have been taken at face value.
Type of report or witness statement produced
In an effort to save time and cost, Streamlined Forensic Reports or SFRs have been introduced. An SFR 1 provides a short factual summary of an analytical result, often written by a police officer, and an SFR 2 gives slightly more detail and, written by a scientist, offers a simple opinion of what the results mean. We have been involved in a number of cases where SFRs have been used inappropriately, sometimes with potentially serious consequences. In one case a drink can from a crime scene was examined by a crime scene examiner who reported that DNA from saliva on the can matched an individual with a statistic of 1 in 250 million. Fortunately a scientist appointed by the defence identified that: no saliva had been looked for let alone detected; the DNA result was a partial profile which could alternatively have resulted from someone just touching the can; no-one had checked the DNA match; the statistic was generic (best guess) and not specifically related to the case; and finally the crime scene examiner had not been trained in DNA reporting and so would be unable to answer any questions about it.
Besides shortcomings in the system that we know about, there will always be other sources of quality failure that are more difficult to predict. Each of the three main forensic science suppliers have suffered from these despite undoubted commitment to quality and quality management systems. In one case, a disposable sample tray was inadvertently re-used during the DNA profiling process, resulting in someone being arrested and charged with rape in a city he had never visited. In another case, the use of non-unique identifiers resulted in a sample from one crime scene being mistaken for one from another. This led to the victim in the first case being arrested and interviewed as a suspect in the second. In a third case, a simple handling error during DNA analysis resulted in someone being arrested and charged for two crimes when any connection was only with one of them. Other examples from accredited facilities relate to eg. incorrect programming or loading of instruments used to analyse toxicology samples leading variously to wrong results, or right results being attributed to wrong samples, and failure to calibrate equipment properly so firearms were wrongly classified as lethal.
So, how can we minimise the risks?
There are several things which would make an immediate improvement to quality in forensic science including ensuring that:
- Scientists are given sufficient background information to understand the significance or otherwise of their results in the context of the specific case at hand
- Scientists contribute to decisions about which items are to be examined and for what, and how their work should be reported
- All organisations providing forensic services – including the police themselves, are accredited to international standards and are not just ‘working in line with them’, and importantly
- Each organisation actively supports its own quality system, using it to embed a culture of continuous improvement, as opposed to merely ticking boxes and caring more about passing proficiency tests than casework itself, and
- Where forensic science evidence is critical to a case, it is always independently checked on behalf of the defence by a properly qualified and experienced, independent forensic scientist – also working in a quality accredited environment. So often problems in forensic science are only exposed as a result of this type of check – the ‘critical last level of quality assurance’ as a former Forensic Science Service Director once described it to me.
Only in these ways can we help ensure that forensic science is as good as it should be and is less likely to feature in miscarriages of justice.
Professor Angela Gallop CBE
Forensic Access Ltd
Angela Gallop is known equally for setting up and running full scale forensic laboratories and for personally leading the scientific teams who helped to solve many of the UK’s most complex and high profile criminal cases including Rachel Nickell, Damilola Taylor and St