KB 2023 004108 - [2025] EWHC 1824 (KB)

In early August 2021, applications were made in the County Court for an order to debar the defendant (acting on behalf of any of the 18 insurer defendants in the Ersan claims) from adducing and relying upon JS1.

The applications to debar were initially advanced on the grounds that JS1 was inadmissible evidence of a quasi-expert nature, was unreliable, and that it ought to be excluded as it contravened the UK GDPR. A skeleton argument (settled by leading and junior counsel) was served in support, which included submissions as to the processing of personal data, including special category data, in JS1, and took issue with the suggestion (made by the defendant) that the processing was necessary for the purposes of establishing, exercising or defending legal rights and thus fell within the exemption permitted by paragraph 5(3) Schedule 2 Part 1 of the Data Protection Act 2018; in addressing the issue of necessity and the litigation exemption, specific reference was made to the fact that the data had not been anonymised.

Addressing these points in a further witness statement (“JS2”), dated 23 September 2021, Mr Stevens provided more detailed explanation as to the reliance placed on the JS1 data analysis and the way this had been compiled, noting that:

“6.

... a. [JS1] reveals an obviously troubling pattern in relation to the presentation of a significant number of claims who are represented by Ersan & Co .... b. More specifically, on its face, JS1 strongly indicates that all of the claims have been cynically managed so as to contrive an outcome whereby in every case, and irrespective of the true circumstances of that case, the Claimant is presenting a claim that they have suffered psychiatric harm as a result of the relevant index event. In other words, it indicates that a cynical “conveyor belt” approach has been adopted in respect of the management of the claims, which approach the Claimants themselves have then readily exploited for the purposes of contriving their individual claims for psychiatric damage. c. These matters must inevitably be treated as a key and integral part of the relevant factual matrix for the assessment of the Defendants’ pleaded case that the claims are by their nature each of them fundamentally dishonest. ...”

And continuing:

“e.

... i. ...[JS1] does not itself reveal any (or certainly any privacy significant) data beyond that which can already be discerned from the content of the publicly inspectable court files in the individual cases and, further ii. the Claimants have at no point sought any anonymity or confidentiality order from the court in connection with their claims.”

As for the evidence gathering process for JS1, Mr Stevens explained:

“33.

All case data relating to matters in which DWF is instructed are initially collated within DWF’s case management system, which ... is stored in a highly secure manner.

34.

... upon the instructions of each insurer client in the defence of legal claims, as a matter of general practice which is carried out across the insurance industry, in order to detect and prevent fraud, DWF carry out some analysis of all claims data in order to identify trends or patterns.

35.

When a new trend or pattern is identified, DWF will notify their insurer clients of the key identifying factors. During this process, the key attractors or behaviours is shared between DWF and the insurer client for the purpose of identifying their own exposure to such claims to assist with the defence of legal claims being made against them. To the extent that data is analysed or shared in this context it is for the exclusive, public interest purpose of safeguarding the insurance industry and thereafter the wider general public against fraudulent conduct.

36.

In relation to these matters, analysis was carried out by ... two individuals ....

37.

It was that cross-claims analysis which revealed that Dr Yahli had been used on a large number of claims where Ersan & Co were the legal representative. This in turn resulted in DWF conducting analysis of the content of the reports produced by Dr Yahli. That analysis was immediately concerning and indeed suggested that the Yahli Reports were entirely lacking in integrity. ...

38.

However, importantly, the analysis suggested that the concerns did not merely begin and end with Dr Yahli. Quite the contrary it indicated that there were serious questions as to the way in which the process across all the claims had been conducted from the outset ...

39.

Importantly, ... no individual person’s data was shared with any insurer client, as DWF took the view that such data did not need to be shared in order to reveal the wider pattern.

40.

Following its sharing of the key indicator data with insurers, DWF received instructions from a number of insurers to investigate these claims further. The majority of these instructions were dealt with by a small group of lawyers at DWF ... .”

Having identified who had had sight of the data at each stage of the process, Mr Stevens continued:

“51.

For the avoidance of doubt, DWF have ensured that the full data set has been reviewed by as few people as possible and only where absolutely necessary for the purposes of the litigation and/or as a result of subsequent correspondence from Ersan & Co Solicitors on behalf of the Claimants.”

Mr Stevens then went on to provide an explanation for each field used in the JS1 spreadsheet. Relevantly (given the way in which the claimants’ cases have developed before me), he addressed the use of the claimants’ names as follows:

“52.

...

...

e.

This column was required to identify the name of the Claimant. DWF’s case management system creates one reference per accident; not per claim. As there are a number of accidents which have resulted in multiple claims from multiple Claimants had the Claimant name not been provided there would have been multiple lines of data with the same reference. This would be confusing and could weaken the reliability of the data. It would also make it more [difficult] for the Claimants’ solicitors to verify the accuracy of the data. Given the people in the same vehicle in the same incident are family members, on a number of cases, using initials would not have been possible as the initials were the same. Therefore, the Claimant’s name (noting that this is information and data held by the Claimants’ solicitors in any event as they are their own clients) was deemed the most appropriate way of being able to enable a fair verification and validation of the data.”

Addressing the question of proportionality, Mr Stevens recounted how an earlier, smaller, similar fact exercise had been carried out, with the resulting statistical data having then been provided to Ersan, which had led to an application for disclosure of each document used to compile the statistical evidence. As Mr Stevens explained:

“58.

... for the purposes of [JS1] ... it was clear that Ersan & Co would equally request the data relating to the wider claims. Rather than provide many hundreds of CNFs and medical reports, which include more sensitive data than is reasonably required for the purposes of my statement (such as addresses, dates of birth, national insurance numbers, defendant names and addresses, information regarding other medical history), the spreadsheet was considered the best way to provide the data so that it could be verified by Ersan & Co on behalf of the Claimants but without providing data beyond that which was proportional and otherwise necessary in all the circumstances.”

The debarring applications in five of the Ersan claims were listed before HHJ Backhouse on 5 October 2021, at which stage the claimants’ interests were represented by a different leading counsel, Mr Coppel QC, acting alone. At that stage, the case for the Ersan claimants was advanced solely on the basis that JS1 amounted to quasi-expert evidence that was inadmissible and/or unreliable; arguments relating to breaches of data protection rights were not pursued, Mr Coppel explaining:

“Every advocate is exhorted, particularly nowadays, exhorted time and again to reconsider and refine their cases and maintaining only those that they consider to be their best. The logic of that, of course, is that if you’re not going to succeed on your best points it’s pretty unlikely you’re going to succeed on your worst.”

HHJ Blackhouse dismissed the application to debar, essentially on the basis that JS1 constituted admissible similar fact evidence and any issues concerning reliability would be a matter for the trial judge. Recording that reliance was no longer placed on any alleged breach of data protection legislation, HHJ Backhouse observed:

“Th[e earlier] skeleton argument is very different to that produced by Mr Coppel today. He explains that he looked at the material de novo and formulated his submissions based on what he thought were the best points and the only points he needed to make to succeed on his application. I make no criticism whatsoever of that approach, but it seems to me that if the claimants have strong grounds for saying that there has been some illegality in relation to data processing, and/or if there were strong grounds for saying that the information or the material is irrelevant, Mr Coppel would no doubt have said so.”

A subsequent appeal against HHJ Backhouse’s decision was dismissed by Freedman J in a judgment handed down on 21 November 2022 (Kerseviciene v Mide Quadri and anor and four other appeals [2022] EWHC 2951 (KB)). Although expressing reservations as to the probative value of JS1, Freedman J accepted that the defendant insurers in the county court proceedings were entitled to run a case relying on what was essentially similar fact evidence, seeking to derive patterns from a much larger body of evidence. There was no attempt to reinstate the data protection points before Freedman J.