The prevailing wind in the battle of statistics over clinical judgment

Chaplin v Pistol 2020 EWHC 1543 (QB)

The case law

Mr Justice Kay has delivered another blow against statistical evidence on life expectancy, reinforcing the approach taken by Master Davison in Dodds v Arif 2019 EWCA 1512 (QB).

The High Court dismissed an application by the defendant to rely on statistical evidence on the issue of life expectancy, a similar application having been rejected at earlier CMC in July 2019. Whilst the statistical machinations are of interest, the broad lesson to be drawn from these two decisions is that statistics will, in the vast majority of cases, be the starting point and that the Court will focus on the clinicians’ views, even if such views will themselves draw on statistical material.

Mr Justice Kay in Chaplin said:

“Evidence from a medical statistician is, in principle, admissible although ordinarily it should be seen as the starting-point for the clinical judgements made by medical witnesses…. In my experience medical experts are usually well able to apply and interpret quite complex statistical evidence which can be admitted as hearsay (particularly if set out in a published paper which has been peer-reviewed) without the need to call probative or explanatory evidence.”

Master Davison in Dodds said:

“For these reasons, it seems to me that bespoke life expectancy evidence from an expert in that field should be confined to cases where the relevant clinical experts cannot offer an opinion at all or state that they require specific input from a life expectancy expert or where they deploy, or wish to deploy statistical material, but disagree on the correct approach to it. This case does not, or does not yet, fall into any of these categories."

The decision in Chaplin was probably influenced by the fact that the statistical material that the expert relied on was as yet unpublished and had not been subject to peer review. The risk of prejudice to the claimant when only some four months from trial was material and the Judge was persuaded that an adjournment would be unfair.

The statisticians concerned, Professor David Strauss and Jordan Brooks, stated that there was unpublished data which significantly reduced the life expectancy data found in the 2015 data series. The suggestion was that whereas the medical experts in Chaplin considered that life expectancy was in the range of 35-44% of normal life expectancy (as a result of the injury sustained) the new unpublished data would reduce the individual claimant’s life expectancy to 27% of normal.

Some background concerning life expectancy in MCS and PVA cases

Previous life expectancy data published in 2007 provided separate specific life expectancy figures for those individuals with the most severe brain injuries (for example those in persistent disorders of consciousness – i.e. in a persistent vegetative state (‘PVS’) or minimally conscious state (‘MCS’).

Subsequent data published in 2015 no longer sought to separate life expectancy estimates for those in PVS / MC, from those with serious (albeit less profound) brain injuries who were unable to ambulate or self-feed. As would be expected, individuals in PVS / MCS are likely to have a significantly reduced life expectancy compared to other individuals with brain injuries who, whilst unable to walk / self-feed, may demonstrate significantly higher levels of awareness, and a substantially increased level of function.

Accordingly, life expectancy estimates for those in PVS / MCS based solely upon the data published in 2015 risks a significant over-estimate (with implications accordingly in respect of life multipliers for the calculation of future losses).

Prof Strauss has indicated that additional data (as yet unpublished) is available, which would deal with this issue by providing updated life expectancy figures solely for those in PVS / MCS (rather than by considering these individuals as part of a cohort which relates to a wider range of individuals with brain injuries on a broader spectrum of severity).

It is unknown whether this additional data will be published. However, if and when it is made available, it is likely to provide significant assistance to medical experts advising on the question of life expectancy in PVS / MCS cases. It is evidently in the interests of both claimants (with such injuries) and insurers that the courts be provided with the most accurate and appropriate data, in order to ensure that appropriate multipliers can be identified for the purposes of quantifying future losses.

Lessons to be learned?

1. If any party wishes to adduce statistical material in addition to the clinical lead, a case must be made out through the clinician as to why the Court would be assisted.
2. The furore concerning statistical issues should not obscure potentially more important clinical comorbidities; these require early identification and specialist comment by relevant expert evidence. A weather eye to admissibility and the Court being persuaded to extend the remit of the experts in the case are critically important.
3. If statistical issues arise the Court will almost certainly require a joint statement on the issue by the opposing experts or very clear dispute in the use of the statistical evidence apparent on the face of the respective reports before hearing an application for permission if contested.
4. Life expectancy and the potential pool of experts required should be dealt with early in the process and as soon as it becomes clear that the reason for divergence is a statistical one, rather than clinical.
5. In case the issue is “parked” at the CCMC it would be prudent to flag the issue and to even include a contingency in the budget. The judge and the other side need to be placed on notice and any chance of negative impact on the timetable avoided and managed.
6. If a Judge refuses permission, then an immediate appeal is probably called for, failing which the Judge on any later application has an easy route to dismiss the application on procedural grounds.
7. Reliance on the 2015 statistical data remains a concern in PVS/MCS cases (and “emerging” cases) but pending publication we have to rely for now on case specific clinical evidence and of course any relevant comorbidities. To assist the court, medical experts should be asked to discuss the 2015 statistical data, and explain how an individual claimant’s level of brain injury is likely to compare to others within the same category from which the data is drawn. This may assist experts in justifying why individualised life expectancy estimates may differ from average figures drawn from the published data.
8. What about Covid-19? As the impact of the pandemic becomes clearer over time we are likely to see a range of statistical and epidemiological material on the impact on life expectancy (unless a vaccine / therapy is developed that might cap the most extreme outcomes on any scale.) If this is an issue that an insurer wishes to pursue, this decision serves as a reminder of the need to build up from the lead clinician, identify the statistical “hole” and apply early in order to lay the case out on clear clinical grounds.

 For more information, please contact Jamie McCabe or Andrew Underwood.