Did the NHS LA mislead parliament?

"The headline-grabbing 30% increase [in clinical negligence claims] is pure fiction and is obviously and readily explained by a change in reporting methods."

Kerry Underwood, 25 August 2011

"In fact the whole of the apparent additional jump last year, over and above the rise of 6% or so, is explained by a change in the reporting system."

Kerry Underwood, 3 August 2012

As two charities teamed up last week to warn that the lapses in standards that lead to the Winterbourne View care home scandal 'could happen again', are similar lapses driving a rise in clinical negligence claims against the health service?

That might be the conclusion drawn from anyone flicking through the NHS Litigation Authority's (NHS LA) recent annual reports, which have seeming shown rises in the number of cases brought against medics.

However this picture was challenged by Kerry Underwood, Chairman of Underwood solicitors, who has argued that the NHS LA's reports are 'misleading parliament' about the scale of the problem.

So what's going on here?

The backstory

Before tackling the figures, it is important to give a bit of background.

In July 2011, the NHS LA released its annual report and accounts for the financial year 2010-11 in which it noted that the number of clinical negligence claims had increased by 30% on the previous year; a figure which was much larger than increases in previous years.

This provoked a blog by Mr Underwood who challenged whether the figure of 30% told the whole story, claiming that it was a gross overestimate and predicting that in 2011/12 the rise would be much smaller.

Sure enough, this year's NHS LA report showed a 5.6% growth in the number of claims, a rate of increase that was slightly lower than previous years. Returning to the topic, Mr Underwood underlined some of the reasons why he thought last year's figure was so anomalous.

Why was the figure of 30% so high?

The first table below records the number of clinical and non-clinical negligence claims that were open in the five financial years from 2007/8 to 2011/12 inclusive, as reported by the NHS LA Report and Accounts 2011-12.

The second table records the above figures as a percentage increase of the previous year — for example, 6088 is 11.3% larger than 5470:

According to Mr Underwood, the jump in 2010/11 of 30.1% that prompted many headlines is misleading as page 12 of the NHS LA's report states that:

"Part of the significant increase in [clinical negligence claims] may be explained to some extent by the requirement for claimants to now send us a copy of the Letter of Claim at the same time as it is sent to the defendant NHS body, at which point we now record the claim, but we are analysing patterns and trends to obtain a better understanding of the reasons behind the increase."

In fact, Mr Underwood goes so far as claim that the only plausible reason for the dramatic increase in the number of claims in 2010/11 was the methodological change for calculating the number of clinical claims between 2009/10 and 2010/11.

We got in touch with the NHS LA to see if they could explain the way the figures were arrived at in more detail.

They informed us that the compulsion for claimants to contact the NHS LA directly reduces the time between the claim being submitted and the claim being recorded by the NHS LA as the middle man (i.e. the defendant NHS body) is cut out.

As such, we would expect to see a one-off increase in the number of claims recorded in the 2010/11 financial year as claims that would normally have been counted in the following financial year's results would instead have been included in that year's figures, due to the quicker turn around.

For example, under the previous system, a claim passed to an NHS trust in March would be unlikely to reach the NHS LA in time for it to be counted for that year's results, whereas under the new system it more than likely would. Hence 2010/11's figures included both those claims that were submitted late in the previous financial year and those submitted late in 2011, producing a one-off jump in the numbers.

However, the NHS LA has questioned whether such a change would have produced an increase as large as 30%.

Although they were unable to say for certain why the figure had increased so much, a spokesperson suggested that it might have been influenced by an increase in aggressive marketing by firms offering conditional fee agreements encouraging individuals to take legal action (although they stressed that this was speculation, despite it also featuring in their annual report).

It is also worth noting that the NHS LA informed us that the methodological change was introduced in October 2010 meaning that it was only in effect for half of the year and it was therefore felt that it would not have a major impact in the claims volume for that year.    

Where does that leave us regarding Mr Underwood's claim?

So far, we cannot definitively know if the dramatic increase between 2009/10 and 2010/11 in the number of clinical negligence claims is entirely attributable to the methodological changes, as suggested by Mr Underwood.

Nevertheless, the change in methodology would not explain why the figure remained so high the year after.

In his blog dated 3 August 2012, Mr Underwood points to the reduction in the rate of increase in 2011-12 as evidence that his theory has been proved. However, by definition, a year-on-year increase takes the previous year as a baseline and thus the 2011/12 figure remains high in comparison to pre-2010/11 figures.  

Considering that the increase was put down to the one-off impact of speeding up the registering of end of year complaints, if the methodology was the whole story then then we might expect to see a decrease in the number of claims recorded in 2011/12 relative to 2010/11. As we can see from the data above, this was not the case: although the rate of increase fell substantially, the actually number of claims rose, which might suggest that other factors were in play from 2010/11 onwards.   

We put this to Mr Underwood who appreciated that we had a point.

However, he did question whether the change in methodology might have had an effect beyond the one-off increase in the number of clinical claims.

Allowing complainants to contact the NHS LA directly rather than through a 'middle man' might increase the number of claims recorded in total, as under the previous system certain claims may not have been passed to the NHS LA, either in error or because they were resolved before they reached the Authority. This effect on the statistics would be sustained year-on-year.

Furthermore, Mr Underwood suggested that the increase in the actual number of clinical claims might be partly attributable to an increase in the caseload of the NHS, meaning that while the raw numbers may have increased, the litigation rate could have remained relatively constant.  

What about Mr Underwood's other evidence?

Mr Underwood does offer some other evidence to support his claim.

In his 2011 entry, Mr Underwood pointed out that the number of non-clinical claims did not rise to the same degree as clinical claims.

Considering that non-clinical claims did not undergo the same methodological transformation as clinical claims, this might suggest that the changes to the ways the figures were recorded are driving the rise rather than an increasingly active 'compensation culture' (although the NHS LA argue that law firms are less active in directing marketing towards non-clinical cases).

Mr Underwood also points to the claims open at the end of the year as another measure of activity, highlighting a rise of 7.51% in the number clinical claims and a 7.52% rise in non-clinical claims between 2009/10 and 2010/11.

There is, however, an arithmetical issue with this claim.

The number of clinical claims open at the end of 2009/10 was 13,320 while the number open at the end of 2010/11 was 14,603 which is, in fact, an increase of 9.63%.

This does not, however, undermine the thrust of Mr Underwood's point. He writes:

"A big surge in new claims will result in an even bigger rise in figures because it will be the older, lower, number of claims falling out whereas the new ones are all still in."          

While it isn't necessarily clear that that old claims will always be sorted more quickly than new claims, implicit in Mr Underwood's suggestion is that an increase in the number of claims within a year would be replicated by an increase in the number of claims left over at the end of the year.

This is supported by evidence from the NHS LA who informed us that the average time for a claim to be completed was 15 months — well over a year.

The figures for clinical and non-clinical claims left over at the end of the year are much closer than the figures for the number of recorded claims in each category.

This would, according to Mr Underwood, bring into question whether the number of clinical claims recorded in 2010/11 was actually so much larger than the number of non-clinical claims.

Conclusions

This lengthy investigation leaves us little closer to understanding the origins of the mysterious figure of 30%.

However, from what we have learnt, there may be other factors in play beyond the methodological changes highlighted by Mr Underwood.

The most likely explanation, as far as we can tell, is that the increase was a multi-faceted phenomenon, although the number of facets and there relative importance is a matter for discussion.

However it may be going too far to claim that the NHS LA mislead Parliament.

We should always be very careful about data comparisons made over a period where there has been a significant change to the way the figures have been produced. This change has been flagged up by the NHS LA in its report, and the outstanding question is more one of the degree to which it is responsible for the spike in its caseload.