High Court of Ireland Decisions

THE HIGH COURT

[2023] IEHC 202

Record No. 2020/5533P

BETWEEN

CATRIONA CRUMLISH

Plaintiff

-AND-

 

HEALTH SERVICE EXECUTIVE

Defendant

 

Judgment of Ms. Justice Mary Rose Gearty delivered on the 25^th of April, 2023

1. Introduction

1.1              The Plaintiff in this case was not successful but has asked the Court to make no order as to costs.  The default order in any costs hearing is usually expressed in the phrase “costs follow the event”, meaning simply that the winner takes it all; the losing party pays her own legal bills and those of the winning litigant.  To make a different order has potential repercussions for litigation generally.  If the losing party might not have to pay all of the costs of the case, speculative litigation becomes more likely; the claimant has less to lose by taking a case. A defendant which is a State body becomes a particularly vulnerable target in that scenario, given the perception that such bodies have access to more, and more reliable, funding than most private entities.  This would have serious implications for the public purse.  More fundamentally, the logic behind the rule is that it is the fairest way to approach the question of costs, leaving aside cases in which there is no outright winner where costs may be much more difficult to assess.

1.2              This Plaintiff could not prove the existence of a 15mm tumour in her right breast in 2017 which was the starting point for her claim that negligent failure to diagnose her cancer led to more extensive treatment and reduced her life expectancy as a result. If there was insufficient evidence of a discernible tumour in May, as the Court found, her treatment at that time was not relevant as it probably did not affect the position in October.

 

2.           Relevant Law

2.1              The default position is that the Defendant, having been successful, is entitled to recover its costs.  The factors to be considered by the Court in exercising its discretion to depart from the default position are set out at Section 169 of the Legal Services Regulation Act 2015.  The Court must have regard to the particular nature and circumstances of the case, and the conduct of the proceedings by the parties, including conduct before and during the proceedings, whether it was reasonable for a party to raise, pursue or contest one or more issues in the proceedings, the manner in which the parties conducted all or any part of their cases, whether a successful party exaggerated his or her claim, and other factors relating to settlement and mediation which are not relevant here. The law did not change in 2015 so, as before, one can depart from the rule that costs follow the event but, if the Court does so, it is an exceptional case requiring reasons for this departure from the usual order. 

2.2              The Plaintiff urged the Court to consider the case of Hanrahan v. Waterstone and others, [2021] IEHC 274, in deciding this issue.  There, a medical expert made a significant error of fact in a letter to another medic, leading that plaintiff and her legal team to pursue the case to a full hearing in circumstances where they might not otherwise have done so.  The defendant in question went so far as to seek only half his costs in acknowledgment of the mistake on his part.  As the defendant’s letter had misled the plaintiff in a significant respect, Cross J was unwilling to award him even half of his costs but also considered that to make no order, leaving each side to bear its own costs, would be unfair to the Plaintiff.  He made no order save as to outlay, to include the experts’ expenses and their fees.  This reflected the position that the Plaintiff had been unsuccessful but acknowledged the significant error by ensuring that the relevant defendant paid his own costs and some of the Plaintiff’s expenses, in particular, the initial costs of hiring the relevant experts.  His comment in this regard was that only deliberate or egregious conduct on the part of the successful defendant would justify an award of costs to the plaintiffs.

2.3              There was no deliberate or egregious conduct in this case, nor was there a significant error.  As set out in the substantive judgment, [2023] IEHC 194, the issue of whether the Plaintiff had an identifiable cancer at her first appointment with the Defendant had been strongly contested but the basis for that dispute only emerged with clarity on day 21 of the 24-day hearing.  The Plaintiff asks the Court to make no order as to costs on the basis that the issue on which the Defendant succeeded only crystallised in the closing stages of this hearing. 

2.4              Hanrahan was very different to this case.  Here, clinical oncology experts considered estimates of tumour growth rates based on data from academic literature and disagreed as to the reliability of such estimates.  The report on which the Defendant relied, Professor Crown’s report, was only served on the Plaintiff after the commencement of the hearing although no complaint was made in this regard and the Plaintiff’s expert was prepared to deal with the report, and commented on it in evidence, having considered it overnight.  There was no conduct comparable to that in Hanrahan.

2.5              The main submission in this case amounts to an argument that the Plaintiff did not predict, and could not have predicted, the true nature of the Defendant’s argument until the end of the hearing.  To assess this submission, it is necessary to re-state some of the facts already set out in the main judgment and to refer to the relevant expert’s report.

 

3.            Relevant Facts

3.1              The Plaintiff’s expert, Professor Bundred, used the concept of “doubling time” to calculate what size the Plaintiff’s tumour was in May of 2017.  All experts agreed that she probably had cancer in May, and it was clear that she had a 34mm tumour in October.  She had been assessed in May and the clinician identified a 15mm lump in the same quadrant of the breast with the radiographer confirming that there were cysts in the relevant area.  The radiographer wrote a contemporaneous report setting out this finding.  One such cyst was 12mm which, the Court accepted, would explain a palpable lump assessed to be 15mm by a clinician.  All relevant experts agreed that her cancer, identified in October, would have been obvious to a radiologist if it was visible in May, as it would have resembled the tumour found in October.  This was an entity that could not have been mistaken for a cyst. 

3.2              Professor Bundred used a mathematical formula to calculate, from the 34mm tumour in October, that he would expect to find a 15mm tumour in May, basing his calculations on results from a study in 1993 in which the “doubling time” for cancer in 289 women was measured and a range of growth rates identified.  It was submitted that had it been faster than the times noted in this data, it would have been “off the scale”.  This referred only to the Peer data, however, and not to other studies, including those involving younger patients, for instance.

3.3              The Plaintiff’s expert used only the figures set out in this paper published in 1993. When invited to consider more recent and more extensive data in direct examination early in the hearing, the relevant expert discounted papers based on those data, pointing out (by way of summary of his answers) that the patients in question were not sufficiently similar to this Plaintiff.   It was clear from his answers in direct evidence and in cross-examination that he was defending his position on the basis that the data underpinning the 1993 paper was reliable and directly applicable to this case.  It was also clear that he argued for this position as he believed that his calculations were correct.  This was, at least in part, because only the figure he had chosen explained the lump she had felt.  This position ignored the flaw in his reasoning which was that that figure became difficult, if not impossible, to justify, or even to explain if the lump in May was a cyst. 

3.4              On day 4, Professor Bundred was asked to comment on passages in Professor Crown’s report in which he, Crown, refers to the Plaintiff’s report as speculative and based on an assumption that her cancer was 15mm in May of 2017.  In other words, the confirmation bias which was discussed in the main judgment delivered in this case was specifically identified in the report.  Professor Crown then comments as follows:

“The ‘80‑day standard doubling time’ is based on a single manuscript which was submitted for publication 30 years ago.  The age of this paper is but one of many shortcomings that make it and Professor Bundred's conclusions from it irrelevant.  The paper by Peer et al was based on measurements of tumour size in serial mammograms.  Most of these, unlike Mrs. Crumlish’s case, were from asymptomatic patients participating in screening programmes.  While the median doubling time for women under 50 was 80 days, the 80% confidence intervals were 44 to 147, i.e, the statisticians could be 80% confident that the doubling time was no shorter than 44 days and no longer than 147 days.  The Peer paper was an honest attempt using the primitive technology of the day, an x‑ray machine, a ruler, and a birth cert) to explain the relatively poor performance of screening mammography in women under the age 50.  The authors’ reasonable hypothesis was that younger women might have faster growing tumours and hence a higher chance of developing an interval cancer in the two-year schedule then in use in the Netherlands national screening service.  All it found was that within very wide confidence intervals, on average, younger women had faster growing tumours than older women.  It was not an attempt to explain how quickly an individual tumour grew.  The extraordinarily wide confidence intervals in this study also demolish any attempt to use it as the basis for reverse extrapolation in an individual case".  The emphasis, in bold type, of the phrase that all the Peer paper found was that cancer grew faster in younger women, was in the original report.

3.5              On the following page, the report contains the following comment:  “In a recent paper Bhattarai et al state that ‘Breast cancer (BC) is a heterogenous disease with tumours exhibiting variable morphology, molecular profiles, behaviour and response to therapy.  Mounting evidence demonstrates that BC shows variable rates of growth, which has important clinical and medicolegal implications.’” Again, the cited phrase was in bold type.

3.6              Before cross-examination, the witness was asked in direct evidence to respond to the possible argument that the Plaintiff’s cancer was indetectable in May.  The relevant part of the question and answer reads:

…presumably by corollary he is saying that Mrs. Crumlish was likely to have been node positive.  So she had cancer, but not only cancer, she had node positive cancer back in May.  To the extent that anyone else on the Defendant's side tries to run a different line of argument that in actual fact Mrs. Crumlish didn't actually have cancer in May, what would you say to that suggestion?

A.  Well you can't have your cake and eat it, can you? You have to back one horse or the other … we have been able to show… that you can palpate the five millimetre cancer that the Claimant claims she had two lumps, that one was peppercorn.  And, as we have seen, a peppercorn is a size you could palpate in an A cup breast in a thin woman and, therefore, she did have cancer and that was the tenet accepted by the Court. 

3.7              Leaving aside that there had been no finding of fact by the Court at that point other than that it was clear that a peppercorn was palpable, inherent in this answer is the assumption, again, that at least one of the lumps palpated in May (or indeed at the end of March) was identified as cancer in October.  The expert does not grapple with the possibility that there was no discernible cancer in May and does not divorce the issue from the palpable lumps in May at any stage.  This pattern is repeated during cross-examination, during which the witness repeatedly links his estimate of the size of the cancer in May to the size of the lumps palpated by the Plaintiff.

3.8              In other words, Professor Bundred did not consider whether his figure might be wrong, but pointed out that a 5mm lump could be palpated, using a peppercorn produced in Court to illustrate this.  This appears to have been to demonstrate that even if the lump was smaller, it should have been palpable, whether in May or in March.  He responded only to the thesis that the Plaintiff had a slower growing cancer, saying that you can't have your cake and eat it.  In other words, he did not consider that it might have been smaller.  As the Crown report set out an argument that the cancer may have been slow-growing, Professor Bundred considered only the defence that it was bigger than 15mm, which must have been visible in May.  The responses in general ignored the section of the report set out above and most of the oral evidence.  The argument that the tumour might have been slower than the estimate was a feature of the report and it is understandable that it attracted attention from both expert and counsel but it was not the only feature of the report and was not even the focus of the question asked.

3.9              Professor Bundred went on to answer Counsel for the Plaintiff about Gompertzian curves to the effect that any slowing of the growth of this cancer would only make the cancer bigger and said that he found that Professor Crown’s note that Gompertzian growth could result in slower growth at large sizes to be extraordinary.   This in fact appears to be exactly what such a curve does, but to say that the tumour was slow-growing does not accord with any version of the facts here.  Professor Bundred repeated that this was as fast growing a cancer as one could find, without further justifying the selection of 45 days as a doubling time. He confirmed his belief that the 15mm lump in May was the same as the lump in October. 

3.10          In the passages from the report set out above, the defence expert did not spell out certain factual conclusions which were the basis for the Court’s decision, namely, that the growth of this cancer was probably faster than those studied by Peer.  However, he did make it clear why he challenged the data.  In evidence, Professor Crown set out his thesis more clearly: there are so many variables that one cannot estimate the doubling time of this tumour with any accuracy using only one figure, 34mm in October, and it was speculation to do so.  It also became clear that, to follow his logic and match it with the observed facts and the features of this cancer, it was likely that the Plaintiff’s tumour grew faster than those observed by Peer et al.

3.11          This Court has not discounted the science on which the Plaintiff’s evidence was based but has examined the methods used by her expert and does not have sufficient evidence that the figure used by him was the correct one to estimate the probable doubling time and hence the probable size of the tumour at any given time in this case.  The argument in respect of costs hinges on the proposition that the Plaintiff could not have anticipated this defence, or refuted it, until near the end of the hearing.

4.       Analysis of The Pleadings, The Report, The Evidence and The Submissions

4.1              Paragraph 3(b) of the defence includes the claim that: “there was an adequate explanation for the lump at the eight o’clock position of the right breast where cysts were identified on ultrasound to explain the S3 lump in the right breast.”  The Plaintiff was put on full proof of all personal injuries alleged, and, in the usual language of such pleadings, the Defendant specifically denied that any personal injury, loss or damage had been caused by negligence or breach of duty on the part of the Defendant.  Causation was clearly in issue.

4.2              The sections of the report quoted above make it clear that the method of calculating the size of the cancer would be challenged, the report highlights the issue of doubling time and specifically challenges the Peer paper data as being of minimal relevance to this Plaintiff.  Further, the reference to variable rates of growth having medicolegal implications is a good signpost as to where the author’s evidence would, and did, go.  Professor Crown went on in his written report to outline one potential result of his criticism of the doubling time chosen, which was the possibility that the cancer was slower growing than Professor Bundred had calculated and, therefore, was in fact larger than the lump palpated by the Plaintiff in May.  However, not only would this theory have been at odds with what was reported by the Plaintiff, it would also be at odds with all the evidence in the case.  While this was unfortunate, it was not the only material in his report nor was it the only possible conclusion to draw from his criticisms.

4.3              At paragraphs 9.1 to 9.17 of the main judgment, the Court set out summaries of the various papers in this field of study.  While I will not repeat them here, one of the significant papers in terms of this witness’s replies in evidence was that of Bhattarai et al, already referred to above in the quotation from Professor Crown’s report.  That paper reported on a study of a number of cancers and the authors constructed a method of predicting tumour size which was notably consistent across a range of different data.  Professor Bundred, however, dismissed this paper in his evidence on the basis that the women the subject of the initial figures examined by the authors had HER2 negative cancer.  This Plaintiff had HER2 positive cancer.  However, the methodology they were using was potentially relevant because it was replicating results in different trials.

4.4              The Tilanus-Linthorst paper, also considered in evidence and a paper with which this expert was familiar, contained the information that those authors, albeit in patients with a genetic pre-disposition to cancer, had recorded doubling times much faster than those set out in the Peer data.  The witness, in other words, did not appear to appreciate, and if he did, he did not engage with, the importance of later articles:  it was not just that the women they studied increased the available data, but the other studies revealed a much bigger range of potential doubling times.  Professor Bundred appeared to conclude that if the women had different characteristics or cancers, this allowed him to dismiss the other papers as irrelevant to this Plaintiff, but did not apply this logic to the Peer paper, on which he relied despite the valid criticisms in Professor Crown’s report.

4.5              A woman such as the Plaintiff might be expected to have a faster doubling time than anyone studied by Peer et al, depending on her particular type of cancer and on her body, her age and numerous other factors.  This was exactly what the defence expert went on to highlight and, having read the Bhattarai article, there is no mistaking one of the conclusions drawn:  doubling times depend on a host of factors, not all of which had been identified and recorded in the Peer data thus leading to the conclusion by the authors of Bhattarai which was set out in the substantive judgment.  The final sentence of their conclusion is repeated here:  Independent validation of these findings in multiple and more diverse cohorts is warranted.

4.6              While Bhattarai’s conclusions did not dismiss Peer’s findings, the authors did not endorse the data from Peer et al as being reliable indicators for other cases.  Their finding was that the concept of extrapolating doubling times for specific cancers from times observed in the recorded data is one that should be validated in more diverse cohorts.  The finding suggests that the concept is more complex than might appear and that it is not sufficient to take the Peer data and use that alone to predict cancer sizes in individual cases.  The Plaintiff’s expert was wrong to dismiss the Bhattarai paper as being irrelevant and Professor Crown’s second report, in which this paper was specifically relied upon, ought to have alerted him to this argument before he gave his evidence.

4.7              It is not reasonable to conclude that the only anticipated defence case was that the Plaintiff had a significant cancer in May, that it was node positive already at that point but that, notwithstanding this, it was not negligent to have failed to find it.  It was clear from the pleadings that the defence argument would include the premise that the lump found in May was a cyst, and that there was no cancer in May or, more accurately, that it was not discernible at that point.  Given that the lawyers had predicted that this line of defence might arise and given the contents of the articles referred to by both sides, it was reasonable to expect the person put forward as an expert on the calculation of growth rates to be able to deal with these arguments.  Professor Bundred repeatedly dismissed Professor Crown’s position as untenable partly on the basis that he, the witness, continued to rely on the assumption that the 15mm lump identified in May was cancer.  It was a matter for him to justify his selection of 45 days as a likely doubling time and, other than to link it with the lump in May, he could not say why this was more likely than any other fast doubling time. 

4.8              In part, this expert’s dismissal of the defence case was also based on an anticipated defence that the cancer was slow-growing.  But it was abundantly clear, after the cross-examination of Professor Crown, that the latter had adopted the much more persuasive argument that the cancer was likely to have grown faster than those observed in Peer et al.  No application was made to re-call Professor Bundred at that point.

4.9              The Defendant met the argument on costs with the reply that the Plaintiff’s case had only crystallised on the 3^rd of May 2022, just before the hearing began, when updated particulars of negligence were served which alleged a failure to discern a cancerous lesion which was at least 15mm on ultrasound.  This clearly identified the cancer in October as the lump felt in May.  However, it seems to this Court that the essence of the claim here was that they were one and the same entity and that this was not an issue that arose for the first time just before the case began.  By the same token, the essence of the defence was identifiable in the pleadings, namely, that the lump felt in May was a cyst.

4.10          The Defendant pleaded that there was no failure to detect a cancer and that the lump in May was explained by the cysts identified on ultrasound.  The dispute about growth rates was identified in a report shared with the Plaintiff on the 3rd of May and explored in questions on day 4 of the hearing.  That dispute was clouded by the introduction of Gompertzian growth as a concept leading to the theory in the report that the tumour might have been even bigger than 15mm in May.  This was unfortunate but an expert in the field should still be able to explain her position from first principles and not just dismiss one hypothetical without addressing the speculative nature of the doubling time chosen, the reliability of the concept in cancer cases generally and the confirmation bias which had been flagged clearly in the written report. 

4.11          The criticism was made of the defence report that Professor Crown had not explained why Professor Bundred’s figures were unreliable.  As the excerpt from the report itself shows, above, while Professor Crown did not offer a convincing counter-theory until he gave evidence, he did specify why the Peer conclusions were not reliable for this purpose and he identified the confirmation bias at work.  It was the Plaintiff’s case to prove and there was no reason for the choice of a 45-day doubling time other than to make the doubling time fit with the 15mm lump that had been observed.  Every other reason offered went no further than to confirm it was fast but if the calculations were to be reliable, there had to be a reason to choose 45 days as a doubling time.  There was no reason to confine the potential doubling times to the range identified in Peer, no explanation in evidence as to why it could not have been faster than those in that range, and no evidence as to the repercussions of calculations based on different doubling times, insofar as these might change the estimated size of the tumour at any earlier point. 

 

5. Conclusions

5.1              The Defendant’s pleadings and report identified its position, which was that the larger entity palpated in May was not cancer.  The Plaintiff relied on the proposition that the lump was a tumour and that it was the same entity as that found in October.  The true picture of the defence case, in respect of the logic of the 45-day doubling time, was clarified on day 21 of the evidence, but this was not an issue that should have been unforeseeable to an expert in this field.  No complaint was made when Professor Crown completed his evidence nor was there a request that Professor Bundred be recalled. 

5.2              The issue of doubling time as a concept was contested in the written report received on day 3 of the hearing and, while the successful argument was made with greater clarity in evidence, this is a normal feature of litigation in that a written statement or report might not spell out all aspects, or even the logical result, of the specific issue or argument.  If this were not the case, oral evidence would rarely be necessary.  If it was difficult, or impossible, to predict the thrust of an argument it might be fair to adjust an order for costs to reflect this and I am satisfied that this was not the case here; the thrust of the argument was not difficult to predict and had been raised with sufficient clarity to allow the expert to consider and deal with it in evidence. 

5.3              There was nothing in the conduct of the Defendant or in the nature and circumstances of the case, to use the words of the Legal Services Regulation Act 2015, which leads this Court to depart from the usual order as to costs.  It would not be fair to refuse the Defendant an award of costs on the basis that the Plaintiff’s expert did not foresee an argument or tackle it in evidence, given the signposts available in the pleadings and the report.  This is particularly so when he could have been recalled, had the expert or the Plaintiff’s legal team considered that the oral evidence, refining the argument, had been unfair or had misled the Court in some way.

5.4              The current system which provides for compensation in a case of negligently caused injury is in place because, as a society, we cannot afford to compensate all of those who are unfortunate enough to suffer serious illness.  That policy has arisen on the basis that where illness or injury is caused or exacerbated by a fault or error on the part of a medical professional, it is unfair to visit the consequences of the fault or mistake on the innocent patient. 

5.5              Where, as here, the damage is caused by cancer and not by medical negligence, it is unfair that the Defendant should have to pay the legal costs of defending the case where the Plaintiff has not shown that her injuries were caused or contributed to by any error on the part of her treating doctors.  In making the claim, it is always for the litigant and her legal team to assess the risks of bringing such a case.  In medical negligence proceedings, there is a requirement that an expert’s report identifying negligence and causation be obtained before the legal action begins.  Here, one of the first questions for such an expert must have been whether or not the Plaintiff had a detectable cancer in May, but this appears to have been assumed, to a large extent, in this case.  The expert went on to comment on allegations of negligent treatment but if there was no detectable cancer, the question of negligence does not arise and it would be unfair to penalise the Defendant by refusing to make the order sought.

5.6              There is no reason to depart from the usual rule in this case.  The successful Defendant is entitled to an order for costs.