Inquiry Evidence 040500 by A4A11h



 1                              THURSDAY 4 MAY 2000
 2                    THE HEARING RESUMED AT 8.30 A.M.

 4                 PROFESSOR McGOOGAN                   (On former oath)

 6   MR HINDLE: Dr McGoogan, the first subject you were asked to consider
 7   was the question of the statistical reports and in particular you said in answer
 8   to madam chair that you felt on a brief review of those documents that they
 9   contained information which was rather more managerially focused than
10   likely to assist in evaluating the programme. Do you recall that exchange
11   ..... yes.

12   Have you now had an opportunity to consider the reports ..... yes.

13   Dr McGoogan has not considered the Maori report in any detail, ma'am. It
14   is the first 3 statistical reports that you have considered ..... yes.

15   Having considered the statistical reports did you find your earlier concern or
16   impresn to be confirmed or not ..... yes, although I think there were attempts
17   to include quality data in the report and perhaps I unfairly didn't recognise
18   those attempts in my first reading of them.

19   Let's start with the third statistical report – Glackin volume 9, exhibit 51. .....
20   this statistical report follows the first and the second, the second having dealt
21   with information up till June 1994, because of unavailability of data from
22   the Auckland cervical screening programme.             It only covered a limited
23   period of time. So this was a report published in 1998, dealing with analysis
24   of data to 31 December 1995, with no other as far as I am aware similar
25   statistical reports delivered for the subsequent years, 95 to 98. it is very
26   interesting but it doesn't tell me what's happening to your programme now.
27   Furthermore, the foreword to this particular statistical report intimates that
28   the Ministry of Health expects the evaluation to commence in 1998,

 1   contracted to the university of Otago to scope an evaluation of the National
 2   Cervical Screening Programme and that the National Cervical Screening
 3   Programme is developing a process … while enrolled in the programme
 4   (reads) in the next reported, which hasn’t appeared as far as I understand. I
 5   think

 6   Do you have any general comment about the reliability of the information in
 7   the statistical

 8   ..... the statement report is the delivery of the data on the register, it tell us
 9   nothing about women on the register who are participating in cervical
10   screening. It only tells us the data held on the register and the report is
11   peppered with the words estimated, under-estimated, caveats, because
12   information is incomplete or unconfirmed or derived from suspect sources.

13   Could you refer us to a couple of examples ..... OK, in section 4, p9, method
14   of data analysis, second paragraph, second sentence “consequently …
15   understated”. first sentence in the next paragraph “the mean … statistics
16   New Zealand … of age specific rates”. And then to my surprise in the third /
17   fourth paragraph “regional … using 91 … by Area Health Board”. So we
18   are using different denominators from different periods of time to assess
19   different aspects of these statistics. Furthermore, in calculating, and I note
20   calculating, or the data relating to the number of previous smears for women
21   in the register was derived from the women’s estimate not from any factual
22   data, the women were asked have you had a previous smear, when was it and
23   how many.

24   What's the problem you see with that ..... well, women are notoriously unrble
25   witnesses as to whether cervical smear was taken during a vaginal
26   examination. They have difficulty knowing the difference between a swab
27   and a cervical smear or indeed whether any material was collected during the
28   vaginal examination. They have difficulty recollecting dates and the data for

 1   previous smears is broken down into 3 year bands.            So I would find
 2   difficulty to rely on that kind of almost anecdotal data.

 3   And with drawing your attention to some of the specific information for
 4   example the number of the incidence of ca reported, did you have any
 5   comment about that ..... yes if we go to p21 yesterday I expressed concern
 6   that repeating abnormal smears might influence the data the %s calculated
 7   for laboratories. We have evidence that this is actually happening, the
 8   second last paragraph on p21 by the end of December 995 a total of 233
 9   smears indicating squamous cell carcinomas had been reported in the
10   National Cervical Screening Programme these results for were for 214
11   women so women had more than 1 smear suggesting cervical cancer. I find
12   that extremely surprising. Why repeat a smear that shows cervical cancer.

13   Perhaps you expressed that sentiment in the form of a question. Why do you
14   find it surprising that there would be a smear after a diagnosis of that sort
15   ..... once the suggestion has been made in a cervical smear report that there is
16   possibility of cervical cancer being present, then the appropriate thing is for
17   the woman to have colposcopy and a gynaecologist opinion not a second
18   smear. The smear test has done its job by identifying a woman who may
19   have a problem. The smear test cannot diagnosis cervical cancer. And if
20   you look at the last paragraph this was not simply a problem in the early
21   years of the register because in the 95 calendar year we have 85 new
22   squamous cell carcinomas from 76 women. And furthermore if you look
23   over the page they did not have histological confirmation whether these
24   smears actually did result in a woman having a positive diagnosis of
25   invasive carcinoma so I have no way of knowing whether these were all
26   false positives or all correct diagnoses.

27   I wanted to ask you a general question about the statistical report, died you
28   have any more detailed comments after your reading of the report ..... the

 1   task the panel set me last night – my skills do not lie in analysis of data,
 2   these are extremely difficult documents to read.

 3   Do you mean statistcl reports genly or these in particular ..... these in
 4   particular are particularly difficult to read

 5   CHAIR: why is that ..... that’s a little like asking me why an elephant is an
 6   elephant, they are full of numbers, %s, totals, age specific rates, age
 7   standardised rates, you have to fully understand the methods of statistcl
 8   analysis to interpret the numbers being presented in table.

 9   Is it fair to say given the population denominatr difficulties identified this
10   morning does that mean the job of reading these reports easier/difficult .....
11   the interpretation becomes more difficult. I have enough knowledge to be
12   cautious about my interprtn because I understand the complications of
13   changing the denominatr in many of these calculations of % to a lay person
14   reading this report I wonder whether any meaningful information could be
15   derived. And after all statiscl reports on a screening programme should be
16   for everyone participating in the programme and comprehensible to women.

17   CHAIR:      when you consider the questions, think of it in the way that this
18   inquiry has responsibility to look into abnormalities in cervical smears in
19   Gisborne prior to march 1996, the inquiry is not an overall evaluation of the
20   cervical screening programme. In that sense its backwd looking, we are
21   looking what happen in 90 to 96 and its to do with whether or not there has
22   been misreading of smears. My understanding of your evidence is that as a
23   pathologist statistical data is helpful to you in determng whether or not your
24   laboratory is providing a quality service in terms of smear-reading is that
25   correct ..... yes that is correct.

26   You've seen the 3 statistical report s, firstly can you tell me as a pathologist
27   are those r- would those reports be helpful to you in deciding whether or not

 1   you were happy with the performance of smear-reading in your laboratory
 2   ..... it wouldn't help me at all.

 3   Secondly, the first report is dated 18 August 92, it was released in August 93
 4   ..... yes

 5   As a pathologist seeking feedback on the performance of your laboratory
 6   and smear-reading, what sort of statistcl data would you expect and how
 7   regularly would you expect to get it ..... of course I would be expecting to
 8   deliver the information that would be returned to me and I would be
 9   expecting all laboratories to do the same. That would allow me to compare
10   the performance of my laboratory with other laboratories

11   How regularly would you expect to receive that information ..... I'm not sure
12   that the actual defined period needs to be exact, annual is an appropriate
13   length of time. It has to be delivered back to you closely enough to the
14   period of tie for which the analysis is made to allow you to make adjustmts
15   if your performance you think may need to be adjusted or improved.

16   Taking that, looking at these reports, the first one rlsd August 93, the second
17   one its analysis of data to 30 June 94, rlsd in October 95, and the third one is
18   analysis of data to 31 December 95, it was rlsd in 98. how helpful to you
19   would receiving statistical information of the nature contained in these
20   reports at the time at which it was received be to you as a pathologist
21   running a laboratory wanting to gauge what the performance of your
22   laboratory was ..... I think my reaction to each of the these statistical reports
23   might be slightly different. The first one is understandably the best that they
24   can deliver at that particulr point in time. And I think most people would
25   have understood the limitations in dlvrg data at that time. It was delivered a
26   year after the period for which the analysis was made, that’s not unrsble.
27   The second report was delivered in October 95

 1   Just over 2 years later ..... yes, but dealing with data to June 1994. so there's
 2   a 15 month delay in delivering the information. Again not too bad, but
 3   drifintg out from what is being helpful if one thinks that a practice needs to
 4   be ijmpd or adjusted. Also if one thinks the statiscl information being
 5   collected should be different, it’s already too late to influence things – you
 6   can only start from where you are, so if this statistical report is delivered in
 7   October 95 you can only start collecting new or better statistics thereafter.
 8   That’s why it is extremely disappointing that the third report, which dealt
 9   with analysis of data up to the end of December 95 took until June 98 to be
10   delivered. Its further dispntg that the quality of the statiscl information
11   leaves a lot to be desired and the authors of the report have done their best to
12   identify the limitations of the quality of the information in the report. While
13   producing the statistics, its simply telling you what the data is on the register
14   but saying it is not perfect so how do you interpret it

15   As a pathologist wanting to measure the performance of your laboratory how
16   helpful are each of these reports ..... not very helpful, particularly the last one
17   is very unhelpful.

18   Can you say in principle ideally what type of statistical information and how
19   regular it wd be received from the perspectv of a pathologist wanting to
20   monitor the performance of your laboratory ..... the source of the statistics
21   needs to be addressed. It seems to me New Zealand is looking to the
22   cervical screening register to fulfil a multiplicity of tasks. Perhaps even
23   managing the programme. Whereas it simply is a database, designed to
24   identify the population participating in the screening programme, which
25   would give you covrge information. Now that you have as I understand it
26   histological information on that register too, am I correct?

 1   It was configrd by 96 ..... you can begin to look at the biopsy smear
 2   correlation, albeit in the broadest of terms, and perhaps even the invasive
 3   cancer rate in the population on the register.

 4   It is interesting to know what the situation is now and to have your
 5   comments on that, but we are looking at the period up until 96 when the
 6   correlation between cytology and histology was not possible, we are trying
 7   to find what information during that period would have been avible to a
 8   pathologist to enable him to address the performance against those of others
 9   ..... the register is simply a database of women being screened, designed to
10   facilitatee the call of the women at the appropriate tie to give for their next
11   smear. It is not a perfect tool for measuring quality of service delivery
12   within a laboratory or allowing comparisons between laboratories in the
13   detailed way a which would be necessary to be of assistance to laboratories.

14   MR HINDLE:        what do you regard as the main pieces of information that
15   you get in your laboratory and upon which you judge your performance .....
16   OK. The pieces of information that are routinely collctd are the reporting
17   profiles of each member of the technical staff, the primary screening staff
18   and the laboratory overall.      In other words the % of smears called
19   inadequate, mild, mod, severe dyskaryosis, within normal limits.           That
20   allows me within the laboratory to monitor individual’s performance against
21   the laboratory total. I also collect the results of rapid review of all negative
22   and inadequate smears for every primary screener which allows me to
23   monitor the performance on a day to day and week to week basis throughout
24   the year of individual primary screeners against the laboratory average. And
25   against their own pattern which is a useful tool . I collect on an annual basis
26   the positive predictive value for all high grade smear reports issued by
27   individual pathologists within my laboratory. So that I can calculate the
28   positive predictive value for each individual pathologist, sensitivity for

 1   screening for each primary screener and I am able to deliver the reporting
 2   profile for the laboratory. The data is collected in a very specific way, which
 3   allows reasonably accurate comparison between laboratories.                The
 4   mathmaticl equations used for calcultg specificity and sensitive are well
 5   documented and followed so we are doing the same things, which means the
 6   published statistics are more meaningful. As with all published statistics
 7   occur, transcription, printing, data errors, so the point of the exercise is to
 8   compare your results against other people’s and investigate when something
 9   looks strange. The statistics do not prove whether you are doing well or
10   badly, they simply tell you whether you are doing something similar or
11   something different to someone else.

12   PROFESSOR DUGGAN:               could you have a statistical report from a
13   laboratory that doesn't look strange and that laboratory may not be a good
14   laboratory ..... absolutely, very simply. All you need is for a laboratory to be
15   issuing the correct number of high grade and low grade results but for the
16   wrong women – in other words you need to have the positive predictive
17   value of high grade results as ell, its not enough to have the rights numbers
18   in each of the boxes in the table, you have to know the reports are actually
19   accurate. You could have a situation where the laboratory has a high false
20   negative rate and a high false positive rate but the numbers turn out correct
21   and the statistical return looks within the normal range but the performance
22   is obviously very poor.


24   Mrs BARRET

25   I know you didn't look at the Maori statistical data, what are your thoughts
26   about that being separate from the generic data provided, do you think it
27   should be separate ..... with most data bases the bigger the database the more
28   accurate the conclusions drawn from it, the smaller the database the more

 1   difficult it is to derive meaningful statistics, it may be of interest to the
 2   Maori people to see what's happening to their people but it may be difficult
 3   to draw meaningful meanings from that compared to the larger database. So
 4   I see no reasons why they shouldn't have their own information but it
 5   shouldn't be excluded from the large database. Any group of women in the
 6   population should be able to have their own statisticl information, the
 7   women of Gisborne, Auckland, wherever, but deriving meaningful
 8   evaluation and comparisons from small numbers is very difficult and can be
 9   dangerous, it may lead you to the wrong conclusions.

10   I think that you would also not agree because that data was at least 4 years
11   old ..... it may have been useful 4 years ago but it doesn't tell you whether
12   things are different, better, worse now, and we need to know what's
13   happening now.

14   MR HINDLE: the second of your tasks for last night was from the statistical
15   reports was to consider some line diagrams. Dr Boyd’s volume 3, tab 13.
16   page 960 of the transcript you were asked by Professor duggan some
17   questions about this document. Do you remember those questions ..... yes.

18   Have you overnight had the opportunity to consider the document in its
19   toality as well as the particular flow diagram ..... yes, I have.

20   Is it your understanding that that is in fact the proposed structure for the
21   proposed screening programme in New Zealand as at the date of the
22   document ..... having read the full document which is a draft report of the
23   National Cervical Screening workshop 1988, on p26 section 4.8 evaluation
24   and quality control, workshop 1, it explains the flow diagram appendix 6
25   shows the points at which quality control and evaluation need to happen
26   (registers are important for evaluation). So this diagram was intended as a
27   flow diagram for quality control and evaluation rather than specifically for
28   management of a programme.

 1   CHAIR: as a flow diagram showing quality control, do you think it would
 2   have been a good idea for it to have been implemented or not, from your
 3   experience of cervical screening programmes ..... I don't think the diagram
 4   goes far enough but with that caveat it would be a good start to
 5   implementing quality control at each of the points of the cross.

 6   Would you see it as setting out a min standard for quality control within the
 7   programme ..... it provides a starting point for quality control, I'm reluctant
 8   to use the word min standard.

 9   MR HINDLE: when you were questioned about this document you gave a
10   number of answers in which you expressed surprise that there weren’t more
11   crosses and more points at which quality control should be considered.
12   Notwithstanding this is not a mgt document, do your comments about
13   quality control remain in force ..... yes they do, those comments remain
14   unchanged.

15   CHAIR:       in terms of your experience of cervical screening programmes
16   what comment would you have to make if you were aware that programme
17   when put in place did not have the quality controls envisaged in this flow
18   diagram ..... I would be extremely disapntd because by the time the New
19   Zealand programme was impltmtd the need for quality control and
20   evaluation for a screening programme of any kind was well recognised.

21   And if they weren’t in place, and this is just the quality control set out in the
22   diagram, when the programme was first ipmltd, how much time do you think
23   should go by before they are implemented ..... well I suspect that some
24   quality control was happening at some of these points

25   I don't want you to suspect, I want you to say, working from the premise that
26   the when the programme first started, the quality control measures were not
27   in place, how much time do you think it is reasonable to have a lapse before
28   the quality control measures in this diagram should be put in place ..... I

 1   would have expected the quality control points marked on this diagram to b I
 2   place within 3 to 5 years of your programme starting, for good accurate data
 3   on these particular points. There's no point doing quality control and not
 4   documenting it.

 5   PROFESSOR DUGGAN: this flow diagram was generated in 1988 and of
 6   the crosses that are present on the diagram if quality control was not in place
 7   – let me rephrase the question – of the crosses present, which one would be
 8   the most vulnerable if it wasn’t in effect, in terms of the success of the
 9   programme ..... there are some come fairly equally with me on this
10   programme. I would look on the smear-reading and the reporting of smear
11   results back to women and provider as equally important aspects which had
12   to be quality controlled from the very early stage of your programme

13   CHAIR: you take the programme starting in 90, what time limit would you
14   put. ..... by the end of your first found a 3 yearly screening programme 1993.

15   MR HINDLE: you were asked also to look at a different diagram which we
16   didn't look at yesterday but which you considered overnight, it is in Ms
17   Glackin’s exhibits, volume 8, tab 40 appendix 1.          before we ask any
18   questions about this document, do you have any general comment to make
19   about your expertise in management of programmes such as this ..... a
20   comment about my lack of management, my expertise lies in training quality
21   control aspects of cervical screening programmes. I'm not an expert in
22   setting up and the formal reporting relationships between different parties
23   within a service structure, particularly not outside the UK

24   Is it fair to say you know how it works in UK and Scotland but you don't
25   offer yourself as an expert in these issues ..... yes

26   You were asked questions about this document, relating to a period of time
27   when

 1   CHAIR:          if this witness is not an expert I fail to see how she can give an
 2   opinion.

 3   MR HINDLE:             it was your question that she look at the document and
 4   comment on it.

 5   CHAIR:           Dr McGoogan as a pathologist working in a country with a
 6   screening programme you have to work within the structures of that
 7   programme and deal with the various difficulties and also the assistance you
 8   might get from the mgt structure. In that role as a pathologist running a
 9   laboratory screening smears for a programme?              you would have been
10   affected by the management of cervical screening programmes would you
11   not ..... yes

12   From the basis of that experience are you able to offer an opinion on a
13   management structure from the perspectv of someone working within the
14   programme whether the management structure is cumbersome or helpful .....
15   I can comment as a pathologist but not an NHS management expert.

16   We heard from Professor Skegg yesterday at one stage in terms of dealing
17   with health management used with word cafcaresque. Obviously someone
18   working in the ? easier or difficult ..... yes.

19   I would like you to comment if you can ..... the diagnose in question was
20   taken from the review of anctdbtys for the National Cervical Screening
21   Programme May 1996 and it was a document produced within the Ministry
22   of Health with no consultation outside the Ministry and in particular no
23   consultn with screening programme staff. In the text of the document it
24   states that this diagram illustrates the complexity of and inherent difficulties
25   with current accountability arrangements and I would have to agree.

 1   If you were a pathologist working in this system how would you find it ..... I
 2   would find it extremely difficult knowing to whom I was actually
 3   responsible for delivering my part of the programme

 4   And if you had concerns about the programme in terms of receiving data
 5   which could be helpful to you would it be clear to you to whom you should
 6   go ..... not from this diagram. The dotted lines and the continuous lines
 7   leave some provider organisations floating. Cancer Registry, the nsr and
 8   women are missing from this diagram. (national screening register) the n n
 9   NHS cervical screening unit is accountable to 3 bodies at one direct with no
10   direct rspby to the Ministry of Health or the ? it seems to be floating rather
11   free.


13   From your knowledge of the system in the UK is that organised in a similar
14   fashion to what you see here ..... I have a diagram of the organisation in the
15   UK if you would find that helpful.

16   Yes.

17   MR HINDLE: is that a line diagram that you've obtained of the – not the
18   management structure but the quality assurance rltnships within the UK
19   National Cervical Screening Programme as at Jan 96 ..... yes.        EM/CA/
20   0004

21   Perhaps I could ask you to take us through the document ..... I suppose the
22   one thing missing from this document is the Department of Health itself
23   which sits above the NHS executive. The Advisory Committee on cervical
24   screening are ministers personally seleted senior expert advisory group to
25   advise the Minister on the cervical screening programme

26   Yesterday when we were discoing the national committee is that the sae
27   committee ..... yes

 1   This was the committee in respect of which you thought it would be
 2   unthinkable for the Ministerial. Not to follow the advice of that committee
 3   ..... I think it is still unthinkable that he would not follow the decision of his
 4   experts. Perhaps I could say I think the confusion arises when listening to
 5   other evidence around the use of the term Advisory Committee. It seems to
 6   me that in New Zealand the term Advisory Committee is more like what we
 7   in the UK would have called a working group or an expert group to deal
 8   with a specific task at a specific point in time rather than a formal body
 9   permanently in existence – therein lies the difference.

10   In the UK with an expert group to what degree is there an expectn that their
11   advice would be followed ..... I f they were working as a working group
12   they would come up with recommendations which may/may not be followed
13   because they would be given the task of looking at a specific problem in
14   isolation and it would up to others to take those recommendations in the
15   context of the whole health service its funding and what is practical in this
16   particular point of time.    The difference with the NHS is their remit is to
17   actually do that, to take the whole programme in the context of the whole
18   health service, its funding, orgnstn, what is practcl what is not practice.

19   Do the advisory groups do cross analysis..... sometimes they are asked to do
20   just that.

21   MR HINDLE: am I right to think that you have an issue about moving to
22   liquid based technology. Would you expect first to have a working party
23   type report find its way to the Advisory Committee and then be dealt with by
24   the advisory’ before going to the Minister ..... yes

25   So the Advisory Committee rely is the executive body for the screening
26   programme – that is the centrally accountable responsible body for the
27   screening programme ..... it is an advisory body, its not an executive body.

 1   But its an advisory committee whose advice gets taken usually ..... usually,
 2   yes. no Minister is oblgd to take the advice of his national advisory group
 3   but it would be remarkable circumstances where that advice is rejected. It
 4   may be that new development was suggested to be implemented in 2 years
 5   and the Minister might say it will take 3 years but it would be unthinkable
 6   for the Minister to say we would not do it at all.

 7   CHAIR: …. Rejection of the advice but nothing is actually done ..... I can't
 8   think of a circumstance that meets that particular set of conditions.

 9   And in your experience if it is said something will be done by a certain time
10   and it isn’t is the issue revisited to see why it hasn’t been done or is it simply
11   left ..... it is definitely revisited and an investigation is carried out as to why
12   it has not been achieved. It may be unrealistic to attempt to achieve it –
13   reasonable causes why not achieved, certainly investigated and new targets
14   set.

15   Has this happened in all cases ..... I hesitate to say there are no exceptions to
16   the rule but I cannot think of one.

17   You haven't encountered one in the health service ..... not in recent years in
18   running cervical screening programmes in the UK. Our programme started
19   in 1988.

20   When you say recent years ..... in the last 10 years

21   You seem to have an exemplary civil service ..... no, quite a number of
22   public incidence requird us to put in place accountabilities and clear lines of
23   responsibility and we have explicit standards against which to measure the
24   delivery of all components of our programme, these are in the public domain
25   and have to be accounted for.

26   MR HINDLE: I would like a word picture from you as to how th3ese lines
27   relates these bodies in the NHS ..... the national co-ordinating team and the

 1   national co-ordinator their role is to develop national policies which I think
 2   are then imposed on the Regional Health Authorities.          And to monitor
 3   performance and quality assurance. The responsibility for the delivery of a
 4   service of qrop quality as defined by the national policy is that of the
 5   Regional Health Authority and in particular it is the personal responsibility
 6   of the region directors of p/h, they are personally accountable       for the
 7   programme in their region. There are the regional quality assurance teams
 8   which ensure that the data collection, the monitoring, evaluation the training
 9   of the programme the local programme within their region meets the
10   required quality standards.

11   CHAIR: I see in your programme that there is a linkage to the ca Registry
12   ..... yes.

13   How important is that ..... that’s very important. You will note we don't
14   have on this diagram a national register because the register as such is
15   simply issued as a call and recall and failsafe database – we do not derive
16   statistics from our register, they are collected by other means.           Ca
17   information and ensuing the quality of the information on the Cancer
18   Registry is very important. Each year I'm given a printout from the Cancer
19   Registry in Scotland of the women with invasive cancer that have been
20   registered during the last year to verify the data. Verify the diagnosis of
21   invasive cervical cancer is correct from my laboratory information. Because
22   coding errors can occur and every year I adjust one or two codes which have
23   been inappropriately applied to that particular individual.

24   MR HINDLE:        the third of line diagrams related to the structure of the
25   period during the era of the Area Health Board – Glackin volume 2, tab 5,
26   p157, you will appreciate here that I showed you first of all our Regional
27   Health Authority structure which came later in time from this structure, the
28   earlier Area Health Board structure that we had until mid 92. so we are

 1   going back in time. Again have you had a chance to consider this document
 2   in its totality overnight ..... yes I have. Again, I don't really see this as a
 3   document showing the lines of accountability and mangrl responsibility, it is
 4   more an information flow and interaction diagram.

 5   CHAIR:     as a pathologist working within the screening programme I ask
 6   you to once again provide an opinion on that basis. Looking at this diagram
 7   at the time the screening programme was introduced there were 14 health
 8   benefits and 14 registers run the health board., as a pathologist working
 9   within that type of structure what impact would the structure have on your
10   work ..... I think I would feel extremely isolated from the programme. This
11   diagram shows the laboratories at the far right with smear results leaving the
12   laboratory and smears entering the laboratory and smear histories entering
13   the laboratory. And that’s sort of it.

14   What else would you expect to see in terms of comnctns with the
15   laboratories ..... I would have expected some line of accountability to the
16   Area Health Board programme managers, some direct line moving between
17   the programme Mrs and the Area Health Board managers. This is more set
18   up as an information flow and interaction .

19   And in terms of information flow do you consider it adqte ..... it is not
20   unreasonable, it depends what is meant by report / data, particularly around
21   the national co-ordinator box. I'm not entirely sure that I'm comftble or
22   understand how the national co-ordinator functions in that particular flow
23   diagram in that they simply report in two directions and don't seem to be
24   accountable to anybody directly. They can't get to the Minister of Health
25   except through the expert group. I was bemused by what WCA stands for
26   and I couldn't find it in the test of the diagram. Again the national co-
27   ordinator seemed to deal with information flowing rather than anything else.

 1   Given that it is described as ? would you expect the Cancer Register to be
 2   part of it ..... yes. all of the diagrams I have looked at, the one group who
 3   appears sometimes but usually, as they do in this particular diagram, at the
 4   bottom of the diagram are the women.           I believe a cervical screening
 5   programme is a partnership between variety of health professionals the
 6   health service funding units and the women, without women participating in
 7   the programme you don't have a programme. In one of the diagrams do I see
 8   a route whereby women can influence the programme or have an influence
 9   on the programme.       I know the south Pacific and Maori women have
10   someone looking after the information but I don't know women feed into the
11   management and the quality of the cervical screening programme designed
12   to serve their needs.

13   This last section relates to the first of the terms of reference and I would like
14   to draw attention to the wording of the first of the terms of reference. Dr
15   McGoogan the first of the committee’s terms of reference is to determine
16   whether there has been an unacceptable level of underreporting (etc). you
17   are aware of that, I'm sure. At the end of yesterday’s hearing you were
18   asked a number of questions by madam chair about how you might set about
19   aud8iting the performance of a particular pathologist working on his/her
20   own and there was this exchange, p989 of the transcript – line 22. “can't be
21   identified”. Is there any one in which you can identify (reads) high. I
22   understand THE PROBLEM behind the question. Can you identify the
23   problem behind the question in dealing with the first terms of reference .....
24   the first terms of reference asks for a subjective assessment of the quality
25   and one measures quality according to how far something conforms to a
26   preset standard. If you don't have a preset standard then my idea of quality
27   and your idea of quality and chair’s idea of quality might be very different,
28   equally valid,. But not comparable. I am unaware that a standard had been
29   set for sensitivity for abnormality detection in laboratories. The second

 1   problem I think is its all abnormalities, the word abnormality is not further
 2   defined in high grade abnormality, low grade abnormality, detection of
 3   infectns, inflammation, I was rather concerned reading the statistcl reports
 4   that infectns and inflammation had been put into a category of abnormality

 5   CHAIR:      the meaning to be given will be ultimately for the committee.
 6   Firstly do you in your laboratory have a means for measuring sensitivity .....
 7   yes

 8   What is that means ..... we rapid review all smears called negative or
 9   inadequate by the first screener before the results is issued by the laboratory.
10   That allows us to detect a proportion of smears wrongly called negative
11   when indeed abnormalities were present. It does not allow us to detect all
12   abnormal smears, but it is our methodology that can be applied routinely in
13   every laboratory and our mathematicl equatn set to come up with a measure
14   of sensitivity and since that is calculated in a similar way with each
15   laboratory comparisons are possible. The standard range for sensitivity of
16   primary screening set for the programme at present is 85 to 95%.

17   Your definition of a false negative is 1 minus the sensitivity rate ..... yes.

18   And how do you determine if you are getting too many false negatives .....
19   the sensitivity as calculated by rapid review is applied to every individual as
20   well as the laboratory overall and one can compare individuals within the
21   laboratory with each other as well as the overall result of one laboratory with
22   another. If the calculation of primary sensitivity falls outside the range 85
23   95 % we are obliged to provide an explanation which is acceptable to the
24   regional quality assurance team as to why we do not meet that particular
25   range. But more than that, if an individual’s sensitivity falls outside the
26   range one would speak to the individual to see were they having a problem,
27   were they unwell, becoming confused about when a slide was normal;’/not

 1   normal, did they require additional training, and in that way we can keep the
 2   performance of individuals within fairly tight variations throughout the year.

 3   PROFESSOR DUGGAN: with these figures of 85 to 95% sensitivity you
 4   would calculate your false negative as 5 to 15% ..... for primary screening as
 5   calculated by this method.

 6   And with this methodology what is your definition of abnormal in terms of
 7   identifying the false negative ..... this was for the identification actually the
 8   question is very valid in that the original document the ABC document
 9   which you have there – exhibit EM/CA/ 0005. if you go to appendix 5, p42,
10   you see that we are calculating the sensitivity of primary screening for high
11   grade lesions or worse.

12   By worse you mean ..... more severe than high grade invasive carcinoma.
13   CIN II, CIN III of invasive carcinoma.

14   Are you restricting that to squamous carcinomas ..... no.

15   Its all carcinomas ..... yes. the number of adenocarcinomas coming through
16   a laboratory are very small.        Laboratories also calculate the primary
17   screening sensitivity for borderline or worse but there is obviously a
18   different range, the standard is applied to the high grade lesions only. There
19   is no standard for the borderline range – we calculate it but it applies to the
20   high grade lesions.

21   Could you clarify for the committee once more what you just said in terms
22   of the standard of 85 to 95%. Is that correct that a false negative for this
23   standard is defined as a high grade lesion or a mod dysplasia or severe
24   dysplasia that is missed by the primary screener and picked up by the rapid
25   review ..... that is correct.

26   Do you also calculate using the same methodology a screening miss that is
27   of lesser degree of abnormality than a high grade ..... yes we do.

 1   And is there a standard for that ..... no there isn’t

 2   There is no national standard ..... there is no national standard for that partly
 3   because the inter and intra observer variation for borderline category is very
 4   high so the standard range would have to be very wide.              It is useful
 5   measurement within a lab comparing individuals within that laboratory
 6   amongst a small number of pathologists comparing a borderline but it is a
 7   less useful measurement between laboratories.             The consistency of
 8   pathologists reporting high grade lesions is high.

 9   MR HINDLE:         this discussions concern standards against which measure
10   false negative rates in the year 2000 in the UK. I will take you back to
11   Gisborne in 96 and the problem we were talking about of trying to assess
12   acceptbty of performance when there is no acceptable standard to put it
13   against.   Accepting that there is no objective standard does there come a
14   point at which the performances is so bad that you could nonetheless as a
15   matter of commonsense and judgement nevertheless describe it as
16   unacceptable ..... yes I think there is.

17   This really is the subject that madam chair asked you to consider overnight.
18   How would you go about trying to make that judgement ..... I think that one
19   must start with the understanding that whatever figure we end up attachg to
20   a false negative rate is going to be a rough estimate and not an exact figure.
21   And from the rough estimate of the false negative rate it may be obvious
22   because this estimated has a false negative rate so large that it would be
23   difficult to see how this would meet anyone’s definition of a quality service
24   or a min standard of service. so I think I don't see how given the data
25   available you can calculate an exact false negative rate and that’s why
26   yesterday I was suggesting that perhaps approaching some measurement of
27   sensitivity for this particular laboratory from a variety of different angles by
28   putting all the information may give one more stable ground on which to

 1   make a judgement as to what its sensitivity rate was and to whether it was
 2   unacceptable

 3   Can we go through a list of possible such investigations.

 4   CHAIR:      you've said that the rate of false negatives in the UK at the
 5   moment can be anywhere between 5 and 15%. At what point ..... as judged
 6   by rapid review, it is important.

 7   Can you extrapolate from that that a % rate much greater than that but to
 8   judged by rapid review could be looked at a means of ascertaining whether
 9   the false negative rate was too high ..... I'm not sure how reasonable it is to
10   apply standards of one health care system to another health care system. I
11   have reservtns about that. the estimation would have to be made the same as
12   the rapid review. It could be done retrospectively. One could attempt to
13   mimic the same circumstances had rapid review been carried out in the
14   Gisborne laboratory at that time to derive sensitivity in the year 200 we
15   could carry out that same task according to the same rules – rough estimate
16   but it might come towards a measure. If the answer came it was 16% you
17   may have a problem, 60% you may have a less false negative rate of 6y%, it
18   comes a matter of a quality judgement at the end of the day whether the
19   calcln is acceptable.

20   Saying 670% is acceptable 16% not so, 40% what does that look like to you
21   ..... I'm reluctant to go along this avenue. I don't think you can in advance
22   put a figure and perhaps if I explore some of the avenues of attempting to
23   arrive at a figure you might see why I am reluctant to do that at the moment.

24   PROFESSOR DUGGAN:             if laboratory in UK has false negative rate of
25   16% what happens to that laboratory ..... there is an investigation to see
26   whether its performance is actually below standard.          There is not an
27   assumption that they are below standard. The standard range is there as a
28   gate, if you fall outside the gate you have to justify why you are outside the

 1   gate. If you fall within the standard range you don't have to justify but its
 2   still no guarantee of quality.

 3   CHAIR:       in terms of the gatekeepers in the UK how readily is the
 4   information available to them for them to see you are at 16% ..... we are
 5   required to deliver it annually

 6   That means if 1 year you were 16 or 17% what would be the time response
 7   before you were asked to account for yourselves ..... the time response, the
 8   first time it happened, might be several times longer than the second time it
 9   happened.

10   The first time it happened how long would it take before the gatekeepers
11   were knocking on your door saying account for yourself ..... they would be
12   asking on receipt of the statistics. It would be unusual for delay several
13   months before someone asked question.

14   When you sent the information into them that is when you would expected a
15   response ..... yes and one would have prepared ones response to their
16   question because we would be aware but questions would be asked.

17   Why is 40% a problem ..... I think the figure you place on different
18   treatments of false negative rate depends on how you measure it. The
19   literature clearly shows that even when one tries to mimic the same
20   circumstances when doing a quality control review exercise by another
21   laboratory the standard of scrutiny of the slides and level of suspicion
22   changes. And I would rather see the results of some attempts at finding a
23   measurement for the false negative rate and then evaluating whether this
24   really was unacceptable or not. It may be that too ways of looking at the
25   false negative rate gave you a false negative rate of 40% but the third
26   approach gave false negative rate of 10%. You have to take the whole thing
27   together, if all 3 gave a false negative rate of 50% you have much stronger
28   evidence that the level being unacceptable.

 1   If you tried 3 different ways and the average results was a false negative rate
 2   40% what would your comment be on that ..... for the period of time 90 to 95

 3   96 ..... I would have anxieties that the quality was of an acceptable standard.

 4   And again working on the basis that there's 3 ways of being identifying to
 5   retrospectively looking back, you take an average of that at 30%, if that was
 6   a false negative rate what would you say ..... I would say the average was
 7   30% but you are pushing me into an area of giving an absolute figure, I
 8   cannot do that. it would be wrong of me to put a figure for that, I would be
 9   able to give a more considered judgement having seen the results and being
10   allowed to scrutiny the results of these different exercise but I would be
11   drawing a figure out of thin air and that would be wrong.

12   There may come a time where the logical consequence of what you are
13   saying is that there are no means of whether or not a false negative rate is
14   acceptable or unaceptble ..... I'm saying it may be difficult to evaluate
15   whether the level of under-reporting was unaceptble or not.           It is not
16   impossible, its worth having a try.

17   MR HINDLE:         Dr McGoogan did you have these sorts of problems in the
18   Inverclyde station ..... yes we did.

19   In the end how did you deal with that ..... in two ways. One we looked at the
20   professional standards and professional       practices within the laboratory
21   under investigation. and we calculated a false negative rate by a
22   methodology published and the false negative rate for high grade lesions was
23   50% and the professional practices in the laboratory under investigation
24   raised many questions about treating professional standards as opposed to
25   quality standards and the two together allowed us to say the quality was
26   unaceptbly low

 1   Am I right to summarise by saying you have a difficulty making a
 2   quantitative judgement if you haven't got objective quantitative information
 3   but you can look at all the available information and in making a true
 4   judgement, a common sense judgement you can also take into account your
 5   analyses of laboratory practice and come up with a judgement ..... yes, may I
 6   read the first sentence from the summary of conclusions of the Inverclyde
 7   report – third exhibit, p9. (reads) I think that sort of approach might help you
 8   in your attempt to fulfil your first remit, looking at whether the sensitivity
 9   was lower than that to be expected from a laboratory of good practice.

10   Can I ask you to think about some specific ways you might tackle this
11   question. Let's start with the proposed ca audit that Professor Skegg talked
12   about. You were here ..... yes

13   Do you have any comment about the study he proposes ..... I have not
14   studied his protocol in detail but listening to his evidence I would
15   wholehrtdly support this particular study.       The audit of the screening
16   histories of women who develop invasive cancer is the most powerful tool
17   we have in auditing the programme.            It may indicate areas where
18   performance of different aspects of the programme is substandard but it also
19   allows us to see the impact of that particular practice in the health of the
20   women in the area and that may help you judge whether a level is
21   unacceptable or not.

22   In all of what follows, we are talking about a level of principle. With
23   relation to Professor Skegg study that relates to the women who've suffered
24   invasive carcinoma in the last 10 years. To draw comparisons is there any
25   other information that you would find helpful .....        as I understand it
26   Professor Skegg only intends to look at the women in this area who
27   developed invasive cancer. It would be a useful comparison to look at other
28   women in New Zealand who developed invasive cancer to see if the same

 1   set of circumstances prevailed in the development of their disease to help
 2   judge whether circumstances here are different than elsewhere.

 3   I am not asking you any questions about whether or not the data is available
 4   in New Zealand to do these things, but whether in principle these
 5   investigations would be helpful. Moving onto another way of tackling the
 6   problem, have you considered at attempt to calculate the positive predictive
 7   value for the laboratory ..... yes I think it is quite clear that the laboratory
 8   under investigation was identifying certain smears as showing high grade
 9   lesions. It would be important to know whether these were being correctly
10   identified       and the women being referred for treatment were correctly
11   referred for treatment. You need to compare the smear reports with the
12   histology derived at the bg of the treatment process.. that would allow some
13   measure of the accuracy with which abnormal reports were being issued
14   from the laboratory.

15   CHAIR: because you would be looking at the number of false positives if
16   any ..... yes.

17   And is it possible as a pathologist wanting to assess your performance, are
18   you reading smears correctly or not in the laboratory, to look at a false
19   positive rate and say if these smears are being read wrongly I can extrapolate
20   there will be a false negative rate which may be higher than acceptable ..... it
21   doesn't naturally flow because you have a high positive rate that you have a
22   high negative rate. You would have to view statisticl returns in a different
23   light if you knew some of the numbers given for high grade squamous
24   intraepithelial lesion were not truly high grade squamous intraepithelial
25   lesion.

26   In a retrospective study it would be a factor to take into account .....
27   absolutely. It would be stronger information if the positive predictive value
28   if the laboratory under investigation could be compared with other

 1   pathologists randomly selected positive predictive value at the same period
 2   in time.

 3   MR HINDLE:         the positive predictive value tests depends on the accuracy
 4   of other data as well ..... yes. The standard range for positive predictive
 5   value for pathologists in the UK is set at 65 to 85%. This at first at face
 6   value looks a very low level of performance but one must remember that the
 7   positive predictive value for a high grade lesion the measurement of this is
 8   influenced by the skill of the colposcopist in identifying the most abnormal
 9   area in the service to biopsy and the skill of the histologist in reading the
10   abnormal present down the microscope and the level has to be set relatively
11   low.

12   PROFESSOR DUGGAN:               what amongst in your country which has well
13   established standards what is the allowed time between the PAP smear
14   report that triggered the referral for colposcopy … for the purposes of these
15   correlations and calctns ..... again since my recollection is rather hazy it
16   would be easier if I could refer to a document, p9 – this is NHS quality
17   assurance for the cervical screening programme and on p9 the acceptable
18   values for the quality assurance of all aspects of the programme are laid out
19   and you will see number 8 to “ensure … within 8 weeks.”

20   So if the time interval between the PAP smear and biopsy is more than 8
21   weeks do you correlate those two ..... yes

22   What is the cut off in time ..... there is no cut off in time

23   The biopsy was done 5t years later ..... it would not be correlated within that
24   statitcs

25   The cut-off is the calendar year ..... yes but you will note the second part of
26   that is the waiting time for colposcopic examination for high grade
27   abnormality is that more than 90% are seen within 4 weeks

 1   CHAIR:      if the colposcopy occurred 11 months later, within your annual
 2   year, for correlation purposes is there any discounting to take account of the
 3   fact there may have been a natural progression of the disease in the
 4   meantime ..... this is factored into the standard range 65 to 85%.

 5   MR HINDLE:         I want to move to a third possibility altogether. EM/CA/
 6   006 – last document. Would it be possible in your view to do some kind of
 7   sample of slides ..... yes.

 8   In comparison with another group of slides ..... yes, this is the way that we
 9   would begin to approach in the UK an investigtn to see whether or not we
10   had a real problem when questions were raised about the particular
11   laboratory or an individual within a laboratory.

12   How would it work in general principle ..... I would suggest that a number of
13   consec smears from perhaps 2 period of time say 500 slides from 2 years
14   apart during the period in question were collected from the laboratory in
15   question say from 1 April in one year and 1 April next year and a similar
16   number of slides collected from 2 or 3 other laboratories from exactly the
17   same period one would expect those slides to contain negative smears true
18   negvs, abnormal smears, perhaps some false negatives and some false
19   positives. If these were collected together by an independent third party
20   renumbered so that the laboratory of origin was not recognisable and
21   allocated randomised numbers so that the slides from each laboratories were
22   randomly dispersed in the batch and these slides were screened again won
23   could calculate whether the performance of the re-screening laboratory or
24   laboratories pref3erably for one of the original laboratories work was
25   different from the other bundle I would caution that the primary screening of
26   this batch of slides the individuals chosen should be chosen randomly by
27   lottery rather than select the best person in the laboratory. Otherwise the
28   slides should be screened according to the protocol in place in the

 1   restructuring laboratory in the period in question.
 2   consecutive slides from 1 April, can you explain why you see it necessary to
 3   use a consecutive batch ..... it is very important not to use any selection of
 4   the slides from the laboratory in question. In paretic you do not want to
 5   select slides which you knew were false negtvs or true positives, its better to
 6   have a continuous number of slides, it removes a lot of the bias.






12   MR HINDLE: I want to move to a fourth alternative. Have you considered
13   whe4ther it might be possible to see what could be done with such data as
14   does already exist ..... the data that exists on the register.   I think that is a
15   source of exisintg data that perhaps could be targeted. I have been critical of
16   the quality of that data but I wonder if someone with an indepth knowledge
17   of the design of the register and with soe software skills that the information
18   could be extracted f m the register in such a way that duplicates and bias
19   could be mininmised and allow a more reasonable comparison between
20   laboratory data on that register. The bias still remains that its only women
21   who have opted on to that register and have not opted off since the change,
22   so it is not a cplmete database. But I wonder if in fact with some tidying up
23   the data might be more helpful in helping to evaluate the system

24   CHAIR:       how could the data be tidied up ..... it is clear from the
25   statement/analysis reports that there are duplics for squamous carcinoma,
26   high grade lesions. Perhaps there wdbe a way of taking the last smear in any
27   calendar year for a woman rather than all the smears for a women. I note

 1   that all the Bethesda codes, and you can code up to 5 different items, are
 2   listed, perhaps there could a way of adjusting that the most severe is
 3   recorded for the purpose of this exercise. I think if someone sat down and
 4   looked at the way the data is put on the register and what exact data we want
 5   for the purposes of our exercise there might be a way of using a software
 6   programme of esdtracting the data in that way – I I is worth a try.

 7   PROFESSOR DUGGAN: would this aprch factor in the varying incidence
 8   of cervical cancer in the ethnic groups in this country ..... I understand that
 9   the register holds all women’s data who have not currently opted off. And
10   therefore should cover all groups within the country. But unfrntly the ias is
11   that the opt off population yay have a different profiles from those on the
12   register and laboratories are reporting smears from both groups of women.

13   CHAIR:     we are looking primarily at the period 90 to 96 and it was only
14   after 93 that the Register became opt off. We've heard from Professor
15   Skegg in his view from 90 to 93 when the register was opt on that the
16   women most likely to opt on are ones in which you would expect a low
17   oincidence of cervical cancer. Firstly would you agree with that or not or
18   are you unable to say ..... unable to comment on that.

19   If that were the case what impact would that have on the statistically data in
20   your view ..... I t would depend on whether the opt on behaviour was similar
21   throughout the country. And I understand that the population coverage rates
22   were relatively uneven in the early stages of ouur programme as far as the
23   registered information is concerned. So I think there would be a bias there.

24   Do you think you could still cary out the exercise that you've talking about
25   ..... yes I think it would be worthwhile carrying out, being aware of thep
26   otentl bias at the end of the day it may not give you accrete information to
27   evaluate the labv in question but it might do.

 1   PROFESSOR DUGGAN:               your comments indicate such a report would
 2   have to be critically evaluated by somebody with knowledge of all the items
 3   you've mentioned ..... that is correct

 4   And if such a report was published in time magazine without any
 5   commentary from an expert in the evaluation, how would that report be
 6   interprtd by the reader ..... I think what we need in this situation is a report to
 7   be presented to a group of experts, those with epdemology knowledge,
 8   staticl knowledge and laboratory knowledge so that all aspects can be
 9   considered. I do not see why one would wish to put such a report in to the
10   public domaoin without guidance attached as to how it could be interpreted.
11   This is atool to try and help the inquiry arrive at a conclusion.

12   CHAIR:      without the p/review of the report how reliable would the report
13   be ..... I can't answer that.

14   As a pathologist looking at any study if the study hasn’t been p/reviewed
15   does that affect the value you place on the study ..... it afedcts the value I
16   place on the study, perhaps it doesn't change the inherent value of the study
17   but certainly it adds value to a study to have it critically assessed by peers.

18   MR HINDLE: I have an impresn that this would be quite a sophisticd thing
19   to undertake. do you see it that way ..... I expect it would be esxtremloy
20   sophisticated.

21   Would you need to be clear before you started exactly what you were
22   looking for ..... yes

23   Would you need to be very clear before you started exactly that the
24   information which you find, if you find it at all, is going to be useful to
25   expedmoolgoists, pathologists and the inquiry ..... actually all of the
26   approaches that we have discussed so far I think have a wider benefit than
27   simply assisting the inquiry in that they may provide start-up quality

 1   standards that may be applicable to the programme albeit min acpetdlbe
 2   standards but a number to attach to these standards to benchmark for the
 3   future. They have value far beyond just this inquiry

 4   Can I move onto a further and 5th possibility. We are interested in the period
 5   in 90/96.    we have already seen the document, the table of Gisborne
 6   laboratory results up to 94. would it be possible to go to the laboratories in
 7   New Zealand and see what information they could give you now about their
 8   pfce in that period ..... yes I think that would be a valuable thing to do in that
 9   I'm anxious that the register contains a select3ed group of women’s
10   information. Whereas the laboratories should hold all the information for
11   the results of the smears issued from that laboratory during that period of
12   time. The archived information retained in laboratories may not allow this
13   to happen but I wonder if approaching a series of laboratories or all
14   laboratories or laboratories with a similar population to the Gisborne
15   laboratory and looking at the sort of information that they have collected
16   themselves as a laboratory, number of smears and the different reporting
17   rates for different abnormalities might allow you to see whether there are
18   significant differences between one laboratory and another. This would of
19   course rqure the co-operation of other laboratories within New Zealand

20   CHAIR:       and dependent on whether or not other laboratories in New
21   Zealand had been getting that sort of data and analysing it in that way .....
22   most laboratories would kp a record of a smear result they had issued. Its
23   good professional practice, I just wonder if going back to those rcrds, not
24   reviewing the slides, there may be a resce you had to delivr to make the
25   records available to the resercher, but you many be able to derifve
26   information from the information from the laboratories which covers a wider
27   group of the women in the population being served by the laboratories. I am
28   slightly anxious that the screening rgster is a limited group of women.

 1   Do you have in mind some of the sortsz of information you would want to
 2   get from the laboratories ..... yes I think we would be looking at very similar
 3   kinds of information that are being collected by the screening registger but
 4   my criticisms of the information in the screening register are valid for this
 5   exercise too. We would not want to be double countoing if that was a
 6   routine practice in one laboratory, moreso than in another laboratory, so they
 7   would have to be sistes put in place to collect similar information from each
 8   of the laboratories. It may not be possible but it is another avenue which
 9   may be helpful to you.

10   Those are 5 posbties and I want to ask you about practicalities. Standing
11   back and looking at those 5 posbties, I know you wouldn't want to exclude
12   any study that might help, but ocan you think of anything else that you
13   would do to try and tackle the proble of judging under-reporting in Gisborne
14   between 90 and 96 ..... Nothing jumps to mind. I am one person, there are
15   lots of other experts in New Zealand with other ideas which are more valid
16   than some of mine. But I can't think of any others at the moment.

17   The Skegg study. We've heard evidence that there may be an insuperable
18   obstacle to that study – namely s74a of our h/act which makes it difficult to
19   collect information on the Cancer Register with information on the screening
20   register so whilst they may want to do that study they may ve prevented from
21   doing that by legislation ..... I think that’s extremely regrettable. As I've said
22   audit of the screening history of women who odvlp invasive cancer is a
23   powerful tool. And as a woman who participles in cervical screening in her
24   own country I would be looking to be able to use that particular kind of tool
25   for the cervical screening in which I am participating to try and identify why
26   some women despite their best efforts and the best efforts of the screening
27   programme have developed invasive cancer. There will always be a few
28   women that the screening programme cannot prevent developing ca but

 1   every situation should deliver a learning should be a learning situation for us
 2   to ensure there isn’t a loophole that’s developed that we can close or a
 3   quality issue that we can address. So as a woman I find that redxtremly
 4   regrettable, as a professional I would find it uancept ble.

 5   CHAIR: of the 5 possible ways that yuovue identified as a means of going
 6   back retrospevtlyu to see whether the level of reporting in Gisborne was
 7   acceptable, how many of those posbt8i9es are depend on being able to corlte
 8   the screening data with the Cancer Registry data ..... the invasive cancer
 9   audit certainly would be facilitated by being able to do that. I would need to
10   think further if that was impossible whether there were ways around it to
11   cvolect the information, but I believe I heard Professor Skegg say it was
12   important to have all the women and I would want to have a selected
13   population of women who developed invasive cancer studies.

14   The first possibility was an audit of the Cancer Registry, so bovsly if you
15   can't access the Cancer Registry to be able to correlate the screening register
16   because of s74a stopping you accessing data on the screening Registry you
17   are not being able to carry out that possibility are you ..... you might be able
18   to identify the women who have developed invasive cancer other than the
19   cregry.

20   The dfvclty in accessing infoi is the screening regry, you can't access data
21   about women on the screening registry ..... women who have dvlped invasive
22   cancer in New Zealand must have had a biopsy reported in a laboratory and
23   in the same way as people keep reports on cervical smears they also keep
24   biopsy they have reported. I am sure they retain the information in the
25   laboratory as well.    I just wonder if there are ways of accessing the
26   information through that routge. I wouldn't wish to give up.

27   As the starting point is whether or not a woman has been screened for a
28   smear test short of actually going to the Cancer Registry identifying women

 1   showing an incidence of ca and interviewing them how would you find out
 2   they had been scrdn ..... I d ustaood the problem was linking the Cancer
 3   Registry with the screening register not linking some other information with
 4   the screening register.

 5   It is the screening register information which is hampered by s74a ..... I don't
 6   have a good enough knowledge to suggest a way around the information but
 7   others might.

 8   PROFESSOR DUGGAN:               one can apply to the Cancer Registry for
 9   information on women who have developed invasive cancer but you cannot
10   go to the cytology registry to find out if these women are enrolled on the
11   screening programme

12   MR HINDLE: it is theoretically possible I suppose to approach the problem
13   by looking for the women first and seeing whether by some campaign you
14   could seek information from the population that would allow you to get
15   consent of the women affected,..... who had ca or their executors or heirs,
16   did you regard that as a possibility ..... I think every means possible should
17   be used to attempt to cary out this exercise because it is such a useful tool.

18   CHAIR: without having to go to extremes, you have listed 5 different ways
19   that you think an exercise could be cxaried out to evaluate whether there has
20   been under-reporting, without going over each one, in summary can you say
21   how many of those esdercises are dependent to access data on the screening
22   Registry and correlate that data with the Cancer Registry as the first
23   approach that you would take ..... using the measurement of positive
24   predictive value can be done without either register by going to the
25   laboratory. Reviewing consec slides from laboratories can be done without
26   either register and trying to derive from laboratories the reporting profiles
27   can be done without going to either registger. The sugestgn I had about
28   attempting to tidy up the staticl information on the register would naturally

 1   require you to go to the register but you wouldn't have to go to the Cancer
 2   Register for that.

 3   MR HINDLE: does it surprise pyou that here we are in the year 2000 trying
 4   to answer this question and it seems that one of the places we can't
 5   information from is the screening register ..... it does surpise me, its because
 6   you want the scvtg register to be all things to all men or women. It is a
 7   database of women bg screened which can be used to ensure they are
 8   recalled at an appropriate time. There surely are other means of clectg
 9   staticl information about your screening programme without relying simply
10   on one register.

11   Is it fair to summarise everying we have been talking about you do these
12   studies, then you look at lasb practice and they are all flags and having taken
13   all of the information into account if you see that the number of the flags or
14   all of them are pointing in a particular direction you can strt to make a
15   commonsense judgement about accountability of reporint in this area ..... yes
16   it is and I think the word commonsense approach is what the panel are going
17   to apply at the edn of the day rather than an exact statiticl answer.


19   MR HODSON:           just on the alternatives.    You may reclct I put it to
20   Professor Skegg on the choice of taking a selection of slides from Gisborne
21   laboratory and a selection of slides from other laboratories and subjecting
22   them to independent review that it might be useful to have cohorts one being
23   a cohort of slieds from other laboratories within a similar size of practice
24   and if possible socio economic background, and then another cohort
25   representative of all laboratories in New Zealand. Can you comment on that
26   ..... you would very have to very carefully design the selection of the slides
27   not to introduce additional bias into your study.         There is an implicit
28   suggestion that laboratories of one size need to be compared with like

 1   laboratories and not laboratories of a larger size.      I'm not sure f that
 2   asujmptinon is a vald one given women don't determine what size laboratory
 3   is going to report their smear and they should be able to expect the same
 4   standard or at least the same basic standard of reporting from any laboratory.
 5   I understand the purpose of the study.

 6   On another topic entirely, in your evidence in chief, paragraph 79, you were
 7   telling us of the role of the cytology laboratory and the National Cervical
 8   Screening Programme and you said in the UK “laboratories … advisory
 9   services .. not simply a results only service.” do you see any merit or would
10   there have been any merit when the programme was getting underway in
11   endeavouring to have laboratories in thecmnty in which rthey serve so that
12   the pathologist or screens could be readily accessible to general practitioners
13   on the one hand and the consultations colposcopoists and so forth on the
14   other to ointeract with the taking of smears and the action arising out of the
15   reports ..... I would have expected pathologists involved in reporting cervical
16   smears to have a Professor relationship with the colposcopoists delivering
17   the copos service so that when discrepys between the grade of disease
18   present on the smear and the inspectn of the cervix arose a discussion could
19   take place as to what should happen nextl. Sometimes review of the slide
20   might change or downgrade the abnormality suggested.              On another
21   occasion review of the slde might maintain the level of abnormality
22   originally reported in the slide an dencrge the colposcopoiste to look again.
23   This ikind of professional ointeractn is good practice. however it can take
24   place over large geographic areas, you don't have to be within walking
25   distance of the other professional.

26   Firstly should there be the same relationship with the smear-takers ..... its
27   very difficult to have exactly the same relationship with smear-takers
28   because of thenbrs inovled. There are likely to be a very small number of

 1   colpposcopoists but a much larger number of smear-takers. Hovr, in the UK
 2   the laboratory has a responsibility to assist smear-takers in auditing their
 3   own performance. And in my laboratory we have smear-takers who ovisit to
 4   see what we do, to better understand what information is included on a
 5   request form, who telephone because they are anxious that the last 5 smears
 6   they took were all reported as inqd and were they doing something wrong or
 7   was it by chance. That kind of professional relationship also exists with
 8   smear-takers.

 9   Do those considerations influence the desirbty of having particular number
10   of slides read in a particular laboratory so that there isn’t too small a sample
11   and not he other hand it isn’t too big and people g3et out of touch ..... I foind
12   that slightly dflt to answer being responsible for oe of the laretgest cytology
13   laboratories in the UK. We have currently about 90000 smears/annum, we
14   have in the past in our busiest year hd wello over 100000 cervical smears. I
15   think its important for a laboratory to ensure that the interaction with other
16   health professionals is maintained whatever its size.

17   In the same paragraph the last sentence it is said “the laboratory should …
18   limited” and I take it that is a reference to having extgernal quality control
19   ..... not entdirely. The external quality control of the NHS cervical screening
20   programme pre-existed our accreditation system for laboratories, the cpa
21   system. external quality assurance involves the profiency testingn scheme
22   where every individual in the UK, oincluding myself, who is involved in
23   iusing a report on a cervical smear must profve their competence by taking a
24   test once a year. And the results of those, the results of the slides evaluated
25   in the test are marked by an edxternal person and returned to the laboratory.
26   And that scheme has been running since the late 80s and we are currently
27   reviewing it to see whether it is still appropriate in the year 2000 or whether
28   it needs to be changed in any way. The laboratory accreditation system is

 1   more    concerned    with    ensuring     that   laboratory    practices   staffing
 2   accommodation are in place in an appropriate way and meeting the set
 3   standards than looking at the specific competencies of oindivids within the
 4   laboratory.

 5   It sounds something like our TELARC ..... I know a olittle about TELARC.
 6   They are not identical accreditation systems but rather similar.

 7   Is shd accurate or is it now must in relation to cpa ..... irtl to cap
 8   accreditation in the UK laboratories are recomdnd to apply for accreditation,
 9   they should maintain accreditation but it is not mandatory except for those
10   laboratories reporting cervical smears where it is mandatory.

11   Can you tell us when that was achieved ..... April 1988 was the date set by
12   which 1998 was the date set by which laboratories must achieve cpa
13   accreditation to continue reporting cervical smears.

14   CHAIR: what is cpa UK limited ..... I have tried to describe it in paragraph
15   88. ..... clinical accreditation UK limited

16   It is a company that specialises in accreditation ..... an indentent company by
17   the Royal College of Pathologists and the professional odies in the UK to
18   cary out accreditation, they have more recently been asked to look after the
19   quality assurance schemes for all laboratories.               Its not specific to
20   histolpaghtyh but all laboratories

21   And the profidciency test who imposes that ..... the prro/test is organised tru
22   the regional quality assurance teams.            Cpa – you will not get cap
23   accreditation unless you show participn in the testing scheme

24   There is the cpa for accreditation for quality assurance, there are also quality
25   assurance teams ..... in each region.

 1   And where do they come from ..... they assist the regional director of public
 2   health personally responsible for the quality of the cervical screening
 3   programme in his region.

 4   Who are these people employed by ..... they are a group of health
 5   professionals employed within the NHS within that region

 6   And in term of your rescrng programme how long have the quality assurance
 7   teams been in existence ..... I would need to check the records, I believe it
 8   was the late 80s

 9   MR HODSON:         I listened yesterday to some discussion of the Australian
10   re-screening of the Gisborne laboratory slides, I have here the contradict
11   between the Health Funding Authority and the laboratory in Australia – it
12   could be properly produced by Tracey Mellor later.

13   MR MURRAY:          this will be produced in the Tracey Mellor bundle of
14   exhibits.

15   MR HODSON:         as we have heard consideration was given to an ovseas
16   laboratory, and a contracted was entgreed into. I ask for your comment on a
17   number of clauses in the contradict. You will aprecite the Australia
18   laboratory and the Health Funding Authority which is the body now
19   responsible for theprog, 2.2 the laboratory “finds itself …. interests of the
20   Health Funding Authority” can you comment on that clause ..... I would ask
21   for alcarifctn as to what exactly is meant by promoting and protecing the
22   interests of the Health Funding Authority. It is not entirely clear to me. I am
23   however not an expert in legal documents or contracts.

24   I think all one can say at this stage, it might raise an eyebrow with you .....
25   only if there is a suggestion that the interests of the Health Funding
26   Authority are in some way different from the interests in the programme.

 1   Schedule 1, p6, top right, the second paragraph “the principal aims of the
 2   investigation … concerned.” ..... yes

 3   Can you comment on whether you would see those two aims as being
 4   entirely congruent ..... I think there is a portentl conflifct between these etwo
 5   aims in that to ensure that women receive the apropl treatment for their
 6   health may lead a laboratory to err on the safe side. To ensure that women
 7   receive the appropriate treatment for their weelbng may not have th same
 8   consqndce one would want to limit in that situation false positives, we don't
 9   want to make women anxious unnecessarily. And the latter part of the
10   sentence “to determine the extent … concerned”, suggest a more scientific
11   exercise than the first part of the sentence would necessarily require.

12   If you turn to p8 please, the directxn to the laboratory and what it is to do,
13   paragraph 4, restaining as necessary, use your recognised screening process,
14   thorough … primary screener and targeted rescrg in accordance with your
15   standard procedeusres” I observe there is nothing in that about quality
16   control, about rapid re-screening or about use of the apt process which the
17   laboratory might be using ..... targeted re-screening is a quality control
18   mecanicm.

19   Can you distinguish that between rapid re-screening ..... yes. targeted re-
20   screening by my understanding is where laboratory protocols require
21   particular smears to be fully read by two individuals before the result is
22   issued. Smears such as those from a woman whose last smear was abnormal
23   but whose current smear is negative might fall into that categryl. Smears
24   from women with symptoms suggestive of invasive cancer post coital
25   bleeding for example might fall into that category. The requirement is that it
26   follows their standard proceedings,l it would be folowoing their protocol in
27   their laboratory. It is a quality control measure within laboratories.

 1   You would expect under standard practice in a good laboratory following
 2   that protocol that abnormal slides would be seen by at least 3 people ..... not
 3   necessarily.

 4   At least two ..... at least two, one who would be a pathologist.

 5   Sometimes 3 ..... yes.

 6   That would be a markedly different procedure to that possible to a laboratory
 7   where all the screening is done by one pathologist ..... yes,

 8   Paragraph 6 “all electronic reporting … New Zealand National Cervical
 9   Screening Programme.        I think you have already commented on the
10   introducen of the atypical squamous cells of undetermined significance
11   report not available to the Gisborne laboratory in the years that it was
12   reporting. That would indicate a very great difficulty in making comparisons
13   ..... yes it makes it difficult to compare the results issued by sidey in the year
14   2000 with those issued by Gisborne in the early 90s.

15   So from the point of view of the health of the women concerned the re-
16   screening exercise would be done in a way that would be likely to be far
17   more beneficial to those women ..... my understanding this exercise would
18   have sacrificed specificity to obtain the best sensitivty possible because of
19   the circumstances in which the exercise was being carried out.

20   It could not be said that the re-screening exercise was done using similar
21   techniques to those used in the original reading ..... no.

22   Thank you.

23   Exhibit EM/CA/ 007 – Contract for services.



 1   Dr McGoogan I know that for the last little while we’ve been dealing with
 2   the problem of how the committee of inquiry deals with terms of reference
 3   one, i.e. determining whether there's been an unacceptable level of under-
 4   reporting. In the course of one of your answers to questions relating to that,
 5   you referred to the issue of professional practices of the laboratory in
 6   question as being a factor which would be taken into account as helpful in
 7   assessing the question. Have I got that right ..... yes.

 8   And with that in mind, I want to take you to some detail of the practices of
 9   this laboratory and the pathologist concerned. Before doing so I want you to
10   understand that its going to be necessary to put some of the material to you
11   on a hypothetical basis because we haven't yet had evidence dealing with the
12   detail of the practices that I've mentioned. So you will understand that I'm
13   going to be putting some material to you which, as yet is hypothetical, but
14   will be put to other witnesses. Do you follow that ..... I think so, I have to
15   assume the information is from a hypothetical laboratory.

16   I am going to ask you to assume that the material I put to you will be
17   evidence that the committee will hear, right. ..... yes.

18   I want to start with the qualifications and training of Dr Bottrill and I would
19   ask you to look at this bundle of documents that has come to me via counsel
20   for the college and for Dr Bottrill, it’s material from the Royal College of
21   Pathologists of Australasia relating to his qualifications. And if you could
22   just take a moment, please, to look at the bundle of material. ..... yes.

23   Now, you will see that that material comprises the Dr’s application for
24   examination for membership, which was lodged in 1964, and he was seeking
25   examination in the four listed categories there, morbid anatomy,
26   haematology, chemical pathology and microbiology ..... yes.

27   In fact, for reasons which I will come to briefly, the Dr did not sit any sort of
28   examination until December or perhaps late, shall we say, 1973, and then If

 1   you look back at about the 5th page of the material that I've given you, you
 2   will see that he passed an oral and practical pass in special anatomical
 3   pathology, do you see that – it’s a document headed, with a handwritten note
 4   at the top “Extract of minutes of Board of Censors 10 December “ ..... yes.

 5   Now, can you assist me in this regard: from your knowledge. Of pathology
 6   generally can you tell us whether an examination in special anatomical
 7   pathology would include any cytology at all and more particularly any
 8   gynaecological cytology. ..... I am not familiar with the examination
 9   structure in 1973 of the New Zealand Royal College of Pathologists of
10   Australasia, I presume

11   Yes ..... so I have no detailed of what the examination at that time would
12   entail.

13   Looking, please, at the third page, which gives Dr Bottrill’s experience – a
14   sort of curriculum vitae really ..... yes.

15   From the time when he qualified as a Dr – that would be MB ChB in July
16   1953, through to, by my calculation, September 61 when he finished his
17   course in pathology which he undertook, as you will see, at Leeds University
18   and the Leeds Hospital Board ..... yes.

19   It is apparent, if you look there, that having done an initial one year’s
20   training, dividing into 3 periods of 4 months, starting in March 57, he then
21   moved on to St James Hospital and then Wakefield Hospital, spending, it
22   appears, 33 months at St James Hospital, correct ..... it would appear so, yes.

23   And at St James the application that Dr Bottrill submitted states that he “did
24   some histology, about 2000 specimens, but there is no reference, is there, to
25   any cytology at all ..... correct .

 1   In fact, you can confirm that, just by looking at it if you would, that in the
 2   entire summary of his training post graduation as a Dr, there's no reference
 3   there at all to any training in cytology is there .....

 4   MR HODSON: not only is it not the entire summary, some of the material
 5   for whatever reason is not now available

 6   CHAIR: I note at the top of the page there is a little (3) and over the page
 7   (4).

 8   MR HODSON: the word continued at the bottom poses the problem.

 9   CHAIR: you say there is a 3a.

10   MR GRIEVE: my friend knows that this issue was traversed by the College
11   and I quote from the letter that he received because I was given a copy of it,
12   as he now is, the original document is a folded form and does not include Dr
13   Bottrill’s working career after 53, there is a suggestion on the curriculum
14   vitae provided that this was included but there is no area on the form where
15   that could be included and no additional information is held. So the college
16   has told us that it doesn't have anything, and I think they went on to say there
17   is no indication that there is anything missing. I will need to get copies of
18   this. [Bundle of documents including the last document EM/CA/ 008] so
19   on the face of it Dr McGoogan, in so far as this material goes, we have no
20   indication that Dr Bottrill’s training in pathology included cytology .....
21   correct.

22   And I come back to this question about anatomical pathology. I know you
23   can't help us with what the Australians examined on, but in a general way
24   would you have expected, hearing the description I've passed in anatomical
25   pathology would you expect that would include cytology – might it ..... in
26   1973 it is likely that it did not include cervical cytopathology.

 1   And that is because that, as I understand it, in the 60s and 70s
 2   gynaecological cytology was – would it be fair to say – probably beyond its
 3   formative stages but still developing ..... yes additionally gynaecological
 4   cytology was often carried out in laboratories organised by gynaecologists
 5   rather than within laboratories where histopathologists were working. There
 6   was a gradual trend over the next 20 years to include cervical cytopathology
 7   within histopathology, but at that time many of the cervical cyto laboratories
 8   were distinct from histopathology laboratories.

 9   I have here for the panel a copy of the relevant transcript in the High Court –
10   it is not proposed to produce the transcript as an exhibit, it is of course –
11   depending on how it goes previous consistent or inconsistent statement and
12   would not be produced.

13   CHAIR:     my view is if you are going to put these questions to Dr Bottrill
14   and elicit the evidence that you got from him in the High Court from him in
15   this inquiry and obviously you anticipate it to be the same, you can put
16   propositions to the witness in hypothetical form – I don't think we need to
17   see the notes of evidence.

18   MR HODSON: I would be obliged to have a copy.

19   CHAIR: in order for the inquiry to determine whether or not there had been
20   under-reporting in Gisborne there need to be a look at the standard of the
21   pathologist’s work in Gisborne at the time. I assumed he wanted to ask this
22   witness as an expert witness of the practice in Dr Bottrill’s laboratory at the
23   time – to suggest that his practices didn't measure up. if that is the line of
24   questioning I see that as very relevant.

25   MR HODSON:           she can only comment on the basis of her knowledge,
26   Scottish, Welsh and English – unless she has some knowledge of New
27   Zealand practices.

 1   CHAIR:     she is certainly a very well qualified pathologist, and there must
 2   come a time at which there is a standard of any country would not be
 3   acceptable.

 4   MR GRIEVE:        you will need to retain that Royal Australasian College
 5   material in front of you Dr McGoogan. If Dr Bottrill tells us that he – as I
 6   understand it Dr Bottrill will tell us that he completed his cytology training
 7   in Yorkshire while he was working as a pathologist in 1960, and he will also
 8   say that during his course of training in pathology, that’s the 4 year course
 9   that you see that he’s referred to in his application, he had an opportunity to
10   study gynaecological cytology with the Dr in the Women’s Hospital, and he
11   will say that he was only just – i.e. the Dr at the Women’s Hospital – was
12   only starting his cervical cytology then and “I spent” he will say “part time
13   about 3 months with him.” Dr Bottrill will say that he went on to Wakefield
14   where he worked in a chest surgery unit, was involved in cytology of sputum
15   and similar material, that he was working as a general pathologist in
16   cytology, which was “although interesting not his main occupation”, and he
17   goes on to say that the range of his practice in Gisborne up until 90 “was that
18   of a general pathologist.”   In relation to his cytology training, he will say
19   that he had done, in terms of examination, “only the oral and practical
20   examinations for fellowship of the Australian College”, and he will further
21   say that after completing his training in the UK in 1961 he came out to New
22   Zealand and until 1968 – that’s a period of about 7 years – he was not
23   practising any cytology. I suggest to you that, and of course we have some
24   difficulty with the precise detail because of the time lag, but I suggest to you
25   that whatever the cytology training he undertook in the UK, in the late 50s
26   up until 1961, was totally inadequate compared to the sort of training that
27   you have described as being a current requirement for those reading cytology
28   smears, would you agree with that ..... I would agree that the training of
29   pathologists in the 50s and 6t0s is very different from the training of

 1   pathologists today. I would agree that this application for examination in the
 2   Royal College of Pathologists of Australasia does not mention the
 3   experience in the women’s hospital while the pathologist in question was in
 4   Yorkshire in 19960 when some gynaecology cytology was done. And that
 5   training is very different from the training which a pathologist now
 6   undergoing training would undertake. very few pathologists before 1980
 7   participated in reporting cervical smears. So the norm was to have very little
 8   or no training in cervical cytopathology because it did not form part of your
 9   normal daily workload.

10   I know we have difficulties transposing New Zealand conditions and
11   comparing them with UK conditions, but hypothetically if you had the
12   situation in the UK pathologist practising on his own as we've heard, doing
13   all the smear reading, primary screening, about 5000 smears/a, and with the
14   qualifications and experience that I've put to you thus far, if that pathologist
15   in that situation had been in the UK at the same time, what retraing would he
16   have been required to undertake in order to qualify him to read cytology
17   smears post 1980 ..... the experience you've described thus far I agree is
18   inadequate for an individual reporting cervical cytopathology in the late 80s
19   90s. had such an individual wished to participate in the reporting of cervical
20   cytopathology for example in my laboratory, then we would have agreed a
21   training course over a period of months where all the work was supervised
22   by another competent cytopathologist until a time was reached where both
23   agreed that the pathologist under training could sign out results on his own.
24   It would be expected that that pathologist would participate in the
25   proficiency testing regularly – that’s what happens today in my laboratory, it
26   happened in 1990, and in 1980, there has been no change.

27   Chair in principle do you think it good practice for a pathologist anywhere to
28   be reading smears in the circumstances in which Dr Bottrill was reading the

 1   smears, given the hypothetical background that HAS BEEN put to you by
 2   Mr grieve ..... no I don't think it would be good practice for a pathologist
 3   without relevant training not to report any type of pathology sample row
 4   unless they were trained and were currently competent.

 5   MR GRIEVE:            do I understand you from earlier answers that the
 6   proficiency tests that UK pathologists are required to undertake must be
 7   passed annually ..... a satisfactory performance in the proficiency tests is
 8   required yes

 9   Annually ..... as often as the region holds those tests they sometimes get a
10   little out of kilter so in some years its every 8mths, others it ma be 15 mtohs,
11   they must pass the test at the next round rather than a specific point in time.

12   And if they fail or don't measure up to the standard of the time what happens
13   ..... an explanation would be sought as to why the performance in that
14   particular proficiency test did not meet a satisfactory standard and if the
15   suggestion or the result of this inquiry led to the suspicion that the
16   competence of the individual might not be of a sufficient standard it would
17   be agreed with the individual that a period of retraining would be undertaken
18   either in their own laboratory or being seconded to another laboratory so that
19   remedial action could be taken. Or it could agreed that that person no longer
20   participated in reporting cervical smears.

21   So subject to the slight variations in period that you mentioned, there is a
22   constant regular quality control on the competence of pathologists reading
23   smears in the UK ..... currently yes

24   Yesterday very early on after you finished reading your brief of evidence
25   madam chair put to you the factual situation that we are here dealing with
26   namely the sole pathologist doing all the reading etc – you know the facts
27   now don't you ..... I think I do.

 1   And you listed 6 criteria that you said would be essentials in order to give
 2   the women of the area concerned some sort of service, although you
 3   qualified it by saying that you were being asked to give a prescription for a
 4   bad service but these were necessary to ensure there was some quality do
 5   you remember that ..... I do remember.

 6   I am going to ask you in relation to some of those about evidence of Dr
 7   Bottrill’s approach to some of the issues,..... and one of them is external
 8   quality assurance and in that regard you will recall saying that the
 9   pathologist in that situation that we are dealing with would need frequent
10   participant in external quality assurance – p940 – line 1. in relation to
11   external quality assurance I should perhaps set the factual scene for you.. in
12   Australasia at the time the Royal College of Pathologists ran from Australia
13   I think an external quality assurance programme that was in the early 90s the
14   subject of some criticism because of, one delays in giving the participants
15   their results back, and two, because of some classification problems as
16   between New Zealand and Australia.

17   MR HODSON: the actual evidence before the inquiry is the 91 document,
18   that in 91 there was no suitable external quality programme available.

19   MR GRIEVE:        the 91 document, which was a publication from the – a
20   report of a recommendation of the Cytology Advisory Liaison Committee
21   whatever the shortcomings of the programme nevertheless recommended
22   that laboratories participate in the Australasian external quality assurance
23   programme – reference Boyd volume 5, tab 22.

24   CHAIR:     if you would be careful when putting the hypotheticals to the
25   witness that subsequently you are going to have to prove them.

26   MR GRIEVE: it is plain, and its before the committee already, that at the
27   time – 91 – Cytology Advisory Liaison Committee was recommending
28   participating in this programme, and as per Boyd exhibit 22. the evidence

 1   will be, Dr McGoogan, that at no stage did Dr Bottrill participate in this or
 2   indeed any other formal external quality assurance programme, and he will
 3   say, I anticipate, it will be certainly put to him that at the time he regarded
 4   external quality assurance as a practice having “no relevance to the practise
 5   of practical cytology”. What is your comment, can you comment on the –
 6   p127 – line 30 through to 35 of the transcript. What is your comment on the
 7   validity of that sort of attitude to external quality assurance ..... I think it was
 8   an unwise attitude

 9   And in the context of a practitioner who was practising alone and needed to,
10   in your opinion, adopt the 6 criteria in order to maintain safe practice

11   MR HODSON: he put the proposition he has just quoted, the answer was
12   from Dr Bottrill, “I don't like your way of putting it”, the judge then said
13   “stripped of the hyperbole, I take it” “yes, it could be”

14   CHAIR:      if it turns out that Dr Bottrill, doing the primary screening and
15   reading approximately 5000 year if it turns out there was no external control
16   applying to his laboratory what comment would you make on his
17   professional practice ..... I would have expected a single pathologist working
18   alone would have sought external quality assurance programme to ensure his
19   competence if only to prove his competence

20   What opinion do you have of a pathologist working in the circumstances in
21   which Dr Bottrill was doing who does not apply external quality control to
22   his practices – is that acceptable, unacceptable, good/bad indifferent ..... it is
23   not good practice, it is questionable practice. In fairness, in the late 80s and
24   early 90s many pathologists did not see the need to participate regularly in
25   external quality assurance schemes.        But I think that a single practising
26   pathologist on their own without well documented evidence of a recent
27   training in cervical cytopathology and with unusually high workload in
28   cervical cytopathology for a pathologist not to participate in well publicised

 1   and easily available external quality assurance schemes was regrettably not
 2   good practice. whether he participated in this particular scheme or whether
 3   he could have chosen a different one if he didn't like the way the scheme
 4   was set up was another scheme but to chose one scheme is not good practice.

 5   From perspective of cervical screening do you think it good practice for the
 6   programme to be structured in such way that someone within the programme
 7   allowed the use of such a pathologist practising in that way to read smears
 8   for the purpose of the programme ..... no It was not good practice.

 9   MR GRIEVE: would you look, please, at some exhibits to be produced by
10   a Mr Mules for the Health Funding Authority, it is going to be volume 2, tab
11   21 – note the date 23 August 1994, a letter from Dr Malpass . ..... yes.

12   The context in which this letter was written was that the Midland Regional
13   Health Authority wrote to all the community laboratories in its area
14   enquiring about TELARC accreditation and you can see that in the first
15   sentence that question is address by Dr Bottrill, but he goes on to say, “I do
16   not participate in any external quality control programmes, I do attended at
17   least one international conference every year, this enables me to keep up
18   with current practice in diagnosis and management”. The point is that this
19   letter brings his attitude a few years further forward into the mid 90s doesn't
20   it ..... yes I think it also identifies that he may be misunderstanding the
21   purpose of external quality control programmes. If attending conferences
22   would balance non participant in external quality assurance. The purpose of
23   external quality assurance programmes is the diagnostic ability of the
24   pathologist rather than keeping up-to-date with current thinking.

25   PROFESSOR DUGGAN:              is proficiency testing equivalent to external
26   quality assurance ..... it is a form of external quality assurance, yes,
27   particularly specialised form of external quality assurance in that it is more
28   of a test and marked as a test than external quality assurance. It may not be

 1   marked in the same critical way that proficiency testing is marked. in
 2   proficiency testing the pathologist is required to define the abnormality in
 3   the slide and get it correct according to the views of the panel who prepared
 4   the slide set. In other external quality assurance schemes it may be that an
 5   area of a slide is marked and pathologists share their opinions on what that
 6   particular area of the slide shows and a pathologist would judge whether his
 7   opinion was in line with other pathologists or out of step with other
 8   pathologists.

 9   CHAIR:       so you wouldn't see attendance at at least one national or
10   international conference symposium of course every year as a replacement
11   of participation and external quality control ..... no I wouldn't.

12   MR GRIEVE:        you will recall that in your 6 criteria to which I've already
13   referred one of them was frequent attendance at meetings of cytologists re
14   cervical screening . what is your comment about the description in that letter
15   of attendance at meetings fitting in with that criterion ..... the letter does not
16   specify the nature of the international conference symposium of course
17   attended each year. They could have been for other aspects of pathology or
18   other aspects of medicine.

19   Subject to further clarification, on its face, not satisfactory ..... it would not
20   satisfy me

21   Another criterion that you mentioned was frequent and good interaction with
22   pathologists in another laboratory involving exchange of work at least from
23   the sole pathologist to the other laboratory and you will see that something
24   of that interaction is mentioned in the letter, the last paragraph beginning
25   “there has always been co-operation between this laboratory and Gisborne
26   hospital.” Just have a look at that paragraph if you would.             The last
27   sentence is interesting. In the context of this inquiry “we show each other all
28   interesting or difficult problems in histopathology and “ and was the author

 1   meaning to include cytopathology “of the type you refer to in your letter.”
 2   Would you consider that interaction of the sort described in that letter was
 3   sufficient to satisfy your criterion as you described it yesterday ..... no, what
 4   I was attempting to describe yesterday was a formal arrangement whereby
 5   slides would be exchanged for the purposes of quality control. This is
 6   admirable practice to discuss a difficult slide with another pathologist who
 7   has an interest in that area and may define your diagnostic abilities but it was
 8   not what I was describing when I discussed it yesterday. Also I note all
 9   interesting or difficult diagnostic problems is a very subjective statement, it
10   implies that all cases which are difficult or interesting were exchange but it
11   simply means only those cases which the observer at that time thought were
12   difficult and interesting.

13   And is my reservation about whether or not this exchange such as it is or
14   was included gynaecological cytology a valid one ..... I see no evidence that
15   it did include gynaecological cytology.

16   So in terms of your 6 criteria the first one was interaction, which we've just
17   dealt with, that was not sufficient. The fourth one was frequent participation
18   in external quality assurance, that was inadequate, the fifth one was frequent
19   attendance at meetings – inadequate, the 6th one was that all laboratory
20   accreditation processes were met and the evidence will be that during the
21   relevant period Dr Bottrill’s laboratory was never accredited by TELARC,
22   so that also is a failure. That leaves internal quality control, and in that
23   regard you mentioned at p939, line 24, the need for well documented
24   processes and data collection to enable internal quality control to be
25   assessed. Now, in that regard, Dr Bottrill will say, as I understand it, that
26   during his time he did a number of quality control checks, which included
27   randomly selecting every smear with a 0 at the end and re-looking at it and
28   re-looking at his report, which meant effectively that he was reviewing

 1   randomly 10% of his slides. What is your comment about that process in
 2   terms of effective internal quality control as you had it in mind when you
 3   mentioned it yesterday ..... was this process documented

 4   There is no evidence of it.. ..... I think this is evidence of an attempt to
 5   quality control the work going through the laboratory. 10% one slide in 10
 6   is the way that many laboratories carried out internal quality control of
 7   course the intention was that the slides would be fully re-screened by a
 8   separate individual and that the result would only be correlated with the
 9   original result after the rescreening had been carried out.

10   CHAIR:        what about with the same individual re-reading his slides ..... I
11   find it hard to see how this could actually qualify as internal quality control.
12   In fact, it is more likely to confirm errors and lead the pathologist into
13   believing that they are doing better than they are because they are just
14   reinforcing the original opinion which may not have been accurate.. I would
15   be more impressed if the pathologist could provide the review result in a list
16   and whether some reports were changed because of this process.

17   MR GRIEVE:          you don't give this criterion a pass mark either, is that
18   correct ..... I think it’s an attempt an internal quality control but it would not
19   satisfy me.

20   Well, the final criterion that you mentioned was a very important one of
21   course, biopsy smear correlation. And we've of course heard a lot about that
22   in the context of the programme generally, the need to correlate histology
23   with cytology ..... yes.

24   We don't have any evidence about what, at this stage, the correlation
25   between the cytology and the histology might be now even, let alone what it
26   was back at the relevant time, and when you mentioned biopsy smear
27   correlation I take it you are referring to a correlation or exchange of
28   information relatively approximate in time to the taking of the cytology

 1   smear are you ..... yes I am. I am speaking about a continual process of
 2   quality assurance continuing throughout all time. I would assume that some
 3   slides in this particular laboratory were reported as abnormal as high grade
 4   squamous intraepithelial lesion, some if not all these women would have
 5   received a biopsy, at least some of these biopsy must have been available or
 6   the report of the biopsy would have b made available for smear around the
 7   time week on week year on year. I'm not speaking of the correlation of
 8   biopsies taken now from women who have been identified because of the
 9   Sydney re-screen g exercise.

10   There is simply no evidence any of that was done at the time., but even if it
11   had, let's say something of that was done hypothetically, we have on my
12   analysis of your answers 5 out of 6 where Dr Bottrill did not comply with the
13   criteria which you nominated yesterday as being essential to give any form
14   of confidence to the women whose slides were being tested that they were
15   getting accurate results – do you agree with that ..... yes.

16   And indeed it may turn out that the failure rate was 6 out of 6 which makes it
17   even worse doesn't it ..... possibly, yes.

18   CHAIR:      you were being questioned by Mr grieve about internal quality
19   control and the circumstance of the screen reader as the person who read the
20   second time. I would like you to comment on paragraph 116 of your brief of
21   evidence where you are talking about experiments by ? and you refer to one
22   at the bottom of the page 38 and paragraph 116 where the interaction
23   “between successive judgements was studied “ . those comments set out at
24   paragraph 116 how do they apply to the circumstances that have just been
25   put to you of Dr Bottrill being the person who was re-reading his 10% of his
26   smears ..... these are exactly the reasons why I have concerns about the same
27   person quality controlling their own work. They simply may reinforce their

1   own mistakes. And develop a belief that their accuracy is being maintained
2   when indeed it is falling off.





 1   MR GRIEVE: Dr McGoogan, before lunch we had got to the point where I
 2   think I had asked you and you had confirmed that the fact that Dr Bottrill
 3   really did not meet at least 5 of your criteria possibly 6, was a matter that
 4   would cause you concern ..... yes.

 5   Bearing in mind the context in which you were asked to specify those
 6   criteria – namely the context in which Dr Bottrill worked, would you agree
 7   that his failure to meet those criteria amounted to a breach of his duty of care
 8   that you mentioned yesterday ..... the practices as described to me so far
 9   would fail to meet the standards I would expect from a laboratory with good
10   practice.

11   And I suggest to you that in failing to adhere to those standards and to
12   comply with the criteria you defined, would amount to a failure on his part
13   to meet the duty of care required of him when delivering this cytology
14   service to patients in this area

15   MR HODSON: this question is not proper.

16   MR GRIEVE:         In using the phrase “duty of care”, I took that from Dr
17   McGoogan’s own evidence, p946, line 25

18   CHAIR: you are referring to Dr McGoogan’s evidence in her evidence.

19   MR GRIEVE: certainly. I am asking you about whether Dr Bottrill met a
20   duty of care in the context or as you defined it or have it in mind when using
21   that phrase yesterday ..... I would have concredrns someone practising in the
22   way descried would not be meeting my standard of care.

23   CHAIR:      could you ourtline what your standards fora duty of care are .....
24   doctors I believe are in a particularly special position in the community.
25   Patients trust them with their health and wellbeing. It is, therefore, important
26   that doctors ensure that as far as possible they can provide a reasonable care
27   to individual patients that other doctors would be providing in similar

 1   circumstances. In order to do that, they have to continually assess their own
 2   performance but be aware of their own areas of weakness and expertise.
 3   And take care to consult with other health professionals when they are in
 4   doubt as to the best course of action.

 5   MR GRIEVE:        and I suggest to you also that in the context in which we
 6   have been discussing these criteria, they are obviously criteria which would
 7   be expected of pathologists based on much of the evidence that we've heard
 8   about standards and so forth ..... yes.

 9   What I am going to put to also is, and it follows from what you said, that Dr
10   Bottrill having failed to meet those criteria is continuing to practice and read
11   gynaecological cytology material was in fact unacceptable practice
12   according to Dr McGoogan’s standard of care.

13   MR HODSON:          which takes into account Dr Bottrill’s state of mind and
14   awareness.

15   MR GRIEVE:         ..... if he failed to meet these criteria then I would have
16   concerns that he was practising to an inadq standnards.

17   Concerns as to whether he was ..... yes.

18   My learned friend has conveniently raised an issue about Dr Bottrill’s
19   awareness of his own shortcomings if they are found to be so, and as I recall
20   ilt you made comment about thaqt very – it was in that very contesdxt, in
21   fact, that you raised the issue of duty of care because in answer to a question
22   from madam chair you referred to this duty of care, and I will remind you
23   what the question is – p946, line 21 of the transcript: do you think as a
24   pathologist that there comes a time when a pathologist should him/herself
25   recognise he/she does not have sufficient skill or is not sufficiently well set
26   up to carry out a task such as reading smears and should therefore refuse to
27   do so”. Did you take that to encompass issues such as awareness of one’s

 1   own shortcomings ..... yes, I did. In laboratories where there are several
 2   pathologists working together, colleague may point out when one is
 3   developing an area of weakness or where one might need do some additional
 4   training, that’s obviously not possible when you are a single practitioner, but
 5   I believe that if you chose to be a single practitioner you have extra
 6   responsibilities to ensure that you are meeting standards.

 7   We are all aware of your evidence about the Sydney re-reading exercise and
 8   the value that that may/may not have in assisting the committee to determine
 9   term of reference one ..... yes.

10   And in putting what I'm about to put to you, I'm not in any way discounting
11   the qualifications and reservations that you have in that regards, but what I
12   put to you about that is that bearing those reservations in mind, I
13   nevertheless suggest to you that if when considering the Sydney re-read
14   results one puts into the mix the nature of Dr Bottrill’s qualifications and the
15   circumstances in which he practised as we have been discussing now for an
16   hour or so – that’s you and I – then, your reservations about his mode of
17   practice have a bearing as to the way in which one would or one can view
18   the Sydney results. In other words, if one was disposed to view the Sydney
19   results as saying something about Dr Bottrill’s smear-reading competence,
20   and assuming that the results said something indicating incompetence, you
21   would be aided in coming to that conclusion by your views about his mode
22   of practice, do you agree with that ..... I believe that the circumstances in
23   which he was practising must be taken into consideration when looking at
24   the attempts to derive evidence as to what his false negative and false
25   positive and positive predictive value were.        Was that what you were
26   asking?

27   Yes, it was. In your brief of evidence you've dealt in some detail with the
28   concept of false negative results ..... yes.

 1   And as I read it at paragraph 117 you deal with the, as I understand it,
 2   accepted position that generally speaking, and I accept this can only be
 3   general, it’s come to be recognised that smears that are known to be false
 4   negatives have particular characteristics ..... that is correct.

 5   And you've referred to the Mitchell and Medley article ..... yes.

 6   And as I understand the thrust of that, and this is from the perspective of a
 7   lay person, but it is to the effect that false negative smears frequently arise
 8   because the number of abnormal cells are few – i.e. the number of cells on
 9   the smear being read, and also because they are few they can be shall we say
10   geographically separate on the smear – in other words there’s not a lot of
11   them and they are spread over a wide area of the smear surface ..... it is one
12   of the features. These of course are attempts to look at false negative smears
13   in laboratories with good practice.        so that certain patterns of difficult
14   diagnostic cases appear in the group of smears which in a retrospective
15   screening are found to be false negatives.

16   You start with known cancers, go back to the smears read as normal and
17   look at them and the studies have attempted to find out why they were read
18   as normal ..... yes, except it was CIN 3 was the starting point in Mitchell and
19   Medley

20   And it was found that the false negative smears generally had a small
21   number of abnormal cells ..... correct.

22   I am sure you are familiar with another study, somewhat more recent than
23   the Mitchell and Medley one, done by Baker O’Sullivan, Hanly and
24   Coleman ..... I have read it in the past, I would need my memory refreshed.

25   First of all, as I understand it, this was a study designed to look at the same
26   issues as Mitchell and Medley but here the starting point started with cases
27   where there was invasive carcinoma ..... CIN III I think..

 1   As the headnote ..... it was a combination of CIN III and cancer

 2   Just using the summary on the first page left hand column, the purpose was
 3   to accurately determine whether there were any features of an abnormal
 4   cervical smear that predispose to a false negative report ..... yes

 5   And the results “false negative smears found to be quantitatively different
 6   from true positive smears, they contained significantly fewer abnormal cells”
 7   and then the article gives the median numbers and goes on “these were more
 8   likely to be .. on the slide. It was possible to predict .. alone” ..... yes.

 9   And if one likes looking at computer diagrams you can go over the page to
10   p359 and the first comparative diagnose is in the left hand column there
11   relate to the numbers and distribution of abnormal cells on the glass slide.
12   And the top computer diagram was for false negative smear ..... both on
13   p359 I believe are false negative compared to the next page which you will
14   find are the true positives.

15   So I take it that this merely, you would accept this is confirmation of the
16   Mitchell and Medley proposition you referred to in paragraph 117 ..... that is
17   correct, it’s a very beautiful pictorial demonstration of the phenomenon.

18   EM/CA/ 009

19   If you could turn to the Health Funding Authority interim report of 6 March,
20   Mellor 85, my learned friend Mr Hindle has taken you to this in part. You
21   will remember appendix 3 is the report from the Sydney laboratory ..... yes.

22   If you look please at the second page of that report, p16 of that bundle, you
23   will see that there's comment there about the nature of the abnormalities seen
24   by the Sydney laboratory ..... which paragraph in particular?

25   The penultimate paragraph, one up from the bottom of p16 – is your p17
26   headed “the results” ..... yes, it is.

 1   The first paragraph that I want to ask you about is the second one. It was
 2   quite apparent from the first batch that both the numbers of abnormalities
 3   and detected appearances were very different from those that we normally
 4   encounter, all the laboratories .. made this observation , the incident of
 5   abnormalities .. correlation” and then she goes on “the project was
 6   remarkable .. more akin to the change originally described by Papanicolaou”.
 7   The point there Dr is that on the basis of that report alone, and of course
 8   again we are into othe hypothetical to same degree ebcse Dr Farnsworth
 9   hasn’t yet given evidence. We are dealing with a general statement, we don't
10   know what proportion of slides this related to, but quite obviously she is thee
11   saying that generally these were not in the usual category of false negative
12   slides i.e. few abnormal cells scattered over the same. That's what she's
13   saying ..... no she is simply saying that the slides which they identify as
14   abnormal more reflect the true positive picture in the paper that you asked
15   me to speak to than the false negative picture. But she doesn't indicate
16   whether or not these particular slides were true positives or false negatives.
17   So I have no way of knowing what exactly the expectation from this
18   paragraph or what the interpretation is from this paragraph. It does not
19   surprise me that the population of slides derived from Gisborne in 1990 to
20   1995 was very different to the population of slides going through Sydney
21   and other Australian laboratories in the year 2000.

22   Well, let's just – I will ask some hypothetical questions here

23   CHAIR: if you look at the heading the results it seems to suggest that these
24   slides that you're being questioned about were false negatives ..... but I
25   understood that the slides were sent to the Sydney laboratory without the
26   results, simply the information on the request form, the fact there is
27   increased incidence of high grade abnormalities in these slides does not
28   surprise me because the incidence of high grade abnormalities in this area is

 1   different from that of other places in Australia. If one presumes these are
 2   false negatives I don't know where you get the evidence to make that
 3   presumptn.

 4   If they were false negatives, what then does this mean ..... it is another piece
 5   of information that the panel might use to assess whether the level of under-
 6   reporting if proven is unacceptable.

 7   But if these slides were false negatives your

 8   MR HODSON:         the whole basis of the Sydney re-screening was they did
 9   not know and the figures relating to the tab was done in New Zealand .

10   CHAIR:       if we can't identify whether the slides descried in 17 are false
11   negatives the answers will have no weight whatsoever. If subsequently we
12   are able to determine these comments are applicable to slides which are false
13   negatives they may of weight. I want to find out what the position is in a
14   hypothetical case and whether it is used or not will depend on later evidence
15   ..... I was merely reflecting my interptn of a report written before the corltn
16   with the original reports was made. If I am wrong in that assumption my
17   comments and opinions must be changed. It is clear it is possible to make
18   the correlation from other information in the document but I was dealing
19   simply with the comments from the Sydney laboratory before the analysis
20   had been made.

21   Given the comments made at p17 in the second and third paragraphs if these
22   comments do apply to false negatives what does that indicate to you about
23   Dr Bottrill’s ability to recognise an abnormal smear ..... if these descriptions
24   apply to the population of false negatives then I think that reflects very badly
25   on the pathologists ability to identify these slides as abnormal. However, I'm
26   aware from my review of the statistical report which I did last night that a
27   surprisingly high number from y experience of smears were rprtd as showing
28   features suggestive of invasive carcinoma from the statistical report so I am

 1   not surprised the Sydney laboratory was finding the same thing. It would be
 2   necessary to go to the individual slides, slide report and check whether they
 3   were correctly reported or were false negatives.

 4   PROFESSOR DUGGAN:               could you go to p7 of that report. This is to
 5   clarify for the committee. Would you take that to mean that in this report
 6   there is both the original results as provided by Dr Bottrill’s laboratory and
 7   the Sydney laboratory re-read results ..... yes, I would understand that, that’s
 8   clear, this is the interim report from the pathology investigation group but I
 9   had understood that Dr F’s letter was written before the analysis was made –
10   there is no evidence that this was written after the analysis was made
11   available to Dr Farnsworth.

12   The two articles quoted here today with relation to false negative features of
13   false negative smears, would you agree that those reports are from high
14   quality laboratories ..... yes, I did try to make that point

15   With good quality control practices ..... from the false negative smears
16   occurring in laboratories with good practice and good quality control tend to
17   have these characteristics described in these publications

18   Are you aware of any publication on the features of false negative smears
19   from laboratories without any quality control ..... its difficult to find a
20   laboratory without any quality control, we do have the Inverclyde report
21   where there is documented evidence o no quality control but no analysis was
22   made of individual smears features in that study. I can't bring to mind that
23   compared good laboratories with poor laboratories false negatives.

24   Are you saying there is no literature on this particular type of laboratory ..... I
25   can't recall any

26   And that one should be cautious in comparing the features of false negatives
27   from laboratories with high standards of quality control from those that have

 1   none ..... one should be cautious of course there is an expectation on the part
 2   of women that smears are being reported in a laboratory that meets the min
 3   standards.

 4   CHAIR: would you expect the smears from a laboratory that doesn't have
 5   quality control and the smears which are false negatives from those
 6   laboratories, would you expected them to be different in character from the
 7   false negatives from the laboratories that are referred to in the literature .....
 8   there are many confounding factors which would affect my answer to that.
 9   the simple answer is yes it is likely they will look different but they may not.

10   PROFESSOR DUGGAN:             can I draw your attention to p16 of this report.
11   Could you read the last line of the second paragraph entitled “the slides” .....
12   yes.

13   Yesterday there was a lengthy discussion on the disadvantages of a small
14   c/slip and its impacted on the complete assessment of the material on the
15   glass slide. Could you comment on the impact of this particular situation
16   where the c/slip is on the wrong side of the slide ..... this may seem strange
17   to people who do not work in laboratories it is not a very unusual situation in
18   laboratory practice to find that the smear-taker has written the individual’s
19   identification details on the slide and then inadvertently turns it over before
20   spreading the material on the slide. at the stage where the material is spread
21   on the slide it is almost transparent, it doesn't have the colours that you see
22   down the microscope once you have stained it with dyes, so on receipt in the
23   laboratory the expectation is that the material is spread on the same slide as
24   the identification has been written, non-Maori so the slides are retained
25   routinely in racks and c/slipped in the same methodology. But generally
26   once the slide reaches the primary screener who finds that they can't focus
27   on the cells because the cells lie on the wrong side of the glass side, the error
28   is detected and the c/slip applied to the correct slide for the material so it can

 1   be visualised down the microscope.              It is very easy to make that
 2   determination. On the other hand as a pathologist I have on a rare occasion
 3   been handed a glass slide by my cytotechnologist to review to find the c/slip
 4   is on the wrong side. Occasionally when the material is very thin it may not
 5   be as easily detected as in most other cases. Had this sentence said a very
 6   occasional slide had the c/slip on the wrong side I would not have been
 7   surprised, however I do find it surprising that the noted the number of slides
 8   requiring re/cover slipping because the cover was on the wrong side. This
 9   would make me concerned that the primary screener was not as vigilant as I
10   would expect p/screeners to be working

11   If she was a pathologist ..... it makes no difference what the qualifications of
12   the individual are, it’s the practice.

13   CHAIR: the fact there were a number requiring recover slipping does that
14   suggest there was something inadq about the smear-taking ..... we are all
15   human. In the best of laboratories slides arrive in your laboratory with the
16   material on the wrong side. If this happens consistently from a smear-taker
17   we are quick to point it out because it’s a great nuisance in our laboratory
18   practice to have to return the slide and have it recover slipped. So it would
19   be interesting to note whether I don't know the actual number, but it would
20   be of interest to note whether these particular slides were from one period of
21   time or one particular location or sporadic.

22   In the UK do you have smear-takers who are non-medical and lay smear-
23   takers – in New Zealand we have nurses smear-takers and lay persons who
24   are smear-takers ..... in the UK in the region of 80% of our cvc smears are
25   taken by nurses in primary care. We do not have lay smear-takers. Some
26   general practitioners and some gynaecologists take smears as well.

27   MR GRIEVE:         I understand that the smear-taker fixes – applies a fixative
28   to the smear, which is alcohol ..... usually an alcoholic fixative.

 1   So that the slide can then be transmitted to the diagnostic laboratory having
 2   dried or sometimes in a fixative solution is that right ..... usually the slide is
 3   transferred dry once it has been fixed in the solution. In days gone by in
 4   some laboratory areas they remained in the fixative with screw top contains
 5   and sent to the laboratory. By and large they are sent fixed and dry to the
 6   laboratory.

 7   So it is the laboratory’s task to apply the c/slip ..... yes.

 8   And did I understand from what you said yesterday about this that in the
 9   absence of a c/slip the material on the slide cannot be accurately viewed
10   under a microscope ..... yes.

11   So that if in this case there were a number of slides requiring recover
12   slipping that raises the question doesn't it of how that number of slides was
13   reported upon at all by Dr Bottrill ..... it raises the question, yes.

14   So that we have a number where, according to Sydney there was no c/slip
15   enabling proper microscopic visualisation ..... that is correct although it is a
16   slightly different circumstance from the one I was describing, when the
17   microscopist is viewing cellular material on the glass side out with the area
18   covered by the c/slip in that situation there is air between the observer and
19   the cellular material. If you turn the glass slide over there is glass between
20   the observer and the cellular material and you can make some evaluation of
21   the material but it is not an accurate evaluation.

22   If that had been done you certainly wouldn't regard that as good practice .....
23   I would not regard that as good practice.

24   In terms of accuracy and the need to deliver accurate readings of the smears
25   – turn it over so you read through glass ..... one would not deliberately do it
26   that way. I'm simply saying on a rare occasion with very thinly spread

 1   material it may be less easy to detect that the cells are on the wrong side of
 2   the glass than in most circumstances where it is very easy to detect.

 3   CHAIR:       what does it say about a pathologist where in some cases he/she
 4   has ben unable to detect that the cellular material is on the opposite of the
 5   glass so it is being looked at upside down ..... the fact that the cells are lying,
 6   the thickness of the glass side away from the observer becomes immediately
 7   apparent when you view any particular area under a higher madgnifictn
 8   because you cannot bring it into focus so it would suggest to me that in these
 9   particular slides no cells had been viewed under higher magnifctn. It is
10   unusual to screen a whole cervical smear without wishing to view some
11   areas of the slide under higher magnifictn.

12   And if that hasn’t occurred what does that indicate to you about the primary
13   screener ..... it would make me question whether the task on that occasion
14   had been carried out in an appropriate way.

15   MR GRIEVE: just still dealing with c/slips. You made comment yesterday
16   about the reference in the report and its in the same paragraph that you've
17   been discussing about the need to recover slip to cover all the material on the
18   slide ..... yes.

19   In this instance we were dealing with Dr Bottrill as the primary screener it
20   would have been his responsibility before reporting on a slide to ensure that
21   all the material macroscopically visible was dealt with by way of c/slip so
22   that it could be microscopically visible, correct ..... yes, it is the
23   responsibility of the pathologist issuing the report to ensure that that report is
24   accurate. It is also the responsibility of the pathologist to instruct technical
25   staff how to prepare material to be presented to the pathologist.

26   We don't know yet who actually put the c/slips on Dr Bottrill’s slides, that’s
27   why I put it to you on the basis if it was someone else they would ensure it
28   was done properly ..... I agree.

 1   As I understand it large c/slips sufficient to cover the material on the slide
 2   were freely available in 91 or from 91 certainly over the relevant period .....
 3   that is correct.

 4   So again it was his responsibility wasn’t it to ensure that not only the slides
 5   had c/slips but that they were of a sufficient size to cover all the material that
 6   he should have been looking at on a particular slide ..... I believe that is true
 7   but I must say in fairness that in the early 90s there were still laboratories
 8   whose routine practice was to use a c/slip less than the full size of the slide.
 9   I still believe it was the responsibility of all pathologists reporting those
10   slides or responsible for slides where they were reported to ensure that that
11   was adqte practice.

12   So this is an area where you are aware that some of your colleagues in the
13   UK weren’t applying best practice ..... yes.

14   But it is still a situation which you would not condone ..... I would not
15   condone it today, but I would have difficulty saying that a laboratory in 1990
16   to 1995 96 using c/slips of that size was not performing to good practice.

17   You will see from Dr Farnsworth’s report that she there refers to cases of up
18   to 20% of the material not captured by a c/slip. Do you regard that as good
19   practice ..... no I don't regard it as best practice nor do I regard it as good
20   practice, however I do not know what size c/slip was used originally, nor do
21   I know what size of c/slip was used by Sydney to recover the slide. 20% off
22   a large area or 20% of a small area. Obviously not all the slides required to
23   be recover slipped, so in many circumstances Sydney considered the original
24   c/slip size to be appropriate even in the year 2000. it was only in a %, I can't
25   remember – it was 50%, they decided that the c/slip was not of adqte size.
26   It’s a large %, I agree.

27   So you have a large % of slides not adqtly covered and of that 50% inadeqtly
28   covered, some of them missing 20% of cellular material ..... yes.

 1   Doesn't obviously lead to accurate reporting, does it ..... its not best practice,
 2   however as I've said before, it was fairly routine practice in laboratories at
 3   that time.

 4   Which time ..... in the years 90 to 95 when I understand these slides were
 5   made. It does however mean that the practice would result in some slides
 6   being called negative because no abnormal cells were under the c/slip when
 7   abnormal cells were present in fact on the glass slide. That is why I say it is
 8   not good practice.

 9   MR HODSON: the smaller size was the British size 42 by 20, larger 50 by
10   20, the difference is the 20% we have been hearing about.

11   CHAIR:       do you want to qualify any of the answers you've just given .....
12   no, that is what I expected to be the case.

13   MR GRIEVE:         so that means he had the larger c/slips available ..... the
14   larger c/slips have been available for decades 0- you chose the size for the
15   task, c/slips are used in histopathology as well as cytopathology.

16   Are we here dealing with cases where on occasions 40 x 20s were used
17   when 50 x 20s should have been used ..... yes we are. But I do not know
18   what the recommended practice in New Zealand was as far as c/slip size in
19   190 to 95. in the UK we did have a standard which said using a c/slip less
20   than 40 x 20 the size used in the lab in question was unlikely to be adqte.
21   And we had a standard that said the c/slip should cover all the cellular
22   material on the glass slide, if you were using 420 x 20 it must be positioned
23   so it covers all material on the glass slide. There are two issues really, one is
24   the size of the c/slip and the second is the positioning of the c/slip. It may be
25   that the c/slip size was adqte but was wrongly placed to cover all material on
26   the glass slide. It may be it was just inadequate in size.

1   I suggest to you that in either of those case it is the responsibility of the
2   primary to fix it either by putting the larger c/slip on ore the smaller or larger
3   c/slip in the right place ..... it the responsibility of the pathologist to ensure
4   that happens, yes.



8   MONDAY 7 MAY 2000

To top