Dr. Tony Holohan:

I thank the committee for the opportunity to address it. The Minister has covered some of the ground and, therefore, I will try to be as brief as I can. The report is set out in a number of sections and it is not available on the Department's website. I have one or two hard copies but I believe copies have been distributed to members.

The report outlines the background, the methodology we used, describes the Midland Regional Hospital, Portlaoise, maternity service and sets out our analysis under two headings: first, quantitative findings which are set out in section 4; and, second, qualitative findings which are set out under seven themes in section 5. It then sets out some overall conclusions and recommendations. There are 42 recommendations throughout the report regarding the section 4 quantitative findings and the section 5 qualitative findings. They are summarised in a sense by the 11 overall recommendations. It is important that all of the recommendations be taken together. We are all aware of the context.

Moving on to the section on methodology, the key question we set for ourselves to try to address and the one which was on everybody's lips was: was the service safe? While we could not answer that question in an absolute sense, we set out to examine the extent to which there was verifiable implementation of the recommendations from the investigations the hospital had conducted into its own declared incidents over a number of years and other relevant national policies. The further along we could say it was in verifiable implementation of these matters, the more satisfied we could be about safety on a continuing basis. That was the approach we took.

We had meetings with a number of organisations as set out in detail in the report in one of the tables in the section on methodology. We met a number of national organisations, including the Medical Council, HIQA, the State Claims Agency and An Bord Altranais. We met representatives of a number of sections within the HSE. In particular, we met representatives of the Midland Regional Hospital, Portlaoise, on three separate occasions, over eight to nine hours in total. We were facilitated significantly by Mr. David Walsh who is present. If it were not for his involvement, we would not have been able to gain access to the information and the response as quickly as we were. We would not have had the report as quickly.

We requested a considerable amount of data from the hospital to give us the basis for some of what we set out in the report. We received all of it in a timely fashion. We also made an examination of national perinatal mortality reporting systems and took a walk around the Midland Regional Hospital, Portlaoise, which I had never seen previously. I will not spend too long describing the services at the hospital, as I imagine members are quite familiar with it and its services. It is a three-consultant service. It is not a training location for the SPR programme in obstetrics and gynaecology or for midwives. It has an agency dependency level of 25% for both medical and nursing staff and, as I will cover, a high vacancy rate, particularly in leadership positions for midwifery.

The slide on section 4 presents national birth numbers. They show a significant increase in national birth numbers between 2000 and the present, in particular between 2000 and 2007. During that time period there was an increase of approximately 30% in the total number of births. It was particularly acute in a two year period when it increased by 17% nationally. The picture in Portlaoise is quite different, with a much more significant increase in the total number of births. In the seven years to 2007, the hospital experienced an increase of approximately 100% in the total number of births, with 50% of that increase occurring in a two year period between 2005 and 2007. That is an interesting and relevant finding.

The next slide shows some of the analytical work we did on the numbers of perinatal deaths at the Midland Regional Hospital, Portlaoise, as derived from its own direct account, from what the national perinatal reporting system states about it and from what the National Perinatal Epidemiology Centre system states about it. Let me take one minute to explain. There are a number of reporting and data collection systems relating to perinatal events, both mortality and morbidity. We have the national perinatal reporting system which was compiled, managed and reported on by the ESRI until the recent health reform changes resulted in that function moving from the ESRI to the HSE. Another is the relatively recently established National Perinatal Epidemiology Centre under the direction of Professor Richard Greene in Cork which gathers its data directly from the individual hospitals and makes reports on both maternal mortality and morbidity rates. We have the General Register Office under the auspices of the Department of Social Protection and the CSO that makes reports in the form of vital statistics. Between each of these systems we see some inconsistencies in reporting and some systematic errors in reporting such that we believe there is an under-reporting of total perinatal mortality rates. It is not sufficiently significant to change our overall ranking on an international basis in terms of how we fare relative to other countries, but it is significant in so far as we believe the correct and absolute measure, the one set out in blue on the slide, would the highest. That is the one we believe would be the best and the one to which the report recommends moving towards.

We are recommending that each national reporting system reports on essentially the same definition. That will require changes to the existing definitions set out in the Civil Registration Act and we are in discussions with the Department of Social Protection on the matter. It will also require some consolidation of both the NPEC in Cork and the NPRS that has come into the HSE from the ESRI. They are two broadly similar systems, collecting similar data and reporting differently. There is room, not just from the point of view of commonality of reporting but also from the point of view of avoiding duplication and the obvious difficulty that arises when two systems give two slightly different answers to the same question. It also reduces the workload for staff on the ground having to report to different systems. Therefore, we have made specific recommendations in that regard.

On the available data we sought and received from the hospital, the next slide shows the numbers of transfers per 1,000 births. These data given to us from the hospital show both paediatric, that is, neonatal, and maternal transfers to other centres, mostly in the Dublin region in the perinatal period. We can see the number increased significantly over time. The increase is not explained by the number of births because these figures are per 1,000 births. That is a relevant finding.

The next slide shows information we received from the hospital on adverse incident reports. Members may be aware that the State Claims Agency requires that each adverse incident that occurs in the health service be reported to it and it compiles and collates the data. We sought all reports from 2006 reported by the hospital. We have set out a graph showing that, overall, it follows the same trend that we can see regarding transfers. Something unusual and different was happening in 2007, which explains that peak and I am happy to talk about it during members' questions, if they so wish.

Figure 4.6 on the same slide shows the number of claims per 1,000 births for the Midland Regional Hospital, Portlaoise, as compared with the average for six similarly sized units in the country. We can see that over the period of time there was a higher number of claims. Overall, the number of claims is not very large - one would not necessarily expect it to be, but there is a significant difference between the two. There would be a lag time in the reporting because of the nature of claims and when they are submitted. That might explain some of the tail-off shown in the graph.

The other thing which is not set out in the slide on which we have reported in the report in the area of quantitative data is the concept the Minister mentioned of "never events". There is a listing which is contained in our forthcoming health information Bill - it is the subject of a recommendation in the report. We know that six such events took place over the time period, four of which related to deaths or significant perinatal injuries that resulted ultimately in the deaths of babies who were essentially normally formed in low-risk pregnancies. There were two others that related to retention of foreign bodies following surgical procedures in that unit over the time period. This is set out in the report. The nature of what is a "never event" and the discipline of patient safety internationally suggests each of these events needs to be treated as though it should not have happened. It is the opportunity to stop and re-examine all procedures to ensure it never happens again.

The qualitative findings are set out in some detail in the report and we have organised them under seven themes, the first of which we deal with is that of patient-centredness. We dealt first with the culture at the hospital. We talked to the families at some length. We met them on a number of occasions through the conduct of this work, as well as at the beginning, with the Minister. Their powerful testimony of their treatment - the things said to them, the behaviour of staff, the interaction of staff with them and each other - and some of the things we heard from staff give us considerable cause for concern about the nature of the culture in the unit.

Members will see much in the literature about the importance of culture in health services and its ability to shape the safety of the service people are receiving. There are ways in which culture can be measured and there are tools that can be used to assess it.

It is important for services to be aware of the culture from a patient safety point of view as it has a big impact on the quality and safety of the service people receive. We have made a specific recommendation, as the Minister stated, on the need for HIQA to conduct an assessment of the culture in the remainder of the hospital. Some of the services dealing with perinatal deaths, in other words, dealing with the family after a death had occurred and dealing with the infant who had died, fell very far short of a reasonable standard of care. This was down to issues as important as the ability to hold or interact with infants after they had died to dress and wash them. This added significantly to the grief of the families and their inability to reach a point of closure on what had happened. Some of this was among the most harrowing evidence we heard about the service.

With regard to the response to patients and families following adverse events, without going on at length about it, we see this as an opportunity for senior clinicians and nursing staff to step up and take additional responsibility in circumstances where something unforeseeable has gone wrong. In these situations families were left significantly on their own in the hands of risk managers not involved in clinical care for the purpose of conducting investigations. Families had questions which were not addressed or answered. At the very moment when senior staff were needed most they were absent in some of the incidents.

With regard to open disclosure, we know from the original programme aired that a number of families were not, even until the programme aired, told about significant adverse events they or their babies had experienced. This is not consistent with the HSE's open disclosure policy, to say the very least. We have examined the HSE's open disclosure policy through the lens of what we have seen and it is a very good policy which stands up. The issue is to ensure it is appropriately and consistently implemented.

Clinical governance describes the arrangements in place for risk management, identifying adverse incidents and appropriately investigating them. Suffice it to say, the extent to which good risk management and adverse incident investigation practices were applied was highly variable. The quality of incident reports completed was highly variable. In some cases, the investigation of incidents did not start until a significant period of time after the incident had taken place, which creates a challenge in making findings on the basis that staff would not clearly remember what had happened. We make recommendations on the need to develop national standards and for HIQA to have national standards for the conduct of investigations. Desktop reviews or more significant investigations differ in their purpose and the period of time within which they must be completed. These will be the subject of the standards we recommend.

The systems in place to provide proper verifiable assurance on the implementation of the recommendations contained in the reports were weak. We believe this explains why many of the same recommendations came up in the hospital's reports without any audit-type information to examine the extent to which they had been implemented. Issues and themes which are set out in the report came up as explaining recurring events and this suggests a weakness in the arrangements and procedures.

Clinical effectiveness relates to clinical guidelines, clinical standards and audit arrangements. Following the HIQA report on the death of Savita Halappanavar, work has started on clinical practice guidance on clinical handovers and the escalation of care to complement much of what has already happened with regard to early warning scores for maternity services and adult and paediatric services. We have made an additional recommendation arising from the recurrence of concerns about the use of oxytocin in the hospital's investigations. We have not second-guessed this finding. We need a national standard on the induction of labour which may include a protocol on the use of oxytocin.

With regard to the escalation of incidents and the role of the HSE, we reflected on the system put in place following what had happened in Portlaoise in 2007. A significant number of missed breast cancer diagnoses had been uncovered as a result of an investigation at the time. The system enabled reporting to take place from local to national level, but this requires the local level to realise there was a difficulty. Other than one individual case, we have not been able to find any other escalation of incidents. The hospital did not know it had a difficulty. This is a passive system of reporting. Beyond this, it appears no other system has been in place since 2007 to detect problems or monitor the extent to which there are ongoing difficulties in the unit.

A number of specific recommendations are made on leadership, staffing and workforce planning. The importance of leadership is recognised from a clinical and midwifery point of view in the delivery of a safe and sustainable service. There are a number of leadership positions in the midwifery unit, the most senior of which is the CNM3 post, a clinical nurse management post, which was approved in 2007 and has never been filled. A CNM2 post has also been vacant for a period of time due to long-standing sick leave and of seven shift leader positions, clinical supervisory positions in midwifery, only one is filled on a permanent basis. This leads us to conclude the officer corps of the midwifery service was largely missing during the time period. This is relevant in the context of some of the hospital's own reports which identified difficulties with clinical supervision during the time period.

We examined a number of issues with regard to infrastructure and equipment. We examined the capital facility, in which we are not experts by any stretch, and identified a challenge posed by the fact that the operating theatre and labour ward are on different floors. This is particularly the case in situations where significant foetal distress requires an emergency Caesarean section. It requires the unit to have arrangements in place to practise evacuation procedures to ensure patients are brought to the operating theatre in a short period of time. It requires teams to put in place arrangements to practise these evacuations. This evacuation is not subject to practise but should be, given the infrastructural deficit.

Another dimension is hospital records. We have made recommendations to ensure implementation of the HSE's national chart and maternity record arrangements.

We heard much about medical devices from families, in particular about deficiencies as they saw them in the way in which equipment was used or the response to alarms on individual machines. We do not doubt any of what we have heard. We do not pass judgment and are not in a position to validate any of it, but we have made a recommendation on national medical devices standards and for assurance to be provided by the HSE for the Department on this issue.

With regard to legal and ethical issues, the coroner process formed a significant element of what families stated to us. The adversarial nature of the process was a big challenge for them. They found that it added to their grief and commented on the extent to which there was significant representation on the service side relative to their own. From our point of view, without any criticism of any individual coroner, the coroner system is operated through local authorities. Reports are collated and collected in local authority offices for the most part. Perhaps the extent to which intelligence is derived from each of these by analysing them at national level to try to identify patterns is more limited than it should be. We have started a discussion with the Department of Justice and Equality on this issue. It is already working on updating a draft 2007 Bill to update the Coroners Act 1962, to which an amendment was made in 2005.

Two other issues we dealt with are consent and confidentiality. Without going on at length about them, some of the dimensions of informed consent, a continuing challenge for health services generally, were deficient.

Without breaking confidences, one of the families we dealt with was under the impression that, as its members entered the operating theatre for an emergency caesarean section, they were signing a waiver, not a consent form. Until we were in a position to correct them, they continued in that belief for a number of years.

The other issue is that of confidentiality, in respect of which we have some findings or recommendations.

The Minister has covered a number of the conclusions, but I will read them as they are set out in the report's summary, section 6. We set out six overall conclusions that relate to the families first, then the service and finally the oversight. Two conclusions relate to each.

Families and patients were treated in a poor and, at times, appalling manner with limited respect, courtesy and consideration. Information that should have been given to families was withheld for no justifiable reason.

Regarding the service, poor outcomes that could likely have been prevented were identified and known by the hospital but not adequately and satisfactorily acted upon. We concluded that the service could not be regarded as safe and sustainable within its current governance arrangements, as it lacked many of the important criteria that were required to deliver, on a stand-alone basis, a safe and sustainable maternity service.

In terms of oversight, many organisations, including the service itself, had partial information regarding its safety that could have led to earlier intervention had that information been brought together. The external support and oversight from the HSE should have been stronger and more proactive, given the issues identified in 2007.

We have set out 11 overall recommendations, which are described over the course of three slides. Since the Minister has covered a number of them, perhaps it would be best if, instead of my reading them all out, members peruse them. I am happy to take questions on them and to explain the background rationale, if doing so would be of interest.

I do not propose to go into detail on the legislative and policy implications, but this report and its findings will have some implications for our ongoing work in both regards. Some examples are set out in the final two slides. In the interests of time, I will conclude now.