Agenda item

Dr. Ankush Mittal

Consultant in Public Health, City of Wolverhampton Council

 

Dr. Jonathan Odum

Medical Director, The Royal Wolverhampton Trust

 

 

[Report is marked to follow]

The Chair welcomed members of the Staffordshire County Council, Healthy Staffordshire Scrutiny Committee to the meeting.  She stated that she understood there had been a Summit held in Wolverhampton recently with the various health agencies on the 10 October 2018 concerning the subject of Mortality statistics. 

 

The Consultant in Public Health presented a report titled, “Learning from Deaths in Wolverhampton and Steps Forward.”  He stated that there had been some constructive discussion on the subject of mortality with health partners.  The Council had recently hosted a summit, led by Public Health, to discuss the subject of Hospital and City wide mortality data with representation from the Clinical Commissioning Group and the Royal Wolverhampton Health Trust.  The purpose of the report was to provide a summary of the data relating to deaths in Wolverhampton at both a City wide level and Hospital level.  The report also assessed the implications of the data and made recommendations on the best approach moving forward.   

 

The Consultant in Public Health stated that hospital mortality statistics worked on a ratio.  The ratio was, observed deaths divided by expected deaths.  The most difficult part of the equation was the calculation of expected deaths.   There were a few different statistics used to measure expected deaths to create the ratio, which all relied on some logical processes.  They looked at what proportion of people admitted nationally die and then applied that proportion to the local numbers of people admitted in any given hospital, to predict the expected mortality rate in the hospital.  If the ratio figure was close to one, it showed that the hospital was on the same level as the national average mortality rate.  If the ratio figure was above or below one, then the hospital was having higher or lower deaths than the national hospital average.  He added that there were a variety of different adjustments made, depending on the measure, used on various elements of the admitted population.  As examples, he cited the age mix of the patients admitted to the hospital, their associated medical conditions and their mode of admission.   

 

The Consultant in Public Health stated that the national average was for men to die around four years earlier than women.  There were a variety of reasons for the difference with one of the main reasons being that men were at a higher risk of suffering from a cardiovascular disease. 

 

The Consultant in Public Health commented that when reviewing hospital mortality data it was important to understand that some hospitals coded conditions differently.  He cited the example of a cough being classed as a chest infection or pneumonia.  The expected number of deaths could be significantly different depending on how the condition had been coded, with pneumonia carrying a higher risk of death than a standard chest infection.  Due to the differences in coding practices, the process of evaluating the data

 

The Consultant in Public Health stated that local care pathways could have an impact on hospital mortality data.  There were therefore many factors other than care quality which could affect the mortality rate of a hospital.  There were two statistics which were commonly referred to in the literature concerning mortality rates.  SHMI stood for Summary Hospital Level Mortality Indicator and HSMR stood for Hospital Standard Mortality Ratio.   The SHMI measure had replaced HSMR as a national statistic in England and had been used quarterly since 2011.  There were however variants of HSMR and a further indicator called RAMI being used in Scotland and Wales.  Internationally there was no agreement on the use of any statistic. 

 

The Consultant in Public Health outlined the detail of the SHMI statistic, which was freely available on NHS Digital.  The SHMI was the ratio between the actual number of patients who died following hospitalisation at the trust and the number that would be expected to have died on the basis of average England figures, given the characteristics of the patients treated there.  The SHMI also looked at the method of admission.  People that were admitted as an emergency were at a higher risk of death than if they were admitted as part of a planned care episode.   The SHMI also looked at associated conditions such as diabetes and cancer.  There was significant variation as to how these associated conditions were recorded across hospitals throughout the country.  It was important to approach SHMI data with a critical mind and to consider all the possible explanations for a lower or higher than expected ratio. 

 

The Consultant in Public Health gave an explanation of City Wide Mortality statistics.  Data on deaths across the city was based on information recorded on death certificates, which was eventually reported by the Office for National Statistics within ONS Mortality statistics.  The most important piece of information to be gained from the death certificate was the actual cause of death.  He considered ONS Mortality data overall to be a more reliable and established dataset around deaths compared to hospital mortality data.  The data had been used by Public Health teams for many years to identify the causes of early death in populations.  There was some limitation to the data, which included the data being affected by population migration and the presence or absence of local end of life care facilities.  People who lived in hospices had a higher chance of death in any one year than someone of the same age who lived at home.  If there was a limited amount of hospices in the local area, then people could move out into a hospice in another locality, which would inflate the mortality figures for that area.  

 

The Consultant in Public Health referred to the graphs in section 5 of the report which summarised the key themes around death in Wolverhampton from a City perspective and were based on ONS Mortality statistics.   Death rates in Wolverhampton spread out over the last twenty years had generally been better than the comparative group.  The comparative group was made up of Local Authority areas which had a similar deprivation rate.   It was the associated aspects of deprivation which leaded to an earlier death, such as obesity.  Over the last 5-6 years there had been an increase in the standardised mortality rate for persons aged under 75 within the Wolverhampton area.  There had been some increase in deaths from circulatory diseases.  Circulatory diseases, cancers and respiratory diseases accounted for the top 3 causes of death in Wolverhampton and shared many of the same key risk factors such as smoking and obesity.  When compared to the West Midlands and the rest of England, Wolverhampton remained significantly high for overall death rates.  The challenge was therefore how to address the vast inequality that existed in the country and more locally.  Between the richest and poorest people there was an eight year gap in life expectancy, which applied both nationally and locally.  Wolverhampton continued to show a high rate of deaths related to alcohol, which had been a persistent trend for the City for several years. 

 

The Consultant in Public Health commented that there was a high percentage of people in Wolverhampton, compared to other areas, that died in hospitals.  He believed this statistic pointed to the structures and processes in Wolverhampton.  Less people in Wolverhampton were dying in care homes compared to the national average in England.        

 

A Member of the Panel asked if there was any particular group within the Wolverhampton area which was suffering more from alcohol related deaths.  The Consultant in Public Health responded that deaths related to alcohol did vary through a number of risk factors including ethnicity.  There were mixed communities effected in Wolverhampton, with Public Health not focusing on anyone particular ethnic group.  There was a particular age group which seemed to have a higher risk of alcohol related death, which was from 40-60.

 

A Member of the Staffordshire County Council, Healthy Staffordshire Select Committee referred to the high number of people from the Indian Sub-continent living in Wolverhampton, who were genetically more predisposed to vascular disease.  He asked what was being done within the primary care system to try and address the issue.  The Consultant in Public Health responded that whilst there was an elevated risk, there was still the same themes of deprivation and inequality.  Over the last 20-30 years the whole country’s profile around weight and physical activity had seen a huge shift in the representation of circulatory diseases.  There was clearly some more work to be done regarding NHS Health Checks.  The uptake in Wolverhampton was relatively low but had improved dramatically in the last couple of quarters.  Early intervention was key to preventing an early death.   The Clinical Accountable Officer for the CCG commented that they were working with GPs on more preventative work.  GPs were being asked to spend a longer time with patients as part of a programme, to discuss topics such as smoking cessation.

 

A Member of the Staffordshire County Council, Healthy Staffordshire Select Committee made reference to the lower than expected mortality rate in Wolverhampton care homes compared to the national average.  She asked if a person was discharged from hospital into a care home and died within 30 days of the hospital admission, whether that death would be classed as part of the hospital mortality figures when using the SHMI indicator.   The Consultant in Public Health confirmed that the death would be classed as part of the hospital’s mortality overall figures.  The Medical Director commented that it was clear there were people entering hospital to spend their last days, who would have had a better end of life experience within the care home they had been staying.   This fact undoubtedly effected the Trust’s mortality statistics.  It was important to record people’s preferred choices for their end of life care.  Nationally this was something which the country was poor at doing.  The Chief Executive of the RWHT commented that they wanted to give the right tools to the care homes for them to have the difficult conversations with residents and relatives about end of life care.  He was considering using the Trust’s transplant nurses, who were used to difficult conversations, to communicate with the care homes about end of life care. 

 

A Member of the Staffordshire County Council, Healthy Staffordshire Select Committee asked why the alcohol related mortality rate had risen during the last 5-6 years, when it had seen a positive decline before this timeframe.  The Consultant in Public Health responded that there was a drug and alcohol service which was commissioned by Public Health.  Due to the numbers being relatively small, there was potentially an element of unfortunate chance in the figures, which could explain some of the variances.  He however accepted that the variance might not all be down to short-term unfortunate chance.  He was happy to reflect and review the data to try and understand the reasoning for the upward trend more fully.  The current drug and alcohol service was moving towards a population screening approach.  The best way to prevent early death through alcohol and drug abuse was early intervention. Some people that died from drug and alcohol abuse had been exposed to them for over 20-30 years.  It was therefore correct to say that groups of people could die at any one time, not because of something that had necessarily happened in their lives recently, but down to potentially sustained abuse over a long timeframe.   He added that alcohol and drug abuse was an area which could be scrutinised in its own right. 

 

The Director for Public Health commented that it was important to think about the context of drug and alcohol abuse within a system based approach.  It was important to do the basics well.  The Wolverhampton alcohol and drugs service was now in the top performing quartile in the country.   Preventative work was also an important aspect of the work related to reducing drug and alcohol abuse.  A huge amount of positive work was ongoing on increasing the amount of health checks within the City.  Wolverhampton had been in the bottom 8% of the country for health checks.  Working in partnership with the CCG and the Trust, in the first two quarters the number had dramatically risen, with 1942 people receiving a health check.  The Council was also using more leverage around how licenses to premises were granted for alcohol. 

 

A Member of the Panel commented that the cause of death on the medical certificate may not reveal an underlying reason such as alcohol abuse, which meant the data on deaths related to alcohol abuse could be underrepresented.

 

The Portfolio Holder remarked that a key ambition for Public Health was to reduce health inequalities and improving the quality of life people had in later life.  This included ensuring people had a preferred choice of their place of death. 

 

The Medical Director of the RWHT commented that the Summit held recently had been very helpful in bringing the various themes together surrounding mortality.  He said that mortality statistics whether they be higher or lower than average were not an accurate reflection of quality of care.  Higher than average mortality statistics were however treated as a smoke alarm by the Trust.  They had not found any significant issues with the care that had been provided by the acute Trust.  He in fact believed that care provision had improved steadily over the years within the Trust.  He cited the seven-day service model as an example of the improvement that had been made.  This meant Consultants presence on weekdays and at weekends was normal.  There had been a considerable expansion of Consultant numbers over the course of the last few years to facilitate the seven-day service provision.  The reviews of the deaths completed had generally shown good quality care, with some omissions, which were appropriately managed.  The CQC had also recently rated the Trust as Good, who were fully informed of the mortality statistics. 

 

The Medical Director of the RWHT said there had been a progressive rise in the SHMI statistic since the opening of the new Emergency Department.  He had not seen any change in care provision over those years, which would account for the rise.  There had been the introduction of a new assessment model within the Emergency Department, which meant acute physicians were now working alongside the emergency physicians.  As a consequence, there had been a significant impact on the statistics in not admitting patients with lesser acuity of medical conditions.  These patients had been discharged with support at home, the community or into hot clinics.  At the same time in England, admissions for the same categories had increased, leading to a lowering of nationally expected death rates.  While SHMI adjusted for primary diagnosis on admission, it did not factor in the severity of the admitting diagnosis.  The number of patients admitted to hospital in the Trust had progressively fallen over the last few years.  This was significantly related to the new assessment process and effected the SHMI calculation.       

 

The Medical Director of the RWHT stated that despite a reducing number of expected deaths at the Trust, local data suggested the population being admitted to the hospital was now increasingly frail and increasingly ill, suggesting that the expected death rate calculation was not adequately adjusting for patient profiles.  Over the last three years there had therefore been a population of patients who were of a much higher acuity than had been previously admitted, which was associated with the rise in the SHMI. 

 

The Medical Director of the RWHT stated they had completed significant work over the last 2-3 years to assess the care that was being provided by the Trust.  There was an established process to investigate all deaths across the organisation.  The Trust’s own investigations and independent investigations had not found any systematic failings in care within the organisations that would account for the deaths. Whilst they did find some omissions in care, which they reviewed through the mortality review processes, those deaths on the whole were occurring in people who were very frail and elderly.  From assessing their admission profiles, the likelihood of their survival had been low.

 

The Medical Director of the RWHT remarked that there was a large amount of people in Wolverhampton who were admitted to hospital with essentially end of life requirements.  He believed this was an area which needed improvement, with more people being able to die at home or within a care home.  The Trust also needed to ensure that the coding of patients across the organisation was fully accurate.  There had been a contraction in the Specialist Palliative Care Team, which had caused the percentage of patients who had contact with the team to fall.  The rest of the Country had increased their Palliative Care Teams.  The decrease in the members of the Specialist Palliative Care Team had impacted on the overall statistics.       

 

The Medical Director remarked that nationally about 3-5% of patients had omissions in their care when resident at an organisation.  Whether the omission had a direct impact on their death was another matter.  The Trust was in line broadly with the national picture in terms of omission in care.  Omissions in care included a failing in communications between teams, which was a widespread national issue.  The second was regarding recognition of a deteriorating patient, which was also nationally being recognised as an area for the NHS to improve.  Across the Country and within Wolverhampton there were cases of sepsis which were managed late, which could have a detrimental impact on the outcome.  The final area was with reference to documentation and clinical records.  The Trust had been focusing on improving their documentation standards.

 

The Medical Director for the RWHT remarked that the Royal College of Physicians had been leading on the development of the National Mortality Case Record Review Programme.  The method rolled out for reviewing deceased patients care was called the Structured Judgement Review (SJR).  The main aim of undertaking the SJR process was for clinicians to learn from aspects of care that could have been improved even when death was inevitable and in addition to identify areas of good practice.  Patients who had died were given a care rating.  Those that were considered to have received poor care went through a much more detailed analysis process.  Leaning from deaths was a critical component of hospital business.  The Trust was implementing a new Medical Examiner Role, which would play a key role in the undertaking of mortality reviews in the future.  He believed this role would add even more analysis to the current assessment process of reviewing deaths.

 

The Medical Director referred to the governance processes in learning from deaths.  The Trust had a multidisciplinary Mortality Review Group which oversaw case note reviews of deceased patients with senior clinical representation from all specialties and was attended by colleagues from the CCG and Public Health for further oversight. In addition, an Executive Governance Group met regularly to oversee the work and provide assurance, particularly in relation to quality of care.

 

A Member of the Panel asked about independent external support to analyse the mortality date and provide assurance.  In response the Chief Executive of the RWHT stated that he was happy for the external experts already at the hospital analysing the mortality date to attend a future meeting of the Scrutiny Panel to give assurance. 

 

A Member of the Staffordshire County Council, Healthy Staffordshire Select Committee recommended to the Panel, the book, “Being Mortal” by Atul Gawande.  The book gave the view that in the western world death had been medicalised to the detriment of dying. 

 

A Member of the Staffordshire County Council, Healthy Staffordshire Select Committee asked why the Palliative Care Team had been reduced at the Trust.  The Medical Director responded that they were looking to expand the team and were currently undergoing a recruitment process.  More generally there were 150 nursing vacancies at the Trust. 

 

A Member of the Panel asked what would happen to a person who faced a certainty of death in the near future but did not wish to leave the hospital for a hospice or care facility.  The Medical Director confirmed that they would respect the wishes of the patient.   

 

A Member of the Panel commented that she was aware of cases where an elderly couple living at home would both see a decline in their health when one of them became ill, due to the extra strain the care needs put on the other.  The Medical Director confirmed that this was a common occurrence and supportive care in the community was clearly something which would help in such an instance.  The Chief Accountable of the CCG stated that she absolutely recognised it as a problem within the City and increasing partnership working was taking place to address the issue.

 

A Member of the Panel asked if the Trust talked to relatives of the deceased during a full SJR Review.  The Medical Director responded that it was not routinely done but if significant issues came out of the review, then it would probably go through a root cause analysis which could involve the family.  The new Medical Examiner role would introduce changes, with them meeting the family after the death to talk to them about any concerns or issues.  If there were concerns it would trigger a review process. 

 

The Director for Public Health commented that after recent negative headlines in the Express and Star about mortality statistics at the Trust, it was important to establish public confidence in the work of the Trust and relay the complexity of the statistics. 

 

The Chief Executive of the RWHT stated that “Dignity in Death” certificates could be issued to residential care homes and more training given to staff. 

 

 

A Member of the Staffordshire County Council, Healthy Staffordshire Select Committee asked for statistics and targets on urgent referral and screening for cancer to be brought before Scrutiny in the future.  The Chief Executive of the RWHT stated that he was far from happy with the performance of the Trust in relation to cancer.  There were major issues with late referrals from other organisations and there was a problem with the target itself, which he thought was wrong.  There was not a single Trust in the country that was meeting the national target.  He thought reform was needed on a national scale, which included a national strategy being drawn up for robotic surgery. 

 

The Chief Executive of the Trust stated that he had been informed that the Telford Accident and Emergency Department was scheduled to close overnight from the 5 December 2018.  The biggest impact would be on Wolverhampton.  He had been informed that 30-80 ambulances a night would be re-directed to Wolverhampton.  He had a concern that the Trust would be dealing with Paediatrics in Wales, where the Trust had no relationship with Social Services.  The relationship would need to be developed.  The overnight closure would also affect transfers across the Black Country, general waiting times and the winter contingency plans.

 

 

Resolved: That the Health Scrutiny Panel:-

 

A)    Notes the uses and limitations of mortality statistics at both a city wide and hospital-level perspective and acknowledge the findings from the learning from deaths work at The Royal Wolverhampton NHS Trust.

 

B)    Supports the shift in focus from ‘hospital death rates’ to ‘healthy life expectancy’ across the city, and the significant opportunity to prevent early deaths using a community Public Health model.

 

C)   Supports the wider ambitions around optimising End of Life care in the city.  This includes people being treated according to their wishes, with the correct documentation in place and their preferred designation of death clearly known. 

 

 

 

      Meeting closed at 2:30pm.