Medical practice

2010 good hospital guide

"Nineteen hospital trusts are today exposed as having alarmingly high death rates in a major report that also reveals how hundreds of people are dying needlessly because of substandard NHS care", reported The Observer.

The newspaper story is based on an annual hospital guide, published today by Dr Foster Intelligence, an independently run health information firm that is part owned by the Department of Health.

The guide is an annual publication that seeks to measure hospital performance in England with a growing series of comparative indicators. These indicators are constructed in partnership with the Imperial College London. They are based on NHS hospital data and are free of political bias.

The Health Secretary, Dr Andrew Lansley, welcomed the report, writing a commentary in The Observer detailing the steps that are being taken to increase transparency and improve hospital safety and performance:

“We need a cultural shift in the NHS; from a culture responsive mainly to orders from the top down to one responsive to patients, in which patient safety is put first.”

What are the main areas of hospital performance covered by the hospital guide?

The guide is broken down into three main parts:

  • an analysis of hospital performance based on different measures of mortality
  • a detailed look at hospital performance in three major areas of care: stroke, orthopaedics and urology
  • an analysis of patient safety in terms of adherence to best practice, infection control and adverse incidents including accidents

There are also new analyses of efficiency, measured by hospital readmission rates and patient experience, measured by patient survey.

What are the main conclusions of the guide?

The guide contains both positive and negative findings. Under the heading ‘Good news’ it reports that:

  • Deaths in hospitals continue to fall, dropping 7% between 2008/09 and 2009/10 in crude terms.
  • The gap between the hospitals with the highest and lowest Hospital Standard Mortality Ratios (HSMRs) has narrowed, with eight fewer hospital trusts’ HSMRs above the expected range.
  • Safety standards have improved, with higher rates of compliance with safety alerts and better reporting of errors.
  • Four hospital trusts – Airedale, Royal Free Hampstead, Ipswich Hospital and East Kent Hospitals – are given accolades for exceptional performance.

     Under ‘Areas of concern’, the guide reports that:

  • Variations in mortality ratios persist, with 19 hospital trusts having high HSMRs.

  • Four trusts have high ratios for the ‘deaths after surgery’ indicator. This means patients who, following surgery, developed another problem such as internal bleeding, and subsequently died. Two of these trusts – University Hospitals Birmingham Foundation Trust and Hull and East Yorkshire Hospitals Trust – also have high HSMRs.
  • Rates of emergency readmissions vary widely between hospital trusts, as do revisions and manipulations following common orthopaedic operations, where three trusts – Frimley Park Hospital NHS Trust, Northumbria Healthcare NHS Foundation Trust and Guy’s and St Thomas’ NHS Foundation Trust – have high rates.
  • Over 27,000 potential medical mistakes (or adverse events) were recorded in hospital data. Dr Foster says that this is almost certainly an undercount due to inconsistent recording.
  • Standards in the treatment of life-threatening conditions such as stroke and broken hips vary widely, with many trusts falling short of best practice.

The hospital guide also makes a call for more data to be made available, noting that there is information ‘we cannot tell you but would like to know’. It highlights a need for better recording of patients developing life-threatening blood clots following treatment, more information about community and primary care services, and better measurement of outcomes in private hospitals providing NHS care, among other things.

What does the guide say about hospital mortality?

It says that Hospital Standard Mortality Rates (HSMRs) are decreasing (i.e. improving) across the NHS in England. Only 19 of the 147 hospital trusts now have ‘significantly high’ HMSRs, compared to 27 in last year’s guide, and 26 trusts have HSMRs that are ‘significantly low’, down from 32 a year ago.

‘The overall improvement suggests greater consistency across trusts, both in terms of data recording and perhaps in the quality of care,’ it says.

The use of HSMRs to measure hospital performance has proved controversial in recent years, with experts pointing out that the measure is ‘imperfect’ and warning that it should not be used to construct simplistic league tables of best and worst hospitals.

Nevertheless, they should not be ignored. Earlier this year, the Department of Health stated: ‘A high HSMR is a trigger to ask hard questions. Good hospitals monitor their HSMRs actively and seek to understand where performance may be falling short, and action should not stop until the clinical leaders and the board at the hospital are satisfied that the issues have been effectively dealt with.’

In a bid to construct a better picture of hospital mortality, the hospital guide published a second mortality indicator measuring deaths after surgery for the first time this year. This looked at surgical patients who had a secondary diagnosis such as internal bleeding, pneumonia or a blood clot, and subsequently died.

The guide reports that there is wide variation between hospitals on this new measure. It says four hospital trusts - Hull and East Yorkshire Hospitals NHS Trust; The Newcastle upon Tyne Hospitals NHS Foundation Trust; University Hospitals Birmingham NHS Foundation Trust; and University Hospital of North Staffordshire NHS Trust - have ‘significantly high’ ratios. Two trusts – Chelsea and Westminster Hospital NHS Foundation Trust and Winchester and Eastleigh Healthcare NHS Trust – have ‘significantly low’ death after surgery results.

‘This measure uses a very different approach from the HSMR, so trusts that have high ratios on both measures – University Hospitals Birmingham NHS Foundation Trust and Hull and East Yorkshire Hospitals Trust - will want to understand the possible causes,’ says Dr Foster.

The Observer reported that sources from the Care Quality Commission have said that they have no concerns about University Hospitals Birmingham on either account or North Staffs on deaths after surgery.

The chief executive of Hull and East Yorkshire, Phil Morley, said, "We are confident that we are providing safe care of a high quality to our patients".

The hospital guide also notes that the focus on mortality ratios in recent years has caused some trusts to revisit how they ‘code’ or report patient deaths. This has resulted in some trusts increasing the number of deaths they identify as occurring in ‘palliative care’. This in turn improves the trusts’ mortality ratio because death is the expected outcome.

In the interests of transparency, Dr Foster now publishes the percentage of deaths coded as palliative care for each hospital trust. These range from less than 1% in some trusts to over 40% in others. Basingstoke and North Hampshire NHS Foundation Trust reports 45.5% of deaths as palliative care and Medway NHS Foundation Trust reports 44.5% of deaths in this way.

Two other trusts – Pennine Acute Hospitals Trust and Royal Bolton Hospital NHS Foundation Trust – are noted in the guide as having been in the ‘higher than expected’ HSMR category for the past six years.

What does the guide say about hospital safety?

Overall, safety standards have improved. Dr Foster says a key way of improving safety is to accurately measure and monitor the way in which it is being addressed. Hospitals were rated on a range of aspects of patient safety in 2009. A comparison with this year’s results shows:

  • Higher rates of compliance with safety alerts compared to 2009. However, three trusts still fall short on the immediacy of their compliance, including Southend University Hospital NHS Foundation Trust, St George’s Healthcare NHS Trust, and the Western Sussex Hospitals NHS Trust.
  • An increase in the number of hospitals routinely screening for and treating patients admitted to hospital for infections, along with a special ‘antibiotic pharmacist’ on staff, from 86% last year to 97% this year.
  • Better reporting of patient safety incidents. Though the average number of incidents has increased (from 5% to 5.7%), this is a ‘positive sign’ because it shows ‘awareness of errors and near-misses and a culture of freedom to report’.
  • More hospitals using ‘track and trigger’ systems, which are regular observations by nurses designed to pick up deterioration in a patient’s condition (79% of trusts compared to 64% last year).

Dr Foster says there is room for improvement in how data is recorded. It lists the available figures for several types of avoidable harm, such as pulmonary embolisms and post-operative sepsis, which it cannot put into context due to the lack of complete data.

It identifies the same problem regarding data on the rate of medical mistakes (or adverse events) that happen in hospitals. The guide says again, that trusts with higher rates of incidents also tend to have more complete records about their patients.

The prevention of blood clots also features highly. The report says that all patients admitted to hospital now must be risk assessed for the risk of venous thromboembolism (VTE, of which DVT [deep vein thrombosis] is a common type).

However, trusts gave varying responses when asked ‘What percentage of patients are risk-assessed for VTE on admission?’ Most trusts were able to report how many patients were risk-assessed, but 15 responded that they were either not assessing patients for VTE or were unable to provide the information.

A spokesperson for the Department of Health (DH) said, “We accept that VTE is underreported, and are taking steps to change that position.

“At a national level, the DH is enabling the NHS to improve accuracy of reporting incidence of hospital acquired VTE.”

What does the guide say about stroke care?

Stroke is the third most common cause of death in the UK, costing the economy an estimated £8 billion a year. Dr Foster says there have been measurable improvements in the way the NHS deals with strokes but still notes ‘a worrying level of variation in care’.

The report identifies six best performing and eight worst performing trusts based on an analysis of six key indicators:

  • the proportion of patients having a brain scan on the same or next day: this ranged from 87% to 42%, with North Middlesex University Hospital NHS Trust having the highest rate
  • the proportion of patients given ‘clot busting’ drugs (thrombolysis) within 24 hours: Rates varied from 0.2% to 17%
  • the proportion of stroke admissions that lead to pneumonia due to swallowing problems: rates varied from 2% to 12%
  • the proportion of patients returning home from hospital within 56 days: rates varied from 55% to 85%
  • the rate of emergency readmissions: this varied from 44% below average to 58% above
  • standardised mortality ratio for stroke (a measure that can highlight preventable deaths): rates varied from 34% below average to 66% above average

What does the guide say about orthopaedic care?

Hip and knee replacements, as well as hip fractures, are a major expense for the NHS. The guide analyses some key indicators of the quality of care in these cases:

  • The percentage of patients readmitted to hospital within 28 days of a hip or knee replacement
    Most hospitals performed as expected. However, for hip replacements, two trusts had high rates: Leeds Teaching Hospitals NHS Trust (75% above average) and Newcastle upon Tyne Hospitals NHS Foundation Trust (63% above average). Two trusts had low rates: Northern Devon Healthcare NHS Trust (67% below average) and Royal Devon and Exeter NHS Foundation Trust (35% below average).
  • The number of patients requiring revision of a knee or hip replacement
    Hip replacement revision rates varied from 0 to 3.5 %. For knees, rates varied from 0 to 2.1%. Sixteen trusts performed particularly well on this indicator, while three had high rates.
  • Standardised mortality ratio for hip fractures
    Hip fractures are the most common reason for orthopaedic admission and about 10% of people with a hip fracture die within a month. All trusts performed as well as expected and Cambridge University Hospitals NHS Foundation Trust performed particularly well, with a mortality ratio 46% below average.
  • Hip fractures operated on within two days
    Operating straight away increases a patient’s chances of survival. However, Dr Foster found only 21% of trusts had rates of delayed surgery that were significantly low. The percentage operated on within two days varied from 34% to 94%.

Dr Foster identified six trusts as the best performers across its orthopaedic indicators and singled out Leeds Teaching Hospitals NHS Trust as the worst performer.

What does the guide say about care in urology?

Surgery for urological cancers should ideally be performed in large hospitals where these procedures are carried out more frequently. NICE guidelines state that pelvic urological cancer surgery should only take place in units where more than 50 procedures are carried out each year.

The guide identifies 19 trusts that carried out high numbers of prostate and bladder cancer operation between 2007 and 2010. It also identifies eight trusts performing high numbers of keyhole prostate operations, which enable quicker surgery and recovery.

Overall, the guide notes that more operations are being performed for prostate cancer, more of these operations are taking place in large hospitals and more keyhole procedures are being carried out.

There is a similar trend towards performing cystectomy (removal of the bladder) in large units. In 2006/07 large trusts performed only 21% of cystectomies but by 2009/10 this had risen to 63%.

The guide notes that operations to treat benign urological conditions are performed in a wider range of units than for cancer with varying quality. Dr Foster looked at the need for repeat surgery following one such procedure, transurethral resection of the prostate (TURP), as an indicator of care quality. The report lists 13 trusts that perform best on this indicator and three that perform worst.

Should I worry about the findings in the hospital guide?

No, the guide shows that overall things are improving. Although there are some trusts that are poor performing relative to the average, the vast majority are in the ‘as expected’ range and there are many batting well above average.

The key is to compare hospitals before making an appointment with a specialist and then to make use of your right to choosing which hospital you go to. For more information on hospital choice go here.

How can I compare hospitals myself?

NHS Choices' Find and choose hospitals function allows you to compare hospitals on a wide and growing range of measures, including:

  • overall quality of service (judged by the regulator)
  • mortality rates
  • waiting times
  • infection rates
  • food quality
  • parking facilities
  • disabled access

NHS Choices also allows patients to record their views on NHS services they have used. Users can leave comments about a hospital and say whether they would recommend it to a friend. They can also rate it on the following:

  • cleanliness
  • how well staff work together
  • whether they are treated with dignity and respect
  • whether they were involved in decisions about their care
  • whether the hospital offers same-sex accommodation

You can view these comments and ratings for any hospital using NHS Choices’ Find and choose services function.

Find more information about your right to choose where you are treated on our page about choosing a hospital.

The hospital guide is published in full on the Dr Foster website.


NHS Attribution