NATIONAL VARIATIONS IN THE PROVISION OF CARDIAC SERVICES IN THE UNITED KINGDOM

Second report of the British Cardiovascular Society working group 2005

Nicholas Boon (chairman)
Michael Norell (secretary)
Jim Hall (England)
Kevin Jennings (Scotland)
Liam Penny (Wales)
Carol Wilson (Northern Ireland)
John Chambers (echocardiography)
Rosemary Weston (BCS development manager)

April 2006

Background

In 1998 the UK government devolved power to elected assemblies in Scotland, Wales, and Northern Ireland. Although the powers of these political bodies differ they all have important freedoms with respect to health care and have pursued different policies. In England, the emphasis has been on targets introduced through national service frameworks. Scotland has concentrated on health promotion and the development of managed clinical networks led by healthcare professionals and patients. Wales has focused on public health partnerships, whilst developments in Northern Ireland have stalled since the elected assembly was suspended in 2002.

The National Services Framework (NSF) for Coronary Heart Disease (CHD) 1 was published in 2000 and has facilitated dramatic and tangible improvements in the provision of cardiac services in England. Amid concerns that the devolved nations had been disadvantaged by exclusion from the NSF the Council of the British Cardiac Society appointed a working group to examine variations in the provision of cardiac services within the UK in 2004. In spite of some methodological difficulties the working group found clear evidence of major differences in service provision, activity and planning that were not related to need 2. The report attracted a lot of interest and an accompanying editorial highlighted the need to monitor progress by conducting further surveys 3. The Council of the British Cardiac Society has therefore decided to conduct a second survey of the provision of cardiac services in the four UK nations.

The working group

The working group included representatives from each of the four devolved nations and the British Society of Echocardiography (see above).

Scope of the report

The group has not attempted to report on every aspect of cardiac services but have instead collected data on a number of core activities, many of which were described in the first report. In order to assess rate of change of data annual statistics have, whenever possible, been presented for the last five years.

Some parameters described in the first report have been omitted (e.g. cardiac surgery, and the number of cardiac catheterisation laboratories) but new ones (e.g. the use of Cardiac Resynchronisation Therapy) have been added. Special attention has been given to the provision of Echocardiography using data provided by the British Society of Echocardiography. Future reports will concentrate on other facets of cardiac services.

Data for each country were collected from the same source using common definitions and have been corrected for the population served using denominators of 50 million for England, 5 million for Scotland, 3 million for Wales and 1.7 million for Northern Ireland 4. Some patients cross national borders to access specialist cardiac care; for example, approximately 700,000 people who live in North Wales rely on English centres for the provision of tertiary cardiac services. The working group have therefore, whenever possible, analysed and presented activity data by place of residence.

Findings

Healthcare Spending (Fig 1)

NHS expenditure has increased rapidly over the last few years. Figures for expenditure on cardiac services are not available but total NHS spending increased by 67% in England between 1999-2001 and 2004-5; in contrast, the increases over the same period were 38% in Scotland, 47% in Wales and 43% in Northern Ireland 5. Nevertheless, England still has the lowest and Scotland the highest per capita expenditure (fig 1) 6; spending is therefore still closely related to disease burden.

Fig 1. Current NHS spending (£) per capita

Mortality from CHD (Figs 2 & 3)

Age-adjusted CHD mortality has been falling across the UK for many years. The proportionate decline in mortality from 1999 to 2003 was similar in all four devolved nations7 (figs 2 & 3). Mortality, which is often used as a crude index of need, remains highest in the devolved nations and the North of England and there is no sign of these gaps closing7. In particular, death rates from CHD in Scottish women are worryingly high (fig 3).

Fig 2. Age-standardised death rates from CHD per 100,000 population for men aged 35-74
. Source: BHF - www.heartstats.org

Fig 3. Age-standardised death rates from CHD per 100,000 population for women aged 35-74
. Source: BHF - www.heartstats.org

Cardiology Consultant Staffing (Fig 4 & Table 1)

The availability of consultant cardiologists is a crucial determinate of the quality of any cardiac service. Analysis of the cardiology workforce has been difficult because accurate data are difficult to find and the definition of a cardiologist is not straightforward. The data presented here are derived from the most recent workforce study conducted by the Royal College of Physicians of London8 and a survey undertaken by the British Cardiac Society for the specific purposes of writing this report.

The data for our survey were provided by the lead clinicians of individual managed clinical networks using a census date of 30th September 2005. We have tried to identify all consultant staff trained in cardiology, on the specialist register and responsible for patients with cardiac disease either full time or in conjunction with a general medical commitment. We have also tried to determine the number of new appointments made in the last three years, the number of long-term vacancies (posts unfilled for more than three months) and the proportion of cardiologists who have a significant general medical commitment (defined as participating in an unselected medical take) and/or academic (university) responsibilities.

The working group was reassured to find close agreement between the Royal College of Physicians data and our own survey.

There has been significant growth in the UK cardiology workforce over the last five years (see fig 4 and table 1). This has been most obvious in England (180 new posts advertised and appointed in the last 3 years), which initially had the fewest number of cardiologists per million population. There is therefore now a more equal distribution of cardiologists across the UK; nevertheless, staffing levels remain well below those reported in comparable Western European countries 9,10 and much lower than those recommended by the BCS workforce committee.

Table 1 indicates that there are a significant number of long-term vacancies. Although it is not entirely clear what geographical or other factors make some consultant posts more attractive than others, most of the long-term vacancies reported to us were in the North of England or the devolved nations; moreover, almost all of these posts included general medical duties.

The proportion of time cardiologists spend on specialist activity seems to vary considerably across the UK. The vast majority (74%) of cardiologists in Northern Ireland are full time specialists; in contrast 65% of Welsh consultant cardiologists have important and apparently time-consuming general medical duties (table 1). These observations may partly explain the national differences in specialist activity (e.g. PCI and device implantation) described below.

Fig 4. Cardiologists per million population by country. Data for 1999, 2000 and 2004 was derived from the Royal College of Physicians London census (there are no comparable data for Scotland in 1999) . Data for 2005 was obtained from the BCS survey.

Table 1. British Cardiac Society survey of UK cardiologists (census date 30th September 2005). * Responsible for unselected general medical take.

Access rates for PCI (Fig 5)

The number of patients treated by PCI has increased dramatically throughout the UK over the last 5 years (fig 5). In 2004 a total of 62,904 PCI procedures were undertaken in both NHS and private institutions throughout the UK (1052 per million population). Procedure numbers for individual countries (NHS only), and rates per million were as follows: England; 50,941 (1019 pmp), Scotland; 5095 (1019 pmp), Wales (allowing for 0.7 m of this population accessing English centres for their procedures); 1632 (725 pmp) and Northern Ireland; 1988 (1169 pmp).

The low PCI rate for Wales (which takes account of Welsh patients being treated in English centres) is a striking anomaly that cannot be explained by differences in need and represents an important health inequality. Moreover, there is no sign that the gap between Wales and the other home nations is closing.

Fig 5. Percutaneous Coronary Intervention procedures per million population by country of residence. Data provided by BCIS (British Cardiac Intervention Society).

Use of Drug Eluting Stents (DES) during PCI (Fig 6)

The marked increase in the use of drug eluting stents illustrated in figure 6 reflects changes in evidence-based practice and the impact of NICE guidance11. However, it is clear that there have been major national differences in the rate at which this new technology has been introduced. For example, Northern Ireland patients were among the first to gain access to this form of therapy whilst DES usage in Scotland has lagged behind the other countries.

Fig 6. Percentage of PCI cases using a Drug Eluting Stent by country. Data provided by BCIS (British Cardiac Intervention Society).

Pacemaker Implantation (Fig 7)

In 2004, there were approximately 31,000 new or replacement pacemaker implants in the UK (518 procedures per million)12. Pacing is now a well-established form of therapy and national implant rates vary less than other clinical activities. Nevertheless, new implant rates are still below the target of 450/million13 in all UK countries (fig 7) and remain lowest in those countries with the highest cardiovascular mortality.

Fig 7. Pacemaker implant rates. New implants only (number per million population) by country of residence (9% of Welsh implants were took place in England). Excludes devices with other functions such as CRT. Data provided by the National Pacemaker Database.

Implantable Cardioverter-Defibrillator (ICD) Therapy (Fig 8)

The implantable cardioverter-defibrillator (ICD) was introduced in 1980 and implants have been registered with the National Pacemaker Database since 1989. NICE published guidance on the indications for ICD therapy in 2000 when it was suggested that an implant rate of approximately 50 per million population would be appropriate for the UK as a whole 14.

An analysis of audit data from the National Pacemaker Database for the period 1998-2002 has shown that the UK as a whole implants far fewer ICDs than most Western European Countries 15. Moreover there are huge regional variations in implant rates across the UK that cannot be explained by differences in the prevalence of heart disease and must therefore reflect differences in clinical practice. Implant rates in Wales (adjusted to take account of the large numbers of patients who live in North Wales but are treated in England) have been consistently lower than those in the other UK countries. Moreover, there was a four fold difference in the mean ICD implant rates for 1998 to 2002 between the UK Health authorities with the lowest (Fife, Scotland – 9.2/million) and the highest (Eastern Health and Social Services Board, Northern Ireland – 39/million) activity 15. Expert commentators have pointed out that these stark differences probably stem from limited access to cardiac catheterisation laboratories and/or a shortage of appropriately trained electro-physiologists/device specialists in many parts of the UK 16.

Figure 8 shows the ICD implant rates for the four home nations up to 2004. Implant rates continue to increase but there is nothing to suggest that the gap between Wales and the rest of the UK is closing.

Fig 8. Implantable Cardioverter-Defibrillator (ICD) implant rates. New implants only (number per million population) by country of residence (9% of Welsh implants were took place in England). Data provided by the National Pacemaker Database.

Cardiac Resynchronisation Therapy (Fig 9)

Cardiac Resynchronisation Therapy (CRT) is a complex form of pacemaker therapy that has been shown to improve the morbidity and mortality of selected patients with heart failure 17,18. The working group has found evidence of a very patchy uptake of this technology across the UK and believes that this reflects a shortage of the relevant expertise and difficulties in funding.

Figure 9, which was prepared with the help of the Arrhythmia Alliance, illustrates huge differences in implant rates between the four home nations. The working group have concluded that the low implant rates in Wales are due a shortage of specialists and a lack of funding from specialist commissioners. The low rates in Scotland seem to be predominantly due to difficulties in agreeing an appropriate funding stream.

Fig 9. All forms of CRT (Cardiac Resynchronisation Therapy) per million population by country. The implant rate for 2005 is a projected figure based on data for the first 8 months of the year.
Data include new and replacement devices and were prepared by the Arrhythmia Alliance with help from CCAD and ABHI.

CRT can been delivered using devices that provide only a pacing function (CRT-P), and those that also incorporate a defibrillator capacity (CRT-D). Interestingly, there are even larger national differences in the use of these subtypes of CRT. Thus, in 2004 the approximate rates per million population for CRT-P implants in England, Scotland, Wales and Northern Ireland, were 17, 6, 4 and 12, whereas the respective rates for CRT-D were 18, 7, 3 and 32.

Echocardiography (Figs 10, 11, 12 and Table 2)

In contrast to our first report we have tried to include non-invasive cardiac investigations, as another comparator of service provision, in this survey and have specifically examined echocardiography facilities, staffing and activity. Data obtained from a comprehensive postal survey of echocardiography facilities conducted by the British Society of Echocardiography are presented in Figures 10, 11, 12 and Table 2 and reveal important national differences in the provision of this vital service.

Figure 10. Annual number of transthoracic echocardiograms per million population by country

Fig 11. Annual number of transoesophageal (TOEs) and stress echocardiograms per million population by country.

The annual number of standard transthoracic echocardiograms ranges from 9,167 per million in Wales to 15,294 per million in Northern Ireland. Once again, it is clear that this disparity is not related to differences in need. There are even greater variations in the delivery of more complex echocardiography (transoesophageal echocardiography and stress echocardiography – fig 11). The annual number of stress echocardiograms performed in Scotland is exceptionally low and offers another example of the slow pace at which new technology has been introduced in that region. In contrast, this sort of specialist activity is particularly high in Northern Ireland.

The number of echocardiograms carried out in each country is, not surprisingly, mirrored closely by the availability of sonographers (fig 12). Paradoxically, waiting times for transthoracic echocardiography (TTE) were longest in the countries with the most sonographers and the highest throughput. Thus, 78% of units in Northern Ireland and 31% of units in England reported waiting times in excess of 18 weeks compared to 14% of the units in Wales and none of the units in Scotland (table 2). This may reflect complex differences in demand that were not addressed by the survey.

Figure 12. Number of whole time equivalent sonographers per million population by country.

Table 2. British Society for Echocardiography postal survey of UK hospitals that provide echocardiography.

Staffing problems were most evident in England where 44% of hospitals reported the loss of 1 or 2, and 13% the loss of more than 3 sonographers within the previous two years. None of the hospitals in other countries reported the loss of more than 3 sonographers in the last two years, but 40% of hospitals in Wales, 33% of hospitals in Scotland, and 22% of hospitals in Northern Ireland reported the loss of 1 or 2 sonographers. Agency staff were used to fill more than 25% of posts in 15% of English units, 8% of Welsh units, but none of the units in Scotland or Northern Ireland. Moreover, more than 25% of posts were unfilled by either permanent or agency staff in 22% of English units, 14% of Welsh units, 8% of Scottish units, and none of the units in Northern Ireland. In addition only 58% of the sonographers working in England were fully trained specialists compared to 68% in Scotland, 91% in Wales and 100% in Northern Ireland

Discussion

Most, if not all, of the national variations in the provision of cardiac services that were identified in our first report 2 are still present. Moreover, there is little evidence to suggest that these differences are diminishing and some new inequalities have been identified.

This report highlights some striking disparities between spending, activity and the burden of disease and includes many examples of the inverse care law (“the availability of good medical care tends to vary inversely with the need for it in the population served”) 19,20.

Cardiac services in the UK are evolving rapidly and it is perhaps not surprising that the most obvious anomalies relate to the use of emerging technology such as drug eluting stents, implantable cardioverter defibrillators and cardiac resynchronisation therapy. The huge (up to tenfold) difference in the provision of some of these services is clearly not related to need and must therefore reflect differences in clinical practice and healthcare policy. At this stage it is not clear whether or not devolution will lead to increasing variations in service delivery, however, characteristic strengths and weaknesses are becoming apparent in each of the four home nations.

Cardiac services in England have developed rapidly since the National Service Framework was published and numerous benefits have been documented 21. In particular, there has been huge investment in front-line services supported and driven by the appointment of 180 new consultant cardiologists in the last three years. A few years ago specialist activity in England was substantially lower than that in Scotland and Northern Ireland but is now at least as high as the rest of the UK. Nevertheless, there are still many gaps in the service with many long-term consultant vacancies in the north of England and wide variations in the delivery of some services such as the provision of implantable cardioverter defibrillators 15. Growing regional inequalities within England will emerge if the patchy uptake of consultant posts (particularly the high number of long-term vacancies in the North) is not addressed. A more detailed analysis of the survey, conducted for this report, will therefore be carried out by the BCS.

Cardiac services in Scotland were amongst the best in the UK five years ago and long established services such as simple transthoracic echocardiography and pacing remain excellent. Unfortunately, Scotland has not achieved the same growth in consultant cardiology staffing or specialist activity as England. For example, the number of PCI procedures has risen relatively slowly and is projected to fall behind the English rate next year (fig 5); this cannot reflect differences in need because cardiovascular morbidity and mortality in Scotland remains the highest in the UK (figs 2 & 3). Moreover, drug eluting stents, cardiac resynchronisation therapy and stress echocardiography have been used sparingly in Scotland compared to the rest of the UK. These procedures are supported by a strong evidence base and Mr Andy Kerr, the Scottish Health Minister, acknowledged that there was a need to review the system for introducing new technology when he addressed the Scottish Cardiac Society in November 2005. Plans to set up a national network to oversee the delivery of revascularisation and specialist electro-physiology services in Scotland 22 were abandoned before the system had become operational and these activities are now managed by three regional planning groups. Unfortunately, regional planning in Scotland has got off to a slow start (the regional planning groups have no budget and can only make recommendations to local health boards) and it seems likely that this accounts for the failure to introduce drug eluting stents and cardiac resynchronisation therapy in a timely and equitable way.

Our first report identified alarmingly low coronary intervention rates in Wales. Even allowing for the substantial number of patients who reside in North Wales but are referred to England for treatment, most forms of specialist activity in Wales remain worryingly low. There seem to be two reasons for this. First of all there have been difficulties in commissioning. The NSF for CHD has never been adequately funded in Wales, and has become aspirational rather than pivotal in health care planning. Tertiary Cardiac and other specialist services are commissioned nationally by Health Commission Wales; their budget has not been increased in line with the emergence of NICE approved evidence based technologies and the new Arrhythmia chapter to the NSF, which was introduced in England more than a year ago, is still to be approved in Wales. Secondly, the consultant cardiology workforce appears to comprise many physicians who have substantial general medical duties and relatively few specialists with skills in coronary intervention and electro-physiology. Unfortunately there seems little prospect of eliminating these inequalities in the near future.

The service in Northern Ireland is provided by a highly specialised workforce that delivers relatively high rates of complex interventional procedures (figs 5, 8 & 9) but may under-perform in other areas such as simple echocardiography (table 2). There are suggestions that some aspects of the Northern Ireland cardiac services are stagnating; for example, 20% of the consultant posts have been vacant for more than three months. Certainly, the lack of an overarching strategy, which might have been provided by an NSF or equivalent document, has forced the four Health and Social Services Boards in Northern Ireland to develop services in isolation and created a form of planning blight.

The phenomenon of postcode prescribing, also known as the postcode lottery, has been well publicised and reflects the many variables, particularly the availability and expertise of staff, that influence small health care units. The regional variations in activity described in this report relate to much larger service blocks and could be construed as a national lottery. They are certainly more likely to reflect differences in health care policy and planning.

The whole purpose of devolution is to allow elected representatives to plan and implement services in accordance with local needs. Whilst, it seems inevitable that local planning will identify different priorities, and therefore create variations in service delivery, the core objectives of NHS in the four nations have not changed and the public rightly continues to expect common standards of care across the UK. There is therefore an obvious conflict between the desire to devolve decision-making and at the same time eliminate postcode prescribing.

The practice of cardiovascular medicine is supported by an extraordinarily large and robust evidence base. The optimum management of most cardiovascular disease is therefore well established. The challenge facing our devolved governments is not what to do but how to ensure that it is done; we are therefore witnessing an intriguing experiment that may provide useful insight into the impact of different health care policies on activity and outcome. The present government has increased health care spending enormously and it will be fascinating to see which policies produce the best return on this investment. The answer will only become apparent through careful audit; the British Cardiac Society will therefore continue to monitor all of the parameters described in this report.

References

1. Department of Health. The National Services Framework for Coronary Heart Disease. London: Department of Health, 2000.

2. Brooks N, Norell M, Hall J et al. on behalf of a working group of the British Cardiac Society. National variations in the provision of cardiac services in the United Kingdom. Br J Cardiol 2005; 12: 192-8.

3. Brooks N. Cardiac services in the UK : are some areas more equal than others? Br J Cardiol 2005; 12: 167-8.

4. www.statistics.gov.uk

5. Christie B. Health spending in England rises faster than rest of UK. Br Med J 2005; 331: 1424.

6. Christie B. Spending in Scotland goes up while activity goes down. Br Med J 2005; 331: 1358.

7. www.heartstats.org

8. The Federation of the Royal colleges of Physicians of the United Kingdom. Census of Consultant Physicians in the UK< 2004. Data and Commentary. Also available at www.rcplondon.ac.uk.

9. Block et al. Manpower in cardiology in Europe. European Heart Journal 2000; 21: 1135-40.

10. Block et al. Manpower in cardiology II in Western and Central Europe. European Heart Journal 2003; 24: 299-310)

11. NICE guidelines on DES. http://www.nice.org.uk/page.aspx?o=TA071

12. Cunningham AD. National Pacing Database/Central Cardiac Audit Database. www.ccad.org.uk

13. Hackett D. Clinical and laboratory cardiac facilities required in the UK : British Cardiac Society 2005. www.bcs.com

14. National Institute for Clinical excellence. Guidance on the use of implantable cardioverter defibrillators for arrhythmias. Technology appraisal guidance no 11. London 2000. www.nice.org.uk.

15. Cunningham AD, Plummer CJ, McComb JM et al. The implantable cardioverter-defibrillator : postcode prescribing in the UK 1998-2002. Heart 2005; 91: 1280-3.

16. Lang CC. Regional variation in ICD implantation rates: the shocking truth? Heart 2005; 91: 1251-3.

17. Cleland JGF, Daubert JC, Erdmann E, Freemantle N, Gras D, Kappenberger L and Tavazzi L. For the Cardiac Resynchronisation Heart Failure (CARE-HF) study investigators. The effect of cardiac resynchronisation on morbidity and mortality in heart failure. N Engl J Med 2005; 35 : 1539-1549.

18. National Institute for Clinical Excellence. Chronic Heart Failure: national clinical guideline for diagnosis and management in primary and secondary care. London, NICE; 2003.

19. Hart JT. The inverse care law. Lancet 1971; 1: 405-412.

20. Watt G. The inverse care law today. The Lancet 2002; 360:252-254.

21. Department of Health. Coronary Heart Disease National Service Framework: Leading the way - Progress report 2005, March 2005.

22. Scottish Executive. Coronary Heart Disease and Stroke. Strategy for Scotland. SEHD 2002.