Case for change. TB services in London

Transcription

1 Case for change TB services in London September 2011

2 List of abbreviations A&E BCG DH DOT ECDC GP HIV HPA HPU IGRA LTBR MDR TB NHS NICE PCT TB WHO XDR TB Accident and Emergency Bacille Calmette Guérin vaccine Department of Health Directly observed therapy European Centre for Disease Control General Practitioner Human Immunodeficiency Virus Health Protection Agency Health Protection Unit Interferon Gamma Release Assay London TB Register Multi-drug resistant TB National Health Service National Institute of Clinical Excellence Primary Care Trust Tuberculosis World Health Organization Extensively drug resistant TB 2

3 Table of contents List of abbreviations... 2 Table of contents... 3 Foreword... 5 Executive summary Background The extent of TB in London Place of birth and ethnicity Social and medical risk factors for TB Rates of treatment completion and drug resistant TB Latent TB International comparisons Western Europe New York Policy overview Patient pathway Challenges with current service provision Delays in detection and referral Community awareness Awareness among healthcare professionals Screening Variability of commissioning Current network arrangements Investment in specialist non-medical staff Commissioning expertise and financial flexibility Accommodation Variability of service provision Availability of specialist expertise Staffing profiles Directly Observed Therapy (DOT) Contact tracing

4 Outreach BCG vaccination Potential fragmentation of TB control Health protection services Public health services Local authorities and the third sector Risks Financial considerations Costs of treatment for TB Activity and finance data challenges Financial modelling Summary of key points APPENDIX A: Summary of progress against London TB Metrics APPENDIX B: Comparative Financial Modelling APPENDIX C: Membership of the TB Commissioning Board and Clinical Working Group

5 Foreword Tuberculosis TB is a serious infectious disease that many people associate with past times. But since the 1980s, it has been making a comeback in London. In 2010 there were 3,302 new cases of TB in the capital, the highest of any major city in Western Europe. This compares to 2,626 new diagnoses of HIV in London in the same period. Despite the best efforts of health and social care professionals, the disease is now a major public health problem for the capital. This case for change provides a compelling set of arguments for the need to improve the care of people with TB in London. It sets out our understanding of the problem, describing the communities that are most at risk of developing TB, how London compares with elsewhere in the world and what has already been done to tackle the disease. Although people generally receive high quality care once they know they have TB, too many people are diagnosed late, when the disease has already begun to cause them damage and too little is done to support people to complete their treatment, avoiding complications in the future. More needs to be done to identify those with latent (inactive) infection who are living with the risk of developing active TB in the future and to offer them treatment to prevent the disease. The document describes specific problems with the way services are currently planned, organised and managed and some concerns about the possible impact on TB control of the proposed changes to the NHS in England. It also highlights examples of good practice in London that are helping to keep the disease under control. The case for change will be followed by a model of care that will set out our proposals to address the TB problem in London. Our aspiration is to reverse the recent trend and reduce rates of the disease as quickly as they have been rising. We would like to thank the many individuals and organisations that helped us develop this case for change and the model of care that will follow it. The members of the clinical working group and TB commissioning board are shown at the end of the document. Details of the many other health and social care professionals, service users and third sector partners who contributed can be found at Nick Relph Chief Executive Outer NW London Cluster and Senior Responsible Officer London TB Project Dr William Lynn Consultant in Infectious Diseases Ealing Hospital NHS Trust and Clinical Lead London TB Project 5

6 Executive summary Tuberculosis (TB) is an infectious disease that is treatable and curable, however it remains a major public health issue for London. The number of TB cases has increased by nearly 50% over the last ten years and as a result, London now has the highest rate of TB of any capital city in Western Europe. The current rate of 43 per 100,000 population exceeds the level considered high by the World Health Organisation (WHO). In some parts of London, the rate has been more than double this for over ten years. The groups that account for the majority of TB in London are those born overseas and those with social risk factors. People born in countries where the disease is more prevalent account for 84% of new cases of TB in London. The majority develop active disease several years after their arrival in the country, making port of entry screening ineffective. TB is also more common amongst people with social risk factors, particularly homelessness and drug and alcohol dependency, because of their tendency to poor immune status and increased risk of exposure to infection. It is important to note that around 80% of people who develop active TB do so as a result of the reactivation of latent TB (TB acquired earlier in life which remains dormant for months or years), rather than through transmission from a person with active disease. The escalating burden of TB in London is set against a background of national guidance and policy. Implementation of some of these measures has contributed to stabilising the rate of TB but has failed to reverse the upward trend. Application of national guidance has been inconsistent in some parts of London. There is currently no systematic approach to detecting and treating latent TB. The effective detection and treatment of TB will reduce the human and financial burden of disease as well as minimising the risk of ongoing transmission. Active TB is relatively inexpensive and straightforward to treat and cure if identified early. But some people become permanently disabled, particularly from brain or spinal disease and about 300 people a year still die from TB. The disease has a real, although poorly quantified, impact on family life, employment and educational attainment for those affected. Awareness of the disease and its symptoms in the general community is poor. Additionally, in high risk communities the presentation of symptoms may be delayed due to the stigma associated with TB. For first line clinicians, TB can be difficult to diagnose and symptoms easily overlooked. Awareness of the complex relationship between multi-faceted symptoms, incidence of the disease in a community, an individual s circumstance, and any one of a number of social risk factors is critical in considering a TB diagnosis. Furthermore, the inconsistent application of screening guidance across London for detecting TB does not produce significant yield in terms of numbers of cases identified. Successful therapy requires adherence to a complex drug regimen over a minimum of six months. If treatment is not taken correctly, or is stopped, there is a higher risk of complications and/or the development of drug resistant TB, which has been identified by WHO as a major threat to global public health. Drug resistant TB is associated with a substantial increase in morbidity and mortality as well as being a considerably more resource-intensive treatment, potentially costing twenty times more than early intervention. While treatment completion rates have improved as a whole across London, they vary considerably across boroughs representing a clear opportunity for improvement. Five TB networks currently manage local service planning, development and protocols for TB across the capital each in different ways. Commissioning for TB services is not always proactive in parts of London as other healthcare issues often take priority. Differences exist in funding arrangements across the networks with some commissioners paying separately for 6

7 services that are included in national tariff elsewhere. While metrics have been developed for TB and are regularly reviewed, active performance management is not comprehensive across the city. The provision of TB services in many parts of London is the result of organic development rather than a planned response to an increasing problem. Staffing mix and grading is notably different across providers and does not necessarily correspond with the incidence of the disease. Specialist expertise of complex cases tends to be focused in a few large hospitals and although there are network arrangements in each sector to share learning, there is no systematic process in place to ensure patients treated in smaller centres always have access to the appropriate level of expertise. Good practice exists, particularly in relation to the use of directly observed therapy and tracing contacts to prevent onward transmission, but it is not systematically adopted across the city. Vaccination guidance for TB currently targets children at risk of exposure, however uptake and the process for giving the vaccination is variable. There is a risk that the control of TB will become more fragmented as the responsibilities for protecting health and procuring services move into new and disparate organisations, most of which will be unable to take a pan-london strategic view of the disease. There are opportunities for London to improve its response to TB and reduce the incidence of the disease. The existing TB service model in London has not impacted on the rates of TB and the capital can learn from models of good practice internationally where cities have been proactive in managing escalation of TB and have had significant successes. Both in New York and European cities, a coordinated, focused, multi-agency approach to tackling TB has led to a dramatic reduction in the TB rate. This document sets out the case for change in more detail and will be followed by a proposed model of care which will set out how services can be improved to achieve the goal of achieving a sustained reduction in TB across London. 7

8 1. Background Tuberculosis (TB) is an infectious disease caused by the Mycobacterium tuberculosis complex group of bacteria. It can affect almost any organ in the body. It is estimated that globally, almost two billion people have been infected and around nine million develop symptomatic disease each year. TB kills 1.5 million people each year. TB is transmitted from person to person through the air. The bacteria are inhaled and pass from the lungs to other parts of the body via the blood stream, airways or lymphatic system. Pulmonary TB is the most common form of the disease occurring in about 80% of cases globally 1. Extra-pulmonary TB affects organs other than the lungs, most frequently the lymph nodes, pleura, bones (including the spine) and central nervous system. This form of disease is not usually infectious. The majority of people (about 90%) who become infected with TB bacteria experience no symptoms and do not pass the disease on to others. This is known as latent TB or latent TB infection. For most, the micro-organisms remain dormant in their body throughout their lives but for some, there is a chance they may subsequently develop symptomatic, active TB disease. It is currently impossible to predict which individuals with infection will go on to develop active clinical disease, although the risk is higher in those with suppressed immune systems, children 2 and in the first five years after initial infection 3. Estimates of the rate of progression from latent to active TB vary from 5% to 15%, but 10% is the figure adopted by the National Institute for Health and Clinical Excellence (NICE). Clinically, latent TB is defined as occurring when an individual is well, has a chest X-ray that does not suggest active TB disease, but a positive test for TB indicating previous infection. The traditional method used for this is the tuberculin skin test (TST), also called a Mantoux test. More recently, blood tests have been introduced (Interferon gamma release assays, IGRA). In March 2011, NICE reviewed the evidence on the newer tests and updated its guidelines on diagnosing latent TB infection 4. If active TB is diagnosed at an early stage of disease, it is curable in around 95% of cases. Treatment comprises multiple anti-tb drugs for a period of at least six months. Treatment for TB carries the risk of unpleasant adverse effects. These, plus the long period of treatment needed for cure, mean that people may not complete their planned therapy. Unfortunately, if the treatment is not taken as prescribed or the treatment regimen fails to take account of previous incomplete treatment, the patient is at risk of developing drug resistant TB. The treatment of drug resistant TB is more likely to result in yet more side effects and a worse outcome (including untreatable disease and death). Drug resistant TB is also considerably more expensive to diagnose and treat. The cost of second line drugs required for treatment is high, additional clinical input is required and treatment takes much longer to complete, often with a reduced chance of cure. 1 Ait-Khaled N, Alarcon E et al. International Union Against Tuberculosis and Lung Disease, Management of Tuberculosis: A Guide to the Essentials of Good Practice. Paris Lalvani A. Diagnosing Tuberculosis Infection in the 21 st Century: New Tools to Tackle an Old Enemy. Chest WHO. Global tuberculosis control epidemiology, strategy, financing National Institute for Health and Clinical Excellence. Clinical guideline 117. Tuberculosis clinical diagnosis and management of tuberculosis, and measures for its prevention and control. March

9 The cost to commissioners of treating drug sensitive TB can be as low as 1,100, based on one new outpatient appointment and seven follow-up appointments. Even in the minority of cases where a brief admission is required, the cost will not exceed 4,000 (based on HRG DZ14B). In contrast, the cost of treating multi-drug resistant (MDR) TB typically exceeds 18,500, based on one admission (HRG DZ51Z), one new outpatient appointment and about 20 follow-up appointments 5. The costs do not take into account the assessment and management of onward transmission of infection from the index case to others. The contact tracing performed has a reasonable yield, such that 10% of those investigated will have latent TB infection and 1% active TB requiring treatment. The longer an individual is untreated, the longer they are infectious and hence able to pass on tuberculosis to others. Further, if diagnosis is delayed, TB can cause irreparable damage to whichever part of the body is affected. In pulmonary TB, this can lead to lung destruction, scarring and chronic infection (bronchiectasis) with other bacteria and fungi. Major surgery can be required, and over the longer term, the patient may develop respiratory failure. In extra-pulmonary TB, the effects can be wide-ranging, including stroke, spinal collapse (with spinal cord compression) and skin disfigurement. Around one third of individuals with spinal TB have chronic back pain after treatment and a number are left paralysed requiring life-long care with costs not only to health care but also to social care. In the UK, the numbers of deaths from TB have fluctuated over the last decade, peaking at 394 in 2003 and dropping to 300 in The most recent data from 2008 shows this figure climbing to 334 premature deaths from TB 6. A significant number of these deaths are avoidable through improved early diagnosis and adequate support to complete treatment. 5 These costs are exclusive of market forces factor and based on clinical advice regarding the HRG4 coding and number of outpatient appointments required. 6 Office of National Statistics. TB mortality data showing numbers and rates of TB deaths in England & Wales from Prepared by TB Section, Health Protection Services Colindale. 9

10 Number of TB notifications 2. The extent of TB in London The World Health Organisation defines a TB rate of 40 per 100,000 population as high 7. The rate in London has exceeded this since the early 2000s, rising by almost 50% between 1999 and In 2010, the rate of TB in London was 42.6 per 100, Figure 1 TB rates in London, Around 9,000 TB notifications are reported in the UK each year 9. These are concentrated in the major cities with London accounting for about 40% of all TB notifications. Figure 2 Number of TB notifications in UK major cities, Birmingham Glasgow London Leeds UK City Leicester City Manchester Sheffield Data courtesy of HPA 7 Tuberculosis in the UK: Annual report on tuberculosis surveillance in the UK. London: Health Protection Agency Centre for Infections. October Annual Report on Tuberculosis surveillance in London, Health Protection Agency 9 Health Protection Agency Centre for Infections. Tuberculosis case reports and rates by country, UK, Health Protection Agency website 10

11 TB cases are widely distributed across London. Healthcare services in all parts of the capital encounter TB on a regular basis (Figure 3). Figure 3 TB rates by primary care trust of residence, 2010 Enfield Barnet Hillingdon Harrow Haringey Waltham Forest Brent Ealing Hounslow Richmond & Twickenham City & Camden Hackney Islington Westminster Kensington & Chelsea Hammersmith & Fulham Lambeth Wandsworth Tower Hamlets Southwark Lewisham Redbridge Barking & Dagenham Newham Greenwich Bexley Havering Kingston Sutton & Merton Croydon Bromley TB rate /100,000 population <20 Source: London Regional Epidemiologist, HPA Although the most recent figures show 13 boroughs with rates above 40 per 100,000, the distribution varies from year to year. Twenty-one boroughs have experienced rates above this level at some point in the last five years. The rates in Brent and Newham have exceeded 80 per 100,000 for the last ten years and continue to rise. Place of birth and ethnicity There is a particularly heavy burden of disease among people born outside the UK. This group accounts for 84% of TB notifications in London and comprises predominantly those of Indian or black African ethnicity (Figure 4). Figure 4 TB notifications and rates by place of birth, London,

12 Number of cases Rate (per 100,000) UK born (n) Non-UK born (n) UK born (rate) Non-UK born (rate) 0 The majority develop symptomatic, active disease several years after their arrival in the UK. Over 80% of people had lived in the UK for two or more years prior to diagnosis and a third for ten or more years. Studies suggest that most cases arise from latent TB infection picked up outside of the UK 10. It is unclear how much TB arises from people travelling to TB endemic areas, or through extended contact with people from high incidence countries in the UK. Social and medical risk factors for TB Anyone can develop TB, although certain groups are at greater risk due to lifestyle or poor immunity. Problem drug or alcohol use, homelessness (and insecure housing tenure), imprisonment and mental health issues are common factors in around 12% of cases of TB in London 11. They are an important population to consider as they are less likely to access healthcare with minor symptoms, consequently presenting late to medical services with worse disease. This means that they are both infectious to others for longer periods (pulmonary TB is more often present in those with social and medical risk factors than in the general population) and have worse treatment outcomes 12. Furthermore, they have higher rates of drug resistant disease, and their contacts are more difficult to trace and screen for possible acquired infection. Rates of treatment completion and drug resistant TB In 2010, the proportion of TB cases successfully completing treatment within 12 months in London was 86.6%, just above the 85% target set by the Chief Medical Officer for England, based on the WHO target. However, as the graph on the following page shows, treatment completion rates vary considerably across the PCTs, with several below the target rate. Figure 5 Treatment completion rates by PCT, Love J, Sonnenberg P, Glynn JR et al. Molecular epidemiology of tuberculosis in England. International Journal of Tuberculosis and Lung Disease vol.13 (2), Health Protection Agency London Regional Epidemiology Unit. Annual report on tuberculosis surveillance in London. London; Public Health Action Support Team. London TB Service Review and Health Needs Assessment. September

13 95% 90% 85% 80% 75% 70% North Central North East North West South East South West Treatment completion is of paramount importance both to avoid the development of drug resistant TB (and the significant additional costs of treating it) and to prevent onward transmission of the disease. Compared to the UK, London TB patients have a higher level of drug resistance with a higher proportion still on treatment after 12 months. Drug resistance continues to be problematic with the proportion of drug resistant cases increasing from approximately 8% in 2008 to just over 10% in 2010 (Figure 6). Figure 6 Drug resistance rates in London, People who are immunocompromised are at increased risk of latent TB progressing to active TB in particular, those with HIV, chronic kidney disease, diabetes, and who are being 13

14 treated with immunosuppressive drugs. Around 2-10% of TB cases in the UK are people who are co-infected with HIV 13. Their risk of progressing to active disease is estimated to be around 20 times that of the general population. London has high numbers of people with each of these risk factors and although they represent a relatively small proportion of the total TB cases, they have a disproportionate impact on transmission and TB control. Latent TB The number of people in London with latent (asymptomatic) TB is unknown. Not everyone with latent TB will become unwell. However it is estimated that approximately 80% of active TB cases in London arise from individuals with long-standing latent TB. There is no systematic approach to identifying people with latent TB in London. Until recently the standard test for latent TB infection was the tuberculin skin test (the Mantoux test). This had a number of operational disadvantages associated with its use, not least of which was the requirement for individuals to return to have the skin test reaction interpreted after 2-3 days. IGRA blood tests often require just a single visit. Recent NICE guidelines 14 recommend the use of IGRA to test for latent TB infection in the following cases: In an outbreak situation when large numbers of individuals may need to be screened (for people aged over five years) New entrants from high incidence countries who are aged16 to 35 years old People who are immunocompromised and most people with HIV New NHS employees who have recently arrived from high incidence countries or who have had contact with patients in a setting where TB is highly prevalent People from hard-to-reach groups Reports from clinical teams suggest that implementation of the guidelines is patchy and uncoordinated across London. This is also supported by evidence from a recent review of services that found the number of cases receiving treatment for latent TB infection did not correlate with the number of contacts or new entrants screened, suggesting different practices in contact tracing and new entrant screening. Uncertainty about the cost effectiveness of IGRA and the lack of a strategic approach are often cited as contributory factors to differences in the management of latent TB. The updated NICE guidance does, however, advocate screening for latent TB and recent research has offered more information about the cost effectiveness of such an approach 15. The prophylactic treatment of latent TB infection prevents the development of active TB. Without treatment, people with latent TB have a lifetime risk of developing active disease and about 10% do so. Preventative treatment can have adverse side effects and may be highly toxic to the liver depending on the dosage and duration of treatment. Therefore, the decision whether to treat is not straightforward and requires consideration of the risks and careful 13 Ahmed AB, Abubakar I et al. The growing impact of HIV infection on the epidemiology of tuberculosis in England and Wales. Thorax,62(8): National Institute for Health and Clinical Excellence. Clinical guideline 117. Tuberculosis clinical diagnosis and management of tuberculosis, and measures for its prevention and control. March Pareek et al. Screening of immigrants in the UK for imported latent tuberculosis: a multicentre cohort study and cost effectiveness analysis. The Lancet: Infectious Diseases, vol 11, June

15 discussion with the patient. Furthermore, once an individual starts treatment for latent TB they need to be monitored closely as many default, which in turn leads to excess costs to service in seeking to re-engage them. If they do not complete their latent TB treatment they remain at continued risk of developing active TB in the future. Key issues London as a whole has very high rates of TB. Over time there has been a significant increase in rates and number of cases of TB. The rate in some boroughs is more than twice that of the definition used by the WHO for high rates. The majority of cases are in people born overseas, although it often takes several years for them to become symptomatic. London has large numbers of socially and medically complex cases of TB. This includes those with HIV infection as well as risk factors for poor treatment completion and onward transmission to others such as homelessness, drug or alcohol problems, a history of imprisonment or mental health issues. The majority of cases in London are caused by the reactivation of latent TB, the identification of which is costly and the treatment of which carries a clinical risk. Many individuals who start treatment for this do not complete their prescribed course. 15

16 Berlin Germany Stockholm Sweden Milan Italy Amsterdam Netherlands Paris France Barcelona Spain Brussels Belgium London UK TB rates per 100,000 population 3. International comparisons Western Europe London has the highest rate of TB of any major city in Western Europe (Figure 7). This represents a significant public health risk, as well as a risk to the reputation of the capital. Given that rates are declining elsewhere in Western Europe, the situation in London is even more striking. Figure 7 London and other Western European cities, comparison of TB rates per 100,000 population, City / Country Source: Dr Ibrahim Abubakar, Health Protection Agency While TB rates in Western European cities are much lower than London, these cities do provide examples of successful initiatives for TB control among groups with social risk factors. For example, the use of a mobile X-ray unit in Rotterdam 16 improved the detection of both active and latent TB and increased the number of contacts traced. In Paris, the TB rate reached 54 per 100,000 in the early 2000s. The epidemiology was similar to London with 79% of cases in those born overseas. The local authorities implemented a series of actions including a mobile X-ray unit and social support team, latent TB testing, widespread vaccination of children and improved contact tracing. As a result, the TB rate decreased to 23.6 per 100,000 over the following seven years. Figure 8 Comparison of TB rate per 100,000, de Vries G, van Hest RA. From contact investigation to tuberculosis screening of drug addicts and homeless persons in Rotterdam. European Journal of Public Health, vol 16 (2)

17 Paris Rotterdam London Source: HPA London New York In terms of TB rates and the number of TB cases, New York in the early 1990s is comparable to London in Figure 9 compares the TB rates per 100,000 population in New York and London since Figure 9 London and New York, comparison of TB rates per 100,000 population,

18 Number of cases TB numbers peaked in 1992 in New York City at 3,811 TB notifications 17 at a rate of approximately 50 per 100,000 of the population. The epidemiology in New York differed somewhat to that now seen in London. There was a higher rate of recent transmission (often in healthcare settings), higher rates of drug resistance and more HIV co-infection. As would be expected, the greatest impact of the New York policy was a reduction in cases arising from recent disease transmission 18. In London, most symptomatic disease is due to reactivation of latent TB in the foreign born (see section 2). The emphasis of TB control in London therefore, needs to differ from that applied in New York and a targeted approach to tackle latent TB is required. Nevertheless, there are lessons to be learned from the New York approach. New York had disinvested considerably in TB programmes and public health infrastructure based on projections that indicated the incidence of the disease would fall to almost negligible levels (Figure 10). Consequently, the cost of tackling the disease exceeded $700million between 1992 and 1996, much of which was required to re-establish its services. Figure 10 TB rates in New York City, Actual Cases Projected Cases Years Source: Chrispin Kambili, MD, Director, Bureau of TB Control, Health Department, New York City There are indications from services that London has also started to disinvest in TB services although on a smaller scale (see section 6). If this is not reversed, the cost of addressing TB in London will be significantly higher in future years. The experience in New York suggests this is likely to occur within the next three to five years. New York adopted a whole systems approach to reversing the rise of TB, including awareness raising, a focus on treatment completion, improved surveillance and outreach 17 Bureau of Tuberculosis Control. Clinical Policies and Protocols, 4 th Edition. New York City Department of Health and Mental Hygiene. March Frieden T, Fujiwara PI et al. Tuberculosis in New York City Turning the Tide. New England Journal of Medicine. 333:

19 services. While the epidemiology is different in London, the problem is equally multifaceted and will require a similarly broad set of interventions. New York brought a range of agencies together to form a TB Control Board, with the financial and political power to bring about change in health services, social care, housing agencies and prisons. London, in comparison, does not currently have a city-wide, multi-agency approach to managing TB. The commissioning of TB services is variable across the city (see section 6). Key issues London has the highest rate of TB of any major city in Western Europe. There have been some significant local initiatives in other European cities that have contributed to a reduction in their TB rate and should inform the model of care for London. New York was in a similar position to London in the early 1990s and brought its TB rate down through investment in services, a multi-faceted strategy and a coordinated, multiagency effort. 19

20 4. Policy overview The care and management of TB in the UK has been driven largely by the following key policies. Stopping Tuberculosis in England: An Action Plan from the Chief Medical Officer (2004) 19 This best practice guidance recognised the re-emergence of TB as a public health problem in the UK and recommended ten actions that were essential to bring TB under control: Increased awareness Strong commitment and leadership High quality surveillance Excellence in clinical care Well organised and coordinated patient services First class laboratory services Highly effective disease control at population level An expert workforce Leading edge research International partnership The actions were recommended with no additional resources for implementation. They prompted the development of pan-london metrics for TB services (see section 6.2) and resulted in some national changes that have contributed to TB care: The national enhanced TB surveillance system is being upgraded DNA fingerprinting and molecular typing is now in place NICE guidelines to support patient care and management were developed and have been updated First class laboratory services are now in place British HIV Association guidelines for the treatment of TB/HIV co-infection (2005, updated in 2011) 20 The British HIV Association (BHIVA) produced specific guidelines on the management of individuals co-infected with TB and HIV. These recommend that care takes place within a multidisciplinary team which includes physicians who have expertise in the treatment of TB and HIV. It also recommended that all TB patients of unknown HIV status should be offered an HIV test. The London TB metric on 19 Department of Health. Stopping Tuberculosis in England: An Action Plan from the Chief Medical Officer. October British HIV Association. British HIV Association guidelines for the treatment for TB/HIV coinfection (awaiting peer review feedback prior to publication)

21 HIV aims to have at least 90% of TB patients offered an HIV test. This is monitored through the London TB Register (LTBR) and overall London achieved 93% for all TB patients aged 16 years or older in In general, joint working between HIV and TB services has been successful in London. NICE clinical guidelines for TB (2006, updated in 2011) 21 The guidance was developed by NICE as a successor to the British Thoracic Society s TB guidelines, which had been used for many years as the principal source of advice on TB management in the UK. It includes recommendations on ways of organising services efficiently to provide the best possible care. The guidance aims to focus NHS resources where they will combat the spread of TB, and where scientific evidence supports it, makes recommendations on service organisation as well as clinical practice. It covers the following subject areas: Diagnosis Management of respiratory TB Management of non-respiratory TB Monitoring, adherence and treatment completion Risk assessment and infection control in drug resistant TB Management of latent TB BCG vaccination Active case finding Preventing infection in specific settings The guidelines were updated in March 2011 to reflect advances in diagnostic techniques. The majority of these changes and some of the original 2006 recommendations have not yet been implemented systematically across London. The resulting variability of service provision is described in section 6.3. Tuberculosis prevention and treatment: a toolkit for planning, commissioning and delivering high-quality services in England (2007) 22 This best practice guidance provides those who commission TB services in England a framework for assessing their local needs and for planning and commissioning high-quality services in order to implement the TB action plan. It also contains models of best practice for the prevention, treatment and monitoring of TB for service providers, including laboratory services. Some of the detailed information in the toolkit is now out of date e.g. the scope of Payment by Results (PbR) has increased and PCTs have been reconfigured. 21 National Institute for Health and Clinical Excellence. Clinical guideline 117. Tuberculosis clinical diagnosis and management of tuberculosis, and measures for its prevention and control. March Department of Health. Tuberculosis prevention and treatment: a toolkit for planning, commissioning and delivering high-quality services in England. Published electronically only. June

22 As a result of this guidance, a Stopping TB in London group was established to coordinate the capital s response. This group developed the London TB metrics (see section 6.2). The guidance included key messages for commissioners, which remain relevant. There is a strong economic case for effective management of TB. As well as the public health imperative, the lack of an effective strategy and poor management of TB can be very costly in the long term. Poor management can lead to the emergence of drug resistant cases which are much more expensive to treat. To secure high-quality services, commissioners need to consider their local TB incidence and population demography, and potential changes to that demography, such as new demands resulting from population migration. Therefore, all PCTs should plan for TB services. Every PCT should identify a named TB lead. TB is best diagnosed and managed by experienced specialists. While primary care clinicians may suspect a diagnosis of TB, a formal diagnosis including treatment and care plans is best made by specialist service providers. Primary care does have an important role in providing support to the patient through the treatment period. Key issues The care and treatment of TB is subject to several important pieces of national guidance, although both compliance with existing guidelines and implementation of new recommendations is patchy across London. 22

23 5. Patient pathway Person has TB symptoms Person enters UK from highincidence country Port Health service screens high-risk person and identifies potential TB infection Person presents at GP surgery, A&E department or other urgent care centre Patient identified by other service - Find & Treat, prison health and other clinical specialists TB suspected and patient referred to TB service Named Case Manager allocated Diagnostic investigations by TB service Patient diagnosed with TB Contact tracing & screening HPU referral for infectious, drug resistant, complex/non-compliant cases use of Public Health Act Treatment Patient followed up and reviewed Treatment completed Patient discharged Individuals with suspected active TB can enter the TB care pathway via a number of different routes. Direct referral to the TB service does occur from primary care, accident and emergency (A&E) or any direct access clinics. Allied hospital services such as respiratory medicine, ENT and orthopaedics or neurology/neurosurgery will also see new presentations 23

24 of disease. For people entering the UK from countries with a high incidence of TB, screening may be performed by Port Health Services. Additionally, the Find and Treat service in London provides a mobile X-ray unit for detecting TB in vulnerable groups, such as homeless people. Similar work, using static digital X-ray equipment is now starting in many of London s prisons. Across London, 30 different hospital outpatient clinic services provide diagnostics and specialist care for TB patients (29 of the 30 services have inpatient as well as outpatient provision). They are distributed widely across London. The model of care provided to TB patients is a traditional one based on the outpatient clinic, typically Monday to Friday, 9am to 5pm. Approximately 1,600 patients are seen in TB clinics across London each week. Following diagnosis of active TB, patients are allocated a case manager, typically a TB nurse from the TB clinic, to support them with administering medication and follow up care. To assess whether infection has spread, screening is offered to close contacts (usually household or workplace contacts) and in some cases, casual contacts of the person with active TB. Treatment for TB is typically a six month course of antibiotics with initially four drugs. Some TB clinics offer directly observed therapy (DOT), to support patients to take their medication as prescribed. DOT involves an enhanced package of care which includes either a health professional or other appropriately trained person observing the ingestion of prescribed medications and recording this in a log. Patients are usually followed up at monthly appointments in clinic to track their clinical progress and response to treatment, ensure adverse effects are minimised, verify that therapy is being taken as planned and to renew prescriptions for medication. Approximately 30% of cases require inpatient care in one of the 29 hospitals across London. Specialist care of children with complex TB disease is provided by several hospitals across the capital. Individuals requiring treatment for drug-resistant or multi-drug-resistant TB are usually cared for in one of the ten hospitals across London that provide treatment for complex cases. Once treatment is completed, discharge takes place. This is generally at six months for noncomplex cases. However for more complex cases, this varies depending on the site and severity of the disease and can extend beyond one year. Post-treatment follow up is often extended if there is ongoing concern regarding risk of TB relapse, such as extensive or drug resistant disease, or post-tb complications such as respiratory compromise or spinal/neurological disease. 24

25 6. Challenges with current service provision 6.1 Delays in detection and referral Delays in presentation, diagnosis and referral of TB can lead to unnecessary suffering, long term disability and increased cost to the NHS and social care services. It also results in increased TB transmission as the index case is infectious for longer. At present, patients often consult their GP numerous times prior to TB being considered as a diagnosis, and patients will present at A&E departments because delays in the recognition of early symptoms lead to a requirement for urgent care 23. TB is often not detected early because of: A lack of awareness of TB and its symptoms by those at greatest risk A lack of awareness among healthcare professionals and resulting failure to consider TB as a possible diagnosis Variability in the interpretation and implementation of screening guidance. Community awareness There are low levels of awareness of TB in the general population, and even in many highrisk groups. In addition, while some communities understand their risk of developing TB, specific individuals may not acknowledge this to the same extent and even defer presentation, given the stigma associated with the disease 24. In some cases, even a positive result may not be believed when given 25. Clarity about the risk, causes and treatment of TB can help tackle both low awareness of TB and the negative issues surrounding the disease. The stigma associated with TB is particularly damaging as it can lead to late presentation, failure to complete treatment, and discourages patients from identifying their contacts for screening. All of these factors can prevent active or latent TB being considered, detected and treated successfully. Perceptions which contribute to stigma include: Belief that TB infection also means co-infection with HIV Fear that one s relatives will be marginalised Belief that TB reflects poor living conditions Belief that TB results from poor hygiene 26. Barriers or perceived barriers to care often prevent presentation for TB treatment. Communities often have inadequate information on how to access health services, a lack of 23 Public Health Action Support Team. London TB Service Review and Health Needs Assessment September Gerrish K, Ismail M, Naisby A. Tackling TB together: a community participatory study of the socio-cultural factors influencing an understanding on TB within the Somali community in Sheffield. Sheffield Hallam University. Project report National Institute for Health and Clinical Excellence. Tuberculosis evidence review. Review of barriers and facilitators. Oct Health Protection Agency. Beliefs and barriers related to understanding TB amongst vulnerable groups in South East London. South East London Health Protection Unit

26 knowledge about entitlement to primary care and that TB treatment is free of charge irrespective of residency status 27. Awareness among healthcare professionals The HPA/TB Alert survey 2010/11 found that primary care health professionals often underestimated TB prevalence in their locality and needed further information about TB screening and diagnosis, referral and management 28. There is evidence that some patients have had to urge their clinician to consider a diagnosis of TB some returning to their country of origin to achieve a diagnosis. Griffiths and Martineau noted that there is a need for GPs and practice nurses...to lower their thresholds for requesting diagnostic tests for TB, and also for referring those with unexplained symptoms 29. As 55% of the TB in London is non-pulmonary, education of primary health care professionals needs to emphasise that TB must be considered in people with symptoms outside of the chest. Some patients (particularly those with social risk factors) tend to access A&E services 30, so heightened diagnostic awareness in A&E doctors would identify people with TB earlier. Screening Screening involves assessments for active TB disease, latent TB infection or both. It seeks to identify cases early. This is for the benefit of both the individual (providing the opportunity for treatment and health education) and the public (reducing the total burden of TB and onward transmission to others). Because more than 80% of TB notifications are attributable to those born outside the UK, the practice since 1971 has been to screen individuals from countries with a TB rate of more than 40 per 100,000 who intend to stay longer than six months when they arrive in the UK. Under the Port of Arrival (POA) scheme 31 individuals who meet the criteria outlined above should be offered a chest X-ray to detect active pulmonary TB, and be referred to TB services for diagnosis and treatment, where required. However, a 2006 audit of the POA scheme 32 found that POA screening was not being applied consistently across the country or systematically within the ports themselves, and variation existed in the provision of medical inspection facilities (for instance, only Heathrow and Gatwick airports had X-ray machines). Chest radiology also does not detect latent TB the source of the majority of TB in recent migrants 33 which would be diagnosed using tests such as IGRA. 27 Jayaweera H. Health and access to healthcare of migrants in the UK. Health briefing papers. Race Equality Foundation National Awareness Survey for TB in Primary Care: National Knowledge Service TB and TB Alert UK, HPA. London, Griffiths C, Martineau A. The new tuberculosis: Raised awareness of tuberculosis is vital in general practice. British Journal of General Practice. 57(535): February Department of Health, Healthcare for Single Homeless People, March Moore-Gillon J, Davies PDO, Ormerod LP. Rethinking TB screening: politics, practicalities and the press. Thorax Srivastava S, et al. New Entrant Screening Strategies for Tuberculosis A worthwhile cause? St George's Hospital, Mayday Hospital. London Tuberculosis in the UK: Annual report on tuberculosis surveillance in the UK. London: Health Protection Agency Centre for Infections. October

27 Systems to screen new entrants at port of entry are fragmented and inconsistent and there is debate regarding its cost effectiveness as a strategy for TB control 34 as the proportion of people with active pulmonary TB coinciding with the time of their arrival in the UK is very small indeed. In 2004, only a quarter (around 70,000) of those who met the eligibility criteria (around 280,000) received a chest X-ray and around 100 active pulmonary TB cases were detected as a result approximately 0.04% of the total number of arrivals 35. Moore-Gillon et al estimated that around 100,000 of those arrivals had latent TB infection and at least 10,000 would develop active TB in the future 36. Where TB is suspected, various factors significantly reduce the effectiveness of follow up for new entrants. Those with suspected TB are referred to local Health Protection Units (HPUs), which in turn, refer on to TB services. In some cases, the addresses given by new arrivals are incorrect and those identified are lost to treatment. A study in 2005 found that follow up of screening by Communicable Disease Control varies considerably due to the low perceived benefits from doing so There is evidence to suggest that targeted TB testing in primary care has a better yield (that is, identifies a higher proportion of both active and latent TB cases relative to the number tested) and is more cost-effective than port of entry screening. For example, in response to high levels of TB in Hackney, Griffiths et al 39 piloted an outreach programme that promoted TB screening for newly registered patients between June 2002 and October Screening was initially verbal and progressed to TB skin testing if the patient s responses suggested they were in a high-risk group. Of the 50 participating GP practices, 25 trialled the programme (intervention practices) and 25 formed the control group. During the trial period, the proportion of active TB cases identified in intervention practices was 13% higher than control practices. In addition, the average number of days from a patient s first GP consultation to their referral to a TB service was lower in intervention (28 days) than control practices (61 days). Overall, the pilot demonstrated that TB screening in general practice was feasible, relatively inexpensive and increased the proportion of active TB cases identified. While the increase in the detection of latent TB was not significant, TB diagnostic technology has evolved since the Hackney pilot and current guidelines for latent TB testing advocate use of interferon gamma release assay (IGRA), which may be more effective at detecting latent disease 40. Screening of homeless people and problem alcohol and drug users for active TB is carried out in London by the Find and Treat service, comprising a mobile X-ray unit and small team of staff. A recent evaluation of this service by the Health Protection Agency (HPA) found that 34 Hogan H, Coker R. Screening of new entrants for tuberculosis: response to port notifications. Journal of Public Health. VoI. 27, No. 2, pp Advance Access Publication. March Srivastava S, et al. New Entrant Screening Strategies for Tuberculosis A worthwhile cause? St George's Hospital, Mayday Hospital. London Moore-Gillon J, Davies PDO, Ormerod LP. Rethinking TB screening: politics, practicalities and the press. Thorax Coker R. Compulsory screening of immigrants for tuberculosis and HIV. British Medical Journal. 328:298. February Underwood BR, White VLC et al. Contact tracing and population screening for tuberculosis who should be assessed? Journal of Public Health Medicine (1): Griffiths C, Sturdy P et al. Educational outreach to promote screening for tuberculosis in primary care: a cluster randomised trial. The Lancet, May 5-May National Institute for Health and Clinical Excellence. Clinical guideline 117. Tuberculosis clinical diagnosis and management of tuberculosis, and measures for its prevention and control. March

28 this service saves the NHS considerably more than it costs to operate. Although the evaluation did not consider alternative models of provision, the findings support those from Holland that suggest this model is a cost and clinically effective approach to identifying active disease in groups with social risk factors. Detailed statistics on the rates of TB in prisons are not yet available, although the recently introduced national Enhanced TB Surveillance System will allow more accurate estimates in future years. Local studies have estimated the TB prevalence rate in London prisons at 208 per 100,000 41, almost five times the rate for London as a whole. The Find and Treat service has traditionally provided screening for London s prisons. But the frequency of planned visits and high daily turnover of prisoners resulted in poor coverage of the prison population. As a result, the Department of Health has funded a programme to install static digital X-ray machines in five of the seven prisons in the city. The aim is to ensure all new and transferred prisoners are screened as part of routine health checks, though the new equipment is not yet fully operational. The Find and Treat service continues to provide an ad hoc response when prisoners display symptoms, but coverage of the prison population is currently poor. Key issues Poorly informed and inaccurate beliefs about TB in the community is delaying early presentation of the disease and increasing the risk of transmission. A lack of understanding of the disease and its symptoms by healthcare professionals often results in delayed diagnosis or misdiagnosis. Current screening guidelines for TB are neither applied consistently across London nor cost effective in detecting the disease. 6.2 Variability of commissioning TB services are predominantly provided by acute trusts and included in cluster acute contracts. Activity is recorded as inpatient spells and outpatient appointments, although the coding of outpatient activity is insufficiently sensitive to identify it from nationally collated data. In some areas, specialist nursing and outreach work is provided as part of a community services contract. Some initiatives (such as the Local Enhanced Service in Hackney) are funded through primary care or other budgets. Current network arrangements Local service planning, developments and protocols are managed through five TB networks, which align to the cluster arrangements in London (one TB network exists for North East London). However, there are significant differences between the networks. In particular, the membership, host organisation, financial support, administrative support and work undertaken vary considerably. Currently, only the North Central London network incorporates a commissioning function, although both North East and South West London have done so in previous years. The current transition from PCTs to clusters will potentially accentuate this as in some areas, it is not yet clear whether new roles will encompass engagement with the network. 41 Story A, Murad S et al for the London Tuberculosis Nurses Network. Tuberculosis in London: the importance of homelessness, problem drug use and prison. Thorax 62:

29 North West London (NWL) Primarily an educational and supportive forum. HPU based network coordinator. The network coordinator administers the network, sets up meetings, writes regular reports and keeps members up to date on developments in and out of the sector. Also supports a range of operational activities such as cohort review and training activities. The network coordinator has no link with cluster commissioners and no influence on the TB budgets at a trust or cluster level. The network is chaired by a lead clinician and the TB lead from the HPU. Members are staff who work in NWL TB services, PCT public health, microbiologists, consultant paediatricians, DPHs, PCT TB leads and HPU. North Central London (NCL) Cluster-employed network manager based in the Strategy and Planning Directorate. Network manager is supported by cluster finance and contracting functions and has responsibility for the TB budget across NCL. Network chaired by a lead clinician and nurse. Clinical teams, local public health services, local commissioners and HPU represented. North East London (NEL) Considered primarily an educational and management forum. No network manager or coordinator since The network management is led by a clinician and nurse with some support from the HPU TB lead. Not coterminous with a single cluster and little direct engagement with Inner North East London (INEL) or Outer North East London (ONEL). Attended by members of staff who work in NEL TB services, PCT public health, other PCT staff and the HPU. South East London (SEL) Considered primarily a provider network. HPU based network coordinator supports the network meetings. The network is chaired by a DPH but has no budget and no direct lines of communication with cluster commissioners. Membership includes TB nurses, consultant microbiologists, 29

30 consultant paediatricians, chest consultants, HPU, PCT public health leads, and patient representatives/advocates. South West London (SWL) Coordinated by the lead HPU consultant with responsibility for TB with some administrative support to organise meetings. Chaired by a clinician and attended by representatives from TB services, microbiology, paediatrics, HPU and public health. No commissioning input and no real performance management role. Investment in specialist non-medical staff Each TB service comprises a multidisciplinary team, including medical and specialist nursing staff. Some also include administrative and social care support staff. The specialist nurses play a pivotal role as case managers for patients, coordinating their care and the involvement of other agencies as required. In some TB services, the specialist nursing service is provided by the acute trust and funded through tariff. There is a TB Nurse Support tariff within the PbR system, which is rarely used. In most of South East London and some parts of North West London, specialist nursing is provided by the local community health provider and commissioned within the community services contract. As a result, the commissioner spend per notification is higher in these clusters than elsewhere (see figure 11). In North Central London, the TB network has established a cluster-wide nursing service, funded partly by recouping a proportion of the tariff paid to acute trusts and partly through additional investment in effect, a partial unbundling of the tariff. Figure 11 Comparison of commissioner spend and provider investment in specialist staff per TB notification, 2010/11 30

31 See section 8 for notes on financial data Figure 11 shows the level of investment in specialist non-medical staff per TB notification. Staff costs are used in this context as a proxy measure for service costs. Although they exclude non-pay, drug, overhead and other costs, they provide a useful illustration of the different levels of internal investment in specialist TB services in provider organisations. Although some variation is to be expected because of different service configurations, skill mix and caseloads, the graph suggests that the capacity of services in some parts of London has failed to keep up with the number of TB cases. A lower proportion of commissioner spend in North West London and North East London has translated into service capacity. Indeed, there is some evidence of disinvestment in North East London, where consultant sessions have been reduced and the network manager post has been lost. In contrast, a much higher proportion of commissioner spend in North Central London has been invested in service capacity. This proactive approach has ensured that the capacity of the service has remained aligned with the level of commissioner spend. It is notable that the only two sectors to have seen a reduction in TB rates in the last seven years are those where the highest proportion of commissioner spend has been invested in specialist staff. There is clearly scope to roll out a similar approach to other parts of London. Figure 12 TB rates ( ) by sector with linear trend lines 31

32 Commissioning expertise and financial flexibility Following the publication of the Chief Medical Officer s Action Plan in , the Stopping TB in London group developed a series of metrics to monitor performance against the Plan across London. The metrics have been regularly reviewed and updated but achievement remains patchy (see Appendix A). They have been used primarily as a measure of progress rather than a tool to manage performance. Effective commissioning of TB care requires some specialist knowledge to appreciate the relevance of the metrics for TB control to re-direct resources to the most effective parts of the care pathway and to manage the relationship between health services, HPUs, social care and housing agencies. Where a dedicated post does not exist at TB network level, the monitoring and performance management of these relatively detailed indicators competes for time and attention with other acute priorities. At present, only one of the clusters (NCL) has a dedicated TB commissioning resource. The current plans to devolve commissioning to clinical commissioning groups may have a further negative effect on TB control. For example, TB outbreaks are likely to be very infrequent within each commissioning group s area or will cross several financial and administrative boundaries. It is therefore unlikely that any single commissioning group will have sufficient budgetary flexibility to make provision for the high cost of cases of infrequent multidrug and extensively drug resistant TB or an unexpected incident such as a school outbreak within their area. Furthermore, the cost-effectiveness of initiatives such as the Find and Treat service is dependent on implementation across a wide geographical area. Accommodation 42 Department of Health. Stopping Tuberculosis in England: An Action Plan from the Chief Medical Officer. October

33 Stable and safe accommodation for homeless TB patients is essential to enable treatment to be completed to reduce the risk of transmission and to protect public health. Approximately 3% of TB patients in London were homeless in Homeless TB patients tend to present with more clinically advanced disease, be more infectious, have more associated co-morbidities such as drug or alcohol use and mental health issues, and are more likely to have drug resistant strains 44. They are also at greater risk of being lost to treatment services and hence not complete therapy. The provision of temporary accommodation during treatment results in better compliance as patients have stable living and contact arrangements, which in turn, enables directly observed therapy (DOT) to be used. This period of stability also allows access to other local health and social support systems. Eligibility criteria for emergency or temporary accommodation vary from borough to borough across London. The majority of local authorities do not currently include TB as a risk priority. They are also reluctant (and sometimes unable, legally) to fund temporary housing for those with no recourse to public funds such as homeless asylum seekers and undocumented migrants. The NHS is obliged to address the risk to public health posed by patients in this situation and some have required detention under section 2A of the Public Health Act. This usually results in a lengthy inpatient spell of six months or more or an ad hoc arrangement for the NHS commissioner to fund or part-fund temporary accommodation. The latter is financially advantageous for the NHS, costing about 250 per week in contrast to about 1,750 for inpatient care 45. Although local arrangements for temporary accommodation have been negotiated from time to time, they are on a single case basis and subject to a relatively lengthy process involving the submission of a business case locally and negotiation of a waiver of standing financial instructions. As well as creating unnecessary delay, this approach is not systematic or guided by any agreed framework to determine eligibility or accommodation requirements. There is scope to formalise the current ad hoc arrangements, reducing the delay and mitigating the financial risk to individual commissioners. A coordinated approach would also provide an opportunity to engage local government at a London level to seek a contribution to these costs. 43 Health Protection Agency London Regional Epidemiology Unit. Annual report on tuberculosis surveillance in London. London; Story A et al. TB and Housing: Meeting the needs of homeless and hard to treat TB patients in London: Personal communication, L Altass, North Central London Network Manager for TB 33

34 Key issues There is significant variation in the configuration and governance of the five TB networks across London. The majority have few links to commissioning clusters and no performance management role. Those managing acute contracts in clusters have little knowledge of TB and in general, poor access to specialist expertise. As a result, the performance management of providers fails to take account of the London TB metrics or the key features of TB control. Although provider trust income increases as the TB rate rises, the capacity of TB services often does not. It is financially advantageous for the NHS to fund temporary accommodation rather than hospital stays for a small group of TB patients. Although an ad hoc process has evolved to facilitate this, it can be lengthy and is not subject to any systematic control. 6.3 Variability of service provision Some variability in service provision is to be expected and is appropriate when planned in response to local need and demographics. However, many of the differences in current service provision cannot be attributed to these factors and appear to be the result of historical precedent and organic development rather than planning. As a result, some TB services are not making the best use of the resources available to them, particularly with regard to the skill mix of staff deployed, the use of directly observed therapy (DOT) and contact tracing 46. The use of BCG vaccination also varies considerably. Availability of specialist expertise TB services are generally led by either respiratory or infectious disease clinicians who dedicate only a small number of sessions per week solely to TB. As a result, clinical expertise in the disease is limited to a handful of individuals in each sector. This is particularly relevant in areas of relatively low incidence, where most clinicians treat very few, if any, complex cases. Although there is little TB-specific evidence of poorer outcomes for people treated by less experienced clinicians, there is a substantial body of evidence relating to similar complex infectious diseases, such as HIV, that shows improved patient outcomes when treated by teams with larger caseloads and greater experience of complex cases. NICE recommends that treatment of complex cases is managed only by physicians with substantial experience in drug-resistant TB in hospitals with appropriate isolation facilities (negative pressure rooms) and in close conjunction with the HPA. Although this happens in the majority of cases, there continue to be instances of complex TB cases managed by relatively inexperienced clinicians. 46 Public Health Action Support Team. London TB Service Review and Health Needs Assessment. September

35 Staffing profiles Decisions about the number, type and grading of staff contribute to variability in the workload and effectiveness of TB services and do not relate to the complexity of TB patients cared for by services. For instance, specialist TB nurses (grades 6, 7 or 8) are often used to deliver DOT in the community a service that could be provided as effectively by social care support staff (grades 3, 4 or 5), community pharmacists or third sector workers at significantly lower cost, while allowing senior nurses to tailor their efforts more appropriately. The following table shows the ratios of TB notifications to senior TB nurse, administrative and social care support posts, averaged across each of the five TB sectors and London as a whole. There is considerable variation between and within the TB sectors. While it appears that across the sectors the NICE recommendation for a ratio of one nurse to 40 TB notifications is being met, the aggregate data present a misleading picture. In North West London for instance, the number of nursing posts ranges from one nurse to 22.5 notifications (St Mary s Hospital) to one nurse to 53.4 notifications (Northwick Park Hospital). The availability of social care support within sectors is also extremely varied. Homerton Hospital in North East London has one post to 107 TB notifications, whereas Newham Chest Clinic (serving an area with particularly high TB rates) has no social care staff. It is notable that the sectors with the highest ratios of TB notifications to staff are also those with the highest rates of TB. Ratios of TB notifications to staff TB sector TB rates per 100,000 population* Nurse Admin Social care North Central North East North West South East South West London *Rates are based on a three year average ( ) Directly Observed Therapy (DOT) In accordance with NICE guidelines, all patients should have a risk assessment for adherence to treatment, and DOT should be considered for patients who have social risk factors such as a history of homelessness, imprisonment or problem drug or alcohol use. The use of DOT improves treatment compliance and completion rates. At present, DOT is not always available to those who would benefit from it, or have been assessed as requiring it. The provision of DOT (as reported in the HPA London TB Annual Report 2009, published January 2011) varies widely across London in relation to who receives it and how it is provided (for instance, by outreach workers or clinic based DOT). There is also variation in when it is provided, with some but not all TB services offering DOT outside standard working hours. 35

36 Across London, 8.5% of TB patients receive DOT significantly lower than the 12% of patients with at least one social risk factor who could benefit from receiving it. This suggests divergent practice in the application of NICE guidelines. The approach to risk assessment has not been standardised and there is no London DOT protocol to achieve treatment compliance. Contact tracing Contact tracing is used to find associated TB cases to detect people infected, but without evidence of disease (latent infection), and to identify those not infected for whom BCG vaccination may be appropriate. It is also undertaken to find a source of infection and any coprimary cases. At present, contact tracing is inconsistently applied across London. Performance data are not collected for London and only the North Central London TB sector collects relevant information for use within the sector. A review of TB services found that although 90% of service providers in London had a contact tracing clinic, the numbers of contacts were not routinely reported and anecdotally, there was considerable variation in the number of contacts traced per notified case. This is a concern as contacts with untreated active pulmonary TB can infect individuals per year and suffer worse health outcomes due to delays in diagnosis and treatment. Outreach Many TB patients have significant social risk factors and often require additional support to complete their treatment (for instance, help to secure appropriate accommodation, access benefits and address drug and alcohol problems). Some TB services employ social care support workers to fulfil this role. As the table on the previous page shows, there is a significant variation in the ratio of these workers to TB patients across the sectors ranging from one social care support worker per 98 TB notifications in South East London to one per 567 patients in North West London. This variation remains significant, even when the proportion of patients with social risk factors in each sector is taken into account. The London Find and Treat team (including the mobile X-ray screening unit) supports London s TB services and people whose lifestyles make it more difficult for them to access health services by providing an outreach and screening service. Referrals to the Find and Treat service vary across TB services and sectors. Some TB services with the highest proportions of TB cases with social risk factors appear not to make full use of Find and Treat. BCG vaccination The BCG immunisation programme was introduced in the UK in 1953 and has undergone several changes since then in response to changing trends in the epidemiology of TB. The programme was initially targeted at children of school leaving age (then 14 years), as the peak incidence of TB was in young, working age adults. In 2005, following a continued decline in TB rates in the indigenous UK population, the schools programme was stopped. The BCG immunisation programme is now a risk-based programme, the key part being a neonatal programme targeted at protecting those children most at risk of exposure to TB, particularly from the more serious childhood forms of the disease. 36

37 A recent review of clinical trials and observational studies confirmed that BCG vaccination provides protection against TB. Studies have shown BCG to be 70 to 80% effective against the most severe forms of the disease, such as TB meningitis in children. The Department of Health Green Book on Immunisation recommends BCG vaccination is offered to all infants in areas where the TB rate is at least 40 cases per 100,000 population. In areas with a lower incidence, it recommends a targeted approach where the vaccination is offered to those whose families come from regions of the world where the TB incidence is 40 cases per 100,000 or greater and are at increased risk of exposure to TB infection. Currently, the TB rate (and therefore the decision whether to offer vaccination) is considered at PCT level. As a result, 13 of the 31 PCTs in London should offer BCG to all children under 12 months of age (based on 2010 TB rates). Actual PCT provision of neonatal universal BCG is higher with some PCTs offering universal BCG even though they have TB rates below 40 cases per 100,000. Provision of neonatal BCG varies across London with a few hospitals administering the vaccine before the baby leaves hospital. Babies who do not receive the vaccination in hospital usually receive it in the community when they are between one week and one year old. Actual uptake however varies considerably in those PCTs offering vaccination to all infants, with an average of 64% across London 47. North Central London 75% average North West London 24%-75% North East London 70%-80% South East London 62%-75% South West London No PCTs offer universal BCG. Key issues Current staffing profiles within TB services often do not take account of local incidence, the characteristics of the local population or the skill level required to deliver different services. The provision of directly observed therapy (DOT) varies, but apparently not in relation to need. The approach to tracing contacts of people with infectious TB varies, with no systematic collection of performance or outcome data. Uptake of BCG vaccination ranges from 24% to 80% in the parts of London where it should be offered universally. 47 Public Health Action Support Team. London TB Service Review and Health Needs Assessment. September

38 7. Potential fragmentation of TB control The control of TB is currently coordinated across NHS commissioner organisations, service providers, public health teams and health protection services. Health protection services The UK s Health Protection Agency (HPA) exists to protect the public from threats to their health from infectious diseases and environmental hazards. Specifically for TB, the HPA provides advice and information to the general public, to health professionals and to national and local government. It supports cluster investigation and the cohort review process, collects and analyses TB surveillance data and is involved in discharge planning for drug resistant patients due for hospital discharge. The agency also refers patients identified by Port Health to their local specialist TB clinic. Consultants in communicable disease control are appointed as Proper Officers by local authorities, who have a formal role in the enforcement of health protection legislation. There are four Health Protection Units (HPUs) in London who deliver the HPA s work at local level, each with a named TB lead, and the London Regional Epidemiology Unit with a TB epidemiologist supported by senior scientists. The overall functions of the HPA, including those related to TB control, are expected to transfer to Public Health England (PHE) in Contact tracing and cluster investigation When a TB incident or outbreak is declared, the HPU, in liaison with local TB nurses, carries out a risk assessment to determine the need for further screening and other actions to prevent ongoing transmission. The HPU liaises with all stakeholders in the management of the incident. London HPUs have developed standard protocols for dealing with TB incidents in institutional settings. An example of HPU TB activity around a school incident is outlined below. Case study: Case of smear positive TB in a school setting A case of smear positive TB in a child attending a secondary school is reported to the HPU. The HPU will contact the local TB nurses to get results of the household contact tracing as well as the possible exposure within the school and assess the need for screening in the school. If screening is required, the HPU TB Lead makes contact with the school and would usually set up a meeting at the school to review the case in detail, the exposure of other students and obtain lists of children who require screening. Letters to the students and staff are sent by the HPU Lead who also informs the PCT and the HPA Press Officer. TB nurses carry out the screening and inform the HPU of the results. Contact tracing and cluster investigation HPUs and TB nurses lead the risk assessment for the tracing of contacts of a TB case beyond the household setting and decisions on any necessary public health action. For clusters of TB cases with identical strain types, the HPA has recently appointed two cluster investigators (one based in London) who work with local HPUs. Reference laboratory services The HPA provides reference laboratory services for TB, including species identification, drug susceptibility determination and molecular strain typing. Its laboratories provide leadership, 38

39 standards and quality control for local TB diagnostic services, including microscopy and culture and the detection of drug resistance, and lead in the development and evaluation of new diagnostics. The HPA Reference Laboratories carry out strain typing on all TB isolates, which identifies clusters and assists in contact investigation. Public health services Local public health teams work closely with commissioners, service providers and the HPA to improve and maintain the health of local communities. They play an important role in raising awareness of TB both within communities at risk of developing the disease and among healthcare professionals in a position to detect it at an early stage. Public health professionals have been key members of the existing TB networks and in some parts of London, have developed training packages to improve the knowledge base of non-specialist clinical staff. In addition, public health teams play a key role in the management of local outbreaks of the disease, often providing the funding to support additional contact tracing, screening and vaccination activities. Local authorities and the third sector Although there are some individual examples of good joint working between TB services, commissioners and local authorities, this does not appear to be the norm. Housing departments, in particular, are not well engaged in TB control activities and could play a more substantial role, particularly in contributing to the management of outbreaks and supporting homeless TB patients. Likewise, environmental health services are not routinely involved, although they often have the most relevant expertise in relation to poor living conditions and could contribute to improved contact tracing through, for example, their local knowledge of houses of multiple occupancy. Social services are already involved with some TB patients, particularly those with drug and alcohol or mental health problems and yet the recent review of TB services in London found little evidence of joint working 48. Similarly, there are some examples of involving third sector organisations to support patients through treatment, either commissioned directly as part of a TB care package or indirectly in relation to other factors. There is scope to engage the third sector more systematically in the care of TB. Risks The overall functions of the HPA, including those related to TB control, are expected to transfer to Public Health England (PHE) as part of the current programme of NHS reform. Local public health services will transfer to local authorities and responsibility for commissioning health care services will transfer to local clinical commissioning groups. As a result, the responsibilities for protecting the public, improving the health of local populations and ensuring healthcare needs are met will sit in a variety of separate organisations. There is a significant risk that this separation could impair the response to TB across London and reduce the system s capability to capitalise on opportunities to improve joint working and coordination. 48 Public Health Action Support Team. London TB Service Review and Health Needs Assessment. September

40 It is unlikely that health protection services will have sufficient capacity to establish and maintain working relationships with all clinical commissioning groups and local authorities. Public health services will have a very local focus and may struggle to respond to an infectious disease that crosses administrative boundaries. Likewise, the capability of individual commissioning groups to plan and procure care pathways that include multiple providers and geographical areas will be limited. Key issues There is a significant risk that the separation of health protection, public health and commissioning responsibilities could result in a fragmented approach to TB control in London. 40

41 8. Financial considerations 8.1 Costs of treatment for TB In terms of resource requirements, treatment for TB generally falls into three categories: uncomplicated, complex and exceptional, although there is considerable variation within each. Uncomplicated An uncomplicated case involves an individual being identified relatively early and receiving a prompt diagnosis. Treatment may include a brief inpatient spell or self-managed isolation at home during the infectious stage of the disease (about two weeks) and a six month course of antibiotic medication. Treatment costs for this group can be as low as 1,100 (based on one new outpatient appointment and seven follow up attendances). Complex In cases where treatment is not completed, patients are at increased risk of developing drug resistant TB, and may require lengthy hospital treatment, the use of specialist facilities and more specialist clinical expertise. Some cases of extra-pulmonary TB are also considered complex, either because of the additional input required to manage co-morbidities or the extent of damage caused by the TB bacteria prior to diagnosis. The range of treatment costs for this group varies considerably but usually exceeds 10,000. Examples of complex cases The treatment of patient AB with Isoniazid resistant TB cost nearly 13,000. This included two visits to the GP, one A&E visit, 34 days isolated in a negative pressure room, nine outpatient appointments with a further five post-treatment follow up appointments. The treatment of patient CD with drug sensitive TB cost over 36,000. This included two visits to the GP, two visits to A&E, 10 days in ITU, 21 days in a negative pressure room, 14 days as a general inpatient and 33 outpatient appointments over the course of a year. Due to a severe side effect of one of the drugs this patient is now registered blind and will require personal care for the rest of their life. Exceptional Additionally, a minority of people with complex TB require such extensive acute treatment and follow up care that the costs of treatment can be considered exceptional. This may be because they have developed extensively drug resistant TB (XDR-TB) or that the damage caused by the infection is very severe. Mortality rates in this group are high and many of those who survive require lifelong care and support. There are a handful of these cases each year in London and costs of treatment often exceed 100,000. The number of multi-drug resistant TB cases is increasing in London (see section 2). Earlier detection and better treatment completion rates would prevent the development of many cases of drug resistance and, therefore, reduce the cost of treatment considerably. 41

42 8.2 Activity and finance data challenges As TB is a notifiable disease, there is a substantial body of data available on rates of infection, treatment completion and the demography of those affected. The primary source of information is the London TB Register (LTBR) and the national Enhanced Tuberculosis Surveillance system (ETS), both managed by the HPA. The HPA produces annual reports based on the data to inform local and national TB control. However, the analysis of financial information is more challenging. Acute trust activity is charged to cluster contracts as outpatient and inpatient activity but is not easily identifiable. There is no discrete tariff for TB treatment. It can be reported as respiratory activity, infectious disease activity or paediatrics and can be hidden completely if patients are seen in another specialty for TB related problems (e.g. neurology/neurosurgery/orthopaedics for spinal or CNS disease or HIV clinics for TB/HIV co-infection). Inpatient treatment can include spells in intensive therapy units attracting a different tariff. The majority of TB patients are treated as outpatients but the national NHS activity data sources (HES and SUS) use insufficiently sensitive coding to identify activity by disease, focusing instead on clinic type. Specialist TB nursing is an integral part of TB care but the way it is organised differs across London. In some areas (e.g. Lambeth, Southwark and Lewisham) the nursing service is provided by the community services provider and funded via that contract. In other areas (e.g. North East London) it forms part of the acute service and is included in tariff. In North Central London, a single specialist nursing services covers the whole sector, funded partly by an unbundling of tariff and partly by additional investment. The Find and Treat service is funded separately by clusters and commissioned on their behalf by London Health Programmes. It focuses explicitly on people with high social risk factors (homelessness, drug and alcohol problems, prison populations), providing screening and re-engagement with services. To complicate matters further, some areas commission services from the third sector either directly or as a sub-contract via the local acute trust and at least one borough have a local enhanced scheme for TB screening in primary care in place. Published reviews of the cost-effectiveness of particular interventions (including those considered by NICE) are predominantly academic in nature and focus on the cost to provider rather than the price paid by commissioners. As individual care packages can vary considerably and national tariffs are calculated using averages across whole services, there is often little correlation between the cost of treatment and price paid when considered at individual patient level. 8.3 Financial modelling It has not been possible to identify actual NHS spend on TB services for the reasons outlined above. A comparative financial model has, however, been developed (Appendix B). The following constructs have been developed from an analysis of HPA surveillance data, a detailed analysis of activity in a few London TB services and Hospital Episode Statistics (HES) data. 64% of TB patients are treated entirely as outpatients and 36% require inpatient treatment. 2% of all TB cases require complex inpatient treatment. 70% of those admitted require five days or less inpatient treatment (24% of the total). 42

43 30% of those admitted require between six and 33 days inpatient treatment (10% of the total). On average, outpatient treatment comprises one first appointment and seven follow up appointments for uncomplicated TB. On average, patients requiring complex inpatient treatment require 20 outpatient appointments. On average, for every confirmed TB case, nine suspected cases will be seen in outpatients. The constructs have been applied to the known TB rates for each borough in London and associated costs calculated. Cross-referencing between data sources has shown that this model is reliable in identifying comparative spend. It does not, however, show actual spend for a number of reasons. Complex cases have been distributed proportionally across all clusters. Although there is no evidence to suggest clustering of complex cases, the actual incidence will be more random. HRG4 tariffs have been used consistently for inpatient and outpatient spells. In reality, provider trusts use a variety of codes with different associated tariffs e.g. respiratory, infectious diseases. The model does not include multiple admissions to different specialties although individual case studies show this is not uncommon. The additional costs of paediatric treatment are not included (all patients are assumed to be adults). This is because it was impossible to correlate national HPA data with local activity data. Paediatric TB is treated within paediatric infectious disease units. The model does not include additional costs for ITU spells. All figures are exclusive of market forces factor. The costs of specialist staff funded outside tariff are shown as actual costs rather than contract prices. As a result, the total spend on TB services shown is an underestimate. A detailed analysis of activity in North Central and North West London suggests actual total spend is in the region of 25million per annum for London as a whole. Figure 13 shows the commissioner spend per notification and the 2010 TB rates. There is little correlation between the two. It might be expected, for example, that areas with a higher incidence of TB would provide additional investment to tackle the problem or that in areas with a low incidence, the costs of maintaining a viable service were higher. Neither of these appears to be true. Figure 14 shows that the majority of cost is associated with outpatient activity. There is no evidence to suggest that a greater proportion of current inpatient activity could be managed as outpatients as only patients requiring isolation or with complications or comorbidities are admitted. Improving treatment completion rates and the subsequent numbers of complex cases would, however, reduce the proportion of activity that requires inpatient treatment. There is also scope to reduce expenditure on community services by adopting a similar approach to unbundling tariffs as North Central London (see section 6.2). 43

44 Figure 13 Comparative commissioner spend per TB notification and sector TB rates, 2010 Figure 14 Proportion of TB spend by service type Expenditure on the pan-london Find and Treat service is considered in more detail in a separate paper. A recent evaluation by the HPA demonstrated the cost effectiveness of this service and showed that decommissioning it would result in a net increase in expenditure for the NHS of between 360K and 640K per annum. 44

45 Key issues The number of multi-drug-resistant cases of TB, and therefore the cost of TB treatment, in London is increasing and is likely to continue to increase unless treatment completion rates and early detection improve. Much of the cost of treating TB is hidden in unattributable outpatient activity, but the total cost is estimated at 25m a year. Those parts of London that have invested in specialist non-medical teams have seen a reduction in TB rates. Improving early detection and treatment completion rates would, in turn, reduce the number of TB cases requiring complex care and the overall cost of TB services. 45

46 9. Summary of key points The extent of TB in London London as a whole has very high rates of TB. Over time there has been a significant increase in rates and number of cases of TB. The rate in some boroughs is more than twice that of the definition used by the WHO for high rates. The majority of cases are in people born overseas, although it often takes several years for them to become symptomatic. London has large numbers of socially and medically complex cases of TB. This includes those with HIV infection as well as risk factors for poor treatment completion and onward transmission to others such as homelessness, drug or alcohol problems, a history of imprisonment or mental health issues. The majority of cases in London are caused by the reactivation of latent TB, the identification of which is costly and the treatment of which carries a clinical risk. Many individuals who start treatment for this do not complete their prescribed course. International comparisons London has the highest rate of TB of any major city in Western Europe. There have been some significant local initiatives in other European cities that have contributed to a reduction in their TB rate and should inform the model of care for London. New York was in a similar position to London in the early 1990s and brought its TB rate down through investment in services, a multi-faceted strategy and a coordinated, multiagency effort. Policy framework The care and treatment of TB is subject to several important pieces of guidance, although both compliance with existing guidelines and implementation of new recommendations is patchy across London. Delays in detection and referral Poorly informed and inaccurate beliefs about TB in the community is delaying early presentation of the disease and increasing the risk of transmission. A lack of understanding of the disease and its symptoms by healthcare professionals often results in delayed diagnosis or misdiagnosis. Current screening guidelines for TB are neither applied consistently across London nor cost effective in detecting the disease. Variability of commissioning There is significant variation in the configuration and governance of the five TB networks across London. The majority have few links to commissioning clusters and no performance management role. Those managing acute contracts in clusters have little knowledge of TB and, in general, poor access to specialist expertise. As a result, the performance management of providers fails to take account of the London TB metrics or the key features of TB control. 46

47 Although provider trust income increases as the TB rate rises, the capacity of TB services often does not. It is financially advantageous for the NHS to fund temporary accommodation rather than hospital stays for a small group of TB patients. Although an ad hoc process has evolved to facilitate this, it can be lengthy and is not subject to any systematic control. Variability of service provision Current staffing profiles within TB services often do not take account of local incidence, the characteristics of the local population or the skill level required to deliver different services. The provision of directly observed therapy (DOT) varies, but apparently not in relation to need. The approach to tracing contacts of people with infectious TB varies, with no systematic collection of performance or outcome data. Uptake of BCG vaccination ranges from 24% to 80% in the parts of London where it should be offered universally. Risk of fragmentation There is a significant risk that the separation of health protection, public health and commissioning responsibilities could result in a fragmented approach to TB control in London. Financial considerations The number of multi-drug-resistant cases of TB, and therefore the cost of TB treatment, in London is increasing and is likely to continue to increase unless treatment completion rates and early detection improve. Much of the cost of treating TB is hidden in unattributable outpatient activity, but the total cost is estimated at 25m a year. Those parts of London that have invested in specialist non-medical teams have seen a reduction in TB rates. Improving early detection and treatment completion rates would, in turn, reduce the number of TB cases requiring complex care and the overall cost of TB services. 47

48 APPENDIX A: Summary of progress against London TB Metrics 48

49 Table : Proportion of new TB notifications in London residents completing treatment within 1 year of notification by year in which they completed treatment by sector of notifying clinic - reported to the London TB Register Notifying Clinic % of all TB cases completing treatment North Central Edgware TB Clinic 86.3% 86.3% 83.1% 91.7% Great Ormond Street Hospital 83.3% 85.7% 61.5% 46.2% North Middlesex Hospital 81.0% 84.4% 84.9% 89.7% Royal Free 88.9% 77.6% 74.4% 86.1% UCLH TB Service 84.6% 80.7% 82.4% 86.7% Whittington 82.8% 88.4% 79.1% 89.2% North Central Total 84.0% 83.3% 81.0% 87.6% North East Havering TB Service 73.8% 77.6% 83.3% 77.3% Homerton 83.6% 88.5% 79.5% 87.2% King George Hospital 83.6% 84.4% 78.3% 85.2% London Chest Hospital 82.9% 84.8% 80.5% 79.1% Newham Chest Clinic 82.3% 86.2% 87.3% 87.3% Whipps Cross University Hospital 87.0% 89.9% 93.7% 95.0% North East Total 83.1% 85.9% 83.7% 85.1% North West Central Middlesex Hospital 88.6% 82.2% 89.9% 88.6% Charing Cross Hospital 82.3% 84.8% 80.4% 85.2% Chelsea & Westminster 84.9% 88.2% 88.5% 90.0% Ealing Hospital 79.7% 75.5% 86.1% 89.7% Hammersmith Hospital (ICH NHS Trust) 78.5% 88.7% 75.0% 84.0% Hillingdon Hospital 74.5% 89.0% 80.5% 82.8% Northwick Park Hospital 88.8% 87.4% 89.4% 88.4% Royal Brompton 20.0% 66.7% % St Mary's Hospital (ICH NHS Trust) 87.6% 87.5% 85.2% 91.0% West Middlesex University Hospital 60.4% 76.9% 61.5% 77.2% North West Total 81.4% 83.8% 83.4% 86.7% South East Bromley TB Service 75.0% 81.8% 93.3% 90.0% Guy's & St Thomas' Hospitals 81.8% 87.2% 82.3% 87.0% King's College Hospital 75.9% 88.9% 91.6% 92.1% Queen Elizabeth Hospital 84.8% 68.4% 86.9% 88.6% Queen Mary's Hospital 92.3% 91.7% 80.0% 100.0% University Hospital Lewisham 92.9% 77.3% 85.1% 87.7% South East Total 82.5% 81.4% 86.3% 88.9% South West Croydon University Hospital 88.9% 79.4% 84.8% 88.1% Epsom & St Helier NHS Trust 87.2% 91.8% 89.2% 84.8% Kingston Hospital 74.4% 71.9% 77.8% 88.2% St George's Hospital 81.5% 83.4% 86.3% 86.2% South West Total 83.6% 82.4% 85.3% 86.9% Non-LTBR Clinics* 42.9% 77.8% 60.0% 63.6% London Totals 82.6% 83.8% 83.7% 86.6% >=85% % <80% 49

50 50