CHRONIC DISEASE IN NHS GREATER GLASGOW & CLYDE Insights from Local Enhanced Services Programme 2012-2014 Dr Anne Scoular, Consultant in Public Health Medicine | December 2014 Contents Index to figures & tables 3 Executive summary 6 Foreword 8 I Introduction 9 II Population need 11 2.1 Registered diagnoses 11 2.2 Disease prevalence 12 2.2 Premature disease 14 2.3 Multimorbidity 15 III Clinical Service Delivery & Key Outcomes 16 3.1 Coronary Heart Disease 17 3.2 Type 2 Diabetes 28 3.3 Stroke/TIA 37 3.4 Chronic Obstructive Pulmonary Disease (COPD) 44 3.5 Left Ventricular Systolic Dysfunction (LVSD) 52 IV Conclusion 62 V Appendix: CDM LES Review Final Report 62 2 Index to figures and tables Figure 1: Total chronic disease diagnoses in NHSGGC: 2012/13 vs 2013/14 (n) 11 Figure 2: Total number of chronic disease diagnoses in 2013/14, by partnership 11 Figure 3: Crude disease prevalence (per 1,000), b y partnership, 2013/14 12 Figure 4: Crude disease prevalence (per 1,000), NHSGGC: 2012/13 vs 2013/14 12 Figure 5: % change in crude disease prevalence, by area: 2012/13 vs 2013/14 13 Figure 6: Age & sex standardised disease prevalence (per 1,000), by area, 2013/14 13 Figure 7: Premature male & female disease prevalence, by partnership, 2013/14 14 Figure 8: CDM LES patient cohort 2013/14: number of co -existent diagnoses 15 Table 1: CDM LES patient cohort 2013/14: details of co -existent diagnoses 15 Figure 9: CHD patients: age distribution 18 Figure 10: CHD patients: ethnicity recording & overview 18 Figure 11: CHD patients: detail of non -white ethnic groups, by partnership area 19 Figure 12: CHD patients: deprivation profile, by partnership area 19 Figure 13: CHD patients: exception coding, by partnership area 20 Figure 14: CHD patients: reason for exception coding, by partnership area 20 Figure 15: CHD patients: smoking status, by residential deprivation quintile 21 Figure 16: CHD patients: current smokers in each par tnership area 21 Figure 17: CHD patients who smoke: smoking cessation referrals, by deprivation quintile 22 Figure 18: CHD patients who smoke: smoking cessation referrals , by partnership area 22 Figure 19: CHD patients: alcohol use 23 Figure 20: CHD patients: lipid control, by deprivation status 23 Figure 21: CHD patients: lipid control, by ethnic subgroup 24 Figure 22: CHD patients: lipid control, by age group 24 Figure 23: CHD patients: lipid control, by partnership area 25 Figure 24: CHD patients: % within target cholesterol range, by practice 25 Figure 25: CHD patients: % with AF, by age 26 Figure 26: CHD patients with AF: % receiving warfarin therapy, by age group 26 Figure 27: CHD patients with AF: therapy by partnership area 27 Figure 28: CHD patients with AF: % receiving warfarin therapy, by practice 27 Figure 29: Type 2 diabetes: age distribution 29 Figure 30: Type 2 diabetes: ethnicity recording & overview 29 Figure 31: Type 2 diabetes: detail of non -white ethnic groups, by partnership area 30 Figure 32: Type 2 diabetes: deprivation profile, by partnership area 30 Figure 33: Type 2 diabetes: exception coding, by deprivation quintile 31 Figure 34: Type 2 diabetes: exception coding, by partnership area 31 Figure 35: Type 2 diabetes: smoking status, by residential deprivation quintile 32 Figure 36: Type 2 diabetic patients who smoke: smoking cessation referrals, by deprivation quintile 32 3 Figure 37: Type 2 diabetic patients who smoke: smoking cessation referrals, by partnership area 33 Figure 38: Type 2 diabetic patients: alcohol use 33 Figure 39: Type 2 diabetic patients: recorded levels of physical activity, by deprivation quintile 34 Figure 40: Type 2 diabetic patients: referrals to physical activity services, by partnership 34 Figure 41: Type 2 diabetes: % cohort with HbA1c values of 48-59 mmol/l, by deprivation quintile 35 Figure 42: Type 2 diabetes: % cohort with HbA1c values of 48-59 mmol/l, by partnership area 35 Figure 43: Type 2 diabetes: % cohort with HbA1c values of 48-59 mmol/l, by ethnic group 36 Figure 44: Type 2 diabetes: % cohort with HbA1c values of 48-59 mmol/l, by practice 36 Figure 45: Stroke/TIA patients: age distribution 38 Figure 46: Stroke/TIA patients: ethnicity 38 Figure 47: Stroke/TIA patients: deprivation profile, by partnership area 39 Figure 48: Stroke/TIA patients: exception coding, by deprivation quintile 40 Figure 49: Stroke/TIA patients: exception coding, by partnership area 40 Figure 50: Stroke/TIA patients: smoking status, by residential deprivation quintile 41 Figure 51: Stroke/TIA patients who smoke: smoking cessation referrals, by deprivation quintile 41 Figure 52: Stroke/TIA patients who smoke: smoking cessation referrals, by partnership area 42 Figure 53: Stroke/TIA patients: alcohol use 42 Figure 54: Stroke/TIA patients: functional problems , by age group 43 Figure 55: Stroke/TIA patients: recorded use of services relative to scale of functional problems 43 Figure 56: COPD patients: age distribution 45 Figure 57: COPD patients: ethnicity recording & overview 45 Figure 58: COPD patients: deprivati on profile, by partnership area 46 Figure 59: COPD patients: exception coding, by deprivation quintile 47 Figure 60: COPD patients: exception coding, by partnership area 47 Figure 61: COPD patients: smoking status, by residential deprivation quintile 48 Figure 62: COPD patients who smoke: smoking cessation referrals, by deprivation quintile 49 Figure 63: COPD patients who smoke: smoking cessation referrals, by partnership area 49 Figure 64: COPD patients: alcohol use 50 Figure 65: COPD patients: MRC grade, by age group 50 Figure 66: COPD patients: referral to pulmonary rehabilitation, by partnership area 51 Figure 67: COPD patients: referral to pulmonary rehabilitation, by practice 51 Figure 68: LVSD patients: age distribution 53 Figure 69: LVSD patients: aetiology 53 Figure 70: LVSD patients: ethnicity 54 Figure 71: LVSD patients: deprivation profile, by partnership area 54 Figure 72: LVSD patients: exception coding, by deprivation quintile 55 Figure 73: LVSD patients: exception coding, by partnership area 55 Figure 74: LVSD patients: smoking status, by residential deprivation quintile 56 Figure 75: LVSD patients who smoke: smoking cessation referrals, by deprivation quintile 56 4 Figure 76: LVSD patients who smoke: smoking cessation referrals, by partnership area 57 Figure 77: LVSD patients: alcohol use 57 Figure 78: LVSD patients: NYHA class, by partnership area 58 Figure 79: LVSD patients: cardiac therapy prescribed within last 12 months, by NYHA Class 59 Table 2: Detail of cardiac drug therapy in preceding 3 months, by NYHA Class 60 Figure 80: LVSD patients: overview of drug therapy over preceding year, by number of agents 61 Figure 81: Percentage of practice LVSD cohort receiving one or more therapies 61 5 Executive summary As at 31 March 2014, 146,461 patients were registered on one or more LES CDM disease registers, an increase of 2.4% compared with the 2012/13 cohort of 143,077 patients. A total of 194,714 chronic disease diagnoses was recorded, representing an average of 1.3 conditions per patient. Of the five disease areas, CHD was the most prevalent, closely followed by Type 2 diabetes. COPD prevalence which showed a threefold variation across the NHSGGC partnerships. Crude prevalence of premature disease showed marked spatial variation and was generally highest in partnership areas with more intense socioeconomic deprivation. The gradient in premature disease prevalence was steepest for COPD and least marked for CHD. For all conditions apart from COPD, the prevalence of premature disease was substantially higher in men than in women. ‘Multimorbidity’ (generally defined as two or more co-existent chronic conditions) was present in 38,411 (26%) of the current cohort. CHD patients represent the largest single disease register in the CDM programme, closely followed by Type 2 diabetes. The proportion of patients who belonged to non-white ethnic subgroups was highest for Type 2 diabetes and CHD, however there was considerable variability at local partnership level. In South Glasgow, for example, approximately 1 in 4 Type 2 diabetic patients belonged to an ethnic minority subgroups. The deprivation distribution of patients with all disease areas differed strikingly across partnerships, but was most polarised towards SIMD quintile 1 areas in COPD and CHD. Exception coding provides some insights into the proportion of patients who were not sought for review. For all disease areas, exception coding rates were higher in residents of more socio-economically deprived neighbourhoods, although there was considerable variability between different partnerships. Overall, the magnitude and distributional patterns of exception coding are likely to be contributing to inequalities in health status within NHSGGC and require further attention in consultation with practices. The proportion of patients exception coded was lowest for patients with LVSD (11%) and highest for COPD patients (19%). Smoking is a major risk factor for all five disease areas within the current CDM programme. High smoking prevalence remains a challenge, particularly in residents of the most deprived neighbourhoods. Smoking cessation support is a powerful clinical intervention in its own right, however relatively low proportions of patients were reported as ‘ready to stop’ and this subgroup had variable recorded rates of referral to smokefree services. However, more encouragingly, there was considerable variability in these proportions at local level and referral rates of patients at the ‘right’ stage of change were consistently higher in residents of the most deprived areas. Overall, however, tobacco continues to generate substantial additional morbidity and mortality within a population already bearing a heavy burden of chronic disease. This situation can be improved by tackling overall rates of smoking in areas of deprivation and further upskilling clinicians with behavioural brief interventions suitable for busy primary care professionals, building on the many novel ideas for improving the health promoting opportunities within the CDM programme identified by general practice staff themselves involved in the last year of the Keep Well programme (see separate report). Specific clinical indicators are presented for each disease area:  In CHD care, achieving target total cholesterol values of ≤5 mmol/l was a NICE/European Cariguideline- informed recommendation in the 2013/14 contract year, however this was achieved in only two thirds of patients overall and an even lower proportion (60%) among CHD patients who lived in the most deprived neighbourhoods compared with residents of more affluent areas (67%). Although ‘treating to target’ has subsequently been the subject of considerable debate, there remain very strong trial data that more aggressive lipid lowering is associated with a continued reduction in mortality and adverse cardiovascular events. 6  For Type 2 diabetes, recommended levels of physical activity (which has a major independent beneficial effect on glycaemic control) were reached by only 38% of the cohort in the most deprived SIMD quintile and 50% in the least deprived and referral rates to physical activity services were uniformly low (2%) across all deprivation quintiles. Overall, a minority (18,760; 36%) of Type 2 diabetic patients had HbA1c values compatible with good glycaemic control, with poorer glycaemic control was poorer in black ethnic minority subgroup and among those with no recorded ethnicity, which may may reflect gaps in structured, systematic care.  In stroke/TIA patients, a high proportion had functional problems which are highly amenable to therapy, however recorded use of services appeared relatively low, although this may be partly explained by incomplete Read code capture of the full range of relevant services.  In COPD, pulmonary rehabilitation should be offered to all people with COPD at MRC Grade 3 and above who consider themselves functionally disabled by their condition. However, only around half of patients in this category are referred, with considerable variation between partnerships and between individual practices.  Clinical indicators for the LVSD programme suggest that a number of areas require attention. Firstly, there are issues concerning the diagnostic precision of Read coding. A separate, detailed audit undertaken within the local CDM programme in 2013 identified considerable variability in Read coding for heart failure. This highlighted two broad types of issues: firstly, the extent to which patients with heart failure were captured on any heart failure register; and secondly, the validity and precision of documented subtypes of heart failure diagnoses. There is a need to clean practice heart failure registers, standardise secondary care discharge letters and enhance support to improve and maintain accurate coding across the whole system in the future.  Functional capacity assessment is a central component of clinical care for patients with heart failure. In NHSGGC, however, over one third of LVSD patients within the primary care CDM programme had no recorded NYHA class, although this proportion varied from 20 to 49% across different partnerships.  Therapeutic management of LVSD patients also requires more detailed audit. 4,600 (41%) patients appeared not to have been recently (ie in the previous three months) prescribed a beta blocker and 3,426 (31%) appeared not to have been prescribed an ACE-I or ARB in the same period. Over a longer retrospective time period (12 months), 3,083 (28%) of patients appeared not to have been prescribed a beta blocker and 2,185 (20%) had not record of having received either an ACE-I or ARB. It is quite conceivable that these apparent prescribing anomalies may be artefactual (as a result of extraction problems) or a result of the inaccurate disease coding previously discussed, however it is a matter of concern that the two main groups of drugs recommended in LVSD may not be prescribed in a sizeable number of LVSD patients and needs more detailed analysis as a matter of urgency. In summary, this report provides local intelligence on the scale and clinical performance of the CDM programme in each of the five disease areas. Specific topics are highlighted for a possible future improvement focus and it is proposed that these are clinically led by the CDM Review and Implementation Group as part of the ongoing programme redesign process established in 2013, following the CDM LES programme review (Final report embedded as Appendix 1). Greater emphasis is required on optimising coverage and delivery of clearly defined, evidence based interventions that hold the greatest potential to change clinical outcomes, together with investment of dedicated clinician time in person-centred consultations, workforce development to ensure that the most effective use is made of that time, further improvements on the recent electronic template changes and ‘whole system’ supportive infrastructure to ensure that a well organised, structured approach to improving outcomes and health status for the increasing numbers of patients who will in the future need effective community based care for chronic disease. Foreword 7 This paper provides insights into the clinical performance of NHS Greater Glasgow & Clyde’s Chronic Disease Management (CDM) programme, which delivers practice based CDM care for patients with five major chronic diseases:  Coronary Heart Disease (CHD)  Type 2 Diabetes  Stroke/TIA  Chronic Obstructive Pulmonary Disease (COPD)  Left Ventricular Systolic Dysfunction (LVSD). The NHS Greater Glasgow & Clyde (NHSGGC) CDM programme is delivered in primary care, but strongly underpinned by a whole population perspective across all aspects of service planning, coordinated by a multidisciplinary planning group, with input from a Consultant in Public Health Medicine, a GP Medical Editor, the Primary Care Support Team, four Managed Clinical Networks (Heart, Diabetes, Respiratory & Stroke) and specialists in Information Management & Technology and in Health Improvement. The programme is delivered via a General Medical Services (GMS) Local Enhanced Services (LES) contractual arrangement, which evolved from a pilot programme initiated in 1999: ‘Glasgow’s Responsive Angina Secondary Prevention Programme’ (GRASPP), targeting patients with coronary heart disease (CHD). With the advent of the new GMS contract in 2004, the GRASPP programme became a CHD LES, along with analogous contracts for Type 2 Diabetes and Stroke/Transient Ischaemic Attack (TIA). These service models were incrementally extended to Chronic Obstructive Pulmonary Disease (COPD) and Left Ventricular Systolic Dysfunction (LVSD) from November 2010. Data extracted primarily for the purposes of LES contracting also generate valuable understanding of the needs and characteristics of patients with chronic disease. Unfortunately, problems with data extraction led to a gap in production of ‘Insights’ reports between 2010 and 2012, however these problems are now resolved and I am pleased to provide, in the report that follows, an overview of these aspects of the CDM programme for the contract year 2013/14. This will be my last report as Public Health lead for the CDM programme, so I would like to take this opportunity to wholeheartedly thank Frances Paton and her excellent team in Information Services for their enormous support with data production for the ‘Insights’ reports over the past seven years. A huge thank you is also due to everyone involved in delivering and supporting the entire CDM programme; every practice nurse, our Primary Care Support team, our Medical Editor and our MCNs have all played a vital role in the programme’s success and this continuing dedication to patient-centred care will ensure that it continues to develop and go from strength to strength in the future. Dr Anne Scoular, Consultant in Public Health Medicine Keep Well/Enhanced Services Data Group Chair December 2014 8 I Introduction Chronic disease management (CDM) is a generic term for systematic delivery of coordinated healthcare for populations with established LTCs. Its overall aim is to transform care for patients with chronic illnesses from acute and reactive to proactive, planned and population-based. CDM has the following key features (Box 1): Box 1: Key features of Chronic Disease Management (CDM) model  Population orientation (ie defined by a condition, not a service)  Comprehensive, multidisciplinary care  Integrated care continuum across entire disease cycle  Effective systems for coordination  Active client–patient management tools (health education, empowerment, self-care)  Structured, evidence-based approach  Use of guidelines, protocols and care pathways  Deployment of information technology and system solutions  Continuous audit and quality improvement The original evidence underpinning CDM came from a review of interventions to improve care for various chronically ill populations. A subsequent Cochrane Collaboration review concluded that several elements work synergistically together to strengthen provider/patient relationships and improve health outcomes. This Cochrane Review concluded that four elements had the most direct relationship to health outcomes: i) increasing provider expertise and skill ii) educating and supporting patients iii) making care delivery more team-based and planned iv) making better use of registry-based information systems. These components now form the basis of the CDM model which has been delivered in NHSGGC for over 15 years. Accumulated evidence continues to support this model as an integrated framework for improving both the processes and outcomes of healthcare for patients with chronic disease. However it is vital that structured approaches (building on those employed in the recently discontinued Keep Well programme; see below) are designed into the CDM system to ‘inequalities-proof’ the basic CDM model, as there is abundant evidence that interventions reliant on individual patients’ motivation to seek and maintain CDM care generally worsen existing inequalities, unless systematic ‘inequalities proofing’ is used. CDM is underused in women and older people with CHD, and those in lower socioeconomic strata are more likely to have angina but less likely to consult their doctor. Keep Well was launched in October 2006, part of the Scottish Government’s 2005 policy ‘Delivering for Health’. Its initial focus was on health inequalities in cardiovascular disease (CVD). Its core element, the Keep Well consultation (or ‘health check’), targeted individual residents in Scotland’s most disadvantaged areas, to whom it offered appropriate interventions to address modifiable CVD risk factors and health behaviours (eg high blood pressure, high cholesterol, smoking, overweight etc). In NHSGGC, the Keep Well health check also sought to address the impact of wider social issues including poverty, employability, literacy and mental health & wellbeing. In March 2010, in advance of publication of its final national evaluation report, the Scottish Government announced its intention to mainstream the programme across NHS Scotland from April 2012. However this decision was overturned a few years later, in December 2013, when the Scottish Government announced its decision to end funding for Keep Well from March 2017. NHSGGC had evaluated Keep Well extensively, which indicated that variation at three critical points largely explained practice-to-practice variations that were likely to curtail the programme’s overall effectiveness: 9  Before engagement (effectiveness of engaging subgroups with greatest need)  At the consultation (effectiveness of initiating changes in health literacy, risk factors, health associated behaviours and use of wider practice systems)  After the consultation (sustained change following the consultation, both at individual patient level and in the overall responsiveness of local health improvement services) An anticipatory care toolkit was developed during 2013 to translate these evaluation findings into practical actions for practices; the toolkit has transferable value in wider aspects of chronic disease management in primary care. In the last year of Keep Well (2014/15), NHSGGC commissioned practices to self-assess their existing CDM programmes against the toolkit and were encouraged to further develop the toolkit in ways that best fitted their own local context and systems. Flexibility, innovation and practical approaches to service delivery was actively encouraged and practices were supported to share details of these activities, both with other practices and with NHSGGC, using simple and clearly structured electronic methods and a highly successful webinar series. This has generated a rich collection of ideas, evidence and actions developed by practices in the context of their own CDM programme, with enormous transferability of organisational learning to other practices that will further strengthen and enrich the CDM programme across the entire NHSGGC area if supported and sustained. A brief summary of the full report from the Keep Well final (2014/15) contract year will be available in early 2015 and this will bring useful transferable learning for all involved in the CDM programme. 10