Evidence-based chronic heart failure management programs: reality or myth?
- 1Monash University, Melbourne, Victoria, Australia
- 2Australian Catholic University, Melbourne, Victoria, Australia
- 3University of Melbourne, Melbourne, Victoria, Australia
- 4Curtin University, New South Wales, Australia AU: please provide city/town data for this affiliation
- 5University of Pennsylvania, Philadelphia, Pennsylvania, USA
- 6Baker IDI Heart and Diabetes Institute, Melbourne, Victoria, Australia
- Correspondence to Professor Simon Stewart, Preventative Cardiology, Baker IDI Heart and Diabetes Institute, 75 Commercial Road, Melbourne, Victoria 3004, Australia;
- Accepted 22 October 2008
Background: Chronic heart failure management programmes (CHF-MPs) have become part of standard care for patients with chronic heart failure (CHF).
Objective: To investigate whether programmes had applied evidence-based expert clinical guidelines to optimise patient outcomes.
Design: Prospective cross-sectional survey was used to conduct a national audit.
Setting: Community setting of CHF-MPs for patients after discharge.
Sample: All CHF-MPs operating during 2005–2006 (n = 55). 10–50 consecutive patients from 48 programmes were also recruited (n = 1157).
Main outcome measures: (1) Characteristics and interventions used within each CHF-MP and (2) characteristics of patients enrolled into these programmes.
Results: Overall, there was a disproportionate distribution of CHF-MPs across Australia. Only 6.3% of hospitals nationally provided a CHF-MP. A total of 8000 post-discharge CHF patients (median, 126; IQR, 26–260) were managed via CHF-MPs representing only 20% of the potential national case load. Significantly, 16% of the case load comprised patients in functional New York Heart Association class I with no evidence of these patients having had previous echocardiography to confirm a diagnosis of CHF. Heterogeneity of CHF-MPs in applied models of care was evident with 70% of CHF-MPs offering a hybrid model (a combination of heart failure outpatient clinics and home visits), 20% conducting home visits and 16% an extended rehabilitation model of care. Less than half (44%) allowed heart failure nurses to titrate medications. The main medications that were titrated in these programmes were diuretics (n = 23, 96%), β-blockers (n = 17, 71%), ACE inhibitors (n = 14, 58%) and spironolactone (n = 9, 38%).
Conclusion: CHF-MPs are being implemented rapidly throughout Australia. However, many of these programmes do not adhere to expert clinical guidelines for the management of patients with CHF. This poor translation of evidence into practice highlights the inconsistency and questions the quality of health-related outcomes for these patients.
Chronic heart failure (CHF), a complex, debilitating and typically fatal syndrome1 2 3 exerts a significant burden within the ageing population both on an individual basis and on the healthcare system. The demands on the healthcare organisations are significant and relate to recurrent hospitalisation in high-risk individuals within the 40 000 or more Australians hospitalised with a primary diagnosis of heart failure each year.1 4 It is within this context that CHF management programmes (CHF-MPs) have been developed to optimise the management of those hospitalised with CHF in order to reduce preventable morbid and fatal events.5 The overall efficacy of these programmes has been established through several meta-analyses6 7 8 in particular multidisciplinary CHF-MP’s either delivered in the community or via specialist clinics to reduce the risk of all-cause hospitalisation or mortality by 26% (95% CI 0.63 to 0.87) and 25% (95% CI 0.59 to 0.96) respectively.7 Consequently, CHF-MPs are now a recommended part of the post-discharge management of patients with CHF throughout the world.5 9 10
Since our pilot survey of the initial distribution of CHF-MP’s in Australia (33 programmes in 2004),11 there has been a rapid growth of CHF-MPs throughout the country; predominantly in metropolitan regions. This has been mainly because of an injection of state government funding into chronic disease management programmes in various states, primarily to avoid unnecessary hospitalisations.12 13 Funding for CHF-MPs has been problematic in Australia with three levels of government having differing power and responsibilities concerning healthcare funding: commonwealth, state and local governments. The commonwealth government is responsible for Medicare, which is a universal taxation-funded health insurance and provides the majority of healthcare funding. A proportion of the healthcare funding is then delegated to the state governments who provide a range of healthcare services of which the majority of funding is spent on public hospitals.
As part of the Benchmarking and Evidence-based National guidelines for CHF-MP (BENCH) study,14 we undertook a contemporary, national audit of CHF-MPs. This was to determine if the previously documented heterogeneity in CHF-MPs11 persists despite supportive evidence and available expert guidelines5 9 10 for these programmes of care. Specifically, we examined if and how current CHF-MPs apply evidence-based recommendations from national5 and international guidelines9 10 for the management of patients with CHF to optimise their health outcomes. An outline of the recommendations that are applicable to this study are listed in table 1 including the level of evidence supporting these recommendations.
A prospective cross-sectional survey design was used to conduct a national audit of all CHF-MPs operating in Australia during 2005–2006. The investigator-developed survey was designed to examine the characteristics of each CHF-MP in respect to the specific components of interventions applied and the type of patients being managed on a routine basis. Survey items were designed to identify evidence-based strategies contributing to programme efficacy. An expert panel of five CHF-MP coordinators who each had more than 5 years experience as a heart failure nurse undertook face-validity of the questionnaire. Ethics approval for the study was granted by Deakin University human research ethics committee.
For the purposes of the BENCH study, CHF was defined as:
“A complex clinical syndrome with typical symptoms (eg, dyspnoea, fatigue) that can occur at rest or on effort and is characterised by objective evidence of an underlying structural abnormality or cardiac dysfunction that impairs the ability of the ventricle to fill with or eject blood (particularly during physical activity).” (5, p. 9).
Similarly, the programmes that managed patients with CHF were defined as “A system of coordinated healthcare interventions prospectively designated for and targeted towards patients with a diagnosis of CHF and where patient self-care plays a vital role.”9 16 This definition has been adopted by American Heart Association and is used in international expert guidelines.
Later in the article we also refer to models of care for CHF-MPs. There were four main models of care for CHF-MPs in Australia. Currently there are no specifically recommended models of care stated in national and international guidelines except for the recommendation that a multidisciplinary team approach be used.5 9 10 The models of care applied in this study were: home visits, extended Cardiac Rehabilitation (CR) model, hybrid model and telemonitoring. A home visit model referred to a programme where the heart failure nurse visited the patient in their home. The extended CR model was similar to the CR model where up to 10 patients with CHF attended education and exercise sessions in conjunction with home visits. Any programme that implemented a combination of home visits and group sessions (as in the extended CR model) was classified as a hybrid model. The final model of care was a telemonitoring model where a remote monitoring device was installed into the patient’s home. This device measured the patient’s blood pressure, pulse, weight and so on depending upon the system installed into their home.
Each programme’s model of care was based on the type of visit with the highest number of visits over a 2-month period, excluding one outpatient clinic attendance. For example, one programme over a 2-month period conducted seven group sessions, four inpatient visits, one home visit, one outpatient clinic appointment and two telephone calls. This programme was classified as a hybrid model as the visits varied from group sessions, inpatient visits to a home visit. However if a programme conducted one outpatient clinic visit and three home visits then it was classified as a home visit model.
All CHF-MPs operating in Australia during 2005 and 2006 were identified from a systematic search of the Australian healthcare system. Of 1286 hospitals in Australia,17 294 were excluded as they were day, psychiatric, cancer or rehabilitation hospitals. The remaining 982 hospitals were contacted by telephone and enquiries made concerning the implementation of a CHF-MP.
In terms of specific attributes of the programme coordinators, all fulfilled the following criteria:
Their programme was a specific disease management programme in addition to routine care.
They enrolled patients with CHF following an acute hospitalisation into a prescribed programme.
Their programme focused on ambulatory patients living in their own home rather than institutional care.
Programme coordinators enrolled more than six patients with CHF annually.
The systematic search of the healthcare system identified 62 potential CHF-MPs from 982 hospitals. Overall, only 6.3% of hospitals nationally were associated with a programme. Three programmes were excluded from the survey because of the small number of patients being managed (less than six patients/annum). Two programme coordinators, in different states, declined to participate and two programmes ceased to operate. Overall, 55 of 62 programme coordinators consented to participate in the audit resulting in an 89% participation rate. There was a subsequent 100% response rate to the audit questionnaire. In addition to the programme audit questionnaire, demographic data was collected on a minimum of 20 consecutive patients per programme (48 programmes, n = 1157). This represented 70% of the total number of patients enrolled in CHF-MPs nationally. Programme coordinators also reported that the group of patients selected from their programme typically represented the overall group of patients participating in their programme.
Data analysis of both descriptive and inferential statistics used the statistical package SPSS for Windows 14.0 (SPSS Inc, Chicago, IL). Data cleaning was performed on all of the data including data range and consistency checks. Level of significance was accepted at the 0.05 level (two-sided). The following methods of analysis were used: descriptive statistics to describe the sample and χ2 for discrete variables.
The results of the survey found that CHF-MPs managed a national total of 8068 post-discharge CHF pts (median, 173; inter-quartile range, 30–216). This represented only 20% of the case load for these CHF-MPs as it is estimated that approximately 40 000 patients with CHF are discharged annually from hospitals.1 Only 8% of CHF-MPs were located within rural regions based on ARIA scores.18
Our survey revealed a disproportionate distribution of CHF-MPs across Australia. New South Wales had 27 (48%) programmes equating to 4.2 CHF-MPs/million population, Victoria had 18 (32%) programmes equating to approximately 3.6 CHF-MPs/million population, South Australia had 1.5 CHF-MPs/million population (four CHF-MPs, 7%), Western Australia had one programme/million population (two CHF-MPs, 4%) and Queensland had 0.56 programme/million population (four CHF-MPs, 7%) (figure 1).19
Distribution of CHF-MPs varied greatly between states as did the location of programmes with over half (31, 55%) sited in an acute hospital. There were 19 (34%) programmes located in community health centres with 89% linked to a metropolitan hospital.
Regardless of the locality of programmes, funding sustainability varied. Overall, 18 (32%) CHF-MPs did not have ongoing funding and programme coordinators had to reapply for funding annually. Of these programmes, 11 (61%) programmes were located within an acute hospital compared with 4 (22%) programmes that were community-based programmes. The main source of unsustained funding was from state government (n = 12, 67%). Figure 2 shows the main sources of funding with state government funding being the predominant source (n = 32, 57%), followed by individual hospital funding (n = 20, 36%).
Programme admission criteria
The majority of programme coordinators (n = 43, 78%) had developed criteria for patients to be enrolled into their programme. Primarily, admission criteria consisted of a recent hospital admission and/or diagnosis of CHF.
The baseline characteristics of this cohort are shown in table 2, which depicts a typically old cohort with multiple co-morbidities and mainly ischaemic CHF. The mean ejection fraction was 31.8 (SD 13.6) with the majority of patients being functional New York Heart Association (NYHA) class II.
Almost all of the programmes (n = 51, 93%) enrolled patients in functional NYHA class II and III with 38 (69%) programmes and 51 (93%) programmes, respectively. Nine (16%) programmes enrolled functional NYHA class I patients representing 10% of their case load. The majority of these programmes were located in the community setting. In the programmes that did enrol functional NYHA class I patients, approximately 20% of functional NYHA class I patients had not had an echocardiogram to confirm their diagnosis of CHF. Of these patients that had not had an echocardiogram, 50% had been participating in a programme for longer than six months. Twenty-eight (51%) programmes enrolled functional NYHA class IV patients and these programmes were operated by an acute hospital with a home visit as their model of care.
Staffing of CHF-MPs
The staff involved in the programme varied greatly. It ranged from one nurse to at least four members of a multidisciplinary team. Table 3 shows the health professionals involved in the team within the programmes.
Twelve (22%) programmes involved a nurse as the sole practitioner in the programme and four (7%) programmes had a nurse and one other health professional such as a cardiologist, physiotherapist, pharmacist and a primary care physician.
The educational preparation of heart failure nurses who acted as sole practitioners also varied. At the time of the survey there were no university-accredited cardiac courses specialising in heart failure in the tertiary education sector. Consequently, many of the heart failure nurses who encountered these patients through the programme had no university qualifications (n = 32, 57%) but the vast majority had greater than four years of cardiac experience (n = 42, 75%) mainly working in a coronary care unit. There were two heart failure nurse practitioners. More than half of the heart failure nurses (n = 32, 57%) were given training prior to commencing the role as a coordinator or sole practitioner. The training consisted of attending a workshop on the management of CHF run by an education provider and visiting other CHF programmes within the region.
Model of care applied to CHF-MPs
It was clear from the literature that a multidisciplinary team approach was not only advocated but also deemed more effective. Nurses were often working as case managers and this was within a model of care that was predominantly based on home visits.
The majority of programmes adopted a hybrid approach to the model of care utilised in the programmes. The most common combinations of models were: outpatient clinic reviews, home visits and telephone follow-up. The only models of care that utilised only one approach were: home visits, extended CR model and telemonitoring (figure 3).
Each of the programmes participating in the study was considered on an individual basis for the type and number of visits conducted over a 2-month period. Table 4 illustrates the classification of programmes according to their model of care and the number of other visits they conducted. From table 4, the hybrid approach to model of care predominated in all areas except the telemonitoring model.
The heart failure nurses and the researcher acknowledged that the number of visits within a programme depended upon the severity of patients’ CHF. For example, a patient considered to have functional NYHA class IV symptoms would have more frequent visits than a patient with functional NHYA class I. For the purpose of the study the number of visits was based upon functional NYHA class III patients as 51 (93%) programmes enrolled such patients compared with 38 (69%) programmes enrolling functional NYHA class II patients.
Titration of medications
Regardless of the model of care applied to CHF-MPs, 24 (44%) programmes supported heart failure nurses titrating patient’s medications. Eight (14%) programmes enabled HF nurses to titrate medications according to a predetermined protocol (approved by a cardiologist) or in consultation with medical staff in 21 (38%) CHF-MPs. Of the programmes that enabled coordinators to titrate medications, 79% were located within an acute hospital and 21% within community health centres. The main medications that were titrated in these programmes were diuretics (n = 23, 96%), β-blockers (n = 17, 71%), ACE inhibitors (n = 14, 58%) and spironolactone (n = 9, 38%).
The national BENCH study has highlighted the rapid growth of CHF-MPs. However this has come at the price of significant heterogeneity. Despite the abundance of evidence supporting recommendations in national5 and international9 10 guidelines concerning the management of patients with CHF very few programmes appear to be implementing evidence-based CHF-MPs. Low-risk patients and many of them with no documented evidence of CHF are being recruited into programmes, less than half of the programmes support heart failure nurse titration of medications and there is an inadequate number of programmes to support the number of patients being discharged from hospital with CHF. This study has also highlighted the rapid growth and heterogeneity of CHF-MPs.
When comparing the growth of CHF-MPs between the current study and a previous study11 in 2003, the number of CHF-MPs had doubled over a 3-year period. During 2003–2004, there were 40 000 hospital admissions for patients with a primary diagnosis of CHF1 however only 6.3% of hospitals had a CHF-MP. This represents a 10-fold difference in the number of programmes compared with Europe, where 63% of hospitals are associated with a CHF-MP.20 This is despite the fact that international and national expert guidelines consider CHF-MPs to be an integral component of post-discharge management of patients with CHF.4 9 10 One reason for the disparity in the number of programmes between Australia and Europe was attributable to the relatively recent introduction of CHF-MPs into the Australian healthcare system. That considered, it still remains an under resourced area considering the burden of disease that heart failure places on the healthcare system.
Not only was the number of programmes inadequate according to guideline recommendations5 9 10 (table 1) but also the type of patients recruited into these programmes were not “high risk” patients. National and international guidelines stipulate that programmes should be aimed at recruiting high-risk heart failure patients (functional NYHA class III and IV)5 9 10 (table 1). Approximately 16% of programmes recruited asymptomatic (functional NYHA class I) patients with no evidence of a prior echocardiography. A recent report evaluating government funded CHF-MPs found there was no difference in hospital admissions and ED presentations between patients participating in these programmes and those not participating in a programme.12 One reason for the poor performance of CHF programmes compared with other disease management programmes may have been the recruitment of inappropriate patients (functional NYHA class I) with just under one fifth of programmes enrolling inappropriate patients with no symptoms of CHF (functional NYHA class I). A study by de Busk et al21 showed that CHF-MPs have little effect on improving patient outcomes in low-risk patients (functional NYHA class I). In asymptomatic (functional NYHA class I) CHF patients 20% had not had echocardiography to confirm their diagnosis of CHF and 50% of these patients had been participating in a programme for more than 6 months. This lack of documented evidence of CHF raises the question as to whether or not these patients were diagnosed with CHF. Expert guidelines stipulate that diagnosis of CHF should not be made solely based upon symptoms and signs and that patients should undergo echocardiography.5 9 10
Another area of poor translation of evidence into practice was the titration of medications as also recommended by international guidelines5 9 10 (table 1). In the light of the poor uptake of these guidelines another strategy to optimise key therapies could be the titration of medications by a heart failure nurse. This strategy has been successfully implemented throughout Europe where 76% of their programmes allow heart failure nurses to titrate medications.22 In the current study, 44% of programmes allowed heart failure nurses to optimise medications including a flexible diuretic regime. Unfortunately, the number of programmes titrating medications was found to be less in the current study than in the preliminary study.11 This was probably because of the increased number of community programmes that had been implemented, these programmes rarely titrated medications. Despite level I evidence demonstrating a dose-dependent improvement in mortality and morbidity in patients prescribed β-blockers and ACE inhibitors 5 9 10 23 very few programmes allowed heart failure nurses to titrate medications.
There were several limitations to this study that need to be considered. One limitation was the method of sampling. Convenience sampling was used as the aim of the study was to include all of the CHF-MPs. The researchers were confident that all of the programmes were identified at the time of the survey. A survey design was used as the aim of the study was to determine the characteristics and interventions implemented within CHF-MPs. A final limitation was that barriers and reasons for non-implementation of national and international guideline recommendations was not investigated as the aim of the study was to determine programme characteristics. Further research in this area will be undertaken.
Many major hospitals have implemented a CHF-MP and these programmes are expanding rapidly into the community sector as recommended in national and international guidelines.5 9 10 However, it is evident that heterogeneity between programmes persists and there is a poor translation of evidence-based guideline recommendations into practice. CHF-MPs continue to enrol low-risk patients with no confirmed diagnosis of CHF and there is a lack of optimisation of doses of key therapies. There are also an inadequate number of programmes to manage the epidemic of CHF. Heterogeneity between programmes and non-implementation of evidence-based guidelines may translate into suboptimal care for patients with CHF. The development of national benchmarks (Stage II of the BENCH study) would reduce the variance in suboptimal care as programme outcomes can be compared against a national benchmark. This would enable a service to evaluate its performance, quality and overall cost-effectiveness in order to maintain excellence and optimise patient outcomes. Education of health professionals concerning evidence-based guidelines in addition to the development of national benchmarks must be undertaken to ensure optimisation of patient outcomes and evidence-based national standards in healthcare delivery.
Andrea Driscoll was supported by an NHMRC postgraduate scholarship and Simon Stewart was supported by the NHMRC of Australia.
Funding This research was supported by a National Heart Foundation of Australia Grant-in-Aid.
Competing interests None.