Abstract
Aim
To determine whether an intensive intervention at a heart failure (HF) clinic by a combination of a clinician and a cardiovascular nurse, both trained in HF, reduces the incidence of hospitalisation for worsening HF and/or all‐cause mortality (primary end point) and improves functional status (including left ventricular ejection fraction, New York Heart Association (NYHA) class and quality of life) in patients with NYHA class III or IV.
Setting
Two regional teaching hospitals in The Netherlands.
Methods
240 patients were randomly allocated to the 1‐year intervention (n = 118) or usual care (n = 122). The intervention consisted of 9 scheduled patient contacts—at day 3 by telephone, and at weeks 1, 3, 5, 7 and at months 3, 6, 9 and 12 by a visit—to a combined, intensive physician‐and‐nurse‐directed HF outpatient clinic, starting within a week after hospital discharge from the hospital or referral from the outpatient clinic. Verbal and written comprehensive education, optimisation of treatment, easy access to the clinic, recommendations for exercise and rest, and advice for symptom monitoring and self‐care were provided. Usual care included outpatient visits initialised by individual cardiologists in the cardiology departments involved and applying the guidelines of the European Society of Cardiology.
Results
During the 12‐month study period, the number of admissions for worsening HF and/or all‐cause deaths in the intervention group was lower than in the control group (23 vs 47; relative risk (RR) 0.49; 95% confidence interval (CI) 0.30 to 0.81; p = 0.001). There was an improvement in left ventricular ejection fraction (LVEF) in the intervention group (plus 2.6%) compared with the usual care group (minus 3.1%; p = 0.004). Patients in the intervention group were hospitalised for a total of 359 days compared with 644 days for those in the usual care group. Beneficial effects were also observed on NYHA classification, prescription of spironolactone, maximally reached dose of β‐blockers, quality of life, self‐care behaviour and healthcare costs.
Conclusion
A heart failure clinic involving an intensive intervention by both a clinician and a cardiovascular nurse substantially reduces hospitalisations for worsening HF and/or all‐cause mortality and improves functional status, while decreasing healthcare costs, even in a country with a primary‐care‐based healthcare system.
Despite survival benefit due to new medical strategies, the prognosis of patients with heart failure (HF) remains poor. Studies consistently show 5‐year survival rates between 35% and 60%.1,2,3,4 A prospective study of patients hospitalised for HF showed that about 50% of early readmissions were preventable, with factors such as poor compliance with medication or diet, suboptimal discharge planning and follow‐up, and inadequate self‐management by patients in case of worsening symptoms of HF being the most important determinants of deterioration.5 HF management programmes could be the answer for this. Many randomised studies of HF management programmes have been performed in the United States, Australia and Europe.6,7 Methodological limitations of these studies include the short follow‐up periods and relatively small sample sizes, whereas heterogeneity in setting and intervention programmes8 hampers the applicability of the results. Of the 21 randomised trials mentioned in a recent review,8 five showed a reduction in the combined end point of all‐cause readmissions and/or mortality,9,10,11,12,13 two studies reported a statistically significant reduction in the combined end point of readmission rates for HF and/or death,14,15 and only 1 reported a statistically significant reduction in total mortality.13 A study on discharge education published later showed a reduction in the total number of deaths and days in hospital.16 A study on telephonic disease management showed a statistically significant survival benefit.17 Overall, multidisciplinary HF management programmes seem to be effective, but they have to be validated for various settings. In several articles,6,18,19 it has been suggested that greater benefit could be expected from a HF management programme if a clinician trained in HF is more directly involved. One trial demonstrated a beneficial effect with an intervention based on a physician‐directed HF clinic assisted by nurses and the patient's primary care physician.20 A HF clinic with an intensive, standardised intervention by a combination of a clinician and a cardiovascular nurse has not been studied yet. This was one of the justifications for our prospective, randomised parallel group trial aimed at estimating the effects of an intensive physician‐and‐nurse‐directed intervention on hospitalisation for worsening HF and/or all‐cause mortality and on functional status. In addition, we wondered whether such a HF clinic would be beneficial in countries such as The Netherlands and the UK, where general practitioners act as gatekeepers for secondary care, with high‐quality guidelines for many chronic diseases, including HF.
Methods
Patients
We performed a parallel group, randomised controlled trial, with measurements at baseline, after 3 months and at the end of the study at 12 months. The local ethics committees of the participating Deventer and Alkmaar hospitals approved the study.
Patients either hospitalised or visiting the cardiology outpatient clinic, with the New York Heart Association (NYHA) class III or IV HF, who gave written informed consent, were eligible for the study. A diagnosis of HF was established by typical clinical signs and symptoms of HF in conjunction with echocardiographic or radionuclide ventriculographic findings of a reduced left ventricular systolic function, left ventricular ejection fraction (LVEF) ⩽45%, or of a diastolic dysfunction with preserved left ventricular systolic function, according to the 2001 guidelines for the diagnosis of HF of the European Society of Cardiology.21 The exclusion criteria were having dementia or psychiatric illness, having been discharged to or staying in a nursing home, having any disease other than HF, with an expected survival of <1 year, participation in another trial, being under ongoing or planned hospitalisation, and undergoing kidney function replacement therapy. After screening, eligible patients were randomised by computer‐generated allocation to either the intervention group or the control group.
Intervention
The intervention, performed in addition to usual care, consisted of an intensive follow‐up of the patients during 1 year at a HF outpatient clinic led by a HF physician and a cardiovascular nurse. The actual intervention commenced within a week after hospital discharge or referral from the outpatient clinic with a telephone call. At the first visit (at week 1) and second visit (at week 3) to the HF clinic, verbal and written comprehensive education was imparted about the disease and the aetiology, medication, compliance and possible adverse events. Patients were advised about individualised diet with salt and fluid restriction, weight control, early recognition of worsening HF, when to call a healthcare provider, and about physical exercise and rest. A patient diary was given. Easy access to the clinic was offered during working hours. An appointment with a dietician was made. The nurse asked the patient about his or her social and medical circumstances, and performed a short physical examination. The physician assessed, after a short review given by the nurse, the clinical condition of the patient, the laboratory results and ECG, performed a physical examination, and, together with the nurse, proposed a treatment regimen. At the regular follow‐up visits at weeks 5 and 7, and at months 3, 6, 9 and 12, the nurse provided counselling, check‐up and reinforcement of the education, and performed a short physical examination. At six of the nine follow‐up visits, the physician assessed the condition of the patient, optimised (medical) treatment and performed an overall assessment together with the nurse. The intervention was described in more detail elsewhere.8
Control group
The cardiologists of the Deventer and Alkmaar cardiology department are known for their special interest in HF. They treated the patients with HF by randomisation to routine care, according to their “usual care”. Their routine care was no doubt largely according to the guideline of the European Society of Cardiology prevailing at that time (version 2001), with optimal application of medical therapy including the target dose or high dose of HF medication (see baseline medication, table 1). As we aimed to compare the intervention with routine care, we decided not to develop a special protocol for the management of the control group of the Deventer–Alkmaar heart failure (DEAL‐HF) study. All cardiologists saw patients from the control group at their outpatient clinic.
Table 1 Baseline characteristics of the study patients.
| Characteristics | Intervention group (n = 118) | Control group (n = 122) |
|---|---|---|
| Demography | ||
| Mean (SD) age (years) | 70 (10) | 71 (10) |
| Male | 78 (66%) | 96 (79%) |
| Living alone | 23 (20%) | 21 (17%) |
| CHF | ||
| Aetiology of CHF: ischaemia | 60% | 65% |
| Prior admissions for CHF | 48% | 51% |
| Mean LVEF | 31% | 31% |
| Systolic dysfunction | 98% | 98% |
| Diastolic dysfunction | 34% | 30% |
| NYHA III | 98% | 95% |
| NYHA IV | 2% | 5% |
| Comorbidity* | ||
| Ischaemic heart disease | 60% | 55% |
| Myocardial infarction | 53% | 56% |
| Current angina | 15% | 16% |
| Prior stroke | 11% | 9% |
| PTCA/CABG | 14%/20% | 16/27% |
| Atrial fibrillation | 25% | 28% |
| Pacemaker | 10% | 7% |
| Hypertension | 39% | 43% |
| COPD | 29% | 28% |
| Current smoker/ex‐smoker | 12%/54% | 14%/52% |
| Diabetes mellitus | 31% | 28% |
| Anaemia | 21% | 12% |
| Hypercholesterolaemia | 54% | 43% |
| Laboratory values | ||
| NT‐proBNP (pmol/l)(pg/ml)†‡ | 262/2216 | 244/2064 |
| Erythropoietin (mU/ml) | 24 | 26 |
| Haemoglobin (mmol/l) | 8.4 | 8.4 |
| hs‐CRP (mg/l) | 11.5 | 13.7 |
| Potassium (mmol/l) | 4.4 | 4.4 |
| Creatinine (μmol/l ) | 123 | 130 |
| Microalbumin:creatinine ratio (mg/mmol) | 23 | 20 |
| Blood urea nitrogen (mmol/l) | 11 | 11 |
| Mean systolic blood pressure (mm Hg) | 123 | 125 |
| Mean diastolic blood pressure (mm Hg) | 73 | 76 |
| Mean heart rate (bpm) | 79 | 78 |
| Medication at entry | ||
| Diuretics | 97% | 96% |
| ACE inhibitor | 84% | 88% |
| ARB | 14% | 8% |
| β‐blocker | 60% | 69% |
| Spironolactone | 36% | 30% |
| Long‐acting nitrate | 19% | 17% |
| Digoxin | 23% | 27% |
| Anticoagulant agents | 62% | 67% |
| Acetyl salicylic acid | 31% | 23% |
| Statins | 44% | 33% |
| NSAIDs | 3% | 5% |
ACE, angiotensin‐converting enzyme; ARB, angiotensin receptor blocker; CABG, coronary artery bypass graft; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CRP, C‐reactive protein; LVEF, left ventricular ejection fraction; NSAIDs, non‐steroid anti‐inflammatory drugs; NT‐proBNP, N‐terminal prohormone brain natriuretic peptide; NYHA, New York Heart Association; PTCA, percutaneous transluminal coronary angioplasty.
*More than one item possible.
†Values are medians.
‡To convert from pmol/l to pg/ml, multiply by 8.457.
Data collection
At baseline, 3 and 12 months, LVEF was measured, NYHA classification assessed and plasma samples for neurohormone tests (NT‐proBNP) taken. Ejection fraction was measured by technicians blinded to the patient's intervention, either with a Philips Sonos 5500 (Philips Medical Systems, Best, The Netherlands) or with a Philips NZE28 Sonos 7500‐Live 3D echo machine (Philips Medical Systems) (biplane Simpson's method), or by radionuclide ventriculography.
In addition, the patients completed quality of life questionnaires at baseline and after 3 and 12 months. Health‐related quality of life was evaluated using the Rand Short Form 36 quality‐of‐life questionnaire,22 whereas disease‐specific quality of life was assessed by means of the Minnesota Living with Heart Failure questionnaire.23,24 Self‐care behaviour was measured by the European Heart Failure Self‐Care Behaviour Scale.25.
Clinical history, physical examination, blood and urine biochemistry, and ECG were also recorded at baseline and after 3 and 12 months. A chest x‐ray was taken at baseline only.
Clinical and demographic data were collected from the patient and from chart reviews.
Hospitalisations during the study period were tracked by means of chart review, hospital databases and patient recall/diary. The cardiologist on call of the emergency room always assessed the need for hospitalisation. He was not aware of the group to which the patient was allocated. Deaths were verified by chart reviews, hospital databases, general practitioner records and family recall. There was no loss to follow‐up.
An external clinical endpoint committee, consisting of three experienced cardiologists and blinded to the allocation status of the patient, judged all causes of hospitalisation and death.
The costs of intervention were based on prospective data collection. Hospitalisation costs were based on the mean daily cost at a specific level of care. Outpatient clinic costs included the nurse's, dietician's and doctor's salary.
Study end points
All study end points were prespecified in the protocol.8 The primary end point was the composite of incidence of hospitalisation for worsening HF and/or all‐cause mortality. Additional end points included the effect on LVEF, NYHA class, quality of life, NT‐proBNP and self‐care behaviour. Furthermore, time to death, utilisation of HF medication and costs of care were assessed.
Statistical aspects
The sample size was based on an incidence of the composite primary end point in the usual care group of 30%, and an expected 50% reduction in this incidence in the intervention group. With an α of 5%, and a discriminating power of 80%, the total number of patients required in each treatment arm was 118.
Statistical analysis was conducted according to the intention‐to‐treat principle. The frequencies of the primary outcome measure “occurrence of hospitalisation for worsening HF and/or all‐cause mortality” were compared, and relative risks (RR) with 95% confidence intervals (CI) and risk difference (RD) were calculated. To adjust for possible confounding arising out of unequal distribution of the baseline characteristics, logistic regression analysis was performed, with the primary outcome measurements as the dependent variable. For the change in normally distributed continuous variables, the Student's t‐test was used. The Mann–Whitney U test was used to test the difference in the not normally distributed continuous variables. The differences in change in quality of life scores were compared by the Wilcoxon rank‐sum test. The differences between the groups were tested by the log rank test. In subjects who died or about whom these data were not available because of hospitalisation for worsening HF, LVEF, NYHA class, quality of life and NT‐proBNP measurements, were assessed with the worst rank assigned. Because NT‐proBNP measurements showed high values and a skewed distribution, natural logarithmic transformation was applied.
Results
Baseline characteristics
We screened 797 patients over a period of 3 years from March 2000 to April 2003 (fig 1). Of these, 221 patients were not eligible according to the exclusion criteria (125 NYHA I–II; 37 terminal illness; 15 participation in other studies; 22 cognitive dysfunction; 22 planned hospitalisation). Among the 797 patients, the reasons that 103 did not participate included the presence of a variety of non‐cardiac disorders, having sick relatives, and sometimes unknown. Of the 473 patients who were eligible, 81 refused to participate mainly because they felt participation in the study would be too tiring and/or the travel distance was too large, and 152 refused because they did not want to participate in a randomised trial at all. Eventually, 240 of the 473 (51% (30% of the 797 screened patients)) eligible patients gave written informed consent and were randomly allocated to the intervention group (n = 118) or to the usual care group (n = 122; fig 1). Of these, 31% were hospitalised due to HF at the time of recruitment and 69% were referred from the cardiology outpatient clinic. The mean age of the patients in the included group was 71 years (male 70.5 years, women 72 years), that for the total group was 72 years and that for the not‐included group was 74.0 years (male 72.6 years, women 76.4 years). The percentage of male patients in the included group was 72%, in the total group 71% and in the not‐included group 70%. In all, 96% of the patients were in NYHA functional class III (table 1). The mean ejection fraction was 31%. The two groups were well balanced with respect to baseline characteristics except for sex.
Figure 1 Flow chart of the trial.
Effect on hospitalisation for worsening HF and/or all‐cause mortality
The incidence rate of this composite end point was 20.7 per 100 patient years in the intervention group and 42.2 per 100 patient years in the usual care group: rate ratio 0.49 (95% CI 0.30 to 0.81; p = 0.001) and rate difference 21.5 (95% CI 0.07 to 0.36) per 100 patient years (table 2). Twelve patients in the intervention group died during the intervention period, and there were 11 hospitalisations for worsening HF in this group, compared with 23 deaths and 24 hospitalisations for HF in the usual care group. Of the 12 deaths in the intervention group, 7 were sudden deaths, 2 were non‐cardiovascular deaths and 3 were terminal HF deaths. In the usual care group, there were 12 sudden deaths, 8 non‐cardiovascular deaths and 3 terminal HF deaths.
Table 2 Effect of a nurse‐and‐physician‐directed heart failure clinic on hospitalisation, death and days in hospital.
| Variable | Intervention group (incidence rate) n = 118 | Usual care group n = 122 | Rate ratio (95% CI) | RD (95% CI; NNT) |
|---|---|---|---|---|
| Hospitalisation for CHF and/or death | 23 (20.7 per 100 patient years) | 47 (42.2 per 100 patient years) | 0.49 (0.30 to 0.81) | 0.215 (0.07 to 0.36; 5) |
| Death (all‐cause) | 12 (10.8 per 100 patient years) | 23 (20.6 per 100 patient years) | 0.52 (0.26 to 1.05) | 0.098 (10) |
| Days in hospital | 359 (324 per 100 patient years) | 644 (578 per 100 patient years) | 0.56 (0.49 to 0.64) | 2.54 (0.4) |
CHF, congestive heart failure; NNT, numbers needed to treat; RD, rate difference.
Ventricular function and NYHA classification
After 3 months, there was no difference (p = 0.22) in LVEF between the intervention and the usual care groups. At 12 months, however, the LVEF had improved in the intervention group, whereas that in the usual care group decreased (p = 0.004; table 3). After 3 and 12 months, the NYHA class had significantly improved in the intervention group compared with the usual care group (p<0.001 for the difference at 3 and 12 months; table 3).
Table 3 Effect of a nurse‐and‐physician‐directed heart failure clinic on left ventricular ejection fraction, New York Heart Association class, N‐terminal prohormone‐ pro‐brain natriuretic peptide, quality of life and self‐care behaviour.
| Variable | Intervention group n = 118 | Control group n = 122 | p Value |
|---|---|---|---|
| LVEF (with the worst rank) | |||
| At baseline | 30.6% | 31.3% | 0.554 |
| At 3 months | 30.6% | 30.0% | 0.220 |
| At 12 months | 33.2% | 28.2% | 0.004 |
| NYHA classification (with the worst rank) | |||
| NYHA III; IV at baseline | 98%; 2% | 95%; 5% | 0.387 |
| NYHA I; II; III; IV at 3 months | 3.4%; 43.6%; 42.7%; 10.3% | 0.9%; 12.8%; 73.5%; 12.8% | <0.001 |
| NYHA I; II; III; IV at 12 months | 10.2%; 50%; 22.9%; 16.9% | 0%; 18.9%; 54.1%; 27% | <0.001 |
| NT‐proBNP (pmol/l)(pg/ml)*† (with the worst rank) | |||
| At baseline | 244/2064 (IQR 101–540) | 262/2216 (IQR 123–520) | 0.677 |
| At 3 months | 198/1666 (IQR 86–643) | 226/1911 (IQR 100–599) | 0.397 |
| At 12 months | 182/1539 (IQR 68–802) | 277/2343 (IQR 96–2242) | 0.089 |
| Rand SF36 | |||
| Total score at baseline | 45.12 | 46.77 | 0.506 |
| Total score at 3 months | 49.63 | 46.41 | 0.131 |
| Total score at 12 months | 49.23 | 41.92 | 0.021 |
| Minnesota Living With Heart Failure questionnaire | |||
| Total score at baseline | 42.5 | 42.6 | 0.958 |
| Total score at 3 months | 28.8 | 36.3 | 0.001 |
| Total score at 12 months | 30.2 | 34.5 | 0.038 |
| European Heart Failure Self‐Care Behaviour Scale | |||
| Total score at baseline | 23.6 | 25.5 | 0.092 |
| Total score at 3 months | 20.8 | 26.3 | <0.001 |
| Total score at 12 months | 23.8 | 30.2 | <0.001 |
| Creatinine levels (μmol/l) | |||
| At baseline | 123 | 130 | 0.144 |
| At 3 months | 124 | 132 | 0.08 |
| At 12 months | 121 | 138 | 0.002 |
LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; NT‐proBNP, N‐terminal prohormone brain natriuretic peptide; SF36, Short Form 36.
*Values are medians.
†To convert from pmol/l to pg/ml multiply by 8.457.
Quality of life
Improvement in the Minnesota Living With Heart Failure Questionnaire (MLWHFQ) scores at 3 months was greater in the intervention group than in the usual care group (p = 0.001), and this difference persisted during the remaining 9 months (table 3). At 3 months, there was no statistically significant difference in the total score of the Rand Short Form 36 (p = 0.131). At 12 months, the change from the baseline in the intervention group compared with that in the usual care was more pronounced (p = 0.021).
Other outcome variables
The differences in median values of the NT‐proBNP measurements at baseline, 3 and 12 months between the intervention group and the usual care group were not statistically significant (Mann–Whitney tests at baseline (U = 6795; Z = −0.416; p = 0.677), 3 months (U = 6019; Z = −0.848; p = 0.397) and 12 months (U = 5604; Z = −1.699; p = 0.089; table 3)). The values of the natural logarithm of NT‐proBNP in the intervention group versus the usual care group at 3 and 12 months were 5.43 vs 5.58 (p = 0.131) and 5.37 vs 5.71 (p = 0.070), respectively.
The mean time to death was 343 days in the intervention group and 333 days in the usual care group (p = 0.06).
The scores of the European Heart Failure Self‐Care Behaviour Scale (EurHFSCBSc) were significantly better in the intervention group than in the usual care group, after both 3 and 12 months of follow‐up (table 3).
There was a statistically significant difference in the prescription of spironolactone in the intervention group compared with the usual care group (60% vs 41%; p = 0.003) after 12 months. No statistically significant differences were observed in the prescription or dose of ACE inhibitors and angiotensin receptor blockers (ARBs) and the prescription of β‐blockers. Importantly, the maximally reached dose of β‐blockers was significantly higher in the intervention group (table 4). Finally, creatinine levels were lower in the intervention group than in the usual care group at 3 months and 12 months (table 3). The mean number of visits of the patients to their cardiologist was 0.79 in the intervention group and 1.43 in the usual care group (p<0.001). The number of days in the hospital constituted the major difference in costs between the two groups. Patients in the intervention group were hospitalised for a total of 359 days compared with 644 days for patients in the usual care group. The difference between the costs of hospitalisation in the intervention group (€65 046 (US$86 849, £44 103) and in the usual care group (€202 728 (US$270 648, £137 338) was €137 682 (US$183 834, £93 279). The total cost for the HF clinic programme (for the salary of the HF nurse, HF physician and the dietician, and for the extra lab and ECGs) was €50 246.00 (US$67 093, £34 038). As a result, the positive balance for the intervention group was €87 436 (US$116 764, £59 238) and the difference in the overall cost of care per patient was €741 (US$989, £502).
Table 4 Utilisation of medication.
| % Receiving drug at baseline | % Receiving drug at 12 months | Maximally reached dose during study | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Usual care (n = 122) n (%) | Intervention group (n = 118) n (%) | p Value | Usual care (n = 99) n (%) | Intervention group (n = 106) n (%) | p Value | Usual care (mg) | Intervention group (mg) | p Value | |
| ACE inhibitors | 88 | 84 | NS | 91 | 83 | 0.067 | 14.2 mg | 14.3 mg | NS |
| ARBs | 8 | 14 | NS | 12 | 25 | 0.008 | 139 mg | 154 mg | NS |
| ACEs and/or ARBs | 94 | 96.6 | NS | 102.5 | 107.7 | NS | |||
| β‐blockers | 69 | 60 | NS | 79 | 78 | NS | 106 mg | 135 mg | 0.005 |
| Spironolactone | 30 | 36 | NS | 41 | 60 | 0.003 | 27 mg | 25 mg | NS |
ACE, angiotensin‐converting enzyme; ARB, angiotensin receptor blocker; NS, not significant.
For the ACE inhibitors, the dosages were converted to an enalapril‐equivalent dose, for the angiotensin receptor blockers (ARBs) to a valsartan‐equivalent dose and for the β‐blockers to a metoprolol‐succinate‐equivalent dose.
Adjustment for the baseline difference in sex between the intervention and usual care groups did not change the results presented above.
Discussion
This 12‐month intervention in an intensive, combined physician‐and‐nurse‐directed HF clinic led to a 51% risk reduction of the primary end point—incidence of hospitalisation for worsening HF and/or all‐cause mortality—in comparison with usual care. Positive effects were also observed for LVEF, NYHA class, prescription of spironolactone, maximally reached dose of β‐blockers, quality of life and healthcare costs.
Compared with most previous HF management studies,9,13,15,19,26,27,28,29 our patients were probably in a slightly worse condition, as 96% were in NYHA class III at randomisation and the mean LVEF was 31%. As much as 69% of our included patients were not hospitalised but were referred by a cardiologist from the outpatient clinic. This is the first time that so many outpatients with NYHA III or IV were included in such a trial, and it is relevant to know that this type of intervention can also be effective for this large target group.
Although the content of the education included in our intervention was similar to those of earlier studies, our approach is unique in its intensive intervention by a combination of a clinician and a cardiovascular nurse, both trained in HF. Several studies have reported collaboration with a cardiologist or a general physician as a consultant, but not in such a standardised manner.11,13,19,26,28,29,30 One study reported a physician‐directed HF clinic assisted by nurses and with a scheduled visit to the general practitioner.20 In addition, our 1‐year intervention with 9 visits at the HF clinic and one telephone call is more intensive than those reported in most previous studies, except the home‐based intervention of Naylor31 and some studies with telemonitoring.9,15,26,32 In a study by Doughty et al in New Zealand,19 regular clinical follow‐up during 12 months was provided, alternating between the general practitioner and the HF clinic, complemented by group education sessions, conducted by the nurse and a cardiologist. Several methodological aspects of this study were comparable to those of our study. The obvious differences with our study are the integrated involvement of primary care and the group education sessions in the New Zealand study and the structural involvement of a HF physician in our study. Interestingly, the study of Doughty did not show a statistically significant effect on the combined end point of hospitalisation or death.
Jaarsma et al18 studied the effect of education and support by a nurse on self‐care and resource utilisation in patients with HF in The Netherlands. The education and support was provided during the hospital stay and at one home visit within a week of discharge. After 1 month, a statistically significant difference in self‐care behaviour was observed in the intervention group compared with the usual care group. No statistically significant differences were found in the mean number of readmission days or with the number of readmissions between the two groups at the end of the 9‐month study period. Jaarsma et al concluded that longer follow‐up and the availability of a HF specialist would probably enhance the effects of education and support. This was applied successfully in our study.
In a recent study by Strömberg et al,13 the HF clinic was staffed by nurses, with delegated responsibility for making protocol‐led changes in medications. If treatment needed to be optimised, a cardiologist was consulted. The first follow‐up visit was planned 2–3 weeks after discharge, and the 106 patients were followed up for 12 months. Most patients visited the HF clinic only once. A major effect on mortality was observed after 12 months (7 vs 20, p = 0.005). The intervention group had fewer admissions and days in hospital during the first 3 months, but there was no long‐term effect. This may have been due to the noticeably high (37%) mortality in the control group. A more intensive follow‐up would possibly have resulted in a more long‐term benefit.
Several limitations of this study should be discussed. First, although we had a reasonable response from 30% of the screened patients (51% of the 473 eligible patients), many suitable patients were not enrolled for various reasons (fig 1). The baseline characteristics, however, show the applicability of this intervention. The modest differences between the included, the total and the not‐included group can possibly be explained by the presence of slightly older women in the excluded group. Second, this study, with a follow‐up of 12 months, does not answer the question of whether and how intensively the intervention should be continued. Third, our results cannot easily be extrapolated to other HF clinics, because most of these do not include a team of a nurse in close, standardised cooperation with a HF physician. Fourth, it should be emphasised that some information bias may have occurred because, inherent to this type of intervention study, patients cannot be blinded to the intervention. We, however, feel that any bias is likely to be limited, because the effects of the intervention on the outcomes most likely to be influenced, such as quality of life measures, were modest.
In the last decade, the attention given to HF management has increased considerably. The standard of care for heart failure in The Netherlands, although not optimal,33,34 is already reasonably good in both primary care and secondary care. This is illustrated by the fact that, at the start of the study, 97% of the patients received ACE inhibitors or ARBs, and 65% received β‐blockers. The justification for our study was the question of whether a HF management programme with an intensive intervention according to protocol, by a combination of a HF clinician and a cardiovascular nurse, would be able to provide additional benefits, even in a country with a primary‐care‐based healthcare system, in which general practitioners act as gatekeepers for secondary care and with high‐quality primary care guidelines for many chronic diseases, including HF. The answer to this question is undoubtedly positive. Such an intensive management programme substantially reduces hospitalisation for HF and/or all‐cause mortality, while improving LVEF, NYHA class, quality of life and self‐care behaviour, and achieving a reduction in costs.
Acknowledgements
We thank Dr Peter Dunselman, cardiologist, Amphia Hospital, Breda, Dr Nicolaas de Jonge, cardiologist, University Medical Centre Utrecht, Utrecht, and Dr AnHo Liem, cardiologist, Oosterschelde Hospital, Goes, all in The Netherlands, for the work they accomplished as members of the Clinical Endpoint Committee of the DEAL‐HF study. We thank our heart failure clinic nurses in Deventer and Alkmaar, Bertie Stoel, Dian Pruijsers, Lammy Pol and Sonja Touw for their dedicated support to the patients, and Nel Rood, research nurse, for coordinating the study in Alkmaar. We acknowledge the supportive grant of Novartis Pharma BV, Arnhem, The Netherlands, AstraZeneca BV, Zoetermeer, The Netherlands, Bristol‐Myers Squibb BV, Woerden, The Netherlands, and also thank Roche Diagnostics, Almere, The Netherlands, for kindly providing the test essays for NT‐proBNP. Finally, we thank all those patients who made this study possible by agreeing to participate.
Abbreviations
ACE - angiotensin‐converting enzyme
ARB - angiotensin receptor blocker
BNP - brain natriuretic peptide
CHF - congestive heart failure
HF - heart failure
LVEF - left ventricular ejection fraction
NYHA - New York Heart Association
Footnotes
Competing interests: None.
Role of sponsors: The funding source for this study played no role in the design or conduct of the study; data management and analysis; or manuscript preparation, review and authorisation for submission.
References
- 1.Ho K K, Anderson K M, Kannel W B.et al Survival after the onset of congestive heart failure in Framingham Heart Study subjects. Circulation 199388107–115. [DOI] [PubMed] [Google Scholar]
- 2.Cowie M R, Wood D A, Coats A J.et al Survival of patients with a new diagnosis of heart failure: a population based study. Heart 200083505–510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mosterd A. Cost B, Hoes AW, et al. The prognosis of heart failure in the general population. The Rotterdam Study. Eur Heart J 2001221318–1327. [DOI] [PubMed] [Google Scholar]
- 4.Gysele S. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure. Eur Heart J 2004251614–1619. [DOI] [PubMed] [Google Scholar]
- 5.Vinson J M, Rich M W, Sperry J C.et al Early readmission of elderly patients with congestive heart failure. J Am Geriatr Soc 1990381290–1295. [DOI] [PubMed] [Google Scholar]
- 6.McAlister F A, Stewart S, Ferrua S.et al Multidisciplinary strategies for the management of heart failure patients at high risk for admission. A systematic review of randomised trials. J Am Coll Cardiol 200444810–819. [DOI] [PubMed] [Google Scholar]
- 7.Phillips C O, Wright S M, Kern D E.et al Comprehensive discharge planning with postdischarge support for older patients with congestive heart failure: a meta‐analysis. JAMA 20042911358–1367. [DOI] [PubMed] [Google Scholar]
- 8.Bruggink‐André de la Porte P W F, Lok D J A, van Wijngaarden J.et al Heart failure programmes in countries with a primary care‐based healthcare system. Are additional trials necessary? Design of the DEAL‐HF study. Eur J Heart Fail 20057910–920. [DOI] [PubMed] [Google Scholar]
- 9.Krumholz H M, Amatruda J, Smith G L.et al Randomized trial of an education and support intervention to prevent readmission of patients with heart failure. J Am Coll Cardiol 20023983–89. [DOI] [PubMed] [Google Scholar]
- 10.Stewart S, Pearson S, Horowitz J D. Effects of a home‐based intervention among patients with congestive heart failure discharged from acute hospital care. Arch Intern Med 19981581067–1072. [DOI] [PubMed] [Google Scholar]
- 11.Stewart S, Marley J E, Horowitz J D. Effects of a multidisciplinary, home‐based intervention on unplanned readmissions and survival among patients with chronic congestive heart failure: a randomized controlled study. Lancet 19993541077–1083. [DOI] [PubMed] [Google Scholar]
- 12.Stewart S, Horowitz J D. Home‐based intervention in congestive heart failure: long‐term implications on readmission and survival. Circulation 20021052861–2866. [DOI] [PubMed] [Google Scholar]
- 13.Strömberg A, Mårtensson J, Fridlund B.et al Nurse‐led heart failure clinics improve survival and self‐care behaviour in patients with heart failure. Results from a prospective, randomized trial. Eur Heart J 2003241014–1023. [DOI] [PubMed] [Google Scholar]
- 14.Bleu L, Lang E, McMurray J J V.et al Randomized controlled trial of specialist nurse intervention in heart failure. BMJ 2001323715–718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.McDonald K, Ledwidge M, Cahill J.et al Heart failure management: multidisciplinary care has intrinsic benefit above the optimization of medical care. J Card Fail 20028142–148. [DOI] [PubMed] [Google Scholar]
- 16.Koelling T M, Johnson M L, Cody R J.et al Discharge education improves clinical outcomes in patients with chronic heart failure. Circulation 2005111179–185. [DOI] [PubMed] [Google Scholar]
- 17.Galbreath A D, Krasuski R A, Smith B.et al Long‐term healthcare and cost outcomes of disease management in a large, randomised, community‐based population with heart failure. Circulation 20041103518–3526. [DOI] [PubMed] [Google Scholar]
- 18.Jaarsma T, Halfens R, Huijer Abu‐Saad H.et al Effects of education and support on self‐care and resource utilization in patients with heart failure. Eur Heart J 199920673–682. [DOI] [PubMed] [Google Scholar]
- 19.Doughty R N, Wright S P, Walsh H J.et al Randomised, controlled trial of integrated heart failure management: The Auckland Heart Failure Management Study. Eur Heart J 200223139–146. [DOI] [PubMed] [Google Scholar]
- 20.Atienza F, Anguita M, Martinez‐Alzamora N.et al Multicenter, randomized trial of a comprehensive hospital discharge and outpatient heart failure management program. Eur J Heart Fail 20046643–652. [DOI] [PubMed] [Google Scholar]
- 21.Remme W J, Swedberg K. Task Force for the Diagnosis and Treatment of Chronic Heart Failure, European Society of Cardiology Guidelines for the Diagnosis and Treatment of Chronic Heart Failure. Eur Heart J 2001221527–1560. [DOI] [PubMed] [Google Scholar]
- 22.Ware J E, Sherbourne C D. The SF‐36 health status survey: 1. Conceptual framework and item selection. Med Care 199230473–483. [PubMed] [Google Scholar]
- 23.Rector T S, Kubo S H, Cohn J N. Patients' self‐assessment of their congestive heart failure. Part 2: content, reliability and validity of a new measure, the Minnesota Living with Heart Failure Questionnaire. Heart Fail 19873198–209. [Google Scholar]
- 24.Wijbenga J A M, Duivenvoorden H J, Balk A H M M.et al Quality of life in chronic heart failure. Validation of the Dutch version of the Minnesota Living with Heart Failure Questionnaire. Cardiologie 19985627–631. [Google Scholar]
- 25.Jaarsma T, Strömberg A, Mårtensson J.et al Development and testing of the European Heart Failure Self‐Care Behaviour Scale. Eur J Heart Fail 20035363–370. [DOI] [PubMed] [Google Scholar]
- 26.Kasper E K, Gerstenblith G, Hefter G.et al A randomized trial of the efficacy of multidisciplinary care in heart failure outpatients at high risk of hospital readmission. J Am Coll Cardiol 200239471–480. [DOI] [PubMed] [Google Scholar]
- 27.Ekman I, Andersson B, Ehnfors M.et al Feasibility of a nurse‐monitored, outpatient‐care program for elderly patients with moderate‐to‐severe, chronic heart failure. Eur Heart J 1998191254–1260. [DOI] [PubMed] [Google Scholar]
- 28.Cline C M J, Israelsson B Y A, Willenheimer R B.et al Cost effective management program for heart failure reduces hospitalization. Heart 199880442–446. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Capomolla S, Febo O, Ceresa M.et al Cost/utility ratio in chronic heart failure: comparison between heart failure management program, delivered by day‐hospital and usual care. J Am Coll Cardiol 2002401259–1266. [DOI] [PubMed] [Google Scholar]
- 30.Rich M W, Beckman V, Wittenberg C.et al A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure. N Engl J Med 19953331190–1195. [DOI] [PubMed] [Google Scholar]
- 31.Naylor M D, Brooten D, Campbell R.et al Comprehensive discharge planning and home follow‐up of hospitalized elders: a randomized clinical trial. JAMA 1999281613–620. [DOI] [PubMed] [Google Scholar]
- 32.Cleland J G F, Louis A A, Rigby A S.et al Noninvasive Home Telemonitoring for Patients with Heart Failure at High Risk of Recurrent Admission and Death. The Trans‐European Network‐Home‐Care Management System (TEN‐HMS) study. JACC 2005451654–1664. [DOI] [PubMed] [Google Scholar]
- 33.Rutten F H, Grobbee D E, Hoes A W. Differences between general practitioners and cardiologists in diagnosis and management of heart failure: a survey in every‐day practice. Eur J Heart Fail 20035337–344. [DOI] [PubMed] [Google Scholar]
- 34.The Study Group on Diagnosis of the Working Group on Heart Failure of the European Society of cardiology Increasing awareness and improving the management of heart failure in Europe: the IMPROVEMENT of Heart Failure initiative. European Society of Cardiology. Eur J Heart Fail 19991139–144. [DOI] [PubMed] [Google Scholar]