Statistics from Altmetric.com
Editor,—To complement the editorial by McMurray and Stewart,1 I present the results of a recent study from the Netherlands in which we randomised 179 patients (mean age 73 years), hospitalised with heart failure to intervention by a specially appointed nurse or to usual care. The intervention was intensive, systematised, and planned education by a study nurse about the consequences of heart failure in daily life, using a standard nursing care plan. During hospital stay, the study nurse assessed patients’ needs, provided education and support to patients (and family), gave patients a card with warning symptoms, and discussed discharge. Within one week after discharge the study nurse telephoned patients to assess potential problems and to make an appointment for a home visit. During the home visit the study nurse reinforced and continued education as warranted by the patient’s situation. If needed, home care was informed in writing about specific patient needs. Between discharge and home visit, patients could call the study nurse in case of problems. After the home visit, the patient was advised to call their cardiologist, general practitioner or emergency heart centre in case of problems. Therefore, the intervention lasted from hospital admission to 10 days after discharge from hospital. Data were collected on resource utilisation and a trend was described (p = 0.06) towards fewer readmissions and visits to the emergency heart centre in the intervention group.2
The main focus of the intervention was education and support by a nurse and follow up of the intervention was limited to 10 days after discharge. The study provides insight in the particular effect of education and support by a nurse. Our results show that this limited intervention is effective to enhance self care, but more is needed to get statistically significant results on readmission. The information is valuable in determining the required “dose” of nursing intervention. This confirms McMurray’s and Stewart’s editorial that describes the importance of determining which aspects of the intervention work. I would like to add two points to the list of issues regarding implementation and achieving optimal cost–benefit mentioned by McMurray and Stewart.1
• There is a huge difference in the populations in the published studies: Rich et al and Stewart et al investigated a high risk sample for hospital readmission.3 4 This means that a specific subgroup (high risk patients) of the very heterogeneous heart failure population can benefit from that specific intervention. Other researchers studied patients from a transplant clinic,5 which also had to be noted before generalising results to a general clinical heart failure population. Comparing all these studies in an overview as provided in the editorial can be helpful, but caution should be used when applying the results to practice.
• End points in effect studies should be standardised as much as possible. There is a great difference between studies reported in the editorial. Some authors used rehospitalisation as a primary end point and others combined this with mortality. Accumulating end points to a “major” variable (such as rehospitalisation and mortality) may increase the power of studies, but it sometimes makes comparison with other studies difficult.
In addition, I would like to support the authors’ plead for inclusion of variables that explain the mechanism of (beneficial) effects of intervention (such as compliance or self care). In this way we might get more insight as to which intervention (and which “dose”) is most appropriate for which heart failure patient.