We have studied the feasibility of an intervention by nurses for people who have survived a cardiac arrest and we found that most patients and caregivers participated and received on average 1.8 consultations. Nurses followed the protocol in most aspects, but the intervention started later after the event than intended, consultations were longer than expected, and self-management was rarely discussed. Overall, the intervention was positively evaluated by patients, caregivers and nurses.
The late start of the intervention, on average 90 days after cardiac arrest instead of the intended 1 month after discharge, can be mainly contributed to the design and organisational aspects of the ALASCA trial as, prior to randomisation, patients had to be informed about the trial, give their consent and perform the first measurements for the study. One can question whether the delayed start of the intervention may have influenced its effectiveness. We think that performing the first consultation earlier, that is more according to protocol, will probably increase the effectiveness because potential problems are addressed earlier, which may prevent some of the future negative consequences. Also the evaluations from patients, caregivers and nurses have shown that it is important to reduce this delay, which seems feasible in case of implementation of this intervention outside a scientific trial.
The duration of most consultations was longer than intended, namely 60 – 90 minutes, instead of the intended 60 minutes for first consultations and 30 minutes for subsequent consultations. However, as the number of consultations per patient was lower than we had expected, the total time spent on face-to-face contact was not exceeded.
The most unexpected deviation from protocol was that the topic of self-management was not frequently addressed: only in 19% of the patients was this topic discussed. Interviews with the nurses revealed that they considered self-management not suitable for this patient group in this phase, which explains why they did not promote it. According to the nurses, the topic of self-management is more appropriate for a later phase, and can be better practised during more extensive interventions. Indeed, most previous studies on self-management have been performed in chronic conditions , and the two effective self-management interventions for survivors of cardiac arrest were much more elaborate and consisted of eight and eleven sessions respectively [6, 7]. We therefore conclude that it is not feasible to address self-management properly in the current intervention and propose to eliminate it as one of the obligatory elements.
Limitations of the process evaluation
A risk in evaluation studies is that participants may tend to give socially desirable answers. We tried to prevent this by providing anonymous evaluations forms to patients and caregivers, and by sending out these forms after the intervention had finished.
The interviews and evaluation forms for the nurses were not anonymous. To limit the risk that the nurses would provide socially desirable answers, the interviews were administered by a researcher who was not involved in the trial. Moreover, the nurses did not have any formal or personal relationships with the researchers before or after the trial.
Suggestions for implementation
This process evaluation has shown that the intervention is sufficiently feasible. However, there can be some tension between ‘a flexible intervention’ and guaranteeing that certain content is sufficiently addressed. The advantage of a flexible intervention is that it can be tailored to the actual needs of the participants and, in case there are no problems or questions, the number of consultations remains limited, which will positively affect the cost-effectiveness. We have noticed structural differences between the nurses with regard to their performance of the intervention, and believe that the feasibility and reproducibility of the intervention can be improved on some aspects by making several of the optional elements of the intervention more obligatory.
First of all, we suggest offering at least two consultations to all patients. When not taking into account the nurse who had only consulted one patient, the two nurses who consulted patients most frequently also received the highest grades for the intervention and their performance. This suggests that conducting several consultations improves the perceived quality of the intervention.
Secondly, we advise administering a formal screening instrument to all patients. The nurses that used the Checklist Cognition and Emotion were very positive about it, and previous research in stroke patients also showed that formal screening significantly contributed to the detection of cognitive and emotional problems . We think that making such a screening instrument obligatory will secure that ‘screening for cognitive and emotional problems’ is effectuated in all patients.
During the trial, most consultations were conducted as home visits. This seemed to be appreciated by the patients, but also demonstrated the disadvantage that it is more time consuming for the nurse. We recommend that home visits should remain possible but suggest combining the consultation with the nurse with regular out-patient consultations with the cardiologist. As such, the extra effort for the patient is limited while it can improve efficiency for the nurse.