Delivering care closer to home has become a priority within the NHS, as it reduces the burden on secondary and tertiary care while maximising convenience to the patient and supporting the ethos of person-centred care.
Home chemotherapy is one example of such care, which allows patients to receive their treatment in the comfort of their home and thereby avoid the stress and inconvenience of hospital visits (Corbett et al, 2015).
Administering chemotherapy at home
Traditionally, home chemotherapy has entailed the delivery of the treatment under supervision by a specialised nurse. With the advent of ambulatory devices, this term has evolved to include ambulatory chemotherapy. The devices used for this, also known as infusion pumps, allow patients to safely receive prolonged infusions unsupervised, with the community nurses visiting the patient at the end of the infusion disconnecting the device.
Specialised community nurses can deliver a range of chemotherapeutic agents of differing complexities at home. Some agents, such as intravenous cytarabine used for the treatment of paediatric leukaemia, are described by community nurses as simple, due to the small volume required to be administered. In contrast, others, such as subcutaneous azacitidine for myelodysplastic syndromes (MDSs), are more complicated due to the need to maintain a cold chain (Murthy et al, 2019).
Infusion pumps can be loosely categorised into mechanical pumps (battery operated) and elastomeric pumps (infusors) (Figure 1). The latter are disposable balloon pumps that are silent and lightweight. These are used to deliver prolonged infusions at home, such as those of 5-fluorouracil (5-FU) used for the treatment of bowel cancer. Patients are usually connected to these devices at the cancer centre and then visited by community nurses at the end of infusion for disconnection. Alternatively, patients can return to the cancer centre to be disconnected.
Problems with administering chemotherapy at home
With the increased use of home chemotherapy services, nurses anecdotally reported facing a previously unencountered problem. When the nurses were presenting to disconnect the elastomeric pumps at the scheduled time, they found that some pumps had not finished infusing all the chemotherapy volume. In such situations, the nurse has two options: to disconnect and dispose of the pump or to allow the pump to continue infusing. Even though the first option follows the prescribed protocol, it translates into patients not receiving the full dose. In contrast, the second option allows the full dose to be infused but translates into increased patient wait times or disrupted schedules for the district/community nurses who have to return at a later time to disconnect the patient.
Variation in the delivery of home chemotherapy via elastomeric pumps was expected but not scientifically measured. Therefore, the authors initially decided to conduct four experiments in the laboratory (Salman et al, 2013), whereby they investigated the flow-rate accuracy and end of infusion time of various commercially available elastomeric pumps. The temperatures of both the flow controller and the actual elastic reservoir were taken into consideration, as well as the viscosity of the diluent. The experiment was simple: the temperatures and viscosity were both varied, and the performance of the devices was monitored. The measured flow rate did not match the specified flow rate, and it varied with temperature and the viscosity of the diluents. However, the temperature around the pump changes significantly from user to user and over time. This temperature fluctuation will cause further inconsistencies, not only in the pump flow rates, but also in the stability of the chemotherapy regimen.
Factors affecting the performance of pumps
The performance of elastomeric pumps is affected by the temperature surrounding the pumps, the viscosity of chemotherapeutic regimes and/or a combination of two. The viscosity of chemotherapeutics is unlikely to be the same as that of 0.9% sodium chloride or 5% dextrose solutions, which are used to calibrate the pumps. The findings of these simple experiments indicated that flow rate and the end of the infusion time of elastomeric pumps were affected by different laboratory-based conditions. However, these factors are unpredictable and uncontrollable in practice. With all the different factors affecting the accuracy of these pumps, it is difficult to duplicate the same use conditions experienced during standard clinical practice. There was a need to evaluate the performance of these pumps in real practice and while used by the patients.
Consequently, the authors conducted a two-phase study (Salman et al, 2017) to observe and evaluate patients receiving home chemotherapy via elastomeric pumps (5FU infused via a 48-h elastomeric pump) at three gastrointestinal medical day units (MDUs). Ethics approval for both phases was sought from and granted by the clinical audit committee (CAC). Phase 1 was an observational cross-sectional study, in which elastomeric pumps connected to patients were observed at connection and disconnection in the MDU or at home with the help of a district nurse. The status of the pump at disconnection time and the volume remaining in the pump were recorded. The action taken in the event of unfinished pumps was noted, to evaluate the hospital policy. Further, the instructions given to patients and nurses' actions in the event of an unfinished pump were recorded, along with the additional time the patient had to wait. The results of this phase of the study showed that, in 50% of the 92 cases/pumps that were monitored—covering 50 cases disconnected at MDU and 42 at home by community nurses—the infusions did not finish on time. This had caused not only disruption to some of the patients and ambulatory home chemotherapy services but also sub-therapeutic dosing, as the pumps were discarded with remaining infusion.
The results revealed that there was no policy that guides nurses in the event of fast or slow pumps, and nursing staff had, therefore, developed different approaches to deal with this unexpected variation in the performance of elastomeric pumps. These approaches included: discarding the pump immediately (73%), asking the patient to wait in the MDU before disconnection (11%), sending the patient home with a request to return the next day (8%) and giving the patient advice over the phone (8%).
Similar results were observed in a 2015 study conducted at another cancer centre by Saleh et al (unpublished), where a significant number of patients were presenting with unfinished pumps. This study also included a survey of the nurses (both hospital and district) to gauge their knowledge of these pumps. Some 50% of the respondents stated that they had no formal training with the use of elastomeric pumps. As part of the survey, the nurses were presented with pictures of pumps containing different volumes and were asked to choose the pumps that they thought were empty. A significant number of nurses misidentified pumps that contained more than 20 ml as empty. These findings highlight the need for further nurse support and training with the use of the elastomeric pumps.
Additionally, the study also included a survey of patients to evaluate their knowledge on handling the pump while at home. The results highlighted gaps in their knowledge, where some patients were placing the pumps under their pillow while asleep, which would increase the ambient temperature of the pump.
It is important to note that, despite all of variations in the performance of elastomeric pumps, patients reported high satisfaction with these pumps, overall.
Conclusion
Home chemotherapy with its various forms has revolutionised patient care and has been linked with improved patient quality of life. Delivering chemotherapy at home has become more common with the advent of elastomeric pumps. However, nursing staff responsible for connecting or disconnecting these pumps at the patients' homes need to be supported and provided with the required training and policies. There are several recommendations to consider to maintain optimal delivery of this theory: