Abstract
Background Electronic care coordination systems, known as the Key Information Summary (KIS) in Scotland, enable the creation of shared electronic records available across healthcare settings. A KIS provides clinicians with essential information to guide decision making for people likely to need emergency or out-of-hours care.
Aim To estimate the proportion of people with an advanced progressive illness with a KIS by the time of death, to examine when planning information is documented, and suggest improvements for electronic care coordination systems.
Design and setting This was a mixed-methods study involving 18 diverse general practices in Scotland.
Method Retrospective review of medical records of patients who died in 2017, and semi-structured interviews with healthcare professionals were conducted.
Results Data on 1304 decedents were collected. Of those with an advanced progressive illness (79%, n = 1034), 69% (n = 712) had a KIS. These were started a median of 45 weeks before death. People with cancer were most likely to have a KIS (80%, n = 288), and those with organ failure least likely (47%, n = 125). Overall, 68% (n = 482) of KIS included resuscitation status and 55% (n = 390) preferred place of care. People with a KIS were more likely to die in the community compared to those without one (61% versus 30%). Most KIS were considered useful/highly useful. Up-to-date free-text information within the KIS was valued highly.
Conclusion In Scotland, most people with an advanced progressive illness have an electronic care coordination record by the time of death. This is an achievement. To improve further, better informal carer information, regular updating, and a focus on generating a KIS for people with organ failure is warranted.
INTRODUCTION
Identifying people who would benefit from a palliative care approach, and ensuring that their needs and preferences are documented and shared across settings, is essential to providing well-coordinated care.1,2 Electronic Palliative Care Co-ordination Systems (EPaCCS) are being developed to enable sharing of patient information, preferences across care settings, and inform emergency and out-of-hours care.3
Across the UK, regional systems include Coordinate My Care (London); the South-West EPaCCS; and the Leeds EPaCCS.3,4 In Scotland, the Key Information Summary (KIS) was introduced in 2013 as an electronic care coordination system for anyone who might have urgent care needs, not just those identified for palliative care.5 A KIS is created in primary care by GPs (in contrast to Coordinate My Care and other EPaCCS that allow users in a variety of settings to create and update a patient record). A KIS allows selected parts of the GP electronic patient record to be shared with emergency services (111 and 999), GP out-of-hours services, accident and emergency departments, ambulance services, hospitals, community nursing teams, and some specialist palliative care services.
A KIS record facilitates ‘anticipatory care’, a person-centred care planning process requiring healthcare professionals to work with individuals, carers, and their families, to discuss and document their goals and priorities for care.6,7 For people with an advanced progressive illness, this includes information on cardiopulmonary resuscitation (CPR) status, preferences for place of care or death, treatment escalation plans, carer details, and anticipatory medicines (see Supplementary Boxes 1 and 2). An evaluation conducted in 2014, examined 605 patient records from nine general practices in one NHS board area and found that 60% of those who died with an advanced progressive illness had a KIS by the time of death.8 Overall, KIS’s were started 18 weeks before death, and those with a KIS were more likely to die in a community setting.
Despite the potential importance of EPaCCS, a recent systematic review found that evidence on EPaCCS is mostly expert opinion, and few rigorous evaluations have been conducted.9 There is some evidence that people with an EPaCCS record are more likely to die in their preferred place compared with the general population.4,10–12 Given the drive to roll out EPaCCS across the UK by 2021, including the KIS across Scotland,13,14 ongoing evaluation is needed.9
| Electronic care coordination systems are being developed in many regions and countries to coordinate care for people with advanced progressive illnesses across settings. These can be generated across a national primary care system and automatically shared daily across emergency and out of hours care settings. In Scotland, an electronic care coordination patient summary is often started early in an advanced progressive illness with the patient’s consent, helping to coordinate chronic disease management and early palliative care. Having an electronic care coordination summary was associated with a greater likelihood of dying in a community setting (home, care home, or hospice). |
This study aimed to describe the evidence on the roll-out of a national electronic care coordination system, with specific reference to palliative care. The authors sought to: estimate the proportion of people with an advanced progressive illness who had a KIS by the time of death and to determine when the KIS was started; examine which elements of anticipatory care planning were most frequently recorded in the KIS; and to explore general practice staff perceptions of the KIS to inform future innovations in electronic care coordination systems.
METHODS
Design
This was a mixed-methods study consisting of a retrospective review of the records of all patients who died in 18 Scottish general practices in four NHS board areas in 2017, and semi-structured interviews with at least one healthcare professional in each practice.
Recruitment of general practices
In Scotland, GPs are trained in generalist palliative care and provide palliative and end-of-life care to their patients, seeking advice and input from a specialist team as required. Average general practice list size in Scotland is approximately 6000 patients, though there is a large degree of variability.15 This study sought to recruit practices ranging in size from just >1000 patients to >10 000 patients. Practices located in urban, suburban, and rural areas were purposively sampled. The authors identified 18 general practices varying in size and rurality across four NHS boards: Lothian, Fife, Ayrshire and Arran, and Tayside. Based on previous studies it was estimated that this would provide a sample of 1200 to 1300 patients.8 Practices were informed of the study by the National Clinical Lead for Palliative Care member of the research team, or were contacted if they had been involved in previous studies with the research team.
Quantitative data collection and analysis; retrospective review of patient records
A GP, currently working in palliative medicine, collected data for all patients in each practice who died in 2017. Patients who died because of cancer, frailty, dementia, or organ failure (including liver, heart, renal, or pulmonary failure) were judged as having an advanced progressive illness. Data collected included dominant trajectory of decline, that is, cancer, organ failure, frailty and/or dementia; place of death; whether specific components of anticipatory care planning were documented in the KIS; and when information was first documented (see Supplementary Figure 1). One author rated each KIS as ‘highly useful’, ‘useful’, or of ‘limited use’, in guiding emergency or out-of-hours care at the end of life. These categories were developed by the multidisciplinary research team for the purpose of this evaluation and are described below:
highly useful — clear plan regarding patient, and ideally family, wishes regarding current care and future planning;
useful — some additional useful clinical information, but no clear wishes regarding current care or future planning; and
limited use — a KIS present but no additional information added, no relevant boxes ticked, no free-text note.
Data were recorded in Excel and imported to IBM SPSS (version 24) for analysis. Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines were followed for reporting purposes.16
Qualitative data collection and analysis
Semi-structured telephone interviews were conducted by one of the authors, a general practice nurse trained in qualitative research methods. Most participants were GPs, as they create and update the KIS, and some worked in out-of-hours primary care services. Data were anonymised, transcribed, and imported into NVivo (version 11) for analysis informed by a framework approach.17 Standards for Reporting Qualitative Research (SRQR) guidelines were followed.18
DISCUSSION
Summary
The evaluation found that 69% (n = 712) of people who died with an advanced progressive illness had a KIS. People with cancer were most likely to have a KIS (80%, n = 288), whereas those with organ failure were least likely (47%, n = 125). Most people who died with frailty and/or dementia had a KIS that was typically started >1 year before death (Table 1).
The existence of an electronic care coordination record indicates that anticipatory care planning, or perhaps even early palliative care, had commenced. Over two-thirds of people with a KIS had their resuscitation status documented (68%, n = 482), and just over half had place of care preferences documented (55%, n = 390). However, for many others this information was not recorded, and there was wide variation in levels of KIS completion across practices. Despite this, the vast majority of KIS’s were judged to be useful, with the free-text section (special notes) providing the most useful information to guide emergency and out-of-hours care.
The existence of a KIS was positively associated with dying in the community (home, care home, or hospice), and those with a KIS were more likely to die in a community setting (Figure 2).
Drawing on these findings, the authors outline several considerations for the development of electronic care coordination systems (Box 1).
Develop systems that can accommodate free-text information and provide guidance on what might be contained within free-text section(s). This might include clinical information, family dynamics, patient access information, and treatment preferences. Optimise interoperability so that relevant clinical information contained within the GP systems can be routinely extracted to populate fields. Enable write-access for all key professionals involved in the patient’s care. Minimise duplication of tick boxes thus reducing inconsistencies. Improve documentation of timing information so it is clear when information was last updated, and when it will next be reviewed. Include a clear section on carer information, to specify, carer(s’) name(s), relationship to carer, and contact information. Brand and promote any new electronic care coordination system so it is viewed by the public as something that is helpful to have for any illness, and that they can request from their GP.
|
Box 1. Considerations for the development of electronic care coordination systems
Strengths and limitations
Overall, 1304 patient records were analysed from 18 diverse general practices across four NHS boards, resulting in a more generalisable sample than is possible in single-site studies. The majority of the 17 GPs interviewed did out-of-hours work, thus offering clear insights on electronic care coordination systems from the perspective of professionals working both in-hours and out-of-hours. Two practice nurses were also interviewed; however, a larger number of practice nurses may have provided more diverse responses. The study steering group was composed of stakeholders from academia, policy, ehealth, general practice, specialist palliative care, and community nursing, thus ensuring relevance to practice.
In terms of limitations, ratings of KIS usefulness had face validity, and were based on subjective judgement. The GP researcher needed to be located within each general practice when categorising each KIS (as patient identifiable information was not collected), and independent rating by a second rater was not practical. To guide the researcher, the usefulness categories were generated through consensus within the research team, and examples of each category were agreed in advance to guide judgements.
The records of deceased patients were examined, so data analysis was based on completed KIS. However, healthcare professionals visiting a patient earlier on in their illness may not have complete KIS information as the KIS is updated iteratively over time. Consequently, data on KIS completeness and usefulness reflect KIS content at the end of life only. Earlier on in a person’s illness, the KIS may contain less information and may not be as useful.
It is possible that only practices that are proactive in the generation of KIS were identified for participation. However, the present study analyses show a wide variation in KIS utilisation across practices, showing that practices with relatively low levels of KIS generation were also recruited.
Comparison with existing literature
The proportion of people with a shared electronic care coordination record has increased from 60% to 69% when compared with the authors’ 2014 evaluation.8 This is also higher than reported in an evaluation of the Leeds EPaCCS, which found that 26.8% of patients dying with chronic advanced diseases between 2014 and 2015 had an EPaCCS.4 The present findings also suggest that KIS are now created earlier than before; a median of 45 weeks before death, compared with 18 weeks in 2014.8
People with cancer were most likely to have a KIS (80%, n = 288), whereas those with organ failure were least likely (47%, n = 125). This reflects previous findings.1,8,21–23 A diagnosis of advanced cancer often triggers anticipatory care planning discussions, but for people with organ failure, the triggers are less clear.24 Different illness trajectories might affect the appropriate time of starting a KIS. Rapidly progressive cancer has a shorter final phase than frailty, so a KIS may be most relevant in the final weeks to months. In contrast, the slow decline experienced by people with frailty and/or dementia suggests that starting a KIS early would be appropriate.
Informal or family carers play a vital role in caring for the person with an advanced progressive illness and enabling them to remain in a home setting,25 however carers were not always identified in the KIS. Challenges to identifying carers of people with a terminal illness have been documented.26,27 General practices could be more proactive in identifying and documenting carer information within the KIS so that informal carers can be contacted quickly in the event of an unexpected or rapid change in the patient’s health.
Significantly, the existence of an electronic care coordination record was found to be positively associated with dying in the community (home, care home, or hospice) (Figure 2). This supports findings from previous evaluations.8,12 In a South West England evaluation based on people who died between 2011 and 2012, Wye et al found that most people with an EPaCCS died in the community (range 87% to 93%), but noted selection bias given that so few patients expected to die within 1 year actually had an EPaCCS record (range 9% to 13%).28 Wye et al suggested that the high proportion of community deaths may be due to the type of patients with an EPaCCS. In the present evaluation, as most people who died with an advanced progressive illness had a KIS, the likelihood of selection bias was much lower. Given that under optimal conditions most people would prefer to die in the community,29 the present results add to growing evidence that the existence of an electronic care coordination record is associated with dying in the person’s preferred place.4,10–12
Implications for research and practice
This evaluation study suggests that the drive to roll out a national electronic care coordination system in Scotland14,30 is bearing fruit. More people have a KIS at the time of death than reported previously,8 and KIS are being commenced earlier, suggesting more time for anticipatory care planning and early palliative care to promote health and wellbeing in the final years to months of life. The KIS was relatively new when the authors’ earlier evaluation8 was conducted. National policy prioritising identification and better care coordination for people with advanced progressive illness, alongside a Directed Enhanced Service (DES) for palliative care, is likely to have supported roll-out.1,14,30 As primary care professionals have gained experience with KIS, they are more likely to create a KIS for patients who may need care out of hours. Local initiatives to commence KIS on admission to care homes have supported identification of those with frailty and/or dementia for a palliative approach. Roll out is set to continue with national incentives to ensure that everyone who would benefit from anticipatory care or a palliative care approach has a KIS.31
People with organ failure are still less likely than people with other conditions to have a KIS (Table 1). Similarly, it can be difficult to identify a trigger point to start a KIS for frail older people at home, as opposed to those in care homes. Approaches to identify these groups more systematically for anticipatory care are required. Healthcare Improvement Scotland has published resources and guidance to support identification of people for a palliative approach,32 and tools such as SPICT are freely available.19 Computerised searches to identify people in primary care based on who may benefit from a palliative approach could support identification, though need to be further evaluated and implemented.33,34
Completeness of KIS could be improved, and more up-to-date information on essential aspects of anticipatory care planning, including patient preferences regarding goals of care, place of care, and treatment escalation, would improve overall KIS quality. The KIS provides a structure to enable coordination, but greater openness to discuss future care needs of those with advanced progressive illness is warranted, alongside regular review and clear documentation of the person’s changing needs and preferences.
Further research and evaluation is essential. Electronic care coordination systems are relatively new and rapidly evolving. Conventional research designs, assuming closed systems and experimental methods, are unsuited for such evaluations given the dynamic and constantly evolving contexts in which such systems are embedded. As argued by Greenhalgh and Papoutsi,35 research designs and methods, such as mixed-method case studies, consider interconnectedness and can provide a rich understanding of how systems come together as a whole. Drawing together different data from multiple sources and future mixed-method evaluations, including case studies, is recommended to understand the contribution of having an electronic care coordination record for a person with an advanced progressive illness, their family, and the wider healthcare system, as well as unintended consequences that might occur when an electronic care coordination record is not in place.
