Discussion

To date, published studies examining the provision of SPC services at a regional, national or international level have largely relied on information provided by one or more national experts.3 9 10 This is problematic because national experts may lack the requisite knowledge to accurately report in detail service provision at the national or subnational level and may be biased in their reporting.11 Our study analysed data gathered from 36 informants directly delivering, or responsible for managing, local SPC services in each of Scotland’s 14 territorial health boards areas, increasing the validity of our findings. Due to their local knowledge and the nature of the telephone survey, most respondents volunteered a narrative around the quantitative data on the level of service provision they provided, which was invaluable in contextualising the study findings. To our knowledge, this is the first study of its kind to use such a robust data collection strategy.

The UK is generally considered to be the reference country against which the development of SPC in other countries is benchmarked.1 Until now, it has not been possible to disaggregate Scottish data from that of the rest of the UK.3 9 10 While we might expect the provision of SPC services in Scotland to be similar to the rest of the UK, this is the first study to provide confirmation. The provision of SPC services in Scotland was found to be well balanced and broadly reflective of that in the rest of the UK, with Scotland’s position being higher than the rest of the UK for inpatient and home care teams but lower for hospital support team. Analysis of coverage against EAPC norms suggests that Scotland has optimal coverage for hospital support teams. More broadly, Scotland was positioned among the top 10 EU countries in hospital and home care resources for SPC. Although the ratio of home care teams per million people (7.1) is better than the ratios for hospital support teams (5.0) and inpatient units (4.3), Scotland shows an optimal coverage of hospital support teams and a lower than recommended coverage for inpatients units (85%) and home care teams (72%). Calculated at both the upper and lower ranges, Scotland scores lower on SPC inpatient bed coverage (65%–82%) than it does on the calculation based on the number of units.

For almost two decades, the strategic narrative for health and social care in Scotland has focused on the delivery of safe, effective, high-quality, patient-centred care delivered at home or in a homely setting.12 13 As a result, over the last decade, the total number of available staffed hospital beds in Scotland has been reduced by almost a fifth.14 The lower than anticipated level of inpatient SPC beds may therefore reflect successful policy implementation rather than deficiencies in the planning and provision of SPC services. Many service providers in our study, especially in remote and rural areas, identified that while they did not have dedicated SPC inpatient beds, they were providing integrated SPC to inpatients that remained under the care of their admitting specialty, through SPC hospital support teams. This flexible model of care ensures effective use of acute hospital beds and continuity of care and enables collaborative working between generalists and specialists that in turn may facilitate the ‘up-skilling’ of healthcare providers from other specialties in delivering ‘generalist’ palliative care. Indeed, in recent years, the role of SPC providers in providing education and support to generalists has been championed.15 16 Increasingly in Scotland SPC clinicians and nurse specialists are jointly appointed to work between the independent hospices, the acute hospital sector and community services to provide integrated care and facilitate early hospital discharge. This change in how and where SPC teams are based is reflected in the optimal coverage of hospital support teams.

The coverage of home care teams in Scotland was less than optimal. This is an important finding. If the policy of shifting the balance of care from hospitals to community settings is to be translated into practice, community services must be adequately resourced.12 There has been a growing policy recognition in Scotland of the importance of achieving preferred place of care and preferred place of death.14 Most people would prefer to die at home or in a community setting.17 It is not possible to measure how well preferences are being met in Scotland because neither preferred place of care and preferred place of death are systematically recorded. In England, an indicator for death in usual place of residence has been developed and is in use. In Scotland, such a measure is not available. A surrogate measure, the percentage of the last 6 months of life spent at home or in a community setting, is in use; in 2014/2015 among decedents in Scotland, 86% of the last 6 months of life were spent at home or in a community setting and 14% in an acute hospital.18 Moreover, people are more likely to die in hospital in Scotland (53%) than they are in England (47%).19 Adequately resourcing community-based services, across health and social care, will be important if Scotland is to realise its policy ambitions; our study provides data to advocate for further developing community-based services. Many other health and social care professionals contribute to the interdisciplinary teams that support people to live and die well in community settings, not least community district nurses, disease-specific specialist nurses and primary care physicians. In Scotland, these key individuals provide generalist palliative care and often act as gatekeepers to SPC services. The availability of high-quality generalist palliative care is likely to directly impact on the demand for and subsequently delivery of SPC services in community settings. At the same time, measuring the availability and delivery of generalist palliative care raised additional challenges. It is important to note that when collecting the primary data for the Atlas, despite clear definitions, there was variation in how service providers conceptualised their services; for example, one territorial health board reported having just one home care team while another, with the same level of staffing across its home care service, reported having multiple home care teams stating that each team covered distinct geographical areas within the board. Our data on home care teams should therefore be interpreted with caution. This does raise questions about what ‘counts’ as a service in any given setting and may have led to an underestimation of the number of home care teams operating in Scotland. It also highlights the importance of common definitions and the challenges in operationalising these, both for service evaluation and for research.

The EAPC White Paper provides norms for the number of services and resources per million people, allowing for the diversity of concepts of SPC and varying cultures across Europe. These norms are based on expert opinion and estimates of demand for services that EAPC considers, if achieved, will translate into the provision of high-quality care. This approach to establishing a range of values for benchmarking is not without limitations. It is common practice where the evidence base is lacking to rely on expert opinion. Underpinning recommendations for service provision using estimated demand for services, rather than need, is problematic; however, estimating palliative care needs is particularly challenging.20 The appropriateness of these measures is likely to change as services evolve and the evidence base grows. The measures may also have differing significance according to the context of the healthcare system within a country. For example, the coverage of hospital support teams in Belgium was over 100%; however, according to Belgian law, every hospital is obliged to have a SPC team. The extent to which ‘generalist’ palliative care principles are effectively embedded within a healthcare system may also influence the required level of SPC services and in part explain an apparent underprovision or overprovision of certain services. It is important to consider that these metrics are not sensitive enough to capture such nuance. The balance between services could also be a result of the distribution of population across different areas of the country. With the exception of the central belt of Scotland, where the population is concentrated into large conurbations, there are many dispersed and rural communities of small population. As in many other countries, the distribution and volume of SPC services in Scotland is the result of historical circumstances, local priorities and enthusiasms, as well as the ebb and flow of government interest and funding. The EAPC norms provide a valuable tool for benchmarking the provision of SPC services between countries or within a country over time, which can be a valuable tool for advocacy and also, over time, for more rational planning. However, these measure require careful interpretation considering the wider sociodemographic, political and economic context within which services are being provided. Importantly, these metrics do not provide any objective measure of the quality of SPC.

The difference in timing of data gathered for Scotland alone and that gathered for the 10 EU countries used as comparators is acknowledged. Data for the EU ranking and coverage studies were collected in 2013, compared with 2015 for the present study. This small time gap is not considered to be significant. Indeed, the proximity of Scottish data to that for the UK as a whole, given the common evolution of SPC provision across the UK, is reassuring.

The primary data used in this study were derived from a novel SPC service mapping study. Prior to this, it was not possible to describe the level of SPC provision at local or national level in Scotland; it is still not possible to measure SPC activity, outcomes or quality of care in Scotland. It is anomalous that data of this kind are not available to inform the design, strategic commissioning, delivery and continuous quality improvement of SPC services. These significant ‘information deficits’ were recognised in a 2015 Scottish Parliamentary Inquiry into Palliative Care,1 and a policy commitment has been made to address them in Scotland’s Strategic Framework for Action on Palliative and End of Life Care later that year by the Scottish Government.15 The national mapping survey was a first step towards this. The study was both labour and time intensive; however, it was necessary to provide reliable information to quantify and characterise SPC provision at local and national level in Scotland. In the study, several informants acknowledged that their service(s) were being redesigned to better meet the needs, preferences and priorities for care in their locality, as informed by local health needs assessments. This highlights the challenge of maintaining accurate and current data on SPC service provision and organisation at a national level. A more sustainable approach to collecting and collating these data is required building on models of SPC directories used elsewhere and making use of multiple providers of data.

Finally, we acknowledge that this is an analysis of indicators for SPC. It does not address the levels and coverage of resources that may be available for the delivery of ‘generalist’ or ‘primary’ palliative care. Such a mapping and assessment is also worth considering but was beyond the remit of the present study.