Article Text

Download PDFPDF

What tools are available to identify patients with palliative care needs in primary care: a systematic literature review and survey of European practice
  1. Elke Arnoldina Theodora Maas1,
  2. Scott A Murray2,
  3. Yvonne Engels3 and
  4. Christine Campbell4
  1. 1Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  2. 2Primary Palliative care Research Group, Centre for Population Health Sciences, University of Edinburgh, Edinburgh, UK
  3. 3Department of Anaesthesiology, Pain and Palliative Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
  4. 4Centre for Population Health Sciences, University of Edinburgh, Edinburgh, UK
  1. Correspondence to Professor Scott A Murray, Primary Palliative Care Research Group, Centre for Population Health Sciences, University of Edinburgh, Doorway 3, Teviot Place, Edinburgh EH8 9AG, UK; scott.murray{at}ed.ac.uk

Abstract

Background It can be difficult to identify when a palliative care approach should be started both in malignant, and particularly, in non-malignant disease, ideally to run alongside disease-modifying care. A structured method or tool may be useful to help general practitioners (GPs) identify patients for early palliative care and trigger assessment and care planning.

Aims To document what tools for identification of patients with palliative care needs are available in the published literature and to ascertain how GPs in Europe currently identify patients for palliative care.

Methods A systematic literature search using PubMed and Embase, and a questionnaire survey among key informants in 14 European countries requesting data on methods used to identify patients with palliative care needs.

Results The literature search identified four tools. The questionnaire survey identified a further three in current use and found that in current practice identification is largely based on a GP's own clinical judgement and information received from the hospital: tools are rarely used.

Conclusions Although several identification tools have been developed, none of these have been validated or widely implemented in Europe. Further collaborative international development, implementation and evaluation of such tools are recommended.

  • Prognosis
  • Chronic conditions
  • Identification Tools
  • primary palliative care
  • primary care
View Full Text

Statistics from Altmetric.com

Background

To optimise quality of life for patients with life-threatening disease, palliative care should run in parallel with potentially curative or disease-modifying treatment.1–3 In studies of earlier integration of palliative care with disease-oriented management, palliative patients have reported improved satisfaction with care, there are less acute interventions and patients are more likely to die at home.4 ,5

Essential steps for good palliative care provision include timely identification of patients, assessment of their care needs and planning of care.6 The patient's general practitioner (GP) is well placed to have an important role in this process. There is some evidence that where the GP is part of a team, palliative care appeared to improve.7 ,8 Yet patients with advanced chronic disease may have a medical specialist actively treating their disease, which makes proactive palliative care planning a challenge for GPs.

Better and more timely identification of non-malignant patients for palliative care has been called for over a decade.9 However, identifying when a palliative care approach should be implemented can be challenging, particularly for patients with non-malignant diseases such as advanced chronic obstructive pulmonary disease (COPD) and congestive heart failure (CHF).10 ,11 Such tools should ideally include the patient's perspective. In a recent study in Scotland, 60% of patients who died from cancer were listed on the practice palliative care register (PCR), while only 20% for patients who had died with other long-term conditions such as heart failure, dementia and COPD had been identified.12 The lack of prognostic indicators and clinical triggers to identify a ‘limited life prognosis’ appeared to be the most important barrier in applying palliative care in primary care.13

There are few specific objective identification criteria or validated tools to support primary care providers in this task of identifying a transition point to start integrating palliative care.14 ,15 In 2008, a literature review did not identify any validated tools that predict the optimal timing to initiate palliative care services.16 However, the use of a structured identification tool has been shown to be an approach by which more patients with non-malignant diseases could benefit from palliative care.17 Therefore, we set out to identify what identification tools were currently known about internationally for GPs to identify patients for a palliative care approach and to ascertain to what extent these have been implemented in practice in a sample of European countries.

Methods

Study design

A mixed methods approach was adopted. First, an international systematic literature review was carried out to identify tools that have been developed to identify patients with palliative care needs for use in a primary care setting. Additionally, a questionnaire-based survey of primary care key informants was carried out among members of the European Association for Palliative Care (EAPC) to identify any unpublished palliative care identification tools being used in practice.18 These informants were experienced practising GPs who had a special interest in palliative care and who attended national and European specialist palliative care conferences. The study ran from April until July 2012.

Systematic literature review

Data sources and searches

A search for studies describing the development of tools to enable identification of patients with palliative care needs was carried out using PubMed/MEDLINE and Embase. Search strategies are provided in appendix 1 (web only). Databases were searched from inception until the end of April 2012. No restriction was put on study design at the search stage or on the setting where the identification tool was used. Additional papers were identified through citation searching.

Study selection, data extraction and quality assessment

Titles and abstracts of identified papers were reviewed independently by two authors (EATM, CC). Final article selection was carried out after reading full papers by the same two authors. Articles selected for final review met the following inclusion criteria: written in English, Dutch or German; and description of an identification tool suitable for use in primary care (even if also appropriate for other settings). Tools designed for use exclusively in secondary care settings or assessment tools for disease monitoring were excluded. Data extraction was carried out by EATM and independently checked by CC. Included studies were assessed for methodological quality by two reviewers (EATM, CC) using the STROBE Statement checklist.19

Data analysis

A narrative approach to data synthesis was adopted. The content and assessment criteria of the identification tools were compared, as well as proposed users and provenance of the tool.

Key informant survey

Participants

One GP member of the EAPC Taskforce in Primary Care from each of the 14 participating European countries was invited to take part. As members of the EAPC, they were assumed to have detailed knowledge about the methods for identification of palliative patients and the delivery of palliative care in their respective countries. In three countries (Albania, Serbia and the Netherlands) where no GP was available via the EAPC Taskforce at the time of the study, we invited, via the personal network of SAM, GPs with palliative care expertise. Participants were contacted by email, and were sent study information, an informed consent form and the questionnaire. Reminders were sent to non-responders after 2 weeks and again after 4 weeks.

The questionnaire

A semistructured English questionnaire was developed by the project team (see online supplementary appendix 2). Participants were asked to answer ten open-ended questions regarding the knowledge and use of identification tools and other methods for identification of patients with palliative care needs in their countries. Additionally, they were asked to describe potential barriers to the introduction of such a tool and to provide their views on any potential benefits for patients when such a tool would be used in their healthcare context.

Analysis

Key elements of the answers on each question were described, compared and analysed according to the research questions and any emerging themes. Additional identification tools obtained through the questionnaire survey were analysed as above.

Results

Systematic literature review

Figure 1 provides details of study selection. We found 572 potentially relevant studies by database searching and three by citation searching, of which 63 were duplicates. Following review of titles and abstracts and reading of full-text articles by two authors independently, five papers were included in this review.20–24 These five articles contained four identification tools: the RADboud indicators for PAlliative Care needs (RADPAC), the Residential home palliative care tool, the Supportive and Palliative Care Indicators Tool (SPICT) and the Early identification tool for palliative care patients (table S1: web only).

Figure 1

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram.

Key informant questionnaire survey

Fourteen questionnaires were received from representatives of the EAPC Taskforce in Primary Care (Albania, Armenia, Belgium, Denmark, Germany, Ireland, Luxembourg, Malta, the Netherlands, Portugal, Serbia, Spain, Switzerland and the UK). The response rate was 100%. Eight respondents stated that there were no identification tools in their country. Three tools, none identified from the literature search, were mentioned: the Prognostic Indicator Guide (PIG) of the UK-based Gold Standard Framework (GSF), the NECPAL-CCOMS-ICO tool developed in Spain and a ‘Quick guide’ developed in England.25–27

Description of tools for identification of palliative care patients in primary care

RADboud indicators for Palliative Care needs

In the Netherlands, Thoonsen et al20 developed the RADPAC. It contains disease-specific assessment criteria for cancer, COPD and heart failure. It is based on prognostic indicators that were developed in a three-step procedure; a literature search, focus group interviews and a modified Rand Delphi study. Currently, further research has been carried out to investigate whether training of GPs, which includes use of the RADPAC tool as well as a structured method of care planning, influences the patients’ quality of life, hospitalisations and place of death.21

Residential home palliative care tool

This tool is an Australian adaptation of the American National Hospice Organisation (NHO)—guideline for identification of nursing home residents with non-malignant diseases.22 In this study, the number of patients that were identified with the adapted tool was compared with the number of patients that scored positive on functional, nutritional and pain assessment tools in order to investigate if this residential palliative care tool can identify non-malignant patients with palliative care needs. The authors concluded, after a 10-week timespan, that the adapted NHO guideline is an accurate approach if it is used in combination with pain scales.

The Supportive and Palliative Care Indicators Tool

SPICT was developed in Scotland and is derived from the American NHO guideline for eligibility of patients for hospice care, combined with a further literature review and expert consensus.3 ,23 SPICT was developed in Scotland and is derived from the American NHO guideline for eligibility of patients for hospice care, combined with a further literature review and expert consensus.3 ,23 SPICT consists of a combination of general indicators, and disease-specific assessment criteria. It is sometimes used with the surprise question. SPICT is a guide to help doctors recognise people who are at risk of deteriorating and dying. http://www.spict.org.uk/

Early identification tool for palliative care patients

Rainones’ criteria for identification for palliative care were developed in North America and were based on expert knowledge.24 They contain the surprise question, general indicators for decline and parameters for advanced stages of illnesses. In the study, the sensitivity and specificity of the tool were determined by application of the tool by GPs on a database of a patient population that met criteria of 75 years and over or having cancer, COPD, heart failure or HIV. The sensitivity and specificity were 94% and 97%, respectively.

Prognostic Indicator Guide

Six of the surveyed GPs mentioned the UK's Gold Standards Framework PIG as a good basis for development of a national guideline. PIG is based on three triggers that suggest that patients are nearing the end of life: the surprise question and general and specific clinical indicators for decline in organ failure, dementia and frailty trajectories.25

Neccesidades palliativas (NECPAL-CCOMS-ICO)

The NECPAL-CCOMS-ICO tool was developed in Catalonia (Spain). Based on the SPICT and PIG of the GSF, it has been adapted to the Latin Mediterranean clinical and cultural context.26 This tool is strengthened by also containing the request of the patient or family for palliative support as a trigger and also pays attention to the psychosocial domain, frailty and geriatric syndromes, and any progression of functional and nutritional decline. Preliminary results show that 1.33% of the total population and 7.0% of the population above the age of 65 years met the NECPAL criteria.26 Ongoing research includes the prevalence of patients with palliative care needs in different care settings and districts using the NECPAL tool.

The ‘Quick guide’

In England, the ‘Quick guide’ has been developed and implemented around London.27 It has been developed by a GP palliative care facilitator with the support of a national charity. It has been created to give simple guidance in daily clinical practice. Triggers for a need for supportive care are the surprise question, general indicators of decline, hospital admissions, weight loss, comorbidities and burden of illness.

Comparison of tools for identification of palliative care patients for use in primary care

The key features of each of the tools are summarised in table 1.

Table 1

Comparison of identification tools

The surprise question is set out in all tools except the RADPAC. The RADPAC, like the Australian tool for nursing home residents, is specific for a limited number of diseases, namely, cancer, COPD and heart failure. All tools contain general indicators for decline such as repeated or prolonged hospital admissions, weight loss and a decrease in functional status. The tools that are related to each other (NECPAL- CCOMS-ICO tool, SPICT and PIG), all extensively summarise disease-specific indicators of decline for a number of conditions. In all tools, the inclusion of psychological, social or spiritual indicators is limited. Rainone et al include frustration of the patients and relatives about deterioration. RADPAC describes the criterion of ‘the feeling that the end of life nears’.

All tools are developed to be used in primary care by GPs; however, the SPICT, PIG and the NECPAL are currently also being tested in hospital settings. The recommended timing for initiation of palliative care differs. The residential home palliative care tool initiates palliative care when treatment is based on relief of symptoms, which often refers to last weeks or days in life, whereas all the other tools are designed to support doctors to start palliative care earlier than terminal care, ideally at a relatively stable earlier phase of the disease trajectory.

Features of identification of palliative care patients in primary care across Europe

Responses from key informants from across Europe highlighted additional features of the process of identifying patients with palliative care needs in their respective countries (table 2). The UK is the only country that advocates tools for use in primary care, with the PIG being well known and SPICT being increasingly used in practice. In the Netherlands, ‘PATz’ has been developed in the area around Amsterdam and is currently being implemented in several other regions.28 Community nurses and GPs do home visits, identify patients with palliative care needs and add them to a database.

Table 2

How palliative care patients are identified in sample European countries

The role of GPs in identifying patients was probed in the questionnaire. Respondents from the Netherlands and Malta indicated that identification of patients eligible for palliative care is based on subjective criteria and therefore dependent of GPs’ experiences in palliative care. Other ways for identification in primary care settings mentioned by respondents were identification based on discharge letters from oncology services, International Classification of Diseases codes belonging to diseases and when patients meet terminal criteria and thus are in the last days of their lives. In Serbia, there are two separate GP services: infirmary care (a GP clinic) and home care (GPs home visits). If a patient is not able to go to the infirmary care, his/her records are transferred to a home care GP. At that moment, the patient is considered to be palliative. In Albania and Armenia, identification was not reported in primary care.

Discussion

We found four palliative care identification tools for primary care through the literature search, and three others via the survey of key informants. All tools contain general indicators for decline such as weight loss and loss of functional status, but only the NECPAL-CCOMS-ICO-integrated patient expressed need. Disease-specific indicators are also mentioned in some. Cancer, COPD and heart failure are included in all the tools. Other common characteristics were the surprise question, declining weight or functional status, and the use of resources such as having had two or more recent hospital admissions.

A number of tools—the NECPAL, PIG, SPICT and the Quick guide—are built on each other. The UK respondent was the only key informant who mentioned widespread use of identification tools, namely, the PIG and the SPICT. Several GPs from other countries mentioned that the SPICT and PIG could be introduced in their country if some adaptations could be made, fitting their culture and healthcare system. We found that in most countries identification in primary care is still largely based on a GP's own clinical knowledge and discharge information provided by secondary care.

The relatively limited number of tools found in our literature search is comparable with the result of the literature review from Quaseem et al.16 They reported that no validated tools were available for use in primary care, although two tools for use in hospital settings were identified.29 ,30 These tools contain many comparable criteria to those listed in the tools we have identified for use in primary care. Two further papers were excluded (see figure 1).31 ,32 The quality of these tools was variable, and further validation is required.

Claessen et al33 recently interviewed Dutch GPs about how they identify palliative patients. Their methods included discharge letters from the hospital, an increase in daily dependence from caregivers and relatives and a decrease in social contacts and activities.

Strengths and limitations of the study

This study benefited from a mixed-methods approach: a literature review supplemented by a questionnaire survey of key informants with an interest in palliative care provision in a primary care setting.

There are a number of limitations: there are limited MeSH terms for use in searching within this field; only two databases were searched (the primary databases for literature in this field). Hence, we may have missed some relevant papers, although we searched reference lists in the included papers. There are insufficient data on the sensitivity and specificity of included tools in primary care settings with a need for trial evidence to inform their usefulness. Tools developed for secondary care were excluded, except the SPICT, as it has been iteratively developed for use in both primary and secondary care.

It was important to note that the PIG tool was considered by a number of respondents as a good platform on which to build an international framework for identification of palliative care needs, but this most widely used tool would have been missed without the key informant survey.

Conclusion

The study documents the growing attempts to implement the WHO definition of palliative care, in which early identification and also palliative care for all conditions are important aspects. Although seven identification tools were identified, none have been validated or widely implemented.

Implications for future research, clinical practice, education and policy

Most GPs consider provision of high-quality palliative care as an important aspect of their profession. The low numbers of patients that are identified early for palliative care, the small number of identification tools and the fact that most of these tools were not spontaneously mentioned by the respondents to the survey show that the WHO call for early and comprehensive palliative care still lacks widespread implementation in clinical practice. Development and testing of tools to identify patients who may benefit from comprehensive palliative care (adapted as necessary for local health service contexts) should be explored. Further collaborative international initiatives, as the SPICT, PIG and NECPAL-CCOMS-ICO tools have started to progress, should be undertaken. Meanwhile, clinicians should be encouraged to use their clinical experience and one of the existing tools for early identification of patients for palliative care as a first step in improving the quality of life of patients currently living and dying with unmet care needs. If further research shows that these tools are useful for screening chronic patients in need of palliative care, there will be great policy implications that palliative care must be a component of chronic care programmes.34

Acknowledgments

We thank Marshall Dozier (University of Edinburgh) for guidance regarding the systematic literature review and the key informants who participated in the questionnaire survey.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Contributors SAM, CC and EATM planned this study. EATM and CC reviewed the publications. EATM, SAM, YE and CC wrote the paper and checked the final draft.

  • Funding This project was carried out during a student research elective carried out at the University of Edinburgh and funded by the European Union Erasmus Programme.

  • Competing interests None.

  • Ethics approval The study was approved by the University of Edinburgh's Centre for Population Health Care Sciences Ethics Committee (May 2012).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.