Abstract
Background The Acute Basic Palliation Concept (ABPC) — a structured end-of-life (EoL) care model — has shown increased quality of EoL care when used in hospital settings.
Aim To measure if the use of the ABPC could increase symptom relief and user satisfaction when applied in general practice.
Design & setting This cohort study was conducted across nine general practices in Northern Denmark with 29 GPs, as well as 10 representatives from the regional out-of-hours GP service.
Method GPs received training in an adapted ABPC and subsequently reported on patient cases managed both with and without the concept. Data on symptom control and satisfaction were collected via questionnaires from relatives, municipal caregivers, and GPs.
Results A total of 74 EoL care trajectories were included in the analysis, with 51% managed using the structured concept. The mean patient age was 84 ±8.9 years; 46% were female. The median duration from EoL care initiation to death was 3.0 (interquartile range 2.0–7.0) days. When using the concept, symptom control increased for pain (97% versus 83%), agitation (93% versus 56%), and discomfort (93% versus 81%; all P<0.05). Shortness of breath was less frequent, with a trend toward improved symptom relief using the concept (81% versus 54%). Most staff were willing to use the concept again (90% GPs and 100% municipal nurses).
Conclusion The structured ABPC model facilitated adequate symptom management and was widely accepted among GPs and municipal nurses, demonstrating promising potential for EoL care in a general practice setting.
How this fits in
End-of-life (EoL) care in general practice can be inconsistent, with GPs reporting variable confidence and few structured procedures. The Acute Basic Palliation Concept (ABPC) has previously improved care quality in hospital settings. This study showed that a structured approach can be applied in general practice without increasing time demands, while improving symptom relief and user satisfaction. These findings highlighted the potential for structured models like ABPC to strengthen community-based EoL care.
Introduction
Palliative care (PC) plays a central role in providing patient-centred health care and ‘is the active holistic care of individuals across all ages with serious health-related suffering due to severe illness’.1 According to the World Health Organization, PC aims to relieve pain and other distressing symptoms. It can be initiated alongside life-prolonging therapies and should begin before the final days.2
EoL is commonly recognised as the final phase of the PC trajectory, marked by progressive physical decline.3 While definitions vary, EoL care refers to comprehensive care for dying patients in the last hours, days, or weeks.4 In Denmark, PC is structured at two levels: specialised and basic.5 Specialised PC is delivered by trained professionals in hospices, hospitals, or at home, targeting patients with complex needs. However, most dying patients in Denmark (>80%) receive basic PC from healthcare professionals without formal PC training, such as GPs.6,7 The GPs form the cornerstone of Danish primary care. The provision of basic PC to non-hospitalised, dying patients is integrated into the general practice framework, with GPs working closely with municipal healthcare staff, including nurses, to deliver EoL care at home.8–10 However, in some regions of Denmark GPs rarely assume primary responsibility for such trajectories.11
Similar findings have been reported in Norway, where GPs struggle to maintain palliative competencies due to limited patient exposure.12 Among Danish GPs, self-reported confidence in EoL care varies widely and only few report structured procedures.11,13 Lack of confidence and training were also noted by Carey et al in a 2019 review.14 Multiple factors influencing EoL care in general practice have been studied, but findings remain inconsistent and non-comparable, and several studies call for further research.15–19 Notably, only 9% of palliative interventional studies are conducted in general practice.20 Challenges extend beyond general practice: Danish hospital doctors describe EoL care as professionally and logistically demanding.21 These challenges were mitigated when applying a structured approach initiated at discharge for patients wishing to die at home, the preferred place of death for most.22 This structured approach, ABPC, was implemented in hospitals across the North Denmark Region during 2022–2025. It markedly improved care quality and healthcare professionals expressed high satisfaction.23–25 Given the central role of GPs in providing EoL care outside hospitals, this study investigated whether a modified ABPC can offer similar benefits in general practice.
The aims of this study were to measure: 1) whether the percentages of patients with good symptom relief for pain, shortness of breath, agitation, and discomfort would increase when using the ABPC; 2) scores of usefulness (0–100) reported by municipal nurses, their willingness to use the ABPC again; and 3) scores of usefulness (0–100) reported by GPs, their willingness to use the ABPC again, and their time-consumption when using the ABPC.
Method
This study was performed as a cohort study between 1 September 2024 and 1 July 2025 in the North Denmark Region, a region with approximately 580 000 citizens. The study has been reported in line with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.26
Creating an ABPC modified for GP
The ABPC was initially developed to be used in a hospital setting. It consisted of: 1) a sealed box containing medicine and utensils covering the expected need for 5 days’ treatment for 95% of the patients; 2) a physician checklist with required tasks to be performed; 3) information pamphlets for the patient and relatives; 4) instructions regarding caring for dying patients for healthcare staff; and 5) a tool for decision support concerning medication administration for municipal nurses. In addition to items within the box, the concept consisted of: 1) electronic instructions for doctors; 2) a standardised text, that was to be added to the medical chart, containing relevant legal information; and 3) a pre-configured medication template within the electronic patient chart (see details in Supplementary Table S1). Based on the hospital version, the research team drafted a general practice-adapted physician checklist, which was reviewed in a 1-hour workshop with five GP registrars. All participants had at least 1 year of experience in general practice and 1 year of hospital-based clinical work.
Following the workshop, the revised checklist underwent external review by two independent specialists in general medicine and one GP registrar, who provided final refinements. Two pilot care trajectories were then conducted to evaluate usability; no further revisions were required.
Description of questionnaires
When testing the ABPC concept in a hospital setting, questionnaires were developed for relatives, doctors, and municipal nurses.23,24 For this study, the hospital-based questionnaires were adapted for general practice by removing items specific to admission and discharge. A few GP-related items were added to the physician questionnaire; no changes were made to the nurse or relative questionnaires.
Identical instruments were used for both the ‘standard care’ and ‘concept’ groups, except that ABPC-specific items appeared only in the latter. The relative questionnaire assessed perceptions of symptom control, reassurance, clarity of information, and involvement in decision making. Healthcare professionals reported on prior experience, information delivery, patient and family participation, symptom management, and confidence in providing EoL care. These caregivers were mainly nurses but also included certified healthcare assistants (authorised to administer medications under the delegation of a nurse). Doctors assessed their confidence in choosing appropriate EoL medications, whether EoL care was an area of special interest, and their experience navigating the required documentation. All healthcare providers were also asked to estimate the amount of time they typically spend on EoL care, the usefulness of the concept (0–100 scale), and their willingness to use the concept again. The distributed questionnaires had items that were answered using either ‘yes’, ‘no’, or ‘don’t know’, and items were answered using a five-point Likert scale with an added ‘don’t know’ option. The Likert scale included: 1) ‘to a very low degree’; 2) ‘to a low degree’; 3) ‘to some degree’; 4) ‘to a high degree’; and 5) ‘to a very high degree’.
Recruitment of GPs
GPs were recruited via the research team’s professional networks and represented the entire North Denmark Region (see Supplementary Figure S1). In total, 29 physicians from nine general practice clinics and 10 representatives from the regional out-of-hours GP service participated. Four clinics were regionally administered, while five were privately owned but publicly reimbursed.
GPs received a 30–60-minute introduction to the ABPC from one or two research team physicians and a senior researcher. Subsequently, each GP identified one to three patients from their own clinic who had initiated and received EoL care within the previous 6 months according to standard practice; these patients comprised the ‘standard care’ group.
Following the introduction, GPs consecutively enrolled eligible patients: adults (aged ≥18 years) with an estimated prognosis of days to a few weeks. Exclusion criteria included EoL care initiated during hospital admission, current treatment by a specialised PC unit, or patient preference for hospital-based EoL care.
A formal sample size calculation was not carried out, as the study was not designed as a randomised controlled trial. However, insights from an earlier study conducted during the development of ABPC indicated that a sample of 25 patients in each cohort was adequate to observe measurable differences.23
Data collection and analysis
Questionnaires were distributed to physicians immediately after the initiation of EoL care, to municipal nurses when they had spent at least 1 day with the patient, and relatives after the patient’s burial. Data were gathered and managed in REDCap, an electronic data capture tool made available by the North Denmark Region.27,28 Statistical analyses were performed using SAS Enterprise Guide (version 8.3). The percentage of positive responses (answers ‘yes’, ‘to a high degree’, and ‘to a very high degree’) were calculated out of all answers, excluding ‘don’t know’. Numerical data were reported either as median with interquartile range (IQR) or as mean with standard deviation (SD), depending on distribution. Differences were tested using the Wilcoxon rank sum test, student’s t-test, or χ2 test as appropriate. A P-value of <0.05 was considered statistically significant.
Outcome measures
The primary outcome was defined as the proportion of responses from relatives indicating adequate symptom relief, that is, responses marked ‘to a high degree’ or ‘to a very high degree’ for relief of pain, shortness of breath, agitation, and discomfort.
Secondary outcomes were defined as the municipal nurses’ and GPs’ evaluation of the concept’s usefulness, measured on a 0–100 visual analogue scale, and the proportion of nurses and GPs indicating willingness to use the concept again. GPs also evaluated their experienced time consumption.
Results
A total of 74 EoL care trajectories met the criteria for inclusion, with 51% managed using the concept. Population characteristics can be seen in Table 1. Four patients lived beyond 3 months; these have been excluded from calculations of ‘median time to death’.
Symptom management
Generally, relatives were inclined towards participating in the study (Table 2). They reported better symptom management in the concept group compared to the standard care group. Specifically, positive ratings for pain management were higher with use of the concept (P = 0.018, Figure 1). Similar statistically significant improvements were observed for the management of agitation (P = 0.002) and discomfort (P<0.001). Slightly higher, but not statistically significant, ratings for ‘Did the patient die a dignified death without suffering?’ was observed in the standard care group compared to the concept group.
Bar chart. X-axis gives four relief categories: pain; dyspnoea; agitation; and discomfort. Y-axis gives per cent from zero to 100. For each category of relief, the number of relatives scoring ‘a high or very high degree’ for their respective patient’s relief during end-of-life care is shown, split into two categories: standard care; and concept used. Pain relief shows 30 responders for standard care and 33 for concept used; dyspnoea relief shows 13 responders for standard care and 16 for concept used; agitation relief shows 27 responses for standard care and 30 for concept used; and discomfort relief shows 27 responses for both standard care and concept used. Pain, agitation, and discomfort relief are marked as statistically significant, with the P-value being greather than 0.05 between groups.
Evaluation by municipal healthcare staff
The municipal nurse questionnaire was answered 53 times. The concept received a high rating regarding usefulness (Figure 2). Like the relatives, the municipal nurses reported favourable evaluations of symptom control in patients managed using the concept (Table 3). All responders indicated that they would be willing to use the concept again. Notably, the participating nurses were generally experienced and confident in providing EoL care (Figure 3).
Three bar charts. Left-most bar chart shows ‘usefulness of the concept for end-of-life care on a visual analogue scale of 0–100’; the x-axis has two categories: municipality staff and GP doctors; the y-axis shows median (mm) from zero to 100. The median response for municipality staff was 92; the median response for GP doctors was 81. The middle bar chart shows ‘willingness to use the concept again (per cent responding ”yes”)’; the x-axis has the same two categories as the left-most bar chart; the y-axis shows per cent from zero to 100. Overall, 100% of municipality staff and 90% of GP doctors responded ‘yes’. The right-most bar chart shows ‘GPs’ perception of time spent on end-of-life care using the concept as opposed to standard care’; the x-axis has three categories: less, same, and more; the y-axis shows per cent from zero to 100. Overall, 56% of responders stated the same amount of time, while 22% stated less or more, respectively.
Four bar charts showing municipal nurses responses to four categories for both standard care and Acute Basic Palliation Concept used. X-axis for all bar charts shows standard care and concept used; y-axis shows per cent from zero to 100. First bar chart shows ‘confidence regarding medicine doses’, with responders giving 81% for standard care and 97% for concept used. Second bar chart shows ‘confidence regarding medicine frequency’, with responders giving 77% for standard care and 97% for concept used. Third bar chart shows ‘felt medicine doses were appropriate’, with responders giving 82% for standard care and 93% for concept used. Fourth bar chart shows ‘feeling professionally confident’, with responders giving 86% for standard care and 97% for concept used. The first, second, and fourth bar charts show statistically significant data, with a P-value greater than 0.05 between groups.
Evaluation by GPs
GPs involved in the project were generally experienced in EoL care and expressed a strong interest in the area (see Supplementary Table S2). Responses from GPs were received 46 times, with 24 being within the concept group. Median usefulness rating can be seen in Figure 2. When using the concept, the GPs in general prescribed more of the recommended five drugs from the start and there were fewer later contacts to the GPs for more prescriptions (Table 1). GPs reported increased uncertainty regarding types, doses, and frequencies of medications when using the concept (see Supplementary Table S2), yet most responders indicated willingness to use the concept again (Figure 2) and, on average, they did not spend more time performing EoL than usual (Figure 2 and Supplementary Table S2). Additionally, most GPs reported that the concept contributed to improving the quality of EoL care they provided (see Supplementary Table S2).
Discussion
Summary
This study examined whether the ABPC, a structured model previously shown to improve EoL care in hospitals, could be adapted for use in general practice. Across nine practices and the out-of-hours GP service, 74 patient trajectories were evaluated, half managed with the concept. Symptom relief was statistically significantly higher for pain, agitation, and discomfort in the concept group. Both relatives and municipal nurses reported higher satisfaction when the concept was applied and nearly all healthcare professionals expressed willingness to use it again. Overall, the ABPC demonstrated feasibility and effectiveness in general practice, suggesting potential to strengthen basic PC outside hospital settings.
Strengths and limitations
The study addressed a core area of general practice. The ABPC was tested under real-world conditions enhancing relevance and applicability, but also creating some methodological limitations. Limitations include non-validated questionnaires, modest sample size, and lack of power calculation. Retrospectively, GP-reported trajectories in the standard care group may have introduced both selection and recall bias. Variation in GP experience, practice type, and established routines may have influenced both the delivery of care and their reception of the concept. The study was carried out within the Danish primary care system, where participating GPs were recruited through professional networks and reported a strong pre-existing interest in PC. These contextual factors may restrict the applicability of the findings to other healthcare systems and to the wider GP population, who might be less engaged with this clinical area. Implementation fidelity was not uniform; not all elements of the ABPC were consistently applied and the timing of initiation likely varied. Specifically, electronic medication records were not always updated as required according to best practice and the provided checklist, and a proportion of relatives reported not having received, or being uncertain whether they had received, the written information material. These deviations illustrated variations in local implementation and highlighted the need for continued support to ensure adherence to the concept. External factors such as municipal nurse staffing and out-of-hours service availability may have affected results independently of the intervention. A higher, though not statistically significant, proportion of nurses in the concept group reported being involved in more than five EoL trajectories. Nevertheless, as EoL care is a core competence among municipal healthcare staff, even fewer experienced nurses are expected to deliver adequate care under supervision from more experienced colleagues.
Comparison with existing literature
Symptom management quality was rated significantly higher with ABPC, particularly for pain, agitation, and discomfort by the relatives. These findings mirror those from earlier studies evaluating ABPC in hospital settings23,24 and findings from the implementation of an integrated PC pathway among Dutch GPs.19 This supports the theory that structured frameworks like ABPC may address known gaps in EoL care and enhance GP-delivered care. A greater proportion of ABPC patients received all five recommended medications prescribed at EoL initiation, similar to the findings by Bell et al.29 Whether increased prescriptions contributed to better symptom control is unclear, though availability of medications is perceived to improve care.30
Municipal nurses rated the concept highly, comparable to hospital settings.23,24 A significant rise in professional confidence among municipal nurses was reported in the concept group (Figure 3). An unmet need for communication and information among municipal nurses has previously been documented.30 Since the concept included information material for both relatives and municipal nurses, the positive acknowledgment of its usefulness may partly reflect the addressing of this need. A qualitative study of municipal health carers’ experiences with the ABPC supported this with key themes including ‘enhanced communication and information flow streamlining workflows’ and ‘effective and timely symptom relief’.31
While all GPs felt confident in their usual practice, they reported lower confidence using the ABPC’s medication template, which recommends all five drugs on an ‘as needed’ (PRN) basis, with very high or no maximum doses. This aligned with Danish10 and international32 guidelines and supported improved care trajectories,33 but may have differed from usual practice and raised concern about overmedication. The PRN nature of the initial prescriptions also required that the GPs relied on the professional competence of the municipal nurses. From an implementation perspective, this finding underscored the importance of targeted training and interprofessional communication to ensure that GPs understand the clinical rationale behind the medication template and feel secure in its safe application. Structured feedback mechanisms between GPs and nursing staff could further strengthen confidence and adherence. Although the GPs expressed lower levels of confidence regarding the medical prescriptions, most of the doctors reported the quality of care to have increased when using the concept. Even experienced and palliative-interested GPs found ABPC useful, suggesting workflow optimisation. Importantly, most GPs reported that initiating EoL care using the ABPC did not require more time than their standard approach, a key finding considering time constraints are frequently cited as barriers to implementing new practices in primary care.34
Implications for research and practice
This study demonstrated that incorporating structured models such as the ABPC into primary care gives high quality symptom control and is strongly endorsed by healthcare staff, hence adopting structured approaches might be beneficial in clinical practice. Integrating the ABPC framework into existing workflows, supported by clear guidance and interdisciplinary communication, could enhance consistency in managing patients during EoL. Future research, with more robust study designs, such as prospective controlled or cluster-randomised trials, should be conducted in order to confirm current findings and explore the long-term sustainability and scalability of the concept, including its integration into routine GP workflows and out-of-hours services.
Notes
Funding
This work was supported by BETA.HEALTH by the Novo Nordisk Foundation, the North Denmark Regional Innovation Foundation, and Lilly og Herbert Hansens Fond.
Ethical approval
The Regional Ethics Committee of Northern Denmark determined that central ethical approval was not required, as the project qualified as a quality improvement initiative. The study received approval from the hospital administrations decentral ethical evaluation staff in 2024 (reference: K2024-084) and extended approval in 2025 (reference: K2025-052).
Provenance
Freely submitted; externally peer reviewed
Data
The dataset relied on in this article is available from the corresponding author on reasonable request.
Acknowledgements
The authors thank the patients and their families for their willingness to participate during a sensitive stage of life. We are grateful to the GPs and municipal nurses in the North Denmark Region who contributed to the study by applying the Acute Basic Palliation Concept in their daily practice and by providing valuable evaluations. We also thank the out-of-hours GP service for their collaboration and support.
Competing interests
The authors declare that no competing interests exist.
- Received November 18, 2025.
- Accepted December 18, 2025.
- Copyright © 2026, The Authors
This article is Open Access: CC BY license (https://creativecommons.org/licenses/by/4.0/)
References
LinkedIn
Jump to section
Keywords
More in this TOC Section
Related Articles
Cited By...
