Introduction

Multimorbidity, defined as the co-occurrence of two or more chronic or acute medical conditions and/or social problems in a single patient [1], is increasingly prevalent in contemporary healthcare settings, particularly as populations age [2, 3]. This growing clinical complexity presents considerable challenges to health systems worldwide, affecting not only clinical decision-making but also the overall quality of life of patients [4, 5].

This global challenge has led to calls for integrating palliative care across all levels of health systems as an essential component of universal health coverage [6]. In primary care, general practitioners (GPs) are frequently at the forefront of managing such complexity, particularly in systems like Portugal’s, where Family Medicine plays a central role in care coordination.

Amongst patients with advanced multimorbidity, the need for palliative care is often under-recognised. Despite strong evidence supporting early palliative interventions to improve symptom control, quality of life, and psychosocial well-being [7], these services are frequently introduced late, if at all.

Contributing factors include limited training in palliative care amongst non-specialists, discomfort in discussing end-of-life issues, and inconsistent criteria for identifying palliative needs, especially in patients whose illness trajectories are not clearly terminal [8].

Understanding how GPs perceive the palliative care needs of patients with multimorbidity is essential to improving early identification and integration of these services. Moreover, little is known about how clinical experience or training in palliative care may influence such perceptions.

This exploratory pilot study aims to fill a significant gap in the literature by examining the perspectives of both trainee and specialist GPs in Portugal regarding the palliative care needs and one-year mortality expectations in patients across key multimorbidity clusters.

Material and methods

Study design and participants

A cross-sectional, exploratory pilot study was conducted amongst GPs working in Portugal. The target sample size of 30–40 participants was determined based on feasibility constraints, and it is consistent with exploratory research using vignette-based methodologies [9]. This sample size allows for descriptive analysis and hypothesis generation, which are appropriate objectives for a pilot study designed to inform future larger-scale investigations.

Data collection instrument

An anonymous online questionnaire was developed specifically for this study, administered via Google Forms from December 2024 to May 2025. The questionnaire consisted of two sections:

sociodemographic and professional data: collected information on age, sex, professional status (resident vs. specialist), years of training, and any prior training in palliative care, regardless of its format or duration,

clinical vignettes: the vignettes were developed by the research team, which includes experienced family physicians and a palliative care specialist.

The questionnaire included 10 clinical vignettes based on five prevalent multimorbidity clusters in Portugal. These clusters were identified based on epidemiological data published previously about Portuguese primary care settings [3]. Each cluster was presented in two versions: low and high severity. The vignettes were reviewed by two senior family physicians and one palliative care specialist for clinical accuracy and relevance before distribution. A pilot test was conducted with five GPs to ensure clarity and comprehensibility. These participants were excluded from the final analysis. Following each vignette, three questions were posed: The Surprise Question (SQ) – Portuguese version [10], recorded as a binary variable. A “yes” response indicated that the GP would be surprised if the patient died within the next 12 months, whereas a “no” response indicated that the GP would not be surprised if the patient died within the next 12 months; and the question: Do you think this patient has palliative care needs? and case complexity: rated on a three-point ordinal scale (low, moderate, and high/very high), where the highest category combined high and very high complexity.

Recruitment

Participants were recruited on a voluntary basis through professional mailing lists, institutional contacts at primary care units, and networks of the Portuguese Association of General Practice and Family Medicine. Approximately 300 invitations were distributed electronically. Invitations were sent once. Inclusion criteria included being a practicing GP (resident or specialist) in Portugal. No additional exclusion criteria were applied. Of the 300 invitations sent, 40 complete responses were received, yielding a response rate of 13.3%. All 40 responses were included in the analysis.

Statistical analysis

Data were analysed using IBM SPSS Statistics software. Given the exploratory nature of this pilot study and the small sample size (n = 40), the analysis was primarily descriptive. Descriptive statistics (frequencies, percentages, means, and standard deviations) were used to summarise participant characteristics and responses. Comparative analyses between subgroups (e.g. specialists vs. residents, trained vs. untrained in palliative care) were not performed due to the limited sample size, which would have insufficient statistical power to detect meaningful differences. This paper reports on the primary descriptive findings and presents preliminary patterns that may inform future hypothesis-driven research.

Results

Participant characteristics

A total of 40 GPs participated, with a mean age of 40.1 years (SD = 12.2). The majority were female (77.5%, n = 31) and specialists (67.5%, n = 27). Most worked in Family Health Units (USF; 87.5%, n = 35). Less than half of the participants (45%, n = 18) had received formal training in palliative care. The low response rate (13.3%) may reflect self-selection bias, potentially attracting participants with greater interest or awareness of palliative care issues. A summary of participant characteristics is provided in Table 1.

Perceptions of palliative needs

General practitioners’ perceptions varied based on the severity of the clinical vignettes. As severity increased, the proportion of GPs who answered “no” to the SQ (indicating they would NOT be surprised by death within one year) and who answered “yes” to the existence of palliative care needs also increased.

Two key findings emerged from the data. First, the most important shift was seen in the obesity/diabetes/hypertension cluster, where the affirmative response to the SQ (indicating surprise if the patient died) dropped from 60% in the low‑severity scenario to just 5% in the high-severity one. Second, in the high‑severity osteoarthritis cluster, most physicians (65%) identified a need for palliative care, even whilst a majority (67.5%) also answered “yes” to the SQ (indicating they would be surprised by death within one year) (Table 2).

Descriptive patterns by training and professional status

In descriptive comparisons, a higher proportion of physicians with formal palliative care training identified palliative care needs and anticipated poorer prognosis in high-severity clusters compared to untrained physicians. Similarly, specialists appeared to demonstrate these patterns more consistently than residents. However, due to the exploratory nature of this pilot study and the small sample size (n = 40), formal statistical testing was not conducted.

Discussion

This exploratory pilot study suggests that the perception of palliative care needs amongst GPs in Portugal is influenced by the severity of a patient’s multimorbidity profile. A pattern emerged across all clinical vignettes: as the described severity of a patient’s condition increased, so did the likelihood of physicians identifying a need for palliative care. These findings should be interpreted as preliminary observations that generate hypotheses for future research rather than definitive conclusions. This was in accordance with physicians’ prognostic assessments; a lower expectation of survival over 12 months, indicated by a “no” response to the SQ, appeared to coincide with recognising palliative needs. However, the predictive accuracy of the SQ remains variable, and overreliance on prognostic tools may delay recognition of palliative needs in patients with longer but uncertain prognoses [11].

For instance, in the high-severity cluster of obesity, diabetes, and hypertension, only 5% of physicians would be surprised by death within a year, and this corresponded with nearly half (47.5%) identifying a need for palliative care.

Interestingly, the findings also suggest that a terminal prognosis is not the sole trigger for considering palliative interventions. In the high-severity osteoarthritis vignette, most physicians (65%) identified a need for palliative care, even whilst 67.5% stated they would still be surprised if the patient were to die within the year. This indicates a nuanced understanding amongst clinicians that the role of palliative care extends to managing severe symptoms and improving quality of life, independent of immediate mortality risk. Furthermore, the study suggests a descriptive trend where formal training in palliative care and specialist experience may enhance this perception, enabling physicians to more readily identify these needs. However, these observations are preliminary and require validation in larger studies with appropriate statistical power.

Our findings provide a granular view into this phenomenon, suggesting that, among patients perceived to have ‘low-severity’ or stable chronic conditions, palliative care needs are almost universally overlooked by the surveyed physicians.

Limitations

The main strength of this study lies in its novel use of clinical vignettes with matched low- and high-severity scenarios, which allowed for a controlled assessment of how clinical judgement is influenced by disease severity.

Furthermore, it addresses a recognised gap in the literature by focusing specifically on the Portuguese primary care context, where Family Medicine has a central coordinating role. As an exploratory pilot study, it successfully demonstrates the feasibility of this vignette-based methodology and generates preliminary insights that can inform the design of larger, hypothesis-driven investigations. Overall, this study provides a valuable foundation for future quantitative or mixed-methods research involving larger and more representative samples of GPs. Future studies should consider broader recruitment strategies, the use of participant reminders and, where appropriate, incentives to improve response rates. They should also formally examine associations between prior training, professional experience, prognostic assessment, and the recognition of palliative care needs.

However, several limitations must be acknowledged. As an exploratory pilot study, the sample size of 40 physicians is small and limits the generalisability of the findings to the broader GP population in Portugal. The study was designed for hypothesis generation rather than hypothesis testing, and as such, statistical comparisons were not performed. The recruitment method, relying on voluntary participation via mailing lists, resulted in a low response rate (13.3%) and may have introduced selection bias, potentially attracting physicians with a pre-existing interest in palliative care or multimorbidity management. This low response rate is a significant limitation that affects the representativeness of the sample and should be addressed in future studies.

Additionally, clinical vignettes, whilst a valuable research tool, cannot capture the full complexity and nuance of real-life patient encounters.

Conclusions

This exploratory pilot study provides preliminary descriptive findings suggesting that physicians’ recognition of palliative care needs in patients with multimorbidity is associated with their perception of disease severity and short-term prognosis. Descriptive findings suggest that whilst clinicians show an ability to identify needs based on quality of life, there may be a tendency to overlook palliative care in less severe or more stable chronic contexts. These preliminary findings support the call for enhanced education and training for GPs to broaden the conceptualisation of palliative care beyond immediate end-of-life care.

Disclosures

1. Institutional review board statement: This study was approved by Ethics Committee of the University of Beira Interior (approval decision no. CE-UBI-Pj-2024-082, dated: 15.10.2024).

2. Assistance with the article: None.

3. Financial support and sponsorship: None.

4. Conflicts of interest: None.

References