Referral criteria for timely palliative care in patients with metastatic cancer: A cross sectional study

Technical Summary: Referral Criteria for Timely Palliative Care in Patients with Metastatic Cancer

Problem Statement
While international guidelines recommend the early integration of specialized palliative care (PC) for patients with advanced cancer to improve quality of life and reduce intensive end-of-life care, resource constraints make systematic referral at the time of metastatic diagnosis unfeasible. Consequently, the field has shifted toward "timely palliative care," aiming to identify the right patients at the right time. Although consensual referral criteria were established via a 2017 Delphi process involving 60 international experts, it remains unclear how frequently patients with metastatic cancer actually meet these criteria and, more critically, how often they receive PC consultations in alignment with those criteria. Furthermore, existing studies on these criteria have primarily focused on outpatients, leaving a gap in understanding their applicability to hospitalized patients who typically present with more advanced disease.

Methodology
This study was a prospective, multicenter, cross-sectional observational study conducted between June 2022 and May 2023 across eight oncology departments in the Île-de-France region of France.

• Population: The study included 355 patients (265 outpatients, 90 inpatients) aged 18 or older with metastatic cancer. Exclusion criteria included lack of medical insurance, inability to understand French, cognitive impairment preventing informed non-opposition, and curable cancers (testicular or ovarian choriocarcinoma).
• Data Collection: Researchers assessed the presence of 11 major international referral criteria (9 need-based, 2 time-based) defined by Hui et al. (2017).
  • Need-based criteria included severe physical/emotional symptoms (assessed via EORTC-QLQ-C30 and SCNS-SF34 questionnaires), requests for hastened death, spiritual crises, need for decision-making assistance, patient/family requests for PC, delirium, brain/leptomeningeal metastases, and spinal cord compression.
  • Time-based criteria included an estimated prognosis of <1 year (assessed via the "surprise question") and progressive disease despite second-line systemic therapy.
• Outcome Measures: The primary outcome was the proportion of patients meeting at least one referral criterion. Secondary outcomes included the proportion of patients who had received a PC consultation in the past 3 months and the factors associated with such consultations.
• Statistical Analysis: Comparisons between inpatients and outpatients were performed using Chi-squared or Fisher's exact tests. Multivariate logistic regression was used to identify factors associated with recent PC consultations, utilizing multiple imputation for missing data.

Key Results

• Prevalence of Criteria: A substantial majority of patients (334/355, 94.1%) met at least one referral criterion. Specifically, 71% of patients met two or more criteria, and 47.1% met three or more.
• Most Common Criteria: The most frequently met criteria were estimated prognosis <1 year (66.2%), need for assistance with decision-making or care planning (60.3%), and severe physical symptoms (41.7%). Rare criteria included delirium (0.3%) and spinal cord compression (0.8%).
• Inpatient vs. Outpatient Disparities: Inpatients met referral criteria more frequently than outpatients (98.9% vs. 92.5%, p=0.048). Inpatients also exhibited significantly higher rates of severe physical symptoms (62.2% vs. 34.7%) and estimated survival <1 year (87.8% vs. 58.9%).
• Referral Gaps: Despite the high prevalence of criteria, only 133 patients (37.5%) had received a PC consultation in the preceding 3 months. The rate was significantly higher for inpatients (52.2%) compared to outpatients (32.5%, p=0.001).
• Predictors of Referral: In multivariate analysis, a performance status of 2 was the only independent factor significantly associated with a recent PC consultation (OR=2.56). Notably, criteria such as delirium, brain metastases, spinal cord compression, and disease progression despite second-line therapy were not associated with increased likelihood of PC consultation, despite their prognostic significance.

Significance and Claims
The authors claim that this study provides one of the first real-world assessments of the alignment between international palliative care referral criteria and actual care delivery in both inpatient and outpatient settings. The primary significance lies in highlighting a substantial gap: while nearly all patients in the study had supportive care needs warranting PC referral, only a minority received recent specialized care.

The paper argues that the current referral process is insufficient, particularly for outpatients, and that strict application of the international criteria in resource-limited settings could lead to saturation. The authors suggest that raising the referral threshold (e.g., requiring 2 or 3 criteria) might be a more sustainable strategy to allocate resources to the most complex patients. Furthermore, the study underscores the need for:

1. Targeted Education: Training oncology teams to recognize specific criteria (such as neurological involvement or disease progression) that are currently under-recognized as triggers for referral.
2. Systematic Screening: Implementing routine screening for palliative care needs to identify gaps.
3. Multidisciplinary Integration: Strengthening collaboration between oncology and PC teams, particularly for inpatients with unstable clinical situations.

The authors conclude that while systematic screening and education are necessary, addressing the discordance between patient needs and available resources requires broader structural changes, including the development of innovative healthcare organizations and stepped-care models.