Opioid prescribing practices for dyspnea across palliative care settings: a retrospective observational study

Technical Summary: Opioid Prescribing Practices for Dyspnea in Palliative Care

Problem Statement
Dyspnea is a prevalent and distressing symptom among patients receiving palliative care, particularly at the end of life. While opioids are recognized as a pharmacological option for refractory dyspnea, their clinical application remains heterogeneous. Existing literature highlights persistent uncertainty regarding efficacy and safety, alongside concerns among healthcare professionals and patients regarding respiratory depression and the misconception that opioids may hasten death. Crucially, there is a scarcity of data describing real-world prescribing patterns for dyspnea across different models of palliative care. This study addresses the gap in understanding how opioids are actually prescribed for dyspnea in routine clinical settings and whether these practices vary based on the organizational context of care delivery.

Methodology
The authors conducted a single-centre, retrospective, descriptive observational study at the University Hospital of Bordeaux, France, covering the period from November 1, 2024, to April 16, 2025. The study adhered to STROBE guidelines and employed an exploratory design without predefined formal hypothesis testing.

• Population: The cohort included 53 adult patients managed by specialist palliative care teams who required the initiation or escalation of systemic opioids (WHO Step II or III) for symptomatic relief of breathlessness. Patients were categorized into three care settings: the Palliative Care Unit (PCU), the Palliative Care Day Hospital (PCDH), and the Mobile Palliative Care Team (MPCT).
• Data Collection: Data were manually extracted from electronic medical records (Dxcare®) by a single investigator. Variables included demographics, comorbidities (Charlson Comorbidity Index), underlying diagnoses, opioid type, route of administration, dosage, and timing of modifications.
• Standardization: Opioid doses were converted to Oral Morphine Equivalents (OME) using a standardized online tool to facilitate comparison across molecules and routes. Dyspnea was identified based on documentation in medical records without standardized measurement tools, reflecting routine practice.
• Analysis: Statistical analysis was performed using Python (3.11). Given the small sample size and exploratory nature, analyses were primarily descriptive (means, medians, interquartile ranges, percentages). No inferential statistics or adjustments for multiple comparisons were applied.

Key Results

• Patient Demographics and Baseline: The mean age of the cohort was 72.8 years. The most common underlying conditions were non-haematological malignancies (43%) and chronic respiratory failure (37%). Notably, 47.1% of patients were already receiving opioids prior to specialist palliative care involvement, primarily for pain management.
• Initiation Practices: Among opioid-naïve patients (n=28), the oral route was the most common method of initiation (78.6%). The median starting dose was 7.5 mg OME/day. Initial doses varied by setting: the PCDH had the lowest median dose (7.5 mg OME), while the PCU had the highest (13.5 mg OME).
• Modification Patterns: A total of 93 opioid modifications were recorded during follow-up. The majority (78%) were dose increases, primarily driven by dyspnea relief (75.3%) or combined indications.
• Variability Across Settings: Significant differences were observed in the frequency and timing of dose adjustments:
  • PCU: Highest frequency of modifications (mean 2.9 per patient) with closely spaced intervals (mean 3.1 days).
  • MPCT: Intermediate frequency (mean 1.6 per patient) with moderate intervals (mean 4.9 days).
  • PCDH: Lowest frequency (mean 0.9 per patient) with markedly longer intervals between adjustments (mean 31 days).
• Correlations: No consistent pattern was identified between the number of dose adjustments and the burden of comorbidities.

Key Contributions
This study provides a detailed, longitudinal description of real-world opioid prescribing practices for dyspnea across three distinct palliative care models within a single institution. Its primary contribution is the documentation of substantial variability in prescribing behaviors, specifically regarding the timing and magnitude of dose titration. The study highlights that while clinical indications (dyspnea) drive the majority of dose increases, the operational context—defined by monitoring capacity and care delivery models—appears to significantly influence the frequency and spacing of these adjustments.

Significance and Claims
The authors claim that their findings underscore the influence of organizational context and monitoring conditions on opioid titration strategies. They posit that the observed variability suggests that prescribing practices are not solely driven by clinical efficacy or safety concerns but are adapted to the constraints and capabilities of the specific care setting. For instance, the day hospital setting, characterized by longer follow-up intervals, demonstrates a more cautious and widely spaced approach to dose adjustment compared to the intensive monitoring of the inpatient unit.

The paper concludes that these descriptive findings highlight the need for further prospective studies incorporating standardized symptom assessments and clinical outcomes to better characterize optimal prescribing strategies. The authors explicitly state that their results do not allow for extrapolation beyond the studied settings and do not assess treatment effectiveness, maintaining a modest stance regarding the generalizability of their observations.