Exploring factors associated with paramedic conveyance decisions for patients with Chronic Obstructive Pulmonary Disease: a retrospective cohort study

Abstract

Background: Paramedic non-conveyance is an established and expanding component of pre-hospital care; however, the processes underpinning these decisions remain incompletely understood. Existing research is characterised by heterogeneous patient populations, reliance on measurable risk predictors, and limited linkage to outcomes, resulting in an incomplete account of how decisions are made in practice. This gap is particularly evident in condition-specific analyses, including chronic obstructive pulmonary disease (COPD), where clinical complexity and baseline physiological variability introduce significant uncertainty into decision-making. Aim: This study examines the associations among patient demographics, physiological observations, service-level factors, COPD pathway suitability and paramedic conveyance decisions, and examines how hospital admission, length of stay and 30-day mortality vary across physiological risk classifications, pathway-defined groups and observed dispositions. Methods: A retrospective cohort study was conducted using routinely collected Scottish Ambulance Service data linked to hospital outcome data. Patients presenting with COPD were identified and analysed with respect to conveyance decisions. Variables included demographics, physiological observations, and service-level factors. Outcomes were examined across physiological risk classifications, pathway-defined groups and observed disposition categories. Results: Respiratory rate and heart rate were strongly associated with conveyance decisions, while oxygen saturation showed a directional but statistically non-significant association. Associations between physiological and pathway classifications and subsequent outcomes varied across the cohort. Patients identified as higher risk were more likely to be conveyed, admitted, and to experience adverse outcomes. In contrast, demographic and service-level factors showed no consistent association with conveyance decisions. However, measurable variables did not fully explain decision-making. Divergence between physiological status and disposition was consistently observed, with both non-conveyance of higher-risk patients and conveyance of lower-risk patients evident in the dataset. Conclusions: The findings suggest that recorded physiological variables and pathway criteria do not fully account for observed variation in conveyance decisions. Although this pattern is consistent with influences beyond the measured variables, including clinical and contextual factors, those influences were not directly examined.