AbstractRisk stratification models in pulmonary arterial hypertension (PAH) rely on World Health Organisation Functional Class (WHO FC). A high proportion of patients are classified as WHO FC III, a heterogenous group which limits the stratification abilities of risk models. The Medical Research Council (MRC) Dyspnoea Scale may allow a more precise assessment of functional status and improve current risk models. We investigated the ability of the MRC Dyspnoea Scale to assess survival in PAH and compared performance to WHO FC and the COMPERA 2.0 models. Patients with Idiopathic, Hereditary or Drug‐induced PAH who were diagnosed between 2010 and 2021 were included. The MRC Dyspnoea Scale was retrospectively applied as derived from a combination of patient notes, 6MWD tests results and WHO functional status using a purpose‐designed algorithm. Survival was assessed using Kaplan–Meier analyses, log rank testing and Cox proportional hazard ratios. Model performance was compared with Harrell's C Statistic. Data from 216 patients were retrospectively analyzed. At baseline, of 120 patients classified as WHO FC III, 8% were MRC Dyspnoea Scale 2, 12% Scale 3, 71% Scale 4 and 10% Scale 5. The MRC Dyspnoea Scale performed well compared to the WHO FC and COMPERA models at follow up (respectively, C‐statistic 0.74 vs. 0.69 vs. 0.75). It was possible to use the MRC Dyspnoea Scale to subdivide patients in WHO FC III into groups which had distinct survival estimates. We conclude that at follow‐up, the MRC Dyspnoea Scale may be a valid tool for the assessment of risk stratification in pulmonary arterial hypertension.