Objectively measured severe physical inactivity (SPI) has been reported as the strongest independent predictor of mortality in patients with chronic obstructive pulmonary disease (COPD). Activity monitoring is not feasible in routine clinical practice; therefore, we set out to determine the utility of simple clinical measures for predicting SPI in patients with COPD. A total of 165 patients with COPD wore an activity monitor for 5 days to define the presence or absence of SPI. Logistic models were generated including the modified Medical Research Council (MMRC) dyspnea grade, spirometry and the age-dyspnea-airflow obstruction (ADO) index. Physical Activity Scale for the Elderly (PASE) and Stanford Brief Activity Scale (SBAS) were also tested for validity and reliability in a subgroup of 67 patients. The MMRC dyspnea grade, PASE score, ADO index and SBAS score were associated with SPI, but general self-efficacy and spirometry were not. An MMRC dyspnea grade ≥3 was the best independent predictor of SPI (AUC: 0.74; PPV: 0.83; NPV: 0.68) followed closely by a PASE score of <111. The combination of MMRC dyspnea grade and PASE score provided the most robust model (AUC: 0.83; Positive Predictive Value (PPV): 0.95; Negative Predictive Value (NPV): 0.63). The results were confirmed using 5000 bootstrapped models from the cohort of 165 patients. MMRC dyspnea grade ≥3 may be the best triage tool for SPI in patients with COPD. The combination of the MMRC and PASE score provided the most robust prediction. Our results may have significant practical applicability for clinicians caring for patients with COPD.