Background: Cognitive reserve (CR) is closely associated with cognitive and functional outcome, disease severity, progression and prognosis in psychiatric patients; however, it has not been extensively tested in mood disorders. This study examined the psychometric properties of the Cognitive Reserve Assessment Scale in Health (CRASH) in mood disorder patients. Methods: Altogether 166 subjects were recruited, 44 with major depressive disorder (MDD), 64 with bipolar disorder (BD), and 58 healthy controls. CR was assessed using the CRASH and the Cognitive Reserve Questionnaire (CRQ). Results: Internal consistency (Cronbach's alpha) was 0.779 for the CRASH. The Receiver Operating Characteristic (ROC) curve analysis revealed an area under the ROC curve (AUC) value of 0.73 (95 % CI: 0.647–0.809). The optimal cut-off score of 51 generated the best combination of sensitivity (0.78) and specificity (0.43) for discriminating between patients with mood disorders and healthy controls. The CRASH score was highly correlated with the CRQ score in both mood disorder patients (r = 0.586, P < 0.001) and healthy controls (r = 0.627, P < 0.001), indicating acceptable convergent validity for the CRASH. Within the mood disorder sample, the CRASH score was associated with functional outcomes (FAST: r = −0.243, P = 0.011). Conclusions: The CRASH is a useful tool to measure CR in mood disorder with acceptable psychometric properties and could be used in both research and clinical practice.