Objective: There is limited literature examining the effect of early electroconvulsive therapy(ECT) on the length of stay(LOS), especially for psychotic disorders. This study aimed to evaluate the association between early ECT and LOS in three main groups of patients with bipolar, depressive and primary psychotic disorders.Method: A retrospective descriptive analysis of 464 patients who received inpatient ECT from May 2017 to March 2021 in a large tertiary psychiatric institution was conducted. Early ECT was defined as ECT done before the mean number of days from admission to initiation of ECT by diagnosis. The main outcome measure was LOS, which was examined by diagnosis between the early and late ECT groups, using a linear regression model with adjustment factors chosen based on univariate analysis.Results: A strong, significant association was found between early ECT and a shorter LOS (beta -28.3, 95 % CI -33.6 to -23.0, p < 0.001), even after accounting for adjustment factors. This association was consistent in all three diagnosis groups: bipolar disorders (beta -14.5, 95 % CI -20.1 to -9.0, p < 0.001), major depressive disorder (beta -18.2, 95 % CI -27.1 to -9.3, p < 0.001) and psychotic disorders (beta -39.0, 95 % CI -46.8 to -31.2, p < 0.001). There was no significant difference in the 30-day readmission rates between the early and late ECT groups transdiagnostically.Conclusion: Early ECT is strongly associated with a shorter LOS across a range of diagnostic indications including major depressive disorder, bipolar disorder and schizophrenia. ECT can be considered earlier in treatmentdecision algorithms and may offer a reduction in LOS.