Background and Purpose: Hydroxychloroquine and chloroquine, alone or in combination with azithromycin, have been proposed as therapies for COVID-19. However, there is currently scant and inconsistent data regarding their proarrhythmic potential in these patients. Moreover, their risk profile in the setting of altered physiological states encountered in patients with COVID-19 (i.e. febrile state, electrolyte imbalances, and/or acidosis) is unknown. Experimental approach: Potency of hERG block was measured using high-throughput electrophysiology in the presence of variable environmental factors. These potencies informed simulations to predict population risk profiles. Effects on cardiac repolarisation were verified in human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CM) from three separate individuals. Key Results: Chloroquine and hydroxychloroquine blocked hERG with IC50 of 1.47±0.07 µM and 3.78±0.17 µM respectively, indicating proarrhythmic risk at concentrations effective against SARS-CoV-2 in vitro and proposed in COVID-19 clinical trials. Hypokalaemia and hypermagnesemia increased potency of chloroquine and hydroxychloroquine, indicating increased proarrhythmic risk. Acidosis significantly reduced potency of all drugs (i.e. reduced proarrhythmic risk), whereas increased temperature decreased potency of chloroquine and hydroxychloroquine but increased potency for azithromycin. In silico simulations across genetically diverse populations predicted that 17% of individuals exhibit action potential durations >500 ms at the highest proposed therapeutic levels, equating to significant QT prolongation. Conclusion and Implications: Significant proarrhythmic risk is predicted for hydroxychloroquine and chloroquine at doses proposed to treat COVID-19. Clinicians should carefully consider the risk of such treatments, and implement long term QT interval monitoring in trials, particularly in patients with electrolyte imbalances.