OBJECTIVES:: The Hypoglycaemia Assessment Tool (HAT) study in Brazil aimed to characterize incidence of hypoglycaemic events and associated health resource use and costs in type 1 diabetes mellitus (T1DM) and type 2 (T2DM) patients among other objectives. METHODS:: Non-interventional, multicentre study with 679 patients, conducted in 22 centres across all Brazilian regions with a 6-month retrospective and a 4-week prospective assessment of hypoglycaemic events. Patients self-reported health resource use related with hypoglycaemia events using a paper-questionnaire. Direct costs included healthcare resources to assist hypoglycaemia episodes (hospitalization, emergency assistance, extra visits to physician/nurse and outpatient administration of carbohydrates or glucagon) whereas indirect costs were related with labour absenteeism due to hypoglycaemia (sick leave and patients’ delay to work) using Capital Human Approach. Prices were obtained from official Brazilian sources (DATASUS, SIGTAP, BPS and IBGE). Yearly costs were extrapolated based on the prospective assessment, in Brazilian Reais, for 2015 from the societal perspective. RESULTS:: From 281 evaluable patients for economic analysis, 47%/53% were T1DM/T2DM, mean age of 35/64 years, 42% were male in both groups and diabetes duration averaged 16/18 years. Yearly average costs due to hypoglycaemia per T1DM/T2DM patient were: hospitalization R$453/R$258, emergency R$27/R$29, carbohydrates or glucagon R$156/R$100 and additional physician and nurse visits R$4/R$3. On average, hypoglycaemia episodes represents a yearly cost to the Brazilian society of R$709 per T1DM patient (range: R$0-R$12.364; direct cost: R$640; indirect cost: R$69) and R$396 per T2DM patient (range: R$0-R$10.431; direct cost: R$390; indirect cost: R$6). CONCLUSIONS:: These results are important to understand the economic burden of hypoglycaemias in patients with diabetes in Brazil. Hospitalizations were the cost driver. Estimated costs for hypoglycaemias in T1DM patients are twice the costs in T2DM probably due to higher incidence of hypoglycaemias. Missing data was a study limitation.