Data from three MONICA centres in Auckland (New Zealand) and Newcastle and Perth (Australia) are used to explore some of the issues involved in comparing event rates and case fatality among MONICA centres. Auckland and Newcastle follow the "hot pursuit" method of identifying and interviewing patients while they are still in hospital. Perth follows the "cold pursuit" method, in which patients are identified by search of computerized hospital records after discharge and all data are abstracted retrospectively from case notes. Fatal cases are identified by the same method in the three centres. The distribution of events by MONICA diagnostic classification varied among centres, with Perth having the highest proportion of definite myocardial infarction events and the lowest proportion of possible myocardial infarction events. These differences appear to be due to the different methods of event ascertainment and data collection, and to variations in post mortem rates between centres. For comparisons among these three centres, the categories of non-fatal definite myocardial infarction and of all coronary heart disease deaths (that is those in the MONICA categories fatal definite myocardial infarction, fatal possible myocardial infarction, and fatal cases with insufficient data) appear to be the most useful.