Comparison of computer-assigned Minnesota Codes with the visual standard method for new coronary heart disease events

The Minnesota Code is the most widely used electrocardiogram (ECG) classifi cation system for epidemiologic studies and has been incorporated into seve ral computer algorithms. The authors compared the Modular ECG Analysis Syst em (MC-MEANS) and NOVACODE computer ECG findings with the Visual coding sta ndard for agreement and prognostic associations with coronary heart disease (CHD) events occurring during follow-up from 1987 to 1995 in 2,116 individ uals participating in the Atherosclerosis Risk in Communities (ARIC) Study. The exact agreement between Visual and computer findings was greater than 90% for all Minnesota Code categories except Q-code, which was 77% for MC-M EANS and 81% for NOVACODE. Approximately 60% of all Q-codes were assigned b y computer methods only. Among the 2,116 participants, there were 246 (11.6 %) new coronary events. Unadjusted relative risks for codes assigned by the three methods were similar. When computer methods disagreed on code severi ty, the GHD occurrence rates for MC-MEANS-detected severer code versus NOVA CODE-detected severer code were 21% and 7%, respectively. This study provid es clear evidence that computers assign more and severer Minnesota Codes wi th similar prognostic importance as does the Visual method; it also alerts researchers to potential problems in pooling Minnesota Code data read by di fferent methods.