Original cohort members of the Framingham Study were studied to determine prognosis associated with unexplained (gradual-onset) Q wave patterns detected on routine electrocardiograms (ECGs). Biennial ECGs were obtained on subjects beginning in 1948. The index ECG was read independently and then serially in relation to all previous tracings for evidence of myocardial infarction. Proportional hazards modeling was used to assess the risk for outcomes in those with unexplained (gradual-onset) Q wave patterns compared with subjects with unrecognized myocardial infarcts (abrupt-onset Q waves) and compared with referent subjects. Unexplained Q wave patterns developed in 53 men and 39 women. Subjects with unexplained Q wave patterns had less than 40% of the risk for myocardial infarction or coronary death as compared with those with unrecognized myocardial infarction after adjusting for risk factors (hazard ratio, 0.38; 95% confidence interval, 0.19, 0.78; P = .008) and a trend toward a lower risk for overall mortality (adjusted hazard ratio, 0.73; 95% confidence interval, 0.46, 1.16; P = .18). Risks for myocardial infarction or coronary death and overall mortality were similar in subjects with unexplained Q wave patterns and referent subjects in the fully adjusted models. Detection of prominent Q waves on an ECG requires comparison with prior tracings to differentiate the unexplained Q wave pattern from the unrecognized infarct, conditions that have significantly different prognostic implications.