Background Cardiac troponin (T and I) are considered the standard markers for detection of myocardial damage and the diagnosis of acute coronary syndrome (ACS) among patients who present to an emergency department with chest pain. However, these markers can be released in other situations and may be associated with short- and long-term clinical outcomes. In this study, we examine late mortality rates among patients presenting with a suspected ACS due to an unstable coronary plaque and those patients having a non-ACS. Methods 4388 patients were hospitalised with suspected ACS, between 14 and 27 May 2012 in the Australia and New Zealand SNAPSHOT ACS study. Those patients were categorised in five diagnostic groups: 1) ST elevation MI (n=419); 2) non-ST elevation MI (n=1012); 3) unstable angina (n=925); 4) non-ACS diagnoses (n=837); and 5) chest pain considered unlikely ischaemic (not otherwise specified, n=1195). Result The respective mortality rates at 18 months in these groups were 16.2%, 16.3%, 6.8%, 12.8%, and 4.8%; Pearson χ2=110 p<0.001. Among non-ACS diagnoses patients (group 4) those with the highest mortality rates (cardiac (14.4%), respiratory (18.2%), sepsis (15.4%) and neoplastic (67%) diagnoses) had the highest rates of elevated troponin levels (48%, 31%, 38% and 67% respectively). By contrast, those with the lowest mortality rates (musculoskeletal (2.9%), gastrointestinal disorders (3.9%) and non-specific chest pain (7.4%)) had the lowest rate of elevated troponin levels (9%, 18% and 15.8% respectively). However, after adjusting for baseline clinical and demographic characteristics, the mortality rate at 18 months for patients with elevated troponin was similar for ACS or non-ACS diagnoses (Hazard Ratio, 95% C.I.0.98–1.07, p=0.333). Conclusions Among patients in the 2012 SNAPSHOT ACS study, non-ACS diagnoses characterised by high rates of elevated troponin levels had high mortality rates similar to those diagnosed with ACS. Therapies known to be effective in ACS patients, including early invasive management, should be examined in these non-ACS patients with troponin elevations within adequately powered randomised trials.