Michael Donnelly1, Gilbert MacKenzie2. 1. , Bishop Auckland, Durham, DL13, UK. 2. Centre of Biostatistics, University of Limerick, Limerick, Ireland. gilbert.mackenzie@ul.ie.
Abstract
This paper deals with the analysis and interpretation of data relating to mortality and survival in the first year of operation of the Erne MICCU study in Co. Fermanagh. AIMS: We aimed to measure in-hospital mortality from AMI, on WHO criteria, identify factors influencing mortality and survival and assess the performance of the MICCU. METHODS: All first admissions of suspected AMI to the CCU from the Fermanagh District in 1983-1984. Some 297 patients were analysed. We recorded demographic data, previous history of heart disease and co-morbidity, status of the current attack, delay to CCU, treatment and outcome. In total, 28 variables grouped as (a) basic risk factors (18) and (b) clinical and treatment risk factors (10) were analysed. OUTCOMES: In-hospital mortality and survival and performance of the MICCU. RESULTS: There were 329 admissions to the CCU of all types of which 297 (90.3%) were first admissions. Of the 297, 170 (57.2%) had AMI on WHO criteria and 42 (14.1%) were dead at discharge. Crude, 28-day, mortality and unadjusted survival were statistically significantly worse in the AMI group. The multi-factor mortality analysis identified 5 variables influencing death at discharge. In relation to multi-factor survival, the MPR Weibull model identified a set of 9 variables in which the treatment variables pre-dominated over basic risk factors. The MICCU delivered patients to hospital statistically significantly earlier (5 h on average) than other modes of transport, but did not prevent more deaths than the ordinary ambulance. CONCLUSIONS: There was no evidence of a direct, statistically significant, beneficial MICCU effect in either of the multi-factor mortality or survival models. However, the performance of the MICCU, measured in terms of crude survival, resulted from an adverse case-mix, which, when controlled for, suggested a small MICCU benefit. The findings relate to the first year of operation of the Erne MICCU study and may improve in later years.
This paper deals with the analysis and interpretation of data relating to mortality and survival in the first year of operation of the Erne MICCU study in Co. Fermanagh. AIMS: We aimed to measure in-hospital mortality from AMI, on WHO criteria, identify factors influencing mortality and survival and assess the performance of the MICCU. METHODS: All first admissions of suspected AMI to the CCU from the Fermanagh District in 1983-1984. Some 297 patients were analysed. We recorded demographic data, previous history of heart disease and co-morbidity, status of the current attack, delay to CCU, treatment and outcome. In total, 28 variables grouped as (a) basic risk factors (18) and (b) clinical and treatment risk factors (10) were analysed. OUTCOMES: In-hospital mortality and survival and performance of the MICCU. RESULTS: There were 329 admissions to the CCU of all types of which 297 (90.3%) were first admissions. Of the 297, 170 (57.2%) had AMI on WHO criteria and 42 (14.1%) were dead at discharge. Crude, 28-day, mortality and unadjusted survival were statistically significantly worse in the AMI group. The multi-factor mortality analysis identified 5 variables influencing death at discharge. In relation to multi-factor survival, the MPR Weibull model identified a set of 9 variables in which the treatment variables pre-dominated over basic risk factors. The MICCU delivered patients to hospital statistically significantly earlier (5 h on average) than other modes of transport, but did not prevent more deaths than the ordinary ambulance. CONCLUSIONS: There was no evidence of a direct, statistically significant, beneficial MICCU effect in either of the multi-factor mortality or survival models. However, the performance of the MICCU, measured in terms of crude survival, resulted from an adverse case-mix, which, when controlled for, suggested a small MICCU benefit. The findings relate to the first year of operation of the Erne MICCU study and may improve in later years.
Authors: W J McIlwaine; M D Donnelly; M Mallaghan; A T Chivers; A E Evans; J H Elwood; A A Adgey; N P Campbell; J S Geddes Journal: Br Heart J Date: 1986-04