The study was performed at the La Princesa Hospital, a 554-bed tertiary-care university teaching hospital for adults, located in Madrid, Spain. The hospital provides services in all the medical specialties and in all fields of surgery.
An observational retrospective study was carried out in a population of patients ≥90 years old admitted for acute myocardial infarction as primary diagnosis ( ICD 9, code 410 ) from January 1, 2003, and January 1, 2014. Patients who were admitted several times during the study period had only their first stay included in the study. The recommendations of the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology)[11] declaration for observational studies …show more content…
These data were collected from in-hospital and outpatient database (Nostradamus program, Madrid community health system).
Statistical analysis Mean and standard deviation or median and interquartile (IQ) range were calculated, depending on distribution of continuous variable. The absolute percentage and relative frequencies were used for categorical variables.
In univariate analysis, א2 test was performed for dichotomous variables. When the variable was quantitative, Student's t test and Mann Whitney U test were performed, according to the parametric or nonparametric distribution.
Those variables with p ≤ 0.25 in the univariate analysis, were included in the multivariate analysis. A cutoff of p value ≤0.25, is a criteria that has been recommended in literature[16] to include potential predictors. The logistic regression model was used thorough the forward method in multivariate analysis of mortality. Goodness of fit was assessed with the Hosmer-Lemeshow test.
Because of ehcocardiography was not performed in the 45.0% of the sample, we compare the distributions of baseline (sociodemographic, risk factors, past medical history, symptoms, data at admission and treatment) and outcome