Association of recent clinical respiratory illness with STEMI and
high-troponin level in blood among AMI patients:
Multivariate log binomial regression analyses were performed to
investigate associations of CRI with occurrence of STEMI and high blood
troponin among AMI patients, controlling for potential confounders. All
variables deemed clinically relevant as well as having potential
confounding effect and also based on P-value were entered in the base
model and sequentially removed using a backward stepwise approach
whereby variables with P values ≤0.25 remained in the final model. We
investigated the aforementioned associations for, 1) entire study period
and 2) independently for either the influenza or non-influenza seasons.
The magnitudes of association were expressed as relative risk (RR) and
adjusted relative risk (aRR) with 95% confidence interval (CI) in
unadjusted and adjusted analyses respectively. The SAS statistical
software version 9.4 (SAS institute, USA) was used to run the analyses.