The Role of Health Preconditions on COVID-19 Deaths in Portugal: Evidence from Surveillance Data of the First 20293 Infection Cases

J Clin Med. 2020 Jul 24;9(8):2368. doi: 10.3390/jcm9082368.

Abstract

Background: It is essential to study the effect of potential co-factors on the risk of death in patients infected by COVID-19. The identification of risk factors is important to allow more efficient public health and health services strategic interventions with a significant impact on deaths by COVID-19. This study aimed to identify factors associated with COVID-19 deaths in Portugal.

Methods: A national dataset with the first 20,293 patients infected with COVID-19 between 1 January and 21 April 2020 was analyzed. The primary outcome measure was mortality by COVID-19, measured (registered and confirmed) by Medical Doctors serving as health delegates on the daily death registry. A logistic regression model using a generalized linear model was used for estimating Odds Ratio (OR) with 95% confidence intervals (95% CI) for each potential risk indicator.

Results: A total of 502 infected patients died of COVID-19. The risk factors for increased odds of death by COVID-19 were: sex (male: OR = 1.47, ref = female), age ((56-60) years, OR = 6.01; (61-65) years, OR = 10.5; (66-70) years, OR = 20.4; (71-75) years, OR = 34; (76-80) years, OR = 50.9; (81-85) years, OR = 70.7; (86-90) years, OR = 83.2; (91-95) years, OR = 91.8; (96-104) years, OR = 140.2, ref = (0-55)), Cardiac disease (OR = 2.86), Kidney disorder (OR = 2.95), and Neuromuscular disorder (OR = 1.58), while condition (None (absence of precondition); OR = 0.49) was associated with a reduced chance of dying after adjusting for other variables of interest.

Conclusions: Besides age and sex, preconditions justify the risk difference in mortality by COVID-19.

Keywords: COVID-19; area under the curve; cancer; cardiovascular; demographics; diabetes; epidemiology; logistic regression; mortality; systemic condition.