The ATHENA COVID-19 Study: Cohort profile and first findings for people diagnosed with COVID-19 in Queensland, 1 January to 31 December 2020

Abstract

Background: To date, there are limited Australian data on characteristics of people diagnosed with COVID-19 and on how these characteristics relate to outcomes. The ATHENA COVID-19 Study was established to describe health outcomes and investigate predictors of outcomes for all people diagnosed with COVID-19 in Queensland by linking COVID-19 notification, hospital, general practice and death registry data. This paper reports on the establishment and first findings for the ATHENA COVID-19 Study. Methods: Part 1 of the ATHENA COVID-19 Study used Notifiable Conditions System data from 1 January 2020 to 31 December 2020, linked to: Emergency Department Collection data for the same period; Queensland Health Admitted Patient Data Collections (from 1 January 2010 to 30 January 2021); and Deaths Registrations data (from 1 January 2020 to 17 January 2021). Results: To 31 December 2020, a total of 1,254 people had been diagnosed with SARS-CoV-2 infection in Queensland: half were female (49.8%); two-thirds (67.7%) were aged 20–59 years; and there was an over-representation of people living in less-disadvantaged areas. More than half of people diagnosed (57.6%) presented to an ED; 21.2% were admitted to hospital as an inpatient (median length of stay 11 days); 1.4% were admitted to an intensive care unit (82.4% of these required ventilation); and there were six deaths. Analysis of factors associated with these outcomes was limited due to small case numbers: people living in less-disadvantaged areas had a lower risk of being admitted to hospital (test for trend, p < 0.001), while those living in more remote areas were less likely than people living in major cities to present to an ED (test for trend: p=0.007), which may reflect differential health care access rather than health outcomes per se. Increasing age (test for trend, p < 0.001) and being a current/recent smoker (age-sex-adjusted relative risk: 1.61; 95% confidence interval: 1.00, 2.61) were associated with a higher risk of being admitted to hospital. Conclusion: Despite uncertainty in our estimates due to small numbers, our findings are consistent with what is known about COVID-19. Our findings reinforce the value of linking multiple data sources to enhance reporting of outcomes for people diagnosed with COVID-19 and provide a platform for longer term follow-up.

Participant characteristics and reasons for non-consent to health information linkage for research: experiences from the ATHENA COVID-19 Study.

“…This study differs from the majority of previous reports as it investigates why participants decline to take part in data linkage research, rather than tending to explore reasons they might agree (17). In a similar study to ours, the activities taken up by people in retirement were explored using a cross-sectional survey (22).…”

“…Participants' ages are reported to affect attitudes to consent. The general ndings are that older respondents are more willing to share or link data than younger participants (17). However, there is some variation, mostly by country in which younger participants have higher consent rates, making direct comparison with Australia di cult (32,33).…”

“…However, there is some variation, mostly by country in which younger participants have higher consent rates, making direct comparison with Australia di cult (32,33). Gender differences in consent rates have not been consistently found, but when found to be a predictor, males were more likely to consent (17,33).…”

“…Concern about privacy is a common theme cited in many reports as to why participants do not wish to share their health data (13,17,30). Speci c reasons include inadequate data security and subsequent exploitation by companies to discriminate and make pro t (30).…”

“…It is therefore imperative that we understand the attitudes of participants whom we are approaching for consent. Whilst numerous studies have examined the motives and experiences of individuals who participate in research, little is known about reasons for nonparticipation, particularly in an Australian context (17). Understanding these reasons is important if we are to attempt to improve consent rates for data linkage for the purposes of research in Australia.…”