Public Health benefits by implementing digital symptom diaries for COVID patients from Cologne

Abstract Background High rate of people infected with SARS-CoV-2 and their contacts in Cologne, Germany required innovative tools for notification, monitoring and reporting. The digital tool for COVID19 (DiKoMa) provides self-service symptom diaries allowing (a) the stratification for prioritized telephone contact by the health authority and (b) training a machine learning (ML) model that predicts infections with prevailing dominant variant (PDV) from early symptom profiles (SP). Methods Pseudononymized SP covering the first week of diary recordings were included for training (16646 index, 11582 contacts). A balanced random forest (BRF) model was trained to differentiate early predictive symptom patterns of different PDV and contact persons. Model evaluation was performed using sex and age stratified cross validation (CV), the model was validated on SP recorded from days 1 and 6. Results From 03/20 to 02/22, 90478 indeces and 75444 contact persons reported symptoms and health status, covering 46% and 42% of all reported cases, respectively. Diaries contained between 1-52 entries (566791, median 2). Daily analysis of entries, prioritized according to age, prevalent co-morbidities and detoriation of symptoms allowed risk adjusted follow up even during phases with high case notification rates. The top 5 predictive factors of the BRF were immunization, cough, dysgeusia and dysnosmia, fatigue, and sniffles to differentiate infection between wildtype, three PDV and contact persons (CV AUC 80.6%, Validation AUC 77.1%). Conclusions The use of digital symptom diary surveillance helps to provide appropriate medical support for patients on a large scale. Machine learning shows potential for symptom based risk assessment to differentiate PDV for future outbreaks and can thus become a valuable tool alongside specific laboratory diagnostics. Key messages • Digital symptom diaries are a powerful and widely accepted tool to attend COVID19 patients in isolation. They allow risk stratification for follow up and are a low-threshold service. • Machine learning supports index case identification by symptom analysis and can thus become a valuable tool alongside specific laboratory diagnostics.


Background:
Non-communicable diseases (NCDs) impose a heavy burden on healthcare systems of countries in the European Union (EU). An estimated 91.3% of all deaths and 86.6% of DALYs in the EU-28 were attributable to NCDs. It is imperative that the EU act on mitigating this challenging health issue and help create trajectories for building resilient health systems. Using qualitative analysis, this study examines the question of how the European Commission 2019-2024 is planning to mitigate the impact of NCDs on health systems, while taking into account the COVID-19 pandemic. Methods: Content analysis was applied to understand how NCDs are framed and how an EU narrative is constructed to mitigate the impact of NCDs on health systems by the European Commission. A total of 44 documents were analysed, including speeches, press releases, newsletters, statements and policy documents. In vivo coding was performed using the software package ATLAS.ti 9. Unique codes were simplified and clustered into descriptive themes with a high level of abstraction.

Results:
This study identified five main themes: 'health plan', 'COVID-19', 'future direction', 'collaboration and solidarity', and 'persuasion'. Themes show that the Commission is emphasising the impact of the pandemic and the relevance of policies tackling NCDs for developing EU-wide resilient health systems. By calling for more cross-and multi-sectoral collaboration, like creating a European Health Union, the Commission hopes to create the right climate for a European framework for cooperation, which can help develop harmonised and resilient EU health systems.

Conclusions:
Although increasing health systems resilience is high on the Commission's agenda, it seems that there are no actionable points for Member States in terms of addressing national health policy. We recommend the Commission looks towards eliminating this observed disconnect and creating actionable points in line with Member States' health systems' capabilities.

Key messages:
Our findings show emphasis is placed on EU-wide health policies, more robust health systems, and the collective power of the EU. However, there are no concrete actions coupled with these concepts. COVID-19 highlighted the limitations of national policy for protecting health systems against cross-border health threats. It tested the Commission's resolve in pushing for more European cooperation.

Background:
High rate of people infected with SARS-CoV-2 and their contacts in Cologne, Germany required innovative tools for notification, monitoring and reporting. The digital tool for COVID19 (DiKoMa) provides self-service symptom diaries allowing (a) the stratification for prioritized telephone contact by the health authority and (b) training a machine learning (ML) model that predicts infections with prevailing dominant variant (PDV) from early symptom profiles (SP).

Methods:
Pseudononymized SP covering the first week of diary recordings were included for training (16646 index, 11582 contacts). A balanced random forest (BRF) model was trained to differentiate early predictive symptom patterns of different PDV and contact persons. Model evaluation was performed using sex and age stratified cross validation (CV), the model was validated on SP recorded from days 1 and 6.

Results:
From 03/20 to 02/22, 90478 indeces and 75444 contact persons reported symptoms and health status, covering 46% and 42% of all reported cases, respectively. Diaries contained between 1-52 entries (566791, median 2). Daily analysis of entries, prioritized according to age, prevalent co-morbidities and detoriation of symptoms allowed risk adjusted follow up even during phases with high case notification rates. The top 5 predictive factors of the BRF were immunization, cough, dysgeusia and dysnosmia, fatigue, and sniffles to differentiate infection between wildtype, three PDV and contact persons (CV AUC 80.6%, Validation AUC 77.1%).

Conclusions:
The use of digital symptom diary surveillance helps to provide appropriate medical support for patients on a large scale. Machine learning shows potential for symptom based risk assessment to differentiate PDV for future outbreaks and can thus become a valuable tool alongside specific laboratory diagnostics.
Key messages: Digital symptom diaries are a powerful and widely accepted tool to attend COVID19 patients in isolation. They allow risk stratification for follow up and are a low-threshold service.
Machine learning supports index case identification by symptom analysis and can thus become a valuable tool alongside specific laboratory diagnostics.
iii432 European Journal of Public Health, Volume 32 Supplement 3, 2022