COVID-19 patient-level analysis

Patient-level data is highly valuable to understand in greater detail which patient groups are likely to be severely affected by COVID-19 and are therefore at higher risk of having to be treated in critical care and at higher risk of mortality.

12 May 2020 | 1 min read

Summary

Patient-level data is a highly valuable resource, which can help hospitals understand which patient groups are likely to be severely affected by the disease and are therefore at higher risk of having to be admitted to an ICU and at higher risk of mortality.

We can analyse your patient-level data relating to COVID-19 and support your trust by:

  1. Analysing a range of risk factors for a COVID-19 hospital admission at trust and/or site level and comparing these to a peer group average
  2. Identifying the most important factors leading to critical care treatment or mortality at a trust and/or site level and comparing these to a peer group average
  3. Quantifying the risk of a patient’s in-hospital mortality and the risk of requiring treatment on a critical care unit after admission with COVID-19

Patient-level Analysis: Identifying risk factors and supporting capacity planning for COVID-19

Dr Foster has been approached to analyse patterns in patent-level data to understand, for example, why more people have died at one hospital even though the cases are lower. The exploratory data analysis includes identifying the most important risk factors for:

  1. admission with COVID-19
  2. treatment in critical care due to COVID-19
  3. mortality after testing positive for COVID-19

At a trust and site level and compared to a peer group average (based on available data)

What risk factors can we analyse?

A detailed exploratory data analysis can identify potential risk factors leading to a hospital admission due to COVID-19 and the subsequent risk of having to be treated in critical care or risk of mortality.

Cases can be broken down by a number of factors.

Further analysis

Trusts can be compared to a peer group to provide further insights into the most prevalent patient characteristics and conditions of patients admitted with COVID-19.

The analysis can also be compared to Dr Foster’s vulnerability maps in the progression dashboard: https://drfoster.com/2020/04/06/uk-covid-19-progression-dashboard/

 

The analyses proposed above can be delivered as a Tableau dashboard available to be read using Tableau Reader or as a static report. The data accompanying Tableau can also be provided separately as an Excel or csv file as required.

[1] https://www.who.int/health-topics/coronavirus#tab=tab_1

[2] https://www.nhs.uk/conditions/coronavirus-covid-19/advice-for-people-at-high-risk/