New tool predicts hospitalization and mortality risk from COVID-19

A new risk tool developed by UK researchers to predict the risk of hospital admissions and deaths from COVID-19 has been published by The. BMJ today.

As cases grow in the UK and elsewhere and winter approaches, a reliable model that predicts the possible course of COVID-19 is urgent to support shield, hospitalization, treatment, and vaccination decisions. Is needed for.

Risk prediction tools (known as QCOVIDs) develop severe illnesses using readily available information about people, such as age, ethnicity, and the presence of certain pre-existing conditions (comorbidities). Identify the individuals at highest risk. It is designed to apply to the entire UK’s general adult population.

This tool provides subtle information about the risk of people with serious illness due to COVID-19 and is designed to be used by clinicians with patients to reach a shared understanding of risk.

As the pandemic progresses and its performance is closely monitored, the tools need to be updated regularly.

Several previous risk prediction models have been developed. They have been identified as having a high risk of bias, raising concerns that these models may be unreliable when applied in practice.

A UK-wide research group has begun developing and validating a population-based predictive model to estimate the overall risk of infection and hospitalization or death from COVID-19. Measures have been taken to mitigate the causes of known biases.

Their findings are based on data from more than 8 million patients aged 19-100 years in 1,205 UK general practice, COVID-19 test results and hospitals and death Registry data.

Developed a 97-day (January 24 to April 30, 2020) model using data from 6 million patients, and two separate periods (2020) using an additional 2.2 million patients We verified the performance from January 24th to April 30th and May 1st). During the first wave of the pandemic (until June 30, 2020).

To develop the model, we estimated the probability and timing of hospitalization or death from COVID-19 using known factors such as age, ethnicity, deprivation, obesity index, and range of comorbidities.

During the study period, 4,384 deaths from COVID-19 occurred in the development group, 1,782 in the first validation period and 621 in the second validation period.

The model worked well and predicted 73% and 74% of COVID-19 time-to-death variability in men and women, respectively.

The top 5% of predicted mortality risks accounted for 76% of COVID-19 deaths during the 97-day study period. The top 20% of people at the expected risk of death accounted for 94% of deaths from COVID-19.

Researchers point out that the model aims to provide risk prediction. It is not intended to provide an explanation of which individual factors have a causal effect on risk and the results should not be interpreted in this way.

The absolute risks obtained from the model vary over time depending on the general COVID-19 infection rate and the degree of social distance measurements performed, and these should also be interpreted with caution. However, the order of individuals with respect to risk is expected to be relatively stable over time, allowing you to identify the individual at highest risk.

According to researchers, QCOVID represents a powerful risk prediction model that may support public health policies, from potential shared decisions to mitigate health and workplace risks to targeted recruitment. I will. Clinical trial Prioritization of vaccinations.

The· model It can be readjusted over different periods of time and may be updated regularly as the pandemic progresses.

QCOVID is specially designed to inform UK health policy and interventions to manage COVID-19-related risks, but has international potential, subject to regional validation. They conclude.

In a linked editorial, researchers at the University of Manchester found that QCOVID and ISARIC (International Severe Acute Respiratory and Emerging Infectious Diseases Consortium) 4C (Coronavirus Clinical Characteristics Consortium) mortality scores improve the quality of the COVID-19 prognosis model. I agree to represent it, but say that care must be taken when interpreting the predictions generated by these models.

Given the rapidly changing nature of the disease and its management, they also emphasize the need to update these models on a regular basis and closely monitor performance over time and space.

They acknowledged that improving data on COVID-19 incident cases would “improve the granularity of forecasts” and, along with these warnings, “support continuous validation and impact assessment of these models.” Stated.