Where's the big COVID data?

The United States has been the most impacted country in the current COVID-19 crisis.

Is COVID, the only common denominator? What else is common to the over 100,000 deaths in the United States, who is collecting and analyzing the death data comprehensively apart from the snapshot of admission and discharge diagnosis from death certificates?

Out of the deaths so far, how many were children, healthy adults, pregnant women, long term facility residents, received anti-malarial or experimental therapies, experienced unusual but similar complications like kidney failure, brain strokes, blood clots, etc.?

Most physicians we spoke to, a reputable journalist and a congresswomen felt Center for Disease Control (CDC) might be collecting it, and some responsible national entity is extensively analyzing the deaths and pooling the data.

The CDC expanded the list of possible symptoms recently that were already known for weeks, does not mention evolving trends like propensity for small blood clots, kidney failure, skin manifestations, etc. Physicians get to know this evolving information from case series and media reports. There is a huge gap for physicians without a single source for a credible and timely interpretation of patient data in this pandemic.

Morbidity and Mortality Weekly Report (MMWR) wing of CDC — often called the “voice of CDC” — is responsible for the publication of timely, reliable, objective, and useful public health information and recommendations. COVID data published for March 1 to 30, 2020, is very preliminary and only provides a snapshot on the date of admission with a big-time lag. Recent publication with the addition of clinical characteristics and outcomes is a welcome addition but has a meager sample size of 305 patients from Georgia.

The National Center for Health Statistics (NCHS) is collecting the provisional COVID-19-related death counts from billing code U07.1 and death certificates. Such information often is not high quality and does not capture all data points of interest leading to the terminal event.

The United Kingdom recently released clinical features of 16,749 people with COVID-19 across166 U.K. hospitals between February and April 2020. Data was collected and uploaded from the start of admission through to completion of the episode of care. Such a networked national database for the United States will fill some of our gaps in understanding and will serve ongoing and future needs.

A national registry can be initiated by the government for it to be orderly, secure, protected health information, centralized access to hospitals receiving federal funds, and provide regulatory approval. It should ideally allocate separate, sufficient budget and personnel to gather data without adding to the burden of existing health care systems. The CDC can also ramp up its existing infrastructure to collect comprehensive data from admission to discharge, on large samples to make it more robust and meaningful.

Without a centralized data source, we will be chasing this devious disease rather than get ahead of it.

Unifying data of over 100,000 deaths, contact tracing to identify clusters, underserved and at-risk population, comparing it to similar cohorts of recovered patients may help shape our future public health policy, differentiate futile versus potential therapies, and identify a pool of patients for vaccine trials and drug development.

While we look forward to innovating and improving on testing, treatments, and vaccines, let us learn from what is behind us, the big data of COVID deaths.

Anuradha Kolluru is a cardiologist. Rakesh Lattupalli is a nephrologist.