Hacker News Reader

Comment by lpolovets | Hacker News Reader

If I'm reading Figure 5 right in the original paper*, it looks like the biggest cardiac impact is on people who were hospitalized or went to the ICU for covid, and the effect on people who were not hospitalized is present but relatively small. Am I interpreting the data correctly?

* https://www.nature.com/articles/s41591-022-01689-3 h/t to @greenyoda for posting elsewhere in this thread.

View on HN

Yes. Also read the supplementary tables that describe the demographic co-variates of the study population:

https://static-content.springer.com/esm/art%3A10.1038%2Fs415...

Mean age of the Covid-19 cohort: 61.42

Percent with BMI over 30: 53.84%

Percent with chronic kidney disease: 19.52%

Lung disease: 15.26%

Diabetes: 32.13%

Hyperlipidemia: 33.97%

Hypertension: 26.7%

Almost all of these factors are significantly elevated over the control groups, and their attempts to normalize the data (ST2) do not eliminate the problem. We're talking about tiny differences in small numbers, and their population is biased toward the sick and the old. It's already clear from the main paper that most of the signal comes from the oldest, sickest cohort, but once you know that pretty much everyone in the study was old and sick, it's clearly difficult to generalize from this.

This thread is awash with 20-somethings pontificating about heart damage based on a paper about elderly veterans. Simply irresponsible reporting.

Relatively small compared to the impact for severe cases, but it looks very relevant if you consider the amount of people that are getting COVID and that many governments and societies are basically throwing the towel and accepting that people should just catch it.

This isn't a representative sample of the general public. Per supplementary tables 1-2, the average age of the participant was over 60, and had multiple co-morbidities. We know that age is the #1 risk factor.

The people in the "hospitalized" cohort, in particular, were quite old and sick prior to the disease.

Covid infections aren't the only thing that happened. Lives changed and access to medical care was reduced. Both could easily explain the slight increase in health risks.

I also didn't see them compare the contemporary control group to 2017. The majority of the contemporary control group probably were infected with Covid. If there were a significant impact they should have different outcomes than the 2017 group.

that's how I'm reading it as well. the Nature article does acknowledge this, but not with a lot of detail.

Reply on news.ycombinator.com