Patterns in Hospital Admissions

Between two and four times a day, the Health Ministry releases updated statistics on the coronavirus in Israel. The frequency of these releases makes it feel like we’re getting real-time detailed insights into what’s going on with the virus and Israel’s attempts to manage the pandemic.

The data includes:

  • the number of tests so far today (and how many of those tests were given to diagnose Covid-19 rather than to confirm recovery),
  • the numbers of new and recovered cases
  • how many people are currently admitted in hospitals
  • How many cases are considered serious or moderate
  • The number of patients on ventilators
  • The number of deaths

All of these numbers are routinely updated to give a national picture. There’s also regional-level data on infections and recoveries.

But the impression of continuous data is actually a little misleading. Israel’s a small country, but it still has four HMOs that do the virus testing at hundreds of national sites. Around 30 hospitals have Covid-19 patients in their wards. All that data has to be collated by people at each of these places, sent to the Health Ministry and fed into their systems.

Patients in hospital with Covid-19 in Israel July/August, according to the Health Ministry

The graph above tracks Israel Health Ministry daily data on the number of hospitalised Covid-19 patients. I’ve also added a mark for each Saturday on the graph.

In July, hospital admissions rose rapidly from 230 at the start of the month (not graphed) to around 750. Since then, throughout August, they’ve bounced around between that level to as high as 875.

Looking closer, you’ll notice that the number of hospital admissions seems to spike around Saturday or Sunday. In August, these early-week spikes have tended to dip again towards the middle of the week before rising again towards the weekend.

If I had to guess, I’d put this down to reporting cycles from the 30 hospitals rather than real spikes. Perhaps some hospitals only report their admission numbers once a week? Or they use Friday and Saturday to find any cases over the week that slip through the cracks?

Or perhaps the admissions spikes are real and there’s some reason to explain them. Perhaps families visit their older relatives on Shabbat, and force Grandpa to go to hospital with that nasty cough, for example. Or maybe it’s a lagging indicator of some infection pattern which happens on a similar cycle.

UPDATE: I missed something important and obvious here: many hospitals don’t do discharges on Saturday or for much of Friday. Patients might even have to wait until Sunday evening to get their clear Covid test and go home. That could also lead to a bulge in admitted patients over the weekend while healed people wait to be allowed to go home. Thanks to Abi who pointed this out. Only thing is, wouldn’t we expect to see that bulge disappear by Monday?

Either way, don’t stress the weekend hospital spikes. They seem to be normal and fade away over the week.

One thought to “Patterns in Hospital Admissions”

Comments are closed.