The numbers for overall mortality in Italy for the
third week of March are out (all deaths of people aged 65 or older). For the whole of Italy, mortality in the third week was 56% above expected, compared to 21.4% for the second week, and 9.5% for the first. (Green curve is average over the past several years, blue curve is current.)
For this
report, Sismg also broke it down by Northern Italy and Central/Southern Italy. This highlights the difference in how hard the north has been hit by excess deaths compared to the south.
So in the north, mortality is close to 2x what was expected for the whole region, while to the south, it's about 1.2x. (For Bergamo City, see my
previous post.) If these trends continue to correspond with the number of reported "coronavirus deaths" in Italy, then it looks like overall mortality in the report for the fourth week of March will increase again, then start falling for the first week of April.
On a side note, I tried finding comparable data for New York City. It doesn't look like they track overall mortality in real time like they do in Italy, but I found the averages in the annual
vitality statistics reports. Around 54,000 people die every year in NYC, and the average for March and April is around 1140 per week. In the last week, NYC
reported 1195 "covid-19 deaths". So it will be interesting to see how much of those overlap with all deaths of other causes.