The week after Easter: more people died in Romania than usual

The rise moves with Easter#

Orthodox Easter falls on a different day each year, between 4 April and 8 May. The rise in deaths moves with it: in 2015 it comes in mid-April, in 2024 in early May.

The week after Easter was above the reference in eight of the nine years, in 2018 by only 1.3%. The exception is 2022, the first Easter after the pandemic years. Without 2017, the year with the largest rise, the average stays at 4.7%; leaving out any other single year, it stays between 5.5% and 6.6%.

Reference: how many deaths that week would have had at the level of the three weeks before Holy Week, adjusted for the time of year.

How sure is the result?#

Nine Easters are few. Romania's weekly data start only in 2015, and 2020 and 2021, with the pandemic waves, cannot be compared with the rest. For 2026 Eurostat does not yet have weeks 2–13, so Easter 2026 cannot be measured like the other years.

To see whether a 5.7% rise could appear by chance, the Easter dates were shuffled among the nine years, 2,000 times: Easter 2017, for example, was moved to the week in which Easter fell in 2024. Each time, the largest rise in Holy Week or in any of the three weeks after Easter was kept, not just in the first. None of the 2,000 shuffled calendars produced a rise at least as large.

Nor does the result depend on how the reference is calculated. Across 32 versions of the calculation (with or without the long-term trend, with the season modelled week by week or as a smooth curve, with or without 2022, with or without the weeks of 2026) the rise stays between 3.0% and 7.6%.

Counting each Easter as one observation, the 95% confidence interval runs from 2.4% to 9.1%. In deaths: roughly 120 to 420 extra in the week after Easter, with a central estimate around 270. The data cannot say whether some of these people would have died a few weeks later anyway.

Two simple explanations do not fit the lines. If deaths over the days off were merely registered late, they would be missing from Holy Week, which includes Easter Saturday and Sunday. Holy Week instead has 2.6% more deaths than is usual for the time of year; and Eurostat counts deaths by the day of death, not the day of registration. If very ill people "held on" for Easter, deaths would fall in the weeks before it. They do not: the three weeks before Holy Week are also slightly above the usual level.

Does it happen in other countries?#

The same calculation, over the same years 2015–2025 and the same Orthodox Easter dates:

Greece shows the same rise: 6.3%, and only 10 of 2,000 shuffled calendars produce a rise at least as large. It is not an independent confirmation, though: Greece's result was seen together with Romania's, over the same years, and if something hit the whole region it would show there too.

Bulgaria is the hard case. Over the same years it has a smaller rise, 3.5%, but also one that chance rarely produces: only 60 of 2,000 shuffled calendars match it. In 2000–2014, though, Bulgaria showed the opposite, 1.5% fewer deaths in the week after Easter.

The rule for the checks was written before the checks, after Romania's result had already been seen: if Bulgaria, Serbia or Hungary showed, on the Orthodox Easter dates, a rise larger than in 95% of the shuffled calendars, the result would not be published. Bulgaria crossed that line. The page appears anyway, with the failed check in plain view, because Bulgaria also keeps Orthodox Easter, so its rise may be the same thing. It may also mean that in these years something hit the whole region in the weeks after Easter. With nine years of data the two cannot be told apart.

Serbia, also Orthodox, shows nothing (−0.2%). Hungary, where most people celebrate the Western Easter, has +2.5% after Orthodox Easter, with no clear rise. Cyprus (+4.5%) has too few deaths a week for its result to say much. Montenegro, a small country with about 125 deaths a week, has +7.2%, with a confidence interval of 0.9% to 13.8%: another Orthodox country leaning the same way.

In Romania, after the Western (Catholic and Protestant) Easter, the rise is 2.5%, and one at least as large appears in 13% of the shuffled calendars. The two Easters fell on the same day in 2017 and 2025, so this result also contains part of the rise after Orthodox Easter. After Pentecost, another holiday with days off, the rise is 3.3%, and one at least as large appears in 30% of the calendars. The evidence for the other holidays is weaker, which does not mean the effect belongs to Orthodox Easter alone.

Who dies in the week after Easter#

The rise appears at every age with enough deaths to measure, at similar sizes: from 5.1% among people over 80 to 8.0% at 50–59.

Below 40, each age group has between 12 and 61 deaths a week, too few for a comparison of a few per cent. The rise is not only among the very old: it is largest at 50–59.

Why would more people die after Easter?#

Nobody knows. There appears to be no study of the week after Easter in the death records of an Orthodox country. Studies in Western countries (Sweden, England, Germany) found no rise after their Easter, and a study in Athens, covering a single year, found no Easter peak in heart-attack deaths. The public data say neither what people died of nor on which day. A few explanations are possible; none is proven:

• The meal after the fast. For those who fast, the Easter meal comes after weeks without meat, eggs and dairy. A Greek study of ten overweight adults who fasted during Holy Week only found total cholesterol down to 146.8 mg/dL at the end of the fast and up to 175.1 a week later, above its level before the fast (163.0). It is a hint, not proof: a small study of blood tests, not deaths.
• The days off. Easter Sunday and Monday are public holidays in Romania, and Good Friday too since 2018; many family doctors' surgeries and outpatient clinics are closed. At the emergency department of the county hospital in Craiova, a model covering 1 January 2025 to 30 April 2026 estimated on average 18.6 more presentations a day on legal holidays, all taken together. Studies of Christmas and New Year in the United States and Russia have also found more deaths around the holidays.
• Accidents. They cannot explain the rise. In the worst year, 2017, 1,951 people died on Romania's roads, some 38 a week. To produce 270 extra deaths, crashes would have to multiply about eight times over in a single week.
• Chance, or something regional. The Bulgarian case leaves this open too.

Romania's National Institute of Statistics holds deaths day by day, with their causes, since 1995, but does not publish them. With nearly 30 Easters instead of nine, the question could be tested much better.

What is certain: the emergency services, 112 and hospital emergency departments, work through Easter.