Download your answer date as a plain text file, or press Calculate again after changing a field.

90 days from today is the date you land on when you add 90 calendar days to today's date, counting every day including weekends.

Rule of thumb: the weekday always moves forward by six, so a Sunday today means 90 days from today is a Saturday. Counting starts tomorrow, so tomorrow is day 1 and day 90 is the answer. If you need working days only, 90 business days from today lands about 126 calendar days ahead instead.

90 days from today is a date, not a length of time. It is the day you arrive at when you take today's date and move forward by 90 single days, one after another, without skipping anything. Weekends count. Public holidays count. Months of different lengths are crossed at their real length, so August gives you 31 days and September gives you 30.

The phrase turns up wherever a rule is written in days rather than months. A job notice period, a tenancy notice, a 90 day return window, a probation review, a visa stay limit, a net 90 invoice, a medical follow up, a court filing deadline. In every one of those the exact day matters, because being one day late is still late.

A worked one-liner

Start on 30 August 2026. One day takes you to 31 August, then 30 days of September, then 31 days of October. That is 62 days used. You still need 28 more, so you land on 28 November 2026, a Saturday. That is the whole method: use up each month at its real length until fewer than a month remains.

90 days from today compared with the nearest alternatives

The three month answer and the 90 day answer differ by two days here. That gap is the single most common source of missed deadlines, so always use the wording of your own document rather than the one that sounds close enough.

▶ Loading your dates into sections 4.1 to 4.5.

4.1 What your answer date means

4.2 The weekday you landed on

The weekday is not random. Ninety days is twelve whole weeks plus six days, so the weekday always moves forward by six places, which is the same as moving back one place. If today is a Wednesday, 90 days from today is a Tuesday. Counting backwards reverses it: 90 days ago from a Wednesday was a Thursday.

4.3 How far away that is in other units

4.4 Weekdays and weekend days inside your span

A span of 90 days always holds either 25 or 26 weekend days. Which one you get depends on the weekday you start from, which is why a tool that assumes a flat 26 will be wrong about a third of the time.

4.5 If your deadline is in working days instead

4.6 What changes the answer

• The counting convention. Counting the start day as day 1 moves the answer one day earlier. This tool uses the standard convention where tomorrow is day 1, and the Advanced options let you switch.
• Calendar days against business days. The same 90 turns into roughly 126 calendar days once weekends are skipped.
• Holidays. They only matter in business day mode, and only if your rule actually excludes them. Many do not.
• Leap years. A span that crosses 29 February covers one more calendar day of the year than the same span elsewhere.
• Time zone. Your device decides what today is. If the deciding office is a day ahead of you, set the start date by hand.

4.7 What to do next

1. Write the answer date in your calendar with the weekday next to it, because a deadline that falls on a Saturday often needs action on the Friday before.
2. Set a reminder at day 60 and day 80, not only on the day itself.
3. Check the wording of your document once more for the words calendar, working, business or clear days.
4. If the date matters legally or financially, confirm it with the person or office who will enforce it.

4.8 Common mistakes this calculator prevents

4.9 When your answer date falls on a weekend or a holiday

Calendar day rules usually stand even if the date lands on a Saturday, a Sunday or a public holiday. Many legal and banking rules add a roll forward step: if the deadline lands on a non working day, it moves to the next working day. Some move it backwards to the previous working day instead. This calculator gives you the plain counted date and tells you the weekday, so you can apply your own roll rule knowingly rather than by accident.

▶ Loading your dates into all five templates.

These are the rules the calculator applies. They are plain arithmetic on the calendar, so you can check any answer by hand.

The worked example running through these steps is a start date of 30 August 2026, which was a Sunday.

1. Read the answer that is already there. The page loads with today's date in the start box and the answer worked out, so if you only wanted the date 90 days from today, you have it before you touch anything.
2. Change the start date if you need to. Type any other date to count from a letter date, a delivery date or a signature date, then press Calculate.
3. Choose the direction. Forward answers "what is 90 days from today". Backward answers "what was 90 days ago from today". In the example, Forward is selected.
4. Keep the day count at 90. The field accepts any whole number from 1 to 10000, so the same tool answers 30, 45, 60, 120 or 180 days as well.
5. Pick the day type. Calendar days counts everything. Business days counts Monday to Friday only. From 30 August 2026 the calendar answer is 28 November 2026 and the business answer is 1 January 2027.
6. Open Advanced options if a rule applies. Switch the counting convention if your document counts the start day as day 1, and paste public holidays one per line if business days must skip them.
7. Press Calculate the Date. The page already shows the answer for today when it loads, so you only need this button after you change a field. Changing a field clears the old answer first, so you never read a stale date.
8. Read the result panel first. It gives the answer date, the weekday badge, where that date sits in its calendar year, and how many days away from today it is.
9. Scroll to Section 2 for the breakdown. You get the weeks, the weekday and weekend split, the months crossed, both charts, and the same date highlighted in the reference tables.
10. Save it. Download TXT gives a one page answer straight from Section 1. Download Doc and Download PDF in Section 2 give the full report, both ending with the StatsUnlock line.

Excel stores a date as a plain number of days, so adding 90 days to today is ordinary arithmetic. That single fact is why date maths in a spreadsheet is easier than people expect, and it is also why the answer sometimes appears as 46354 instead of a date: the cell is formatted as a number rather than a date. This section covers the calendar answer, the backward answer, business days, holiday lists and the errors you will meet, with the spreadsheet shown at each stage. The same formulas work in Google Sheets unless noted.

8.1 The functions you need

8.2 Step by step

8.3 The awkward cases

Your date will not add. If =A2+90 returns #VALUE!, the cell holds text that looks like a date. Select the column, then Data, Text to Columns, Finish. That forces Excel to re-read every cell as a real date.

The answer shows as a number. That is correct behaviour, not an error. Excel stores 28 November 2026 as the serial number 46354. Press Ctrl+1 and pick a Date format.

Holidays in business day mode. Put your holiday dates in a column, then reference it: =WORKDAY(B2,90,$H$2:$H$12). Lock the range with dollar signs or it will slide when you copy the formula down.

A weekend that is not Saturday and Sunday. Use WORKDAY.INTL with a weekend code, or a seven character string of ones and zeros starting on Monday. "0000011" is the standard weekend and "0000110" is Friday and Saturday.

Counting the start day. If your rule counts the start date as day 1, subtract one: =B2+B3-1. Decide once, write it in a note in the sheet, and stay consistent.

8.4 Why your Excel answer might differ from R or Python

Google Sheets note. Every formula above works unchanged, including WORKDAY and WORKDAY.INTL. The only real difference is that Sheets recalculates TODAY() on a timer rather than only on open, so a shared sheet can change under you at midnight.

8.5 Charting the countdown in Excel

1. Build two cells: days elapsed, =TODAY()-B2, and days remaining, =B6-TODAY().
2. Select both, then Insert, Charts, 2-D Bar, Stacked Bar. You now have a progress bar for the 90 day span.
3. For the weekday and weekend split, use =SUMPRODUCT(--(WEEKDAY(ROW(INDIRECT(B2+1&":"&B6)),2)<6)) to count weekdays, then chart that against the remainder as a doughnut.
4. Add a data label showing the answer date so the chart reads without the axis.
5. Set the axis maximum to 90 by hand, or the bar will rescale every day and the shape will stop meaning anything.

Chart 1 in Section 2 of this page draws the weekday, weekend and holiday split of your own span automatically, and Chart 2 shows the days counted in each month.

8.6 Excel errors you will probably hit

R treats a date as a real Date object, so adding 90 days is one plus sign: Sys.Date() + 90. The script below is complete and runs top to bottom in RStudio, VS Code or plain Rscript with no packages to install. It prints the calendar answer, the backward answer, the business day answer, the weekday split of the span and a plain English verdict, then saves one 300 dpi figure.

9.1 The complete script

R, base only, saves days_from_today.png

# ---------------------------------------------------------------
# 90 days from today, in base R
# Gives the calendar date, the backward date, the business day date,
# the weekday, the weekday and weekend split, and one 300 dpi figure.
# Change ONLY the two lines marked CHANGE ME.
# ---------------------------------------------------------------

start_date <- Sys.Date()   # CHANGE ME: or as.Date("2026-08-30")
n_days     <- 90           # CHANGE ME: any whole number of days

# Optional holiday list used only for the business day answer.
holidays <- as.Date(c("2026-12-25", "2027-01-01"))

# --- 1. The calendar answers -------------------------------------
fwd  <- start_date + n_days
back <- start_date - n_days

# --- 2. Weekday and week arithmetic ------------------------------
wd_fwd  <- weekdays(fwd)
weeks   <- n_days %/% 7
rem     <- n_days %% 7

# --- 3. The business day answer, weekends and holidays skipped ---
business_offset <- function(from, n, hol) {
  d <- from
  counted <- 0
  while (counted < n) {
    d <- d + 1
    wd <- as.POSIXlt(d)$wday          # 0 = Sunday, 6 = Saturday
    if (wd != 0 && wd != 6 && !(d %in% hol)) counted <- counted + 1
  }
  d
}
bus <- business_offset(start_date, n_days, holidays)

# --- 4. What the calendar span is made of ------------------------
span     <- seq(start_date + 1, fwd, by = "day")
wdays    <- as.POSIXlt(span)$wday
n_wend   <- sum(wdays == 0 | wdays == 6)
n_week   <- length(span) - n_wend

# --- 5. Plain English verdict ------------------------------------
cat("Start date        :", format(start_date, "%d %B %Y"), "\n")
cat("Days counted      :", n_days, "=", weeks, "weeks and", rem, "days\n")
cat("Answer, forward   :", format(fwd,  "%d %B %Y"), "(", wd_fwd, ")\n")
cat("Answer, backward  :", format(back, "%d %B %Y"), "(", weekdays(back), ")\n")
cat("Business days     :", format(bus,  "%d %B %Y"), "(", weekdays(bus), ")\n")
cat("Span contains     :", n_week, "weekdays and", n_wend, "weekend days\n")
if (wd_fwd %in% c("Saturday", "Sunday")) {
  cat("VERDICT: the answer date lands on a weekend. If your rule rolls a\n")
  cat("         deadline to the next working day, use", format(business_offset(fwd - 1, 1, holidays), "%d %B %Y"), "\n")
} else {
  cat("VERDICT: the answer date is a working day, so no roll forward is needed.\n")
}

# --- 6. One 300 dpi figure ---------------------------------------
png("days_from_today.png", width = 2000, height = 1200, res = 300)
par(mar = c(4.5, 5.5, 3.5, 1.5))
counts <- c(n_week, n_wend)
bp <- barplot(counts,
        names.arg = c("Weekdays", "Weekend days"),
        col = c("#2563eb", "#f59e0b"), border = NA,
        ylim = c(0, max(counts) * 1.25),
        ylab = "Days in the span",
        main = paste0(n_days, " days from ", format(start_date, "%d %b %Y"),
                      " ends ", format(fwd, "%d %b %Y")))
text(bp, counts, labels = counts, pos = 3, font = 2)
box(bty = "l")
dev.off()

cat("\nFigure saved as days_from_today.png\n")

9.2 What each part of the script does

9.3 What the figure shows

The bar chart splits your 90 day span into weekdays and weekend days, with the counts printed above each bar and the start and end dates in the title. For a start of 30 August 2026 it shows 65 weekdays and 25 weekend days. That split is the quickest way to see why a working day deadline lands so much later than a calendar day deadline.

9.4 Problems and fixes

9.5 One-liners worth keeping

Python gets the answer from the standard library, so nothing needs installing for the date part: date.today() + timedelta(days=90). The script below adds the backward answer, a business day function, the weekday and weekend split, a plain English verdict and one 300 dpi figure. Only the figure needs an extra package.

10.1 The complete script

Python 3, saves days_from_today.png

# ---------------------------------------------------------------
# 90 days from today, in Python
# Gives the calendar date, the backward date, the business day date,
# the weekday, the weekday and weekend split, and one 300 dpi figure.
# Change ONLY the two lines marked CHANGE ME.
# ---------------------------------------------------------------

from datetime import date, timedelta
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

start_date = date.today()   # CHANGE ME: or date(2026, 8, 30)
n_days     = 90             # CHANGE ME: any whole number of days

holidays = {date(2026, 12, 25), date(2027, 1, 1)}   # used for business days only

# --- 1. The calendar answers -------------------------------------
fwd  = start_date + timedelta(days=n_days)
back = start_date - timedelta(days=n_days)

# --- 2. Weekday and week arithmetic ------------------------------
wd_fwd = fwd.strftime("%A")
weeks, rem = divmod(n_days, 7)

# --- 3. The business day answer, weekends and holidays skipped ---
def business_offset(from_date, n, hols):
    d, counted = from_date, 0
    while counted < n:
        d += timedelta(days=1)
        if d.weekday() < 5 and d not in hols:
            counted += 1
    return d

bus = business_offset(start_date, n_days, holidays)

# --- 4. What the calendar span is made of ------------------------
span    = [start_date + timedelta(days=i) for i in range(1, n_days + 1)]
n_wend  = sum(1 for d in span if d.weekday() >= 5)
n_week  = len(span) - n_wend

# --- 5. Plain English verdict ------------------------------------
print("Start date        :", start_date.strftime("%d %B %Y"))
print("Days counted      :", n_days, "=", weeks, "weeks and", rem, "days")
print("Answer, forward   :", fwd.strftime("%d %B %Y"), "(" + wd_fwd + ")")
print("Answer, backward  :", back.strftime("%d %B %Y"), "(" + back.strftime("%A") + ")")
print("Business days     :", bus.strftime("%d %B %Y"), "(" + bus.strftime("%A") + ")")
print("Span contains     :", n_week, "weekdays and", n_wend, "weekend days")

if fwd.weekday() >= 5:
    roll = business_offset(fwd - timedelta(days=1), 1, holidays)
    print("VERDICT: the answer date lands on a weekend. If your rule rolls a")
    print("         deadline to the next working day, use", roll.strftime("%d %B %Y"))
else:
    print("VERDICT: the answer date is a working day, so no roll forward is needed.")

# --- 6. One 300 dpi figure ---------------------------------------
fig, ax = plt.subplots(figsize=(6.6, 4.0), dpi=300)
counts = [n_week, n_wend]
bars = ax.bar(["Weekdays", "Weekend days"], counts, color=["#2563eb", "#f59e0b"])
ax.bar_label(bars, padding=3, fontweight="bold")
ax.set_ylabel("Days in the span")
ax.set_ylim(0, max(counts) * 1.25)
ax.set_title(f"{n_days} days from {start_date:%d %b %Y} ends {fwd:%d %b %Y}")
ax.spines["top"].set_visible(False)
ax.spines["right"].set_visible(False)
fig.tight_layout()
fig.savefig("days_from_today.png", dpi=300)
print("\nFigure saved as days_from_today.png")

10.2 What each part of the script does

10.3 What the figure shows

The same chart as the R version: weekdays against weekend days for your own span, labelled with the counts, titled with the start and answer dates. For 30 August 2026 it prints 65 and 25, and both scripts agree with the numbers shown in Section 2 of this page.

10.4 Problems and fixes

10.5 One-liners worth keeping

Eight lookup tables for the 90 days from today question. The weekday table and the unit table are true for any year. The dated tables are worked for 2026 and 2027 so you can check the calculator against a printed answer.

Table 11.1 The weekday you land on, 90 days from any start day

What to conclude: the weekday of the answer never depends on the month or the year, only on the day count. Because 90 leaves a remainder of 6 when divided by 7, the answer weekday is always six places forward, which is the same as one place back.

Table 11.2 Common day counts, converted

What to conclude: 90 days is 12 weeks and 6 days, about 2.96 months, and roughly 126 calendar days once you count it as working days instead. The highlighted row is the one this page is about.

Table 11.3 90 days from the 1st of each month, 2026 into 2027

What to conclude: the answer never lands on the 1st of a month, because 90 days is not a whole number of months. The weekend count is 25 or 26 depending on the start weekday, never a fixed 26.

Table 11.4 90 days ago from the 1st of each month, 2026

What to conclude: this is the table to use for a look back window, such as "activity in the last 90 days from today". The earlier date is the first day of the window if your rule includes it.

Table 11.5 90 business days compared with 90 calendar days

What to conclude: ninety working days always lands about 34 to 36 calendar days later than ninety calendar days, before any public holidays are removed. If a deadline looks a month later than you expected, this is usually why.

Table 11.6 What 90 days is in other units

What to conclude: quote 90 days as 12 weeks and 6 days when someone asks for weeks, and never round it to 3 months in anything that has a consequence.

Table 11.7 February length and the leap year rule

What to conclude: a 90 day span that crosses 29 February ends one calendar date earlier than the same span in a common year, because one extra real day is consumed inside February. The calculator applies this rule automatically.

Table 11.8 Where a 90 day period is commonly used

As of 2026. These are the common patterns, not a statement of law in any country. Notice periods, return windows and filing deadlines are set by your own contract and by local rules, both of which change. Confirm the exact wording and the current rule with the organisation that will enforce it. What to conclude: get the counting rule from the document first, then set this calculator to match it.

Eight worked examples, all using the same style of counting as the calculator above. They cover the plain case, the working day case, the backward case, a borderline month end, an extreme count, a leap year span, an invalid entry and a real invoice.

Almost every wrong answer to "what is 90 days from today" comes from a wrong input, not wrong arithmetic. These ten steps fix the inputs before you press Calculate.

1. Find the real start event. A clock starts on a delivery, a signature, a posting, an invoice or a first day at work. Find the document that names the event, and use that date, not the date you happened to notice it.
2. Check whether the start day counts. Look for the words "from", "after", "beginning on" and "clear days". "After" and "clear days" normally exclude the start day, which is the setting this calculator uses by default.
3. Confirm today's date on the device you are using. A phone with the wrong date or a laptop set to another region will give you a confident wrong answer. The start date box shows what the tool read, so check it.
4. Watch the date order. 03/08/2026 is 3 August in most of the world and 8 March in the United States. The date field on this page uses your browser's own format, so read it back before calculating.
5. Decide calendar or working days once. Read the clause, choose the mode, and write down which one you chose. Switching halfway through is how five week errors happen.
6. Get the right holiday list. Holidays only matter in business day mode, and only the ones your rule recognises. A national list, a state list and a company list are three different things.
7. Type holidays in YYYY-MM-DD. One per line, no commas and no words. Any line that is not a date is ignored, and the field will tell you.
8. Use whole days only. Part days do not exist in this kind of counting. If a rule says 90 days and 12 hours, the extra hours belong to a separate time of day rule, not to the date count.
9. Ask which time zone decides. If a deadline is judged in another country, set the start date to that country's current date rather than yours. A day either way is the most common cross border mistake.
10. Write down what you entered. Start date, direction, day count, day type, convention and holidays. Section 2 exports all six with the answer, which is exactly what you want if anyone questions the date later.

Every input, with its type and range

Four worked situations

Decision guide

1. Day 1 is the day after the start date. If your rule counts the start day itself, every answer is one day too late until you switch the convention in Advanced options.
2. Today comes from your device. If the device clock or time zone is wrong, the answer is wrong by a day in that direction. Type the start date by hand to remove the risk.
3. The Gregorian calendar is assumed throughout. Dates before its adoption in your country will not match historical records. The error can be as much as thirteen days.
4. Weekends are Saturday and Sunday. Where the working week is Sunday to Thursday, the business day answer will be wrong. Use calendar mode and adjust by hand, or use a tool with a custom weekend setting.
5. Holidays are only what you type. No public holiday list is built in, because holidays differ by country, region and employer. Missing holidays make the business answer too early.
6. No roll forward is applied. If the counted date lands on a weekend or a holiday, the tool reports it as it is and names the weekday. Rules that move a deadline to the next working day must be applied by you.
7. Whole days only. Times of day, hours and daylight saving shifts are outside the calculation. A span quoted as 2160 hours is nominal, since one day in the span may be 23 or 25 hours long.
8. Months are never approximated. Real lengths are always used, so any answer that assumes a 30 day month will disagree with this one. That is the other tool being wrong, not this one.
9. The day count is capped at 10000. That is about 27 years, and it keeps the business day loop instant. Longer spans need a different approach.
10. This is a calendar tool, not advice. It cannot know your contract, your jurisdiction or your regulator. For anything with legal, medical or financial consequences, confirm the date with the party that will enforce it.

Where a limitation matters, the better tool is named: a month based date calculator for periods written in months, an age calculator for years, months and days, an hours calculator for times of day, and a business day calculator with a custom weekend for working weeks that are not Monday to Friday.

If the date supports a legal, financial or medical decision, cite the governing document as well as this tool, and record the counting convention you used.

All six links point to live StatsUnlock pages. No other domain is ever linked from this section.

Twenty peer-reviewed sources behind the methods on this page: calendar arithmetic algorithms, temporal representation in software, the standardisation of civil time, how people judge and miss deadlines, and the effects of clock changes on the days being counted. Statutory notice periods and holiday rules are not literature, so they are handled in Section 11 with a date and a caution instead.

1. Dershowitz, N., & Reingold, E. M. (1990). Calendrical calculations. Software: Practice and Experience, 20(9), 899–928. https://doi.org/10.1002/spe.4380200905
2. Reingold, E. M., Dershowitz, N., & Clamen, S. M. (1993). Calendrical calculations, II: Three historical calendars. Software: Practice and Experience, 23(4), 383–404. https://doi.org/10.1002/spe.4380230404
3. Fliegel, H. F., & Van Flandern, T. C. (1968). A machine algorithm for processing calendar dates. Communications of the ACM, 11(10), 657. https://doi.org/10.1145/364096.364097
4. Allen, J. F. (1983). Maintaining knowledge about temporal intervals. Communications of the ACM, 26(11), 832–843. https://doi.org/10.1145/182.358434
5. Snodgrass, R., & Ahn, I. (1986). Temporal databases. Computer, 19(9), 35–42. https://doi.org/10.1109/MC.1986.1663327
6. Dyreson, C. E., & Snodgrass, R. T. (1993). Timestamp semantics and representation. Information Systems, 18(3), 143–166. https://doi.org/10.1016/0306-4379(93)90034-X
7. Hobbs, J. R., & Pan, F. (2004). An ontology of time for the semantic web. ACM Transactions on Asian Language Information Processing, 3(1), 66–85. https://doi.org/10.1145/1017068.1017073
8. Bettini, C., Wang, X. S., & Jajodia, S. (2002). Solving multi-granularity temporal constraint networks. Artificial Intelligence, 140(1–2), 107–152. https://doi.org/10.1016/S0004-3702(02)00223-0
9. Zerubavel, E. (1982). The standardization of time: A sociohistorical perspective. American Journal of Sociology, 88(1), 1–23. https://doi.org/10.1086/227631
10. Zerubavel, E. (1977). The French Republican calendar: A case study in the sociology of time. American Sociological Review, 42(6), 868–877. https://doi.org/10.2307/2094572
11. Levine, R. V., & Norenzayan, A. (1999). The pace of life in 31 countries. Journal of Cross-Cultural Psychology, 30(2), 178–205. https://doi.org/10.1177/0022022199030002003
12. Ariely, D., & Wertenbroch, K. (2002). Procrastination, deadlines, and performance: Self-control by precommitment. Psychological Science, 13(3), 219–224. https://doi.org/10.1111/1467-9280.00441
13. Buehler, R., Griffin, D., & Ross, M. (1994). Exploring the "planning fallacy": Why people underestimate their task completion times. Journal of Personality and Social Psychology, 67(3), 366–381. https://doi.org/10.1037/0022-3514.67.3.366
14. Roy, M. M., Christenfeld, N. J. S., & McKenzie, C. R. M. (2005). Underestimating the duration of future events: Memory incorrectly used or memory bias? Psychological Bulletin, 131(5), 738–756. https://doi.org/10.1037/0033-2909.131.5.738
15. Read, D., Frederick, S., Orsel, B., & Rahman, J. (2005). Four score and seven years from now: The date/delay effect in temporal discounting. Management Science, 51(9), 1326–1335. https://doi.org/10.1287/mnsc.1050.0412
16. LeBoeuf, R. A. (2006). Discount rates for time versus dates: The sensitivity of discounting to time-interval description. Journal of Marketing Research, 43(1), 59–72. https://doi.org/10.1509/jmkr.43.1.59
17. Zauberman, G., Kim, B. K., Malkoc, S. A., & Bettman, J. R. (2009). Discounting time and time discounting: Subjective time perception and intertemporal preferences. Journal of Marketing Research, 46(4), 543–556. https://doi.org/10.1509/jmkr.46.4.543
18. Kantermann, T., Juda, M., Merrow, M., & Roenneberg, T. (2007). The human circadian clock's seasonal adjustment is disrupted by daylight saving time. Current Biology, 17(22), 1996–2000. https://doi.org/10.1016/j.cub.2007.10.025
19. Janszky, I., & Ljung, R. (2008). Shifts to and from daylight saving time and incidence of myocardial infarction. New England Journal of Medicine, 359(18), 1966–1968. https://doi.org/10.1056/NEJMc0807104
20. Roenneberg, T., Wirz-Justice, A., & Merrow, M. (2003). Life between clocks: Daily temporal patterns of human chronotypes. Journal of Biological Rhythms, 18(1), 80–90. https://doi.org/10.1177/0748730402239679