Home / Science / Health and medicine / Epidemiology / Epidemic curve and attack ratesTool #464

Epi curve maker for outbreaks: epidemic curve from onset dates, incubation and food attack rates

Draw an outbreak epidemic curve from onset dates (or dates and counts) in bars of hours, days or weeks, with the peak, median onset and incubation from a shared exposure. Add a food-history table for attack rates, attack-rate ratios with intervals and Fisher's exact p. PNG and CSV downloads.

Keep it one tap away for everyday use.

Bookmark Epidemic curve and attack rates?

Press Ctrl + D now. Your browser opens its own bookmark box, where you choose the name and folder.

  1. Onsets
  2. Epi curve
  3. Attack rate
Nothing is uploadedWorks offline once open

Onset dates

For study and teaching, not medical advice. Check with a qualified professional.

The example outbreak is made up.

A bar a quarter to a third of the usual incubation period wide shows the shape best. With a shared exposure (a meal or event), the page works out the incubation periods.

Attack rates by food or exposure

Download

What it does

The epidemic curve is the first picture drawn in any outbreak investigation: the number of cases by date (or hour) of symptom onset. Its shape hints at how the outbreak spread, when people were exposed and whether it is over. This tool draws it from a plain list of onset dates, and works out the attack rates in a food-history (or exposure-history) table, the classic way to find the meal or item behind a food-borne outbreak.

  • Epidemic curve: bars of 1, 6 or 12 hours, 1 to 3 days, or weeks starting Monday or Sunday, with empty bars kept so gaps show.
  • Summary: number of cases, first and last onset, the peak bar and the median onset.
  • Incubation: with a shared exposure time (a meal or event), the shortest, median and longest incubation period, and a marker on the curve.
  • Attack rates: for each item, the share of eaters and non-eaters who fell ill, the attack-rate ratio with its interval, and Fisher’s exact p.

For study and teaching, not medical advice. Check with a qualified professional.

How to use it

  1. Paste one onset per line as YYYY-MM-DD (add a time such as 2026-06-10 14:30 if you have one), or a date, a comma and a number of cases. Or open a CSV or Excel file.
  2. Choose the bar width. A width of about a quarter to a third of the usual incubation period shows the shape best: hours for a toxin with a short incubation, days for most bacteria and viruses, weeks for long outbreaks. Hour bars need onset times.
  3. Optionally type the shared exposure. With dates only, incubation is counted in whole days.
  4. In Attack rates, paste one row per item: ate and ill, ate and well, did not eat and ill, did not eat and well.
  5. Download the curve as PNG, the counts or attack rates as CSV, or a report as an image, PDF or Word file.

The formulas

The attack rate is ill ÷ (ill + well) in each group. The attack-rate ratio is the attack rate among eaters divided by that among non-eaters, with limits exp(ln RR ± z·SE), SE = √(1/a − 1/(a + b) + 1/c − 1/(c + d)). Fisher’s exact test gives the probability of a table at least as unusual as the one seen if the item made no difference.

Worked example

The made-up outbreak has 44 cases from 2026-06-08 to 2026-06-17. With daily bars the page shows a peak on 2026-06-12 with 12 cases and a median onset of 2026-06-12. With the shared meal on 2026-06-07, the incubation runs from 1 to 10 days with a median of 5 days. Weekly bars from Monday give 41 cases in the week of 8 June and 3 in the next.

In the made-up food table, 38 of 58 people who ate the chicken salad fell ill (65.5 %) against 6 of 42 who did not (14.3 %): an attack-rate ratio of 4.59 (95 % interval 2.14 to 9.85), Fisher p < 0.001. Rice, fruit and ice cream show no clear link.

Tips

  • Use the onset of symptoms, not the date of the test or report.
  • Keep empty days in the picture: gaps are information.
  • The item behind the outbreak usually has a high attack rate among eaters, a low one among non-eaters, and accounts for most cases.

Frequently asked questions

Which date format can I use? Year first: 2026-06-10 or 2026/06/10, with an optional time. Day-first dates such as 10/06/2026 are refused because they are easy to misread.

Why are times shown without a time zone? Times are taken as typed and never shifted, so a case at 23:30 stays on its own day.

Is my data stored? No. Everything is worked out in your browser and nothing is sent anywhere.

Where the numbers come from

  • Methods: Centers for Disease Control and Prevention, Principles of Epidemiology in Public Health Practice, 3rd ed. (2006, textbook), lesson 6 on epidemic curves and food-specific attack rates; Katz et al. (1978, Biometrics 34:469) for the risk ratio interval; Fisher (1935, J R Stat Soc) for the exact test.
  • The example outbreak and food-history numbers are made up for teaching.

Standard scientific values, such as constants, atomic weights and melting points, are facts found in every textbook; we name where ours were checked. If you are an author or publisher and want a value, a credit or a table changed or removed, tell us and we will act quickly.

Last updated 2026-11-09.

Related tools

Rates and age standardisation #463 Incidence rates, cumulative incidence and prevalence with confidence intervals, directly age-standardised rates with Fay-Feuer gamma intervals, and indirect standardisation: expected events and the SMR with exact Poisson limits. Paste data by age group or open a file. NNT calculator #465 Number needed to treat or harm from events in a treated and a control group, with the absolute risk reduction, relative risk, relative risk reduction and odds ratio, all with 95 % intervals. Shows the NNT interval the Altman way when it crosses zero. Odds ratio and risk ratio calculator #466 Odds ratio and risk ratio from a 2 × 2 table with Woolf, exact and mid-p intervals, chi-square and Fisher exact tests, or several strata pooled by Mantel-Haenszel with the Breslow-Day test and a forest plot. Case-control or cohort. Sample size and power calculator #32 How many subjects, samples or survey responses do you need? Sample size for comparing two means, paired data, two proportions, or a survey's margin of error, with power curves. Matches R and G*Power. Publication plot maker #27 Paste columns or open a CSV or Excel file and get a clean scientific figure with error bars, sized for a journal column. Download vector SVG or PDF, or PNG at 300 or 600 dpi. Your data stays in your browser. Diagnostic test calculator #461 Sensitivity, specificity, predictive values, likelihood ratios and accuracy from a 2 × 2 table, with exact or Wilson 95 % intervals. Turn a pre-test probability into a post-test probability and see it on a Fagan nomogram.