Mortality Rate Calculator

Four fractions, and the denominator does all the work

Every calculation on this page is deaths divided by something, multiplied by a round number. That is it. The arithmetic is the easiest on this entire site, and you could do all four on the back of a receipt.

Which is exactly why they go wrong so often. When the sum is trivial, all the meaning migrates into the definitions, and the definitions are where the traps are. Deaths divided by what, over what period, counting which deaths? Get the denominator wrong and you produce a number that is arithmetically flawless, looks entirely plausible, and means nothing at all.

So this page is mostly about denominators. What each of the four rates actually counts, which multiplier convention goes with which, the label on this tool that will mislead you if you take it literally, and the one deep flaw that makes crude death rates dangerous to compare between countries.

The four it calculates

Each has a fixed, conventional denominator and a fixed, conventional multiplier. They are not interchangeable, and using the wrong pairing is the fastest way to produce a figure nobody can read.

Look at the third and fourth rows. Both divide by live births, not by population, and that is the thing people get wrong most. Infant and maternal deaths are measured against the number of births, because births are what generate the risk. Dividing infant deaths by total population would tell you about a country's birth rate as much as its child health.

Note also that the tool bundles crude and cause-specific into one option. The arithmetic is identical, since both divide by population, but they answer different questions, and cause-specific rates are conventionally quoted per 100,000 rather than per 1,000, because deaths from any single cause are rarer.

A worked example

A district records 240 infant deaths against 30,000 live births in a year:

IMR = (240 ÷ 30,000) × 1,000 = 8 per 1,000 live births

Read that as: for every thousand babies born alive here, about eight died before their first birthday. That phrasing is the sanity check, and it is worth doing out loud every time. If your sentence does not make sense, your denominator is wrong.

Same district, 12 maternal deaths against those 30,000 live births:

MMR = (12 ÷ 30,000) × 100,000 = 40 per 100,000 live births

One thing to watch here. The tool defaults the multiplier to per 1,000, which suits infant mortality but not maternal. Run those same maternal numbers at per 1,000 and you get 0.4, which is arithmetically the identical fact and completely unusable, since no source on earth quotes maternal mortality that way. Switch the multiplier to per 100,000 when you select maternal, and your figure will be comparable to WHO's.

Rate, ratio, proportion: the distinction nobody explains

Three words that get used interchangeably in conversation and mean genuinely different things in epidemiology. Since this tool produces one of each, here they are.

• A proportion is a fraction where the numerator is part of the denominator. Proportionate mortality is a proportion: cancer deaths are a subset of all deaths. It can never exceed 100 percent.
• A ratio compares two quantities where the numerator is not part of the denominator. Maternal mortality is a ratio: maternal deaths are not a subset of live births. They are two different things being compared, which is precisely why it is called a ratio and not a rate.
• A rate, strictly, involves time: events divided by the person-time at risk. A true death rate counts deaths against how many people were alive and exposed, for how long.

Now the fun part. By that strict definition, most of the famous "mortality rates" are not rates. WHO says so openly about its own indicator: the infant mortality rate is, in its words, strictly speaking not a rate, because the denominator is live births rather than an estimate of the person-time at risk. It is kept because it is enormously convenient. You can compute it from simple counts of births and deaths without needing population estimates by age, and that matters in places where the birth register is solid and the census is not.

So the names are historical rather than precise. Use them, because everybody else does, but know that "infant mortality rate" is a term of art rather than a description, and that "maternal mortality ratio" is the one that got named honestly.

Infant mortality is not child mortality

Worth being precise about, because the distinction is easy to lose. The infant mortality rate counts deaths before the first birthday, and nothing else. Child mortality is a different measure over a different span.

These are three separate, internationally defined indicators, and mixing them up will make your figure roughly three times too big or too small:

So when you use the infant setting, enter deaths of babies under one year old, whatever the label says. If what you have is deaths under five, you are calculating the under-five rate, which is a different indicator with its own place in the development goals, and you should label your output accordingly.

The two are not remotely the same number. Globally, under-five deaths substantially exceed infant deaths, because the under-five figure includes the infants and then everyone who died in the following four years.

What the maternal ratio actually counts

The tool labels this field "Deaths During Pregnancy", which is the short version and undercounts the real definition considerably. The internationally agreed definition, the one WHO and the ICD use and the one every published figure follows, is wider in one direction and narrower in another.

A maternal death is the death of a woman:

• While pregnant, or within 42 days of the pregnancy ending, however it ended and however far along it was. So the six weeks after birth count, and that window contains a great many maternal deaths. Restricting yourself to deaths "during pregnancy" would miss most of them.
• From any cause related to or aggravated by the pregnancy or its management. That includes indirect causes, meaning an existing condition made worse by being pregnant, not just obstetric emergencies.
• But not from accidental or incidental causes. A pregnant woman killed in a car crash is not a maternal death. This is the narrowing clause, and it is the one people forget.

And the naming, which is unusually precise here and worth respecting:

• Maternal mortality ratio = maternal deaths ÷ live births × 100,000. This is what the tool computes, and it correctly calls it a ratio. It approximates the risk attached to a single pregnancy.
• Maternal mortality rate = pregnancy-related deaths ÷ women of reproductive age, usually 15 to 49. A different denominator answering a different question, since it folds in how often women in that population get pregnant at all.

Confusing those two is a classic error, and it matters: a country can improve its ratio while its rate stays flat, or the reverse, purely through changes in fertility. If you are quoting a figure, say which one you mean.

Why the multiplier is 1,000 or 100,000

Purely so humans can read the answer. The multiplier changes nothing about the underlying fact; it just moves the decimal point somewhere the eye can cope with.

Our maternal example was 0.0004 maternal deaths per live birth. True, useless. Multiply by 100,000 and it becomes 40 per 100,000, which you can hold in your head and compare to another country.

The rule of thumb: the rarer the event, the bigger the multiplier. Deaths overall are common, so per 1,000 is enough. Infant deaths are less common, but measured against births rather than the whole population, so per 1,000 still works. Maternal deaths are genuinely rare in absolute terms, so they need per 100,000 to avoid drowning in zeros. Deaths from one specific cause are likewise usually quoted per 100,000.

The only real rule is to state your multiplier every single time. A bare "40" is not a mortality figure, it is a number. And never compare two figures without checking they share a multiplier, which sounds too obvious to say and is a mistake published regularly.

Crude rates lie when you compare populations

This is the most important section here, and the one that separates people who calculate mortality from people who understand it.

Take two places. One is a retirement region: lots of people over 70. The other is a university city: mostly people in their twenties. The retirement region will have a dramatically higher crude death rate. Not because it is unhealthy, not because its hospitals are worse, but because old people die at higher rates than young people everywhere, and one place has more of them.

The crude death rate has no idea any of this is happening. It divides all deaths by all people, and the age structure of the population is invisible to it. So it will faithfully report that the retirement town is more deadly, and it will be technically correct and completely misleading.

This is why serious comparison uses age-standardised rates. The method is to take the age-specific death rates from each population and apply them to one shared reference population, so you are asking: if these two places had identical age structures, how would their mortality compare? That answer is meaningful. The crude comparison is not.

This tool computes crude rates, which is correct for what it is, and they are perfectly good for one population's trend over time, or for a genuinely quick estimate. But if you are about to compare a crude death rate between two countries, or two regions, or two decades in which the population aged, stop. You are probably about to measure demography and call it health.

Proportionate mortality is a slice, not a risk

This is the odd one out, and the most commonly misread figure of the four. It answers: of the people who died, what share died of this?

Take 180 cancer deaths out of 900 total deaths: (180 ÷ 900) × 100 = 20 percent. One death in five here was from cancer.

Notice what that does not tell you: anything at all about your risk of dying from cancer. Proportionate mortality has no population in its denominator. It only knows about people who already died.

Which produces a genuinely counterintuitive trap. Imagine a country that dramatically reduces its deaths from infectious disease. Cancer's share of deaths rises, possibly steeply, and someone will write a headline about a cancer surge. But the cancer death rate, measured against the population, may not have moved at all. The slice grew because the pie shrank elsewhere. Every death has to be from something, so the shares always add to 100 percent, and pushing one down pushes the others up by definition.

So use proportionate mortality to describe what a population dies of. Never use it to say whether a disease is getting worse. For that you need a rate with a population underneath it.

One practical note: since the numerator is a subset of the denominator here, your deaths figure should never exceed your total deaths. The tool will not stop you, and it will cheerfully report 166 percent, which is your cue that the inputs are the wrong way round.

Small numbers, big noise

A last caution, especially if you are calculating for a small area or a short period.

Mortality figures are counts of rare events, and rare events are noisy. A village with 2 infant deaths one year and 5 the next has not seen its infant mortality rate rise 150 percent in any meaningful sense; it has seen three more tragedies and a lot of statistical noise. The smaller the numbers, the wider the uncertainty around the rate, and the more likely you are reading randomness as trend.

There is a neat illustration of the scale of this: a maternal mortality ratio of 500 per 100,000 carries roughly the same sampling error as an infant mortality rate of 5 per 1,000, because both rest on similarly small counts of actual events. The tidy-looking number hides how few deaths it is built on.

So if you are working with small denominators, quote confidence intervals, pool several years, or say plainly that the figure is unstable. A rate calculated from a handful of deaths is a headline waiting to be wrong.

Questions people ask

How do I calculate infant mortality rate from 240 deaths and 30,000 births?

Divide 240 by 30,000 and multiply by 1,000, giving 8 per 1,000 live births. Make sure your 240 is deaths of babies under one year old. If it is deaths under five, you have calculated the under-five mortality rate instead, which is a different indicator.

Should I use per 1,000 or per 100,000?

Follow the convention for the indicator, not your preference. Crude death rate and infant mortality are quoted per 1,000. Maternal mortality and cause-specific death rates are quoted per 100,000. Using an unconventional multiplier does not make your number wrong, it makes it incomparable, which in practice is worse.

What is the difference between maternal mortality rate and ratio?

The ratio divides maternal deaths by live births, per 100,000, and approximates the risk of a single pregnancy. The rate divides pregnancy-related deaths by women of reproductive age, and reflects both that risk and how often women get pregnant. This tool computes the ratio, which is what WHO publishes and what almost everyone means.

Can I compare two countries' crude death rates?

Not meaningfully. A country with an older population will show a higher crude death rate regardless of how good its health system is, because the measure cannot see age structure. For comparison you need age-standardised rates, which apply both countries' age-specific rates to a common reference population. Crude rates are fine for tracking one population over a short span.

If cancer is 20 percent of deaths, is cancer increasing?

You cannot tell from that number, and this is the classic misreading. Proportionate mortality has no population in its denominator, so it only describes the mix of causes among people who died. If deaths from other causes fall, cancer's share rises automatically without a single extra case. To know whether cancer is increasing, you need the cause-specific death rate, measured against the population.

References

Where the definitions come from. The maternal mortality ratio as maternal deaths per 100,000 live births, and the underlying ICD definition of a maternal death as one occurring during pregnancy or within 42 days of its termination from causes related to or aggravated by the pregnancy but excluding accidental and incidental causes, follow the WHO indicator specification. The definition of infant mortality as deaths before one year of age per 1,000 live births, its distinction from the under-five rate, and the note that it is strictly speaking not a rate, come from the WHO Global Health Observatory indicator metadata.

1. World Health Organization. Maternal mortality ratio (per 100 000 live births), SDG indicator 3.1.1. WHO Data. data.who.int
2. World Health Organization. Infant mortality rate (per 1000 live births): indicator metadata registry. Global Health Observatory. who.int