Skip to content
Omnikalc

Standardized mortality ratio calculator: are there more deaths than expected?

Enter the observed and expected deaths to get the standardized mortality ratio, its confidence interval, the excess deaths and an approximate p-value.

Updated on

Educational tool, not medical advice: this page does not diagnose and does not replace a doctor, nurse or pharmacist. It has not been reviewed by a healthcare professional. Editorial and medical policy

Your inputs

Value out of range

Value out of range

Results

Standardized mortality ratio (SMR)

1.32

Not significantly different from the reference
Lower confidence limit
0.97
Upper confidence limit
1.74
Excess deaths (observed − expected)
11.5
Approximate p-value (two-sided)
0.069

This calculator is for general information and education. It is not medical advice and does not replace the advice of your doctor or a registered dietitian.

The standardized mortality ratio, or SMR, compares the deaths seen in a group with the deaths expected from a reference population. The standardized mortality ratio calculator takes the observed deaths, the expected deaths and a confidence level, and gives the SMR, its confidence interval, the excess deaths and an approximate p-value.

How the calculation works

  • Count the observed deaths in your group, and compute the expected deaths by applying the death rates of a reference population to the age, sex and time at risk of your group.
  • SMR = observed ÷ expected. A value of 1 means the same mortality as the reference, 1.5 means 50 percent more deaths than expected and 0.8 means 20 percent fewer.
  • The confidence interval uses Byar’s approximation to the exact Poisson limits: lower = O × (1 − 1 ÷ (9O) − z ÷ (3√O))³ ÷ E, and upper = (O + 1) × (1 − 1 ÷ (9(O + 1)) + z ÷ (3√(O + 1)))³ ÷ E.
  • The excess deaths are observed minus expected, and the p-value comes from a normal approximation, (|O − E| − 0.5) ÷ √E, with a continuity correction.

Example

In a group of workers there are 48 deaths while the reference rates predict 36.5. The SMR is 48 ÷ 36.5 = 1.32, with a 95 % confidence interval from 0.97 to 1.74, and 11.5 excess deaths. The interval includes 1 and the p-value is about 0.069, so the excess is not statistically significant at the 5 % level.

Why it matters

Raw death counts cannot be compared between groups of different ages or sizes. The SMR adjusts for this and shows at a glance whether a group, such as a workforce, a hospital’s patients or a town, has more or fewer deaths than expected.

Good to know

The calculation takes the expected number as given and treats it as fixed, which is reasonable when the reference rates come from a large population. The interval is an approximation that works well even for small counts. Because the SMR is based on indirect standardization, SMRs of different groups should not be compared with each other directly.

The relative risk calculator compares risks in two groups, and the case fatality rate calculator gives deaths among those with a disease.

Frequently asked questions

What is a standardized mortality ratio?

It is the ratio of the number of deaths observed in a group to the number expected if the group had the death rates of a reference population, after allowing for differences in age and sex. An SMR above 1 means higher mortality than the reference, and below 1 means lower.

How do I get the expected deaths?

Multiply the reference death rate for each age and sex band by the person-years your group spent in that band, then add up the results. The reference might be the national population, a regional one or a similar group of workers. The method is called indirect standardization.

Why use a confidence interval with so few deaths?

Deaths are counted, so they vary by chance, following roughly a Poisson distribution. With few deaths, the SMR can be far from its true value just by chance. The interval shows the range that is consistent with the data, and if it includes 1, the result may be due to chance.

What is the healthy worker effect?

People who work are on average healthier than the whole population, which includes those who are too sick to work. An occupational group may therefore show an SMR below 1 even if the job carries some risk. Compare with a suitable reference to avoid this bias.

Was this calculator helpful?

Thanks for your feedback!

Results are estimates provided for information purposes. They do not replace professional advice.