The absence rate measures the share of scheduled working time lost to absences within a defined scope. To calculate it without distortion, use hours lost in the numerator and comparable scheduled working hours in the denominator. Excused absences, unexcused absences and extended leave should not be added together indiscriminately: inclusion rules must be decided and documented before calculating.
What is the absence rate formula?
Absence rate (%) = hours of absence included in the metric ÷ scheduled working hours for the same group and period × 100. If there are no eligible scheduled hours, the rate should not be calculated. The unit and reporting window must match on both sides of the calculation.
For a management dashboard based on scheduled hours, the denominator is the sum of hours each person was expected to work during the period, after calendar adjustments and any planned time off the policy excludes, such as holidays or rostered days off. Hours associated with an included absence stay in the denominator: subtracting them there as well would artificially reduce the rate. With different work patterns, add up individual scheduled hours rather than multiplying the whole workforce by one standard schedule.
The Office for National Statistics (ONS) specifically measures sickness or injury absence as a percentage of hours lost and uses its own construction of total hours, combining actual hours for people without sickness absence with usual hours for those with it. The scheduled-hours management formula shown here is an illustrative operational choice, not an automatic reproduction of the ONS series.
Spreadsheet example: scheduled hours and hours lost
Consider four people in the same month, each initially scheduled for 160 hours. This example is fictional. Previously scheduled holidays and days off are deducted from eligible hours; sickness and unexcused absence remain in those hours.
| Person | Initial hours | Excluded holiday/time off | Eligible hours | Sickness | Unexcused |
|---|---|---|---|---|---|
| A | 160 | 0 | 160 | 8 | 0 |
| B | 160 | 8 | 152 | 4 | 0 |
| C | 160 | 0 | 160 | 0 | 8 |
| D | 160 | 16 | 144 | 16 | 0 |
| Total | 640 | 24 | 616 | 28 | 8 |
If the internal metric includes both absence types, the result is 36 ÷ 616 × 100 = 5.84%. On the same base, the sickness-only rate is 28 ÷ 616 × 100 = 4.55%, and the unexcused absence rate is 8 ÷ 616 × 100 = 1.30%. Small differences when adding rounded rates come from rounding. These three views should carry distinct labels on the dashboard.
How should absences, excuses and extended leave be classified?
- Sickness or injury: record the hours actually lost and the length of the episode. Medical documentation may excuse the absence, but it does not turn those hours into time worked. If the dashboard measures sickness, those hours belong in that metric's defined numerator.
- Unexcused absence: record it separately. It belongs in the combined rate only if that rate's published definition includes this type of event; never mix it into a sickness-only rate.
- Partial absence: count hours not worked, rather than automatically treating it as a full day. A two-hour early departure and an eight-hour absence carry different weights in a time-based rate.
- Holidays, public holidays, rostered days off and previously excluded leave: remove their hours from the denominator and do not record them as hours lost. Document the rule for other types of leave and approved time off.
- Extended leave: if the scope includes long-term sickness absence, count only scheduled hours falling within the period and retain those hours in the base. If a leave category is excluded, remove it from both sides under a consistent rule. Do not add calendar days or hours outside the roster.
A classification table should specify reason, excused status, start and end dates, affected scheduled hours, and inclusion rule. The CIPD guidance on absence measurement and management emphasizes clear policies and support for employees' health needs.
Rate, frequency and duration measure different things
The rate shows the proportion of hours lost. Frequency counts episodes: consecutive days relating to one occurrence may count as one episode if that is the adopted rule. Average duration is hours lost divided by the number of episodes within the same scope. In the example, assuming one episode per person, there are four episodes and an average duration of 36 ÷ 4 = 9 hours. A team with a few long absences may have the same rate as another with many short ones; frequency and duration help distinguish these patterns.
When comparing frequency between teams, also state the population size and period, for example episodes per person per month. For duration, do not combine sickness episodes with unexcused events when the question specifically concerns health.
How can teams and months be compared fairly?
- Use the same absence definition, included population, unit of measurement and leave rules for every comparison.
- Account for part-time schedules, shifts, public holidays and changes in staffing: compare eligible hours rather than only the raw number of absences.
- Handle overtime explicitly and consistently; do not add it to one month's base merely to dilute the rate.
- Keep monthly and annual comparisons distinct and consider seasonality, line of work and exposure differences. External rates are comparable only where methods align.
- Flag every definition or recording-system change in the historical series. A drop may reflect reclassification rather than fewer hours lost.
What can, and cannot, this metric guide?
The rate helps estimate lost hours and cover requirements, but it cannot explain causes on its own or measure an individual's commitment. Read it alongside frequency, duration, working conditions, occupational risks and support needs. In guidance for the UK context, Acas recommends recording absence without discrimination, considering disability- or pregnancy-related circumstances and protecting health data; the applicable legal requirements depend on the jurisdiction.
For an auditable report, publish the period, eligible hours, hours lost by category, treatment of leave, and methodological changes alongside the rate. This helps guide action on work organisation and health instead of using the percentage in isolation as a punitive measure.
