Years of Potential Life Lost: YPLL Formula

Q: What does YPLL stand for?

YPLL stands for years of potential life lost, sometimes written potential years of life lost (PYLL). Both describe the years lost to deaths that occur before a chosen cut-off age.

Q: What cut-off age does YPLL use?

There is no single required cut-off. The CDC reports cited here used age 65, CDC reports in 1990 and 1992 also compared results with age 85, and other analysts use 70 or 75. Always state the cut-off you used.

Q: Why does YPLL give more weight to young deaths?

Because each death is weighted by how far before the cut-off it occurs. With a cut-off of 65, a death at age 5 counts 60 years and a death at age 60 counts 5, a 12-fold difference.

Q: How do I compare YPLL between two populations?

Convert totals to a rate per 1,000 or 100,000 persons younger than the cut-off age, because raw YPLL depends directly on the size of the group.

Q: How is YPLL different from life expectancy?

Life expectancy summarizes average survival. YPLL adds up the years lost by deaths that occur before a chosen cut-off. The CDC also describes an alternative form of YPLL that emphasizes the life expectancy remaining at the time of death.

Years of potential life lost (YPLL) is a measure of premature mortality: for each death that occurs before a chosen cut-off age, it counts the years between the age at death and that cut-off. Add those years up across all deaths and you have the YPLL for a cause, a population or a year. Because a death at age 5 counts for far more than a death at age 60, YPLL ranks causes of death very differently from a simple count of deaths. This guide gives the formula, a fully worked midpoint-method calculation, a reading table, reporting wording and code you can paste into Excel, R or Python.

What are years of potential life lost (YPLL)?

The US Centers for Disease Control and Prevention (CDC) defined YPLL in its Morbidity and Mortality Weekly Report as the number of years of potential life lost by each death occurring before a predetermined end point. In that report the end point was set at age 65, so a death at age 24 contributes 65 − 24 = 41 years, and deaths at age 65 or older contribute nothing. YPLL is therefore a summary of how much life is lost before the cut-off, not an estimate of how long any individual would have lived.

The weighting is the point of the measure. Using a cut-off of 65, a death at age 5 counts 60 years and a death at age 60 counts 5 years, so the earlier death carries 12 times the weight. A cause that kills many older adults can top a death-count ranking and still contribute little YPLL, while injury or infant deaths can contribute a lot. Related ideas are covered in what is epidemiology and what is public health.

YPLL formula and the midpoint method

For individual death records the formula is:

YPLL = Σ max(0, L − aᵢ)

Σ: sum over every death i in the group you are studying.
L: the cut-off age you choose (65 in the CDC reports cited here).
aᵢ: age at death of person i. A death at or after L adds 0.

When you only have deaths grouped into age bands, use the midpoint method described by the CDC:

Count the deaths in each age band.
Find the years lost per death in that band: L − the midpoint of the band.
Multiply deaths by years lost per death.
Sum the products across bands.

In the CDC’s midpoint method the 15–24 band has a midpoint of 20, so each death in that band counts 45 years; 1,000 deaths in the band would give 1,000 × 45 = 45,000 YPLL. Following the same convention, deaths at ages 55–64 count 5 years each, and deaths at 65 and over are not counted.

The CDC also notes that YPLL depends directly on the number of births or the size of the group, so totals from populations of different sizes are not comparable. Convert to a rate before comparing, as shown below.

Worked example with real arithmetic

Illustrative data: the death counts and population below are hypothetical numbers made up to show the arithmetic. They are not real mortality statistics.

Suppose a region records 600 deaths before age 65 in one year, with 250,000 residents under 65. Cut-off L = 65.

Age band	Midpoint	Years lost per death (65 − midpoint)	Deaths	YPLL (deaths × years lost)
15–24	20	45	40	1,800
25–34	30	35	60	2,100
35–44	40	25	90	2,250
45–54	50	15	150	2,250
55–64	60	5	260	1,300
Total			600	9,700

Arithmetic: 1,800 + 2,100 + 2,250 + 2,250 + 1,300 = 9,700 YPLL.

YPLL rate: 9,700 ÷ 250,000 = 0.0388, which is 38.8 per 1,000 persons under 65 (3,880 per 100,000).
Average per death: 9,700 ÷ 600 ≈ 16.2 years lost per premature death.
Why it differs from a death count: deaths under age 35 are 100 of 600 (16.7%) but account for 3,900 of 9,700 YPLL (40.2%). Deaths at 55–64 are 260 of 600 (43.3%) but only 1,300 YPLL (13.4%).

How to interpret YPLL and YPLL rates

Result	What it tells you
Larger total YPLL for cause A than cause B (same population, same cut-off)	Cause A removes more years before the cut-off, which can happen with fewer deaths if they occur at younger ages.
YPLL per death is high	Deaths from that cause are concentrated at young ages (the maximum per death is the cut-off age).
Deaths rank high but YPLL ranks low	Deaths are clustered close to the cut-off; with a cut-off of 65 the 55–64 band counts only 5 years each.
YPLL rate per 1,000 (or 100,000) under the cut-off age	Allows comparison between groups of different size, because it divides by the population at risk.

Choosing a cut-off age (65, 75, 85)

The cut-off is a decision, not a constant. The CDC reports above used 65; CDC reports in 1990 and 1992 also compared results with an upper cut-off of 85, and a 1997 editorial note lists upper limits such as 65, 70, 75 and 85 in use. Moving the cut-off up gives more weight to deaths in middle and older age and counts more deaths overall. State the cut-off in every table title and never compare YPLL figures calculated with different cut-offs. The CDC also describes an alternative form of YPLL that emphasizes the life expectancy remaining at the time of death rather than a fixed end point; it gives different numbers, so label which one you used. Population-level context for these choices appears in what is demography and what is actuarial science.

Assumptions and common mistakes

Counting deaths at or above the cut-off. They contribute zero, not a negative number. Use max(0, L − age).
Comparing totals across populations of different size. Use a rate per 1,000 or 100,000 persons under the cut-off age, as the CDC reports do.
Mixing cut-offs or methods. A midpoint-method result and an individual-record result will differ slightly; say which you used.
Dividing by the whole population. The denominator for a YPLL rate is the population younger than the cut-off.
Treating YPLL as a prediction. It is a descriptive weighting scheme with an arbitrary cut-off, not an estimate of individual lost lifespan. For cost-based measures see what is health economics.

How to report YPLL (APA example)

State the cut-off, the method, the data source and the denominator. Example wording in APA 7th edition style:

Years of potential life lost before age 65 (YPLL-65) were calculated with the midpoint method (Centers for Disease Control, 1986). The region recorded 9,700 YPLL-65 in 2025, a rate of 38.8 per 1,000 residents younger than 65.

Reference entry (see the Sources section for the exact report): Centers for Disease Control. (1986, December 19). Premature mortality in the United States: Public health issues in the use of years of potential life lost. Morbidity and Mortality Weekly Report, 35(2S), 1s–11s.

How to calculate YPLL in Excel, R or Python

Excel

Put deaths in B2:B6 and band midpoints in C2:C6, then use a cut-off in F1 (65):

=SUMPRODUCT(B2:B6, F1 - C2:C6)

For individual ages in A2:A1000: =SUMPRODUCT((A2:A1000<F1)*(F1-A2:A1000))

R

deaths <- c(40, 60, 90, 150, 260)
mid    <- c(20, 30, 40, 50, 60)
cutoff <- 65
ypll   <- sum(deaths * (cutoff - mid))   # 9700
rate   <- ypll / 250000 * 1000           # 38.8 per 1,000
# individual ages: sum(pmax(0, cutoff - ages))

Python

deaths = [40, 60, 90, 150, 260]
mid = [20, 30, 40, 50, 60]
cutoff = 65
ypll = sum(d * (cutoff - m) for d, m in zip(deaths, mid))   # 9700
rate = ypll / 250_000 * 1000                                # 38.8 per 1,000
# individual ages: sum(max(0, cutoff - a) for a in ages)

Frequently asked questions

What does YPLL stand for?

YPLL stands for years of potential life lost, sometimes written potential years of life lost (PYLL). Both describe the years lost to deaths that occur before a chosen cut-off age.

What cut-off age does YPLL use?