1751P - Re-ranking cancer mortality using years of life lost

Background

Standardized incidence and mortality rates are the default measures to describe cancer trends while survival assesses advances in cancer detection and treatment. Mortality compounds both incidence and survival but does not fully address age at death or long-term survival, both essential to quantify the cancer burden on society.

Methods

Using Swedish National Cancer and Cause of Death Register data together with publicly accessible projections of life expectancy, we calculated years of life lost (YLL) as the difference between age at cancer death and expected age at death for an individual of the corresponding age, sex, and birth year in the general population. All adults (18-95 years) with death reported as caused by one of the 10 solid tumors causing most deaths (lung, colorectal, prostate, pancreatic, breast, hepatobiliary, urinary, central nervous system, gastric cancer, or melanoma) in Sweden from 2010-2019, were included.

Results

When YLL were related and compared to mortality estimates in 2019 (Table), lung cancer (43,152 YLL) and colorectal cancer (32,340 YLL) remained at the top. Pancreatic cancer was up-ranked from 4th to 3rd place (22,592 YLL) and breast cancer from 5th to 4th place (21,810 YLL), while prostate cancer was down-ranked from 3rd to 5th place (17,380 YLL). In the table, a traditional approach ranking the 10 most common causes of cancer death in number of deaths and mortality rate (A), and new ranking using years of life lost approach (B), Sweden 2019.

Table: 1751P

Assessing YLL trends over the period from 2010-2019, women consistently lost more life years due to lung and pancreatic cancer than men. A downward trend in colorectal cancer mortality was reflected as a YLL decline only among women but not men.

Conclusions

Although all vital statistics have inborn advantages and disadvantages, this study demonstrates that YLL is simple to calculate, intuitive to interpret, and is as a meaningful measure to be added to the statistical toolbox box when presenting and comparing cancer trends within and between populations.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Region Stockholm (grant number: FoUI-963792).

Disclosure

All authors have declared no conflicts of interest.