A recent BMJ Oncology paper (Zhao et al. 2023) reported that “Global incidence of early-onset cancer increased by 79.1% and the number of early-onset cancer deaths increased by 27.7% between 1990 and 2019.”.
Those are stagering numbers, so let’s take a look at how they make their calculations.
Incidence
In page 3, they state: “In 2019, the incidence number of early-onset cancer was 3.26 million, a 79.1% increase from 1990”.
So, they are using the absolute number of early-onset cancers in 2019 (3,260,000) and comparing it with the 1999 number (1,821,229).
If you calculate the increase, it really gives you a 79% = 3,260,000 / 1,821,229.
Deaths
Also in page 3: “Besides, the number of early-onset cancer deaths in 2019 was 1.06 million, which was an increase of 27.7% from 1990”.
Again, they are using the absolute number of early-onset cancer deaths in 2019 (1,060,000) and comparing it with the 1999 number (830,070).
If you calculate the increase, it gives you a 28% = 1,060,000 / 830,070.
Context
The human population grew quite a bit in the last 30 years. According to worldometers, it was 5,316,175,862 in 1999 and increased to (7,764,951,032) in 2019. That is a 46% increase.
Something I do not understand well is that in the paper, the authors talk about incidence, which is usually defined as a “measure of the probability of occurrence of a given medical condition in a population within a specified period of time”, but give an absolute number: “In 2019, the incidence number of early-onset cancer was 3.26 million, (…)”.
Redoing the calculations
Instead of the absolute numbers, how many people got or died of an early-onset cancer, we can normalize those numbers, and make them relative to the population at that point in time.
Indicence
For 2019, dividing the incidence number (3,260,000) and dividing by population (7,764,951,032) gives us 0.042%. This is the % of the population with an early-onset cancer diagnosis in 2019.
For 1999, dividing the incidence number (1,821,229) and dividing by population (5,316,175,862) gives us 0.034%. This is the % of the population with an early-onset cancer diagnosis in 1999.
Finally, if you calculate the increase: (0.0004198352 / 0.0003425825), we get a 22.550%. This is an increase of 22.5% in ‘incidence’ between 1999 and 2019.
Mortality
For 2019, dividing the mortality number (1,060,000) and dividing by population (7,764,951,032) gives us 0.014%. This is the % of the population that died of an early-onset cancer in 2019.
For 1999, dividing the mortality number (830,070) and dividing by population (5,316,175,862) gives us 0.016%. This is the % of the population that died of an early-onset cancer in 1999.
If you calculate the increase: 0.0001365108 / 0.0001561404 , we get a -12.572% . This is a reduction of 12% in mortality between 1999 and 2019.
Conclussions
First of all, I might be missunderstaing something or making a mistake here. I would be very happy to correct it.
Assuming the logic above is reasonable, we go from the reported 79% increase in ‘incidence’ to an increase of 22.6%. This is way easier to explain in terms of improvements in the medical tests, screening programs, etc.
For deaths, we go from the reported 27% increase to a decrease of -12.6%. This is also makes more sense given all the medical advances in detection and treatment of the last 30 years.
References
Citation
@online{navarrete2023,
author = {Navarrete, Gorka},
title = {Cancer Increase in Young Adults?},
date = {2023-09-27},
url = {https://gorkang.github.io/GN-web/posts/2023-09-27-incidence_mortality/},
langid = {en}
}