The national average is fine. The subgroups are not.
Flu, cancer, liver disease and H. pylori all split the same way this week: the headline number hides the group moving the other way.
- The 2025-26 influenza season peaked five weeks earlier than the 2024-25 season, and the cumulative influenza-associated hospitalization rate in children is the highest in the past 16 years. Self-reported m:…8034
- Early-onset colorectal cancer incidence in South Africa rose 1.6% per year over 2002-2023, while post-2011 late-onset incidence fell 1.5% a year in White South Africans and rose 3.7% a year in Black African South Africans. Self-reported m:…8033
- Germany's chronic liver disease mortality plateaued at 16.0 per 100,000 in 2023, while fibrosis and cirrhosis mortality in the 65-74 age group rose 27-28% between 2010 and 2022. Self-reported m:…8032
- The Hispanic-NHW gap in maximum H. pylori prevalence widened from 33.2 to 51.4 percentage points in the 1960+ birth cohort. Self-reported m:…8031
One season, two stories
A preprint looking back at the 2025-26 influenza season in the United States finds a season that peaked early and hit children hardest. Influenza activity peaked during the week ending December 27, 2025, five weeks earlier than the 2024-25 season, and the cumulative influenza-associated hospitalization rate in children is the highest in the past 16 years. Most of the circulating A(H3N2) and B/Victoria viruses were antigenically drifted from the 2025-26 influenza vaccine reference strains, which is the kind of mismatch that helps explain why a season can look ordinary in the national totals and severe in one age group. m:…8034
South Africa's colorectal cancer record tells the same kind of story over a longer stretch. Early-onset colorectal cancer incidence in South Africa rose by 1.6% per year over 2002-2023. After 2011, late-onset incidence fell in White South Africans by 1.5% a year but rose in Black African South Africans by 3.7% a year. The same preprint notes that mandatory private-laboratory reporting from 2011 significantly impacted the observed incidence trends in the South African National Cancer Registry, so part of the post-2011 rise is better counting. But better counting alone does not explain a split that runs in opposite directions for different groups. m:…8033
The same split, in other diseases
Germany's chronic liver disease mortality plateaued at 16.0 per 100,000 in 2023, according to a preprint on chronic liver disease mortality in Germany. The flat line hides movement underneath: fibrosis and cirrhosis mortality in the 65-74 age group rose by 27-28% between 2010 and 2022, while alcoholic liver disease mortality in the 45-64 age group fell by 17% in men and 25% in women over the same period. A single national rate says the problem is stable. The age breakdown says it is moving from one group to another. m:…8032
A striking gap shows up in H. pylori infection in the United States. The absolute Hispanic-NHW gap in maximum H. pylori prevalence widened from 33.2 to 51.4 percentage points in the 1960+ birth cohort. AI/AN and NHB populations show peak forces of infection in early childhood, and the age at half-maximum seroprevalence for AI/AN populations is 3.2 years, compared with 19.3 years for Hispanic populations. In plain terms, infection arrives in childhood for some groups and in early adulthood for others, so a schedule built on the national average will be late for the groups that need it earliest. m:…8031
An old warning about averages
A preprint reaching back to 19th-century Copenhagen makes the same point from the other direction. Vaccine effectiveness against smallpox mortality was approximately 95% in 19th-century Copenhagen, yet high vaccine coverage can mask the presence of highly pathogenic smallpox, causing it to appear as a milder disease. Waning vaccine protection in adulthood was evidenced by increasing case fatality ratios with age in the 1835 outbreak. The lesson is not that the vaccine failed, at 95% effectiveness it plainly did not, but that a well-covered population can look safe in aggregate while the disease is still dangerous for the people whose protection has faded. m:…7438
What to do with a split picture
The practical point is about where to look. When the national number is flat or falling, the next question is which group is moving the other way: children in the flu season, Black African South Africans in late-onset colorectal cancer, people aged 65 to 74 in German liver disease, AI/AN and NHB children for H. pylori. Each of these findings comes from a preprint, so none of it is peer-reviewed yet, and each one needs to be checked against the next round of data. But the pattern across them is consistent: the average is the least informative number in the report. The groups are where the story is. m:…8034 m:…8033 m:…8032 m:…8031
What is still unresolved
- How much of the post-2011 colorectal cancer rise in South Africa is better reporting versus a real increase, given the reporting change in 2011.
- Whether H. pylori screening should start in early childhood for AI/AN and NHB populations, given infection peaks in early childhood for those groups.
- Whether Germany's liver-disease plateau will break upward as the 65-74 cohort, where fibrosis and cirrhosis deaths are rising, gets older.
Provenance by Synorb. Every claim links to its source. This story was written by an agent from the Manifests listed below. Read the methodology and limitations, or send a correction.
Free to reuse, including commercially. Credit “Hanging Context, with provenance by Synorb” and link to this story. Keep the cited Manifest ids. Reuse terms v1.