Every year it happens the same way. The rains come in June. By August the hospitals fill up. By October it eases off.
I always thought that was just a feeling. It is not. It is one of the most reliable patterns in Philippine health data.
Where the Data Came From
The Department of Health's Epidemiology Bureau publishes dengue counts. Their file went up on HDX, a humanitarian data site.
It holds 32,702 rows. Each row is one province in one week. It runs from 2016 to 2021, covering 81 provinces and 17 regions.
For the newer years I used DOH press releases, up to mid-February 2026. Those give national totals only, not province detail. I have marked which years come from which source, because mixing them quietly would be dishonest.
The Mosquito Calendar
Dengue follows the rain. Cases climb from June. They peak in August and September. Then they fall through the last months of the year.
August is about six times worse than April. And this shape repeats every single year in the data.
That regularity is useful. June is the month things turn. So spraying and clean-up drives that start in May do better than ones that start in July. By July you are already chasing it.
Where It Lands
Three regions carry 35.9% of all cases: CALABARZON, Central Luzon, and Western Visayas.
At the province level, Cavite leads with 44,169 cases from 2016 to 2021. Then Laguna with 42,013. Then Cebu with 39,957.
But I want to be careful here. Those are big, crowded provinces. More people means more cases. Per person, they are not unusual. The map is mostly showing where Filipinos live.
Two Numbers That Need Explaining
From 2016 to 2020, about 1.47 out of every 100 people with reported dengue died. That figure is called the case fatality rate. It has been falling as care improved.
Two things in the data look strange, and both have the same lesson.
First, 2017. The death rate jumped to 2.96%. That is high for a year with only about 150,000 cases. The likely reason is not a nastier virus. Late 2017 was the Dengvaxia vaccine controversy. Trust in the health system fell, and people waited longer before going to a hospital. Waiting is what kills with dengue.
Second, Quezon City. Its death rate reads 3.42%, about nine times the national figure. That almost certainly is not worse care. Big city hospitals record deaths more completely, and they treat patients who live elsewhere. The city gets counted for deaths that belong to a wider area.
Both cases say the same thing. A strange number usually means a strange way of counting, not a strange disease.
Recent Years
2020 dropped 79% from 2019. Part of that was real, because lockdowns kept people home. Part was simply fewer people going in to be counted.
2024 came roaring back. 269,467 cases by October alone. That is 37.8% more than all of 2023 — and October is not the end of the year. It likely finished near 300,000.
2026 has started slow. The health department says it is slower than last year, and I believe them. But I cannot tell you by how much. Nobody has published the matching count for the same weeks of 2025, so I am leaving that number out instead of guessing at one.
And I would not celebrate yet. It might be a genuine low year in the cycle. It might also be late paperwork. August will tell us which.
What I'd Do Differently
The biggest gap is that province-level detail stops in 2021. Everything after that is one national number. So I can see that 2024 was bad. I cannot see which towns it hit.
I would also like to lay rainfall on top of the cases, week by week, town by town. The link is obvious in outline. Nobody has drawn it here in detail.
Want to see all the charts and data tables?
View the Full Analysis →