Eleven years of cases, deaths, seasonality, and geography
32,702 province-week observations from DOH-Epidemiology Bureau via HDX (2016-2021), combined with national totals from DOH press releases through mid-February 2026. Covers the 2019 epidemic peak (442K cases), the COVID-era collapse, the 2022-2024 resurgence, and the early-2026 reversal.
The 2019 dengue epidemic remains the modern record (442K cases) — but 2024 came close, and dengue's seasonal shape (peaks in Aug-Sep, troughs in Apr-May) is remarkably consistent year over year.
Annual reported dengue cases follow a roughly 3-year boom-and-bust cycle. The 2019 epidemic dwarfed everything before it.
2021 incomplete (Jan only); 2024 covers Jan-Oct; 2025 covers Jan-Mar; 2026 covers Jan to mid-Feb only.
Deaths declined faster than case counts as treatment improved
2017 spike likely Dengvaxia-controversy era effect
Dengue seasonality is one of the most consistent patterns in Philippine epidemiology — cases rise with the rains in June, peak in August-September, then taper through Q4.
Cumulative across all six years — same shape repeats every year
CALABARZON dominates total burden — three of the four top provinces (Cavite, Laguna, Batangas) sit there. Visayas hot spots include Cebu, Iloilo, and Negros Occidental.
Reported cases × 1,000
Cumulative cases by administrative region
Cumulative reported cases and deaths · case fatality rate
| Province | Cases | Deaths | CFR | Region |
|---|---|---|---|---|
| Cavite | 44,169 | 170 | 0.38% | CALABARZON |
| Laguna | 42,013 | 108 | 0.26% | CALABARZON |
| Cebu | 39,957 | 688 | 1.72% | Central Visayas |
| Iloilo | 37,155 | 287 | 0.77% | Western Visayas |
| Quezon City | 35,280 | 1,208 | 3.42% | NCR |
| Pangasinan | 31,310 | 87 | 0.28% | Ilocos |
| Bulacan | 30,898 | 90 | 0.29% | Central Luzon |
| Negros Occidental | 30,323 | 778 | 2.57% | Western Visayas |
| Batangas | 28,465 | 108 | 0.38% | CALABARZON |
| Rizal | 27,686 | 216 | 0.78% | CALABARZON |
| Pampanga | 27,267 | 73 | 0.27% | Central Luzon |
| Bukidnon | 27,216 | 244 | 0.90% | Northern Mindanao |
Five takeaways for public health planning, vector control, and epidemic preparedness.
3.42% case fatality is 9× the national average — likely a reporting artifact (better death registration, residence-vs-hospital geocoding) rather than worse outcomes. Worth disaggregating before drawing clinical conclusions.
CFR of 2.96% in 2017 is unusually high for a year with ~150K cases. The Dengvaxia controversy (Nov 2017) likely caused care-seeking delays. CFR returned to 0.4-1.3% from 2018 onward.
163K cases in 2016-2021 (15% of national) — a function of population density and urbanization, not unusual transmission. Per-capita rates are similar to other dense regions.
June is the inflection month every year. Vector control programs that ramp up by May (one month ahead) consistently show better July-September outcomes than reactive June-July deployments.
14,907 cases through mid-February vs 43,732 in the same period of 2025 — a ~66% drop. Could be true epidemiology or just delayed reporting; the August peak will tell.
Case counts say how many people got sick. They do not say which dengue. There are four serotypes, and which one dominates matters: a second infection with a different serotype carries a much higher risk of severe dengue. This section pulls every Philippine dengue sequence deposited in the European Nucleotide Archive — 1,591 of them — and asks what has been circulating, year by year.
Percent of sequenced samples, ENA portal API
This measures sequencing effort, not incidence
DENV-2 led the early 2000s, DENV-3 took over around 2006-2009, DENV-1 dominated 2011-2016, and DENV-3 returned from 2017. That rotation is the mechanism behind big outbreak years: when the circulating serotype changes, a population with immunity to the old one is newly vulnerable to severe disease.
2023 produced 380 sequences from 195,603 reported cases. 2019 — the worst dengue year on record, with 441,902 cases — produced 116. And 2022, with 252,700 cases, produced five. The count chart maps who was paying for sequencing, not where the disease was.
In 2019 and 2023 every serotype appears in double digits at once. Co-circulation is the norm here, not the exception, which is why a single-serotype vaccine strategy proved so fraught.
Behind the sequences sit only 147 raw sequencing runs across 5 registered studies, and one project (PRJNA1009983) contributes 127 of them. Philippine dengue genomics rests on a very small number of efforts.
All data is open. The province-week granularity stops at 2021; 2022-2026 are nationwide totals only.
doh-epi-dengue-data-2016-2021.csv from data.humdata.org. 32,703 rows, 5 columns (loc, cases, deaths, date, Region). HXL-tagged second row skipped on read.
Annual nationwide totals from DOH press conferences via PNA, Manila Times, Statista, vax-before-travel.com. No province-level granularity for these years in open form.
2021 only has January data in HDX. 2024 is Jan-Oct. 2025 is Jan-Mar. 2026 is Jan to mid-Feb. The 2024 full-year total is likely ~310-330K based on Q4 seasonality.
PIDSR is passive surveillance — it captures hospitalized cases. Active community surveillance studies (e.g., Cebu City PMC paper) suggest true incidence is 2-4× reported. The shape is right; the level is a floor.
Annual sums use ISO calendar weeks. Province totals match HDX; some provinces appear with leading whitespace (" NEGROS ORIENTAL") — stripped before grouping.
Query tax_tree(12637) AND country="Philippines" against ebi.ac.uk/ena/portal/api — free, no key. Serotype is read from scientific_name, which is populated far more reliably than the serotype field. Sequence counts measure sequencing effort, not incidence; only the within-year composition is epidemiologically meaningful.
ph_dengue_annual.csv, ph_dengue_top_provinces_2016_2021.csv, ph_dengue_by_region_2016_2021.csv, ph_dengue_monthly_seasonality.csv — all source-tagged.
I wrote a companion post — “Dengue Arrives Every August, and the Data Knows It” — in simple, everyday words.
Every figure on this page traces to one of these, through a CSV in
data/ph-dengue/. Each is checked against its source query on every
build.
Province-week cases and deaths, 2016-2021 — PIDSR surveillance redistributed by the Humanitarian Data Exchange
Nationwide annual totals, 2022-2026 — Reported via PNA, Manila Times and Statista; doh.gov.ph is not reachable by script
Dengue virus sequence and run metadata for the Philippines — Queried as tax_tree(12637) AND country=Philippines