Urban health is where a city's data finally has to answer a health question: did anyone get treated, cooled, reached or spared — and did the people who needed it most actually use the thing? It is the one field in this atlas that routinely publishes the answer no.
This guide covers 23 documented entries across 18 cities in 18 countries, from Mexico City's automatic air network of 1986 to Rio's heat decree of 2024. Seven are graded peer-reviewed, the strongest evidence base of any topic here — not because health cities are better run, but because health researchers publish negative findings. Sensing networks belong to the air quality guide; this page stays on the service and the outcome.
| City | Intervention | What the peer-reviewed work found |
|---|---|---|
| Bangkok | Telemedicine, 2022 | Of 389 outpatients, 271 (69.7%) accepted it — 78.2% of millennials, but 1 of 13 aged 60 or over |
| Bogotá | Cable car to a hillside district, 2018 | Against a control district: travel time 110.0→90.2 minutes, quality of life +5.81 for women, not significant for men, no change in physical activity |
| Singapore | Free step trackers with rewards, 2015 | 696,907 registered adults raised mean daily steps by about 1,580 during the challenge phase |
| Manila | Court-ordered river monitoring, 2008 | Sewage is the dominant fecal source — 38% downstream, 22% upstream, animal sources under 1% |
| Cape Town | Shielding customers from load-shedding, 2015 | Independent synthetic control: about 547 premature deaths averted among over-65s, 2015–19 (95% CI 94–1,002) — in areas holding 81% of white but 27% of black African residents |
Because the technology usually works and the uptake usually doesn't, and only health-grade evaluation catches the difference. Bangkok is the clearest case. Its platform is real, it satisfies those who use it (81.2% satisfied, 72.5% intending to continue, same 389-person sample), and it reached almost nobody who was old. Alongside the survey, the hospital's deputy director reported over 30,000 app registrations against about 381 of roughly 2,000 regular endocrinology patients consulting by video. Registration is not use.
Barcelona answered that problem eight years earlier from the other end: staff-facilitated tablet groups rather than a clinical app, 3,067 users by 2022, with a 2024 peer-reviewed study of 255 of them finding it a key channel for social connectedness through the COVID lockdowns. What closes the age gap is a person, not a better interface. Osaka's passive version — meters in 30 mostly single-elderly homes, watched for consumption patterns suggesting something was wrong — carries no published evidence at all, and is graded none.
Miami-Dade created what it and its funder both call the world's first Chief Heat Officer in 2021; Athens named Europe's first weeks later, on 23 July. The post won something rare: a National Weather Service pilot lowering the county's heat-advisory threshold from a heat index of 108°F to 105°F for two hours, the only such pilot in the country. Then Florida's HB 433 barred counties from requiring heat protection for outdoor workers, and a $400M budget gap cut the post. The heat season still runs; nobody holds the title.
Rio shows the softer failure: five precise heat tiers whose decree words every consequence permissively and routes each through a committee and the mayor — while its own favela sensors have read almost 10°C hotter than the official station the level is computed from. Singapore's TraceTogether is the rare deliberate ending: 70% adoption, roughly 600,000 quarantine orders supported, switched off in February 2023 and the data deleted in 2024, the 2021 police-access controversy left in the record rather than out of it.
The practical takeaway: health is the one corner of this atlas where somebody usually checks — and the finding is almost never about the technology. It is that the tool works, and the people who need it most use it least. Budget for the facilitator and the control group before the platform; they are what turn a service into an outcome.
Rarely, on the best-measured evidence in this atlas. A peer-reviewed survey of 389 outpatients in Bangkok's Dusit Model area found 271 (69.7%) accepted telemedicine, but acceptance fell sharply with age: 78.2% of millennials, 46.7% of Generation X, and just 1 of 13 respondents aged 60 or over. Barcelona's answer to the same problem is not a better app but a facilitator — staff-led tablet groups for isolated residents over 65.
A senior official whose job is to coordinate a city's response to extreme heat. Miami-Dade created what it and its funder both call the world's first in 2021, and Athens named Europe's first weeks later, on 23 July. The Miami post achieved a National Weather Service pilot that lowered the county's heat-advisory threshold from a heat index of 108°F to 105°F for two hours — then Florida's HB 433 barred counties from requiring heat protection for outdoor workers, and a $400M budget gap cut the post. The heat season still runs; nobody holds the title.
Very few. Bogotá's TransMiCable cable car is the clearest: a natural experiment surveyed 2,052 adults against a comparable hillside district with no cable car, and found travel time fell from 110.0 to 90.2 minutes and quality of life rose 5.81 points for women but not significantly for men — while physical activity did not change at all. Cape Town's load-shedding protection has an independent synthetic-control estimate of about 547 premature deaths averted among residents over 65 between 2015 and 2019, with a wide confidence interval of 94 to 1,002.
Usually nothing is written down in advance. Singapore's TraceTogether is the exception in this atlas: the Bluetooth contact-tracing system reached 70% adoption and supported roughly 600,000 quarantine orders, was deactivated in February 2023, and had its data deleted in 2024. The 2021 controversy over police access to that data remains part of the record rather than being removed from it.
All three measure bacteria in urban water. What separates them is what the reading is allowed to do — post a warning, open a gate for the day, or file a report with a court for eighteen years running.
| Project | Since | Status | Outcome | Evidence | Run by | Cost | What the reading does |
|---|---|---|---|---|---|---|---|
| Copenhagen Bathing Water Forecast & Overflow Warning SystemCopenhagen, 🇩🇰 | 2002 | Live | Succeeded | Official | City government, Utility or public operator | not public | Forecasts rather than measures after the fact: rainfall and sewer-overflow data estimate risk and raise a red flag before anyone gets in. |
| Seine Bathing-Water Sensing & Public Swimming ReturnParis, 🇫🇷 | 2025 | Live | Succeeded | Official | City government | not public | Decides the day. Sites open or close each morning on that morning's water-quality result, under health-authority control. |
| Pasig River: Court-Ordered Water Quality MonitoringManila, 🇵🇭 | 2008 | Live | Ongoing | Peer-reviewed | National government | not public | Reports to a court, indefinitely, and the river still fails — a court can compel measurement for decades; it cannot compel the sewers that would change the reading. |