Public health surveillance: America the backward

by revere, cross-posted from Effect Measure

A good story by the AP's Lauran Neergaard yesterday highlighted the need for better public health surveillance and the efforts being made to improve it so as to keep track of possible rare side effects from the swine flu vaccine. This is an issue we've talked about a lot here, most recently in the context of not being able to fully test any vaccine for rare adverse outcomes prior to deployment. There's more involved than that, but first here's Neergaard's lede:

The U.S. government is starting an unprecedented system to track possible side effects as mass swine flu vaccinations begin next month. The idea is to detect any rare but real problems quickly, and explain the inevitable coincidences that are sure to cause some false alarms."Every day, bad things happen to people. When you vaccinate a lot of people in a short period of time, some of those things are going to happen to some people by chance alone," said Dr. Daniel Salmon, a vaccine safety specialist at the Department of Health and Human Services. (Lauran Neergaard, AP) 

No pre-deployment clinical trial could hope to pick up a risk of 1 in 100,000 but even risks that small will occur when you are vaccinating more than 100,000,000 people. But we've discussed this quite a bit, most recently here. So here's another part of the article, one that might surprise you:

Then there's the glare of the Internet â where someone merely declaring on Facebook that he's sure the shot did harm could cause a wave of similar reports. Health authorities will have to tell quickly if there really do seem to be more cases of a particular health problem than usual.So the CDC is racing to compile a list of what's normal: 25,000 heart attacks every week; 14,000 to 19,000 miscarriages every week; 300 severe allergic reactions called anaphylaxis every week.

Any spike would mean fast checking to see if the vaccine really seems to increase risk and by how much, so health officials could issue appropriate warnings. 

The implication here -- and it is a correct -- is that we don't have these elementary facts instantly available, facts like how many heart attacks occurred last week (or even last month), where did they occur and to whom. Consider what CDC is trying to jury rig for the vaccine issue:

• Harvard Medical School scientists are linking large insurance databases that cover up to 50 million people with vaccination registries around the country for real-time checks of whether people see a doctor in the weeks after a flu shot and why. The huge numbers make it possible to quickly compare rates of complaints among the vaccinated and unvaccinated, said the project leader, Dr. Richard Platt, Harvard's population medicine chief. 
• Johns Hopkins University will direct emails to at least 100,000 vaccine recipients to track how they're feeling, including the smaller complaints that wouldn't prompt a doctor visit. If anything seems connected, researchers can call to follow up with detailed questions.  
• The Centers for Disease Control and Prevention is preparing take-home cards that tell vaccine recipients how to report any suspected side effects to the nation's Vaccine Adverse Event Reporting system. (from the same AP article) 

Every other industrialized country has a national health care system that makes keeping track of these elementary facts possible. The US doesn't. We have a lot of electronic medical records, all right, but they are mostly devoted to billing and insurance. And there are a lot of different proprietary software systems that can't be easily adapted, altered or modified and can't talk to each other. One of Obama's initiatives to control costs is Electronic Medical Records (EMR), but the economic benefits he touts are almost certainly being oversold. It won't save us that much money.

But what a decent system could do -- and the system that we might get might be very, very far from a decent one from the provider and patient perspective -- is provide the kind of surveillance information that would make assuring the safety and efficacy of vaccine programs and a myriad of other things possible.

As we've noted too many times to count, nobody cares about public health surveillance until they suddenly need the information it produces and then it's too late. Surveillance is essential public health infrastructure but frequently is a casualty when budgets get cut. Surveillance systems are invisible to the public and politicians, unless they serve some kind of economic benefit for a powerful industry and then you can be sure some CongressThing or Senator will have gotten a federal agency to foot the bill. We spent a long post yesterday giving the details of one part of CDC's cobbled together patchwork flu surveillance system, but there are many more systems out there, hanging by slender threads, that alert us to problems with consumer products, violent deaths, unintentional injuries and much more.

Surveillance systems not only alert us to problems, they allow us to know something isn't a problem. That can save a vaccination program and thereby save lives. One of the many benefits of a national health system would be the required EMR that would go with it. And that would make possible the kind of continuing surveillance and special disease tracking that CDC is desperately scrambling to set-up on an ad hoc basis even as it launches an ambitious vaccination program for the first influenza pandemic of the 21st century.

The anti universal health care talking point that this country has the best health care system in the world is a joke. A sick joke. Sick, like our health care system. We can't even count up how many people are sick and from what on a timely basis. America the backward.