Let’s have a kiki… I mean, a study

They keep asking for it. Antivaccine activists keep asking for a vaccinated vs. unvaccinated study. But they don’t want just any study. They want a study that has as much validity as the most valid epidemiological studies out there: Randomized Clinical Trials (RCTs). RCTs are incredibly difficult to set-up.

To set up a proper RCT you have to do several things. (I know because I had to help set one up for my master’s degree.) You’re looking to compare intervention A (e.g. a drug, a vaccine, or some thing like that) to intervention B. You get a group of people and you randomly assign them to group A or group B. Here’s the thing; you don’t tell them which group they’re in. This is called “blinding.” Otherwise, the people who are getting A and believe that it works might feel better or do things to make it work. It’s happened in real life. A group of people who had diabetes knew that the drug they were being given was not the best drug. So they started watching their carb intake, or continued to use their other medication, or exercised. Needless to say, that kind of skewed things a little. That is called “bias.”

But it’s not just the participants that you keep in the dark. You also have to keep the researchers conducting the study in the dark about which group is which. Say, for example, that you have been paid mad money to conduct the study for drug B. If it fails, you’re out of a job. You’re human, have a family to feed, and really don’t want to go back to waiting tables for a living. If you see that the people on B are not doing too well, you might be inclined to either fudge the results a little bit or put them on an additional drug to make them feel better.

Randomization and “blinding” keeps bias at bay.

The rest of the study is pretty standard. You check for predetermined outcomes like a change in blood pressure values, fasting blood glucose, days of school or worked missed… Things like that. Finally, you compare the two groups for outcomes. Some number crunching later, you determine which of group A or group B is more likely to have the outcome you were measuring, and whether this difference in likelihood was statistically significant.

So why can’t we do this with vaccines?

First of all, we’ve done this with vaccines. Here is the data on what was done to get the HPV vaccine approved. It has been done with other vaccines as well, but you know how anti-vaccine activists are. They don’t want facts to get in the way of their ideas.

Second, we can’t keep vaccines known to work away from children. Why? Because we’d be leaving those in the control group (the unvaccinated group), unprotected from diseases… From some very serious diseases.

Third, anti-vaccine parents will never go for this because there will be a 50/50 chance that their children are immunized. Remember, we randomize the groups into the vaccine or no vaccine groups. We have no control over it, and neither do they. Further, should the anti-vaccine parent find out that their child was vaccinated, do you think that they won’t exaggerate any ailment as being caused by the vaccine or minimize any good outcome — like not missing school due to the chickenpox — and not report it? Bias is a hell of a drug.

Of course, anti-vaccine advocates have come back and said that there are thousands of children who are not vaccinated. Why can’t we go back and look at their medical records and compare those records to vaccinated children?

We can, but what we get from it would be hard to interpret. It would be hard to interpret because these children will have gone to different providers in different healthcare systems at different times and in different locations. Lots of bias there. If we look at only one location, the sample size becomes too small to detect any “signals” in all the “noise.” And then there’s the very real possibility that anti-vaccine parents will tend to be anti-medicine and thus not take their children to licensed healthcare providers, opting for “alternative” providers like chiropractors or wizards, or homeopaths (who might as well call themselves wizards).

So how do we know that vaccines are safe, then?

We know because vaccines have to go through a stringent approval process that includes the agreement of continuing surveillance for adverse events even after the vaccine is “released into the wild.” The vaccines go through several phases of evaluation before they’re licensed to be used. And then, once they’re licensed, healthcare providers, epidemiologists, and scientists are all on the lookout for adverse events. And, contrary to what the anti-vaccine forces will tell you, the adverse events are far and few in between, and they are very rarely catastrophic or even deadly.

To quote a friend, “Anti-vaxxers must think that they win the Power Ball lottery every time they play” because of the way they exaggerate the odds of rare events.

Then again, if you still want to do this study, make sure that you are well-funded. (I’d ask the companies selling alternative medicine supplements and products for the cash.) You’re going to need it if any kids in your control group catch a deadly vaccine-preventable disease and you knowingly kept the vaccine away from them.

Advertisements

7 thoughts on “Let’s have a kiki… I mean, a study

  1. "The HPV vaccine used as a placebo was the same vaccine however non-L1 specific."Fail again. There were no antigens whatsoever in the placebo. There were also no adjuvants. You would never want anything that would elicit an immune response and confuse your results. But that shows how much you know.

  2. In case anyone beside the troll missed it, the study he points to — and the part he bolded — clearly show that controls were not vaccinated. They got a placebo, an inert substance that have them no protection as the vaccine is intended. You can always bet and still lose. The HPV vaccine used as a placebo was the same vaccine however non-L1 specific. Hence, some subjects in the placebo group had the infection caused by vaccine antigens (used as placebo) other than L1.

  3. Accuses me of lying, and then posts the evidence that I'm not. In case anyone beside the troll missed it, the study he points to — and the part he bolded — clearly show that controls were not vaccinated. They got a placebo, an inert substance that have them no protection as the vaccine is intended.Dear troll, your IPs are getting recorded when you comment. I know who you are, and that's why I know you don't know what you're talking about. ;-)Go post this all over anything you want, then go home and finish your homework. It's late in the semester.

  4. I'm curious as well as to why you find my complaining to be invalid. What do you think a placebo should contain? Certainly not another vaccine. Do you disagree?It is quite disingenuous to label subjects as unvaccinated in a vaccine study when they are also intended to receive vaccine-containing "placebos".

  5. Anonymous, I'm curious. You're complaining about what's in the placebo for a double blind study. What would you put in the placebo that would keep it double blind?Chemmomo

  6. Remember, we randomize the groups into the vaccine or no vaccine groups. We have no control over it, and neither do they. Oh really. How many times are you going to lie about placebo/control groups?The quadrivalent HPV-6/11/16/18 L1 VLP vaccine (GARDASIL/SILGARD, Merck and Co., Inc., Whitehouse Station, NJ) has been described.[22] The placebo used in this study contained identical components to those in the vaccine, with the exception of HPV L1 VLPs and aluminum adjuvant, in a total carrier volume of 0.5 mL. Vaccine and placebo were visually distinguishable.Safety and Persistent Immunogenicity of a Quadrivalent Human Papillomavirus Types 6, 11, 16, 18 L1 Virus-Like Particle Vaccine in Preadolescents and Adolescents: A Randomized Controlled TrialKeith S. Reisinger, MD, MPH, Stan L. Block, MD, Eduardo Lazcano-Ponce, MD, Rudiwilai Samakoses, MD, Mark T. Esser, PhD, Joanne Erick, BA, Derek Puchalski, BS, Katherine E. D. Giacoletti, MStat, Heather L. Sings, PhD, Suzanne Lukac, BS, Frances B. Alvarez, RN, BSN, Eliav Barr, MDDisclosuresPediatr Infect Dis J. 2007;26(3):201-209. Don't worry I'll post this one over at RI.

Comments are closed.