Vaccination and the omissions bias.

Vaccination has been in the news recently, as an outbreak of measles hits California. The US virtually eliminated measles around the turn of the century, but it has made a comeback. A big factor in that comeback has been ‘conscientious objection’ on the part of parents, who refuse to have their children vaccinated for religious or ‘philosophical’ reasons. Media reports often focus on the ignorance or confusion of these parents. And there’s plenty of both on show. Prominent anti-vaxxers continue to push the long discredited vaccination causes autism line, while the California conscientious objectors seem to have embraced an ill-informed ‘no chemicals’ line. I want to suggest that these views may be motivated, to some extent and in at least some parents, by the omissions bias.

The omissions bias is the tendency to see harms brought about by actions as more significant and more blameworthy than similar harms brought about by inaction. It plays a role in the moral thought of ordinary people: it is the reason why many people think it would be morally monstrous for a doctor to hasten the death of a terminally ill suffering person but permissible and perhaps even obligatory for the same doctor to withdraw treatment so that the person dies more rapidly. Now think about vaccination from the perspective of the omissions bias.

Experts recommend vaccinations because the benefits of vaccinations very significantly outweigh the costs. There are genuine risks associated with vaccinations. Most side effects are mild and transitory; less common ones, like febrile seizures, are not mild but usually do no lasting harm. Anaphylactic shock is probably the most common serious side effect. Treated promptly it is completely reversible but is life threatening if untreated. Deaths plausibly linked to vaccination are almost non-existent, but they may nevertheless occur. On the other hand, deaths from measles, while not common, are orders of magnitude more common than severe side effects of the vaccines.  Measles left about twice the number of people killed with permanent disabilities from severe brain swelling.

Suppose you are a parent who accepts all of this. You know that if you vaccinate your child, there is an extremely low probability that they will suffer a severe adverse event. On the other hand, there is a very much higher probability that if you don’t your child will contract measles and suffer a severe harm. Expected utility tells you to vaccinate. But vaccinating is acting, whereas not vaccinating is omitting to act. Due to the omissions bias, you will weigh these options quite differently.

Think about how you would feel if you vaccinated your child and she had a severe adverse event that left her disabled. You would blame yourself. “If only I hadn’t allowed her to be vaccinated”, you might say. If you don’t vaccinate her, though, and she contracts measles you are less likely to blame yourself: you might blame God or fate or nature, or the health system. Because acting and omitting to act are psychologically very different, you are likely to weigh the risks of vaccinating much more heavily, and the risks of not vaccinating much more lightly. I suggest that this may be part of the reason why the scare stories about vaccination continue to be given credence.

If that’s right, then one way to respond to the vaccination problem (in those countries that can’t or won’t make vaccination mandatory or impose sanctions, like exclusion from school, on children who are not vaccinated) is to reframe the options so that vaccinating occurs in the normal course of events. When the options are reframed like that, intervening to stop the vaccination looks like acting and allowing the vaccination to occur is what looks like inaction. This might be a perfect opportunity for advancing a policy goal by nudging: using social science to reframe options so that people are more likely to make the choices they should make, but without removing the effective option of dissenting and therefore without violating their freedoms.

  • Facebook
  • Twitter
  • Reddit

4 Responses to Vaccination and the omissions bias.

  • Lucas Collins says:

    It appears there’s some evidence for this approach. Nation Public Radio in the United States aired a story recently about a study comparing two methods by which doctors can raise the issue of vaccinations: either by presuming that the child will be getting the vaccination or offering the parents a chance to discuss whether the child should be vaccinated. When the doctor simply acted like it was the expected action, about 30% of the parents objected. When the doctor asked parents to discuss the decision, 83% of the parents objected. (Thirty percent still seems pretty high to me, but the difference is clear.)

  • Paul Salkovskis says:

    We were looking at the issue of omission around the immediate post Wakefield period. We wanted to use our focussing intervention, but dept of health said they couldn’t be seen to be seeking to subtly influence this!
    A few papers, eg

  • Nick Fitz says:

    Great post, Neil. This reminds me of the evidence that suggests that, in this case, changing policy is certainly more effective than changing minds. Nyhan et l al (2014) examined multiple strategies for communicating about the safety and importance of vaccines. None of them worked: In interviews about the study, Nyhan said, “I don’t think our results imply that [we] shouldn’t communicate why vaccines are a good idea, but they do suggest that we should be more careful to test the messages that we use, and to question the intuition that countering misinformation is likely to be the most effective strategy.” And that “other policy measures might be more effective.” It appears (to me) that it’s too easy to opt out of having your kids vaccinated; in some states in the US, it requires little more than a onetime signature on a form. I’ve been meaning to run some pilots that test effective messages in other science/technology culture wars (e.g., in-vitro meat, enhancement, drug criminalization, etcetera).

  • Dave Frame says:

    Excellent idea. I also think we should emphasize that vax-dodging is a particularly nasty form of free-riding. Basically it’s like tax dodging, but much worse. If I don’t pay my taxes, I free ride on public services paid for by others. The benefits are concentrated and the costs are diffuse (everyone else has to cover the cost of the same services). But in vax-dodging, the benefits (such as they are) are concentrated, but so are the costs. And those costs are borne by the most vulnerable members of society: infants, children with conditions such as leukemia which compromise their ability to cope with vaccination. We take a dim view of tax dodging. We should take an even dimmer view of vax dodging.


Subscribe Via Email