Alternative Schedules

Ethical and Scientific Reflections on Studying Alternative Vaccination Schedules

Study Groups

Vaccinated - ACIP Schedule
Vaccinated - Alternate Schedule(s)
- Problem of heterogeneity: How does one deal with the fact that there are more than one alternate schedule? This group may not be homogenous. How should we include children who have missed one or more vaccine (either intentionally or unintentionally) and were administered vaccines using a “catch-up” schedule? Would these children be considered on an “alternate schedule”? (Could an RCT impose uniformity?)
- Estimates of parents who have elected an alternate vaccination schedule range from 13 to nearly 22 percent. (See Dempsey 2011 and Smith 2010.)
Unvaccinated – Dempsey (2011) found that only 2% of parents surveyed refused all vaccinations

Herd Immunity

The level of herd immunity in the general population is an important variable impacting on the risk of vaccinepreventable disease in the study population.
Existing herd immunity in the United States would make it extremely unlikely to observe an increase in vaccinepreventable disease during an RCT of any feasible design.
The level of herd immunity in the general population may have an impact on all three study groups, including those infants and children who are being vaccinated using the ACIP recommended vaccination schedule.
How do we account for this important variable?

Efficacy as an Outcome

When administered according to the recommended dosing regimen, vaccines are known to be safe and effective in preventing the diseases for which they have been approved and licensed.
The ethical concern is that the use of an alternate vaccination schedule would result in an increase in vaccine-preventable diseases.
The risk of disease depends broadly on two factors - the individual child's degree of immunity and that child's exposure to the vaccinepreventable disease (e.g., travel, child care, school, herd immunity).
Presumably the hypothesis of the proposed study would be that an alternate vaccination schedule offers a safety advantage (i.e., a

reduction in vaccine-associated severe adverse events) while not resulting in an unacceptable increase in vaccine-preventable disease.

Thus, the study endpoints must include the occurrence of vaccinepreventable diseases in addition to any vaccine-related adverse events.

Safety as an Outcome

The safety and efficacy of new vaccines are established against the backbone of the recommended vaccination schedule. In other words, infants and children who are enrolled in clinical trials of new vaccines have received the other routine vaccinations that are recommended for that age group. In effect, studies of new vaccines should be considered “add on” trials of the new vaccine along with the recommended vaccination schedule in use at the time of the trial. Thus, the safety of the new vaccine is, in effect, a measure of the overall safety of the recommended vaccination schedule.
The perception that the overall safety of the vaccination schedule has not been tested is, strictly speaking, not accurate.
That being said, there are limits to the power of prospective controlled clinical trials to detect exceedingly rare vaccine-related adverse events

Randomization And Binding

Randomization independent of parental preference would not be ethical nor feasible.
Absent randomization, blinding the parent is not an option.
Absent randomization and blinding, prospective observational studies are subject to confounding and bias (as are retrospective epidemiological studies).

Retrospective Study?

Can one perform an epidemiological study based on health records, such as the Vaccine Safety Datalink, using de-identified data from participating managed care organizations (more than 8.8 million people annually, representing nearly 3% of the United States population)?
Concern: Under-immunization is geographically variable (i.e., not uniform) and may be clustered such that some data sets (e.g., based on HMO participation) may not be representative of parents who elect alternate vaccination schedules and thus may under-estimate the risks to local populations. (See, for example, Omer SB et al 2008.)

Prospective Observational Study?

The list of possible outcomes to evaluate may require standardized assessment protocols (rather than relying on clinically generated data alone). This approach would require a pre-specification of those outcomes of particular interest.
The observational study may need to over-sample children whose parents have decided to either forgo vaccination or to follow a delayed (or reduced) vaccination schedule. The overall sample size would need to be large enough to detect exceedingly rare adverse events.
As health outcomes should also include vaccine-preventable diseases, an observational study will be confounded by issues such as disease exposure and the level of herd immunity in the population to which any given child is exposed.

Good Ethics starts with Good Science!

The interventions and procedures that likely would be included in a prospective observational study (e.g., blood tests, neurodevelopmental assessments) would be considered minimal risk or, at most, a minor increase over minimal risk.
The ethical justification of such an observational study depends on the quality of the science.