(RxWiki News) The issue of vaccine safety has long plagued parents as much as it has researchers. How do we really know if a vaccine "caused" a condition or not?
The answer is that researchers compare the rate of illnesses in different groups.
If a certain condition shows up more often than it normally would in a group of kids after a new vaccine is introduced, there's a better chance the vaccine is related to that condition.
But where do they get those numbers in the first place? They get them from studies of populations.
"Follow the CDC vaccination schedule."
A new study from Denmark provides an excellent reference for future research vaccine safety by calculating how often a range of different conditions would be expected to occur in a large group of children.
The study, led by Thomas A. Rasmussen, a resident at the Department of Infectious Diseases within Aarhus University Hospital in Denmark, uses data from Denmark's entire population for certain years.
All babies born after January 1, 1980 were included in the study, which concluded on the last day of 2009. This included a little over 2.3 million children.
The children were tracked until they were admitted to the hospital for one of the particular conditions the researchers were looking for or until the child died, moved out of the country or turned 18.
The average follow-up time per child was until the child was almost 17 years old.
The children who were vaccinated were exposed to the standard child vaccination schedule used in Denmark. Vaccination rates in Denmark range from 82 to 93 percent for this study period.
The Danish schedule includes the DTaP (diphtheria-tetanus-pertussis), Hib, polio, hepatitis B, HPV, MMR and pneumococcal vaccines. It differs from the U.S. schedule in that the CDC schedule also includes the chicken pox, rotavirus, hepatitis A, meningococcal and flu vaccines.
The conditions the researchers looked for included the following:
- Guillain-Barré syndrome
- acute transverse myelitis
- optic polyneuritis
- facial nerve palsy
- anaphylactic shock
- seizure
- multiple sclerosis
- autoimmune thrombocytopenia
- type 1 diabetes mellitus
- juvenile and rheumatoid arthritis
- narcolepsy
- death from unknown cause
Using these rates, the researchers predicted how many of these conditions would be expected to occur in every group of 1 million children who were vaccinated, for up to 182 days after being vaccinated.
They calculated how frequently the conditions would show up in "patient years." Patient years are the number of patients times the number of years followed.
Using patient years lets researchers looked at the frequency across people and time simultaneously for more accurate figures.
The researchers found that the conditions they looked for would be expected to occur in a group of vaccinated children at rates ranging from 0.32 to 190 diagnoses per 100,000 patient years.
That means that the researchers established how many children would be likely to get a specific condition, such as narcolepsy, diabetes or seizures.
Over 100,000 patient years, for example, approximately 190 children should be expected to have a seizure.
(An example of 100,000 patient years could be 10,000 children followed over a ten-year period. In 10 years, 190 seizures would be expected to occur among those 10,000 children.)
Knowing these numbers is valuable for future research into the safety of vaccines. Researchers can compare the rates of these conditions from this study to the rates in the future with new or different vaccines to find out if there is any difference between them.
If the condition shows up more in children receiving a different vaccine — in clinical trials or in the general population after a vaccine is introduced — than it did in this study, there is a stronger possibility that the new vaccine is linked to that condition.
The study was published September 17 in BMJ. The research was funded by the Department of Infectious Diseases at Aarhus University Hospital in Denmark.
The authors declared no association with organizations or companies who may have an interest in this work. No pharmaceutical companies contributed to this research or are connected with the authors.
