November 1999
SAN FRANCISCO - In a recent study, Haemophilus influenzae type b (Hib) and hepatitis B (HepB) vaccines were not associated with an increased risk of developing type 1 diabetes. Additionally, the timing of HepB vaccination was not associated with risk.
"Our study had two objectives: to determine the relative risk of developing type 1 diabetes mellitus associated with vaccines for Hib and for HepB; and to determine if the risk associated with HepB vaccination varies by timing - that is, if risk is decreased if the vaccination occurs at birth as opposed to increased if occurring at 2 months of age or later," said Frank DeStefano, MD, at the recent 39th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) held here.
Recently, controversy has developed surrounding allegations that vaccines may cause type 1 diabetes. "This controversy has been fueled predominantly by some analyses of follow-up data from a large clinical trial that was conducted in Finland. An American researcher interpreted 10-year follow-up data from that trial as indicating a significant increased risk of diabetes associated with Hib. Finnish researchers, on the other hand, analyzed the same data and concluded that there is no association," said DeStefano, who is the Vaccine Safety Datalink Project director at the Centers for Disease Control and Prevention (CDC).
Another area of this debate concerns the hypothesis that timing of vaccination may have different effects on diabetes risk. The hypothesis is that vaccination at birth is actually protective or decreases risk, whereas waiting to give a first vaccination later in life may increase risk.
"The hypothesis of a protective effect at birth has been supported by some studies in animal models. However, support for increased risk for later vaccination is weaker and is primarily based on ecologic analyses. These analyses compare diabetes rates in different countries or at different times in which different vaccination schedules have been in effect. We're all familiar with the limitations of these kinds of analyses," he said.
![[bar]](/art/gradred.gif){kind=small}
In the United States, the timing hypothesis is primarily relevant to HepB vaccinations, because they are the only vaccines for which there is the option to vaccinate at birth. "However, there have been no published epidemiologic studies evaluating HepB vaccination and diabetes risk or the timing hypothesis," DeStefano said.
Therefore, he and his colleagues conducted a matched case-control study. The data were derived from three large HMOs located on the West Coast of the United States. All three participate in the CDC's Vaccine Safety Datalink Project. The study was restricted to children born in the HMOs between 1988 and 1997.
"We made this restriction so that we could have a complete vaccination history from birth during the years in which the vaccines of interest have been in use," he explained.
Cases were identified from diabetes registries that are maintained by each of the HMOs and were confirmed using World Health Organization (WHO) criteria. Three controls were matched to each case according to HMO, date of birth and gender. A standardized chart review was conducted to verify case status. Additionally, vaccination history and information on other risk factors were obtained for both cases and controls. Conditional logistic regression was used to estimate the relative risk while accounting for the matching strategy and adjusting for race and family history of diabetes.
"We identified 140 confirmed cases of type 1 diabetes and 418 matched controls. The proportion of cases and controls that were vaccinated with each vaccine was similar. The results of the regression analyses revealed adjusted relative risks of 1.11 for Hib vaccine and 0.72 for HepB vaccine. The analysis of timing of HepB vaccine revealed an adjusted relative risk of 0.60 for children who were vaccinated at or up to 14 days after birth, compared with children who had not received HepB vaccine. Children vaccinated at 8 weeks of age or later had a relative risk of 0.78 compared with nonvaccinated children," he said.
This study found that neither Hib nor HepB vaccine increased the risk of developing type 1 diabetes in children. Additionally, the risk did not vary by timing of HepB vaccination. In fact, no evidence was found of an increased risk if first vaccination occurred at 2 months of age or later.
"We think these results are reassuring, but we would like a larger sample size with confidence intervals that are less wide. A fourth HMO is currently collecting data. When we have data from all four HMOs, we will have more precise results," DeStefano added.
For more information:
- DeStefano F, Okoro CA, Mullooly JP. Hepatitis B and Hib vaccinations and type 1 diabetes. Abstract LB-11. Presented at the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy. Sept. 26-29. San Francisco.
You can express your
views on this article, or other relevant themes, in the
Infectious Diseases in Children
Specialty Forums.