“All currently available evidence suggests that DTP vaccine may kill more children from other causes than it saves from diphtheria, tetanus or pertussis.”
—Dr Peter Aaby et al., of the Statens Serum Institute, Denmark. Quote is from study reviewed in this article.
Healthy user bias (HUB) is a serious problem in studies of vaccine safety. HUB is created when people with health problems avoid vaccination. When this occurs the unhealthy, unvaccinated subjects are used as controls. Consequently, the vaccinated group has better health at the outset. The better health of the vaccinated is erroneously attributed to the vaccine. The vaccine gets credit for improving health, when in fact it is causing harm.
Vaccine safety studies typically use administrative data (data collected by government agencies, HMOs and insurance companies), and HUB is present in this type of data. HUB occurs because, with administrative data, researchers can not control who receives the vaccine and who does not. HUB creates the appearance that vaccines have dramatic, diverse and implausible beneficial effects. For example, HUB is almost certainly responsible for the dramatic and implausibly amazing mortality reductions (about 50%) associated with the influenza vaccine. The HUB phenomena instead suggests that subjects getting the influenza vaccine pursue more “health seeking” behavior in general.
Fortunately, there is growing interest in performing studies that are not affected by HUB. The study described below (Mogensen et al 2017) is specifically designed to avoid HUB. It reports a 5-fold higher mortality associated with the DTP vaccine in infants. This result indicates a serious safety problem with the DTP vaccine, and serious problems with prior studies that do not account for HUB.
DTP, Death, and HUB
For decades there has been controversy and concern that the DTP vaccine may cause death, for example sudden infant death syndrome (SIDS). There are many reports of infants dying shortly after the DTP vaccine. However, studies using administrative data consistently find that DTP vaccine is inversely associated with SIDS. That is, children receiving a DTP vaccine were less likely to die (in a short period after vaccination) than unvaccinated controls. CDC researchers Dr Paul Fine and Dr Robert Chen (of the CDC) wrote an excellent paper on SIDS and healthy user bias in 1992. It is described here: http://vaccinepapers.org/healthy-user-bias-why-most-vaccine-safety-studies-are-wrong/
The Fine and Chen paper provides this table summarizing DTP-SIDS studies.
Above: List of studies of DTP vaccine and SIDS deaths from Fine et al, 1992. Notice that most of the odds ratios (ORs) are less than 1. An OR less than 1.0 indicates that the vaccinated subjects have a lower risk of death than the controls. Is this a real effect of the vaccine, or healthy user bias?
Most studies reported odds ratios (ORs) less than 1.0, which means that infants receiving the vaccine had lower risk of SIDs compared to unvaccinated controls. The low odds ratios (ORs) obtained in studies of DTP and SIDS should not be reassuring. This is because the low ORs suggest the presence of healthy user bias. It is not plausible that the DTP vaccine reduces risk of death in the days following vaccination. Of course, its impossible to know how strongly HUB is affecting the results. But HUB may be concealing a risk of death from the DTP vaccine.
The Fine and Chen paper states:
“Review of the literature on SIDS, encephalopathies, and DPT suggests that a large number of factors are associated with both a tendency to avoid or delay vaccination and an increased risk of SIDS and other serious neurologic events. That failure to control for such factors may lead to spurious negative associations between vaccination and adverse events is evident in several published investigations. Examination of the logic underlying this relation reveals that failure to control for such factors in analyses may mask true associations between vaccinations and certain adverse outcomes under certain conditions…” (Emphasis added)
The Mogensen et al. study was specifically designed to avoid HUB. It says the following about healthy user bias in prior studies of DTP and mortality:
“…the “unvaccinated” children in these studies have usually been frail children too sick or malnourished to get vaccinated, and the studies may therefore have underestimated the negative effect of DTP.”
NOTE: “these studies” = prior studies of DTP and mortality (or SIDS)
Until now, I have not been able to find studies of the DTP vaccine and mortality specifically designed to avoid healthy user bias. The Mogensen et al. study appears to be the first.
A Clever Study Design
The Mogensen study uses data collected during the “Bandim Health Project” (during 1980-1983) in the west african nation of Guinea-Bissau. In the Bandim Health project, the community of Bandim was followed for several years and health services were provided. New babies were identified and tracked for health outcomes. On a quarterly basis (once every 3 months), new babies received DTP vaccine. The quarterly vaccination schedule is critical to the study design.
Normally, in developed countries, babies are vaccinated according to a schedule based on a baby’s age. DTP vaccine is given at ages 2, 4, and 6 months according to the US CDC schedule for example. But this was not possible in the Bandim project because vaccination could only be done quarterly, when workers visited villages. All children (of appropriate age range of 3-6 months) in a visited village received vaccine on the same calendar day, but not at the same age.Consequently, babies were vaccinated at varying ages, depending only on their date of birth. This creates an excellent opportunity for research, because it means there is no HUB. The age at vaccination is determined only by birth date, and not health status, decision to vaccinate or delay, or any other confounder. Babies that did not receive DTP vaccine were not included in the study. So the Mogensen study avoids HUB, even though it was not prospectively randomized. Very clever.
“…the children were allocated by birthday to receive vaccines early or late and the “unvaccinated” were therefore not frail children.”
“The present analysis assessed DTP and child survival in a “natural experiment” in which the children were allocated by the timing of their birth and community weighing sessions and the group allocation was therefore not influenced by the usual selection biases to the same extent as most other studies of DTP.”
NOTE: “selection biases” = healthy user bias
“Though not individually randomized, the present study is a natural experiment with limited bias in group allocation: With 3-month intervals between weighing sessions, children were allocated by their birthday to receive their first vaccinations early or late between 3 and 5 months of age. We therefore compared 3–5-month-old children who had received DTP vaccinations early with children who had not yet received these vaccinations.”
“Sick children were not vaccinated, in the main analysis we therefore censored ‘unvaccinated’ children who attended a weighing session but did not receive a vaccination. Since the censoring of sick children could have introduced a bias, we also conducted an intention-to-treat analysis in which the censored children were transferred to the DTP group. Hence, in this analysis we compared the mortality of the intended-DTP-vaccinated group and the not yet DTP-vaccinated group.”
NOTE: Censor= Eliminate from the study.
So, Mogensen compared health outcomes for 1) subjects receiving the vaccine early, and 2) not-yet-vaccinated subjects receiving the vaccine late. The groups were well-matched because only the date of birth (not health conditions or decision to vaccinate) determined age at vaccination. The Mogensen study design definitely avoids HUB, as the researchers intended.
The primary result was that the DTP vaccine was associated with a 5-fold or 10-fold higher mortality rate. Detailed results are shown below. Receipt of the oral polio vaccine (OPV) seemed to reduce the mortality associated with DTP. DTP without OPV was associated with a 10-fold increase in mortality. However, the association with OPV is not statistically significant, so we cannot be confident that OPV truly affects mortality.
Above: Receipt of DTP vaccine in infants (at ages 3-6 months) is associated with 5-fold higher mortality. This study was designed to avoid healthy user bias (HUB). From Mogensen et al. 2017.
Mogensen et al. make strong statements about the inadequacies of prior studies and apparent dangers of the DTP vaccine:
“The negative effect of DTP was much worse in this natural experiment than has been reported in previous studies of DTP. This is presumably due to the “unvaccinated” control children in previous studies having been a frail subgroup too frail to get vaccinated. Previous studies have not been able to compare DTP-vaccinated children with “normal” controls. Hence, most previous studies have probably underestimated the negative effect of DTP.”
“DTP was associated with 5-fold higher mortality than being unvaccinated. No prospective study has shown beneficial survival effects of DTP…It should be of concern that the effect of routine vaccinations on all-cause mortality was not tested in randomized trials. All currently available evidence suggests that DTP vaccine may kill more children from other causes than it saves from diphtheria, tetanus or pertussis.”
Implications for Other Vaccine Safety Studies
These results have profound implications for other vaccine safety studies. Many vaccine safety studies may have dramatically underestimated the risks of vaccination because of HUB. A review paper (Shrank 2010) on HUB and related biases states:
“Clinicians should be skeptical when interpreting results of observational studies of preventive services that have not accounted for healthy user and related biases.“
DTP, OPV, and Th1-Th2 Balance
It is biologically plausible for the OPV to reduce mortality caused by the DTP vaccine. Specifically, the DTP and OPV vaccines induce opposite types of immune activation: DTP vaccine induces Th2 polarization (since it contains Al adjuvant), and OPV induces Th1 polarization (citation: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1906307/pdf/cei0127-0495.pdf). By inducing Th1, the OPV reduces the Th2 polarization created by the DTP vaccine. In infants, Th2 polarization is harmful and Th1 polarization is beneficial.
Research in animals show that Th2 polarization from the Hep B vaccine is harmful, and Th1 polarization from the BCG vaccine is beneficial for brain development: http://vaccinepapers.org/two-vaccines-opposite-effects-brain/