Author: Peter C. Gøtzsche, Professor, DrMedSci, MSc.
Summary:
A WHO systematic review from 2014 assessed the effect of three vaccines on total mortality in infants and children: BCG (Bacille Calmette-Guérin), DTP (diphtheria, tetanus, and pertussis) and measles.
I updated the literature searches and found two highly relevant studies where the researchers had improved on their previous research in response to the criticisms raised in the WHO report. They found that the DTP vaccine doubled mortality, confirming their previous findings. In one of their studies, which represents the best available evidence, they explained that the criticisms raised in the WHO report were either not relevant, or they had taken it into account. They found that all the documented biases favored the vaccinated group, i.e. they had likely underestimated the harmful effect of the DTP vaccine on mortality.
The researchers provided a statistical summary (meta-analysis), of the only three studies of the introduction of DTP in Guinea-Bissau. They found that DTP vaccination was associated with a hazard ratio of 2.14 (1.42 to 3.23) compared with DTP-unvaccinated children. They also found that all studies of DTP, which analyzed existing data sets collected for other purposes, suffered from substantial frailty and survival bias that lead to underestimation of the harms of the vaccine.
There were major problems with the WHO report. Although it found that most studies showed a deleterious effect of DTP, the authors concluded that the results were inconsistent because two studies showed a beneficial effect. However, they did not find a significantly beneficial effect on mortality, and they were so seriously biased that they should not have been taken into account.
The authors did not provide summary estimates because the WHO Working Group had requested that meta-analyses not be done.
This is an unacceptable interference with research by a body that includes people with numerous financial conflicts of interest in relation to vaccines. Furthermore, the reasons offered for not performing meta-analyses were invalid. It is difficult to explain unless one assumes that the WHO did not want to run a risk of receiving a systematic review that suggested that the DTP vaccine increases total mortality.
WHO’s experts that advised against using meta-analysis wrote, after having seen the WHO report, that the data suggested that both the BCG vaccine and the measles vaccine reduce all-cause mortality. However, when I did meta-analyses of the randomized trials, I did not find significant reductions in mortality. Therefore, the experts could not conclude that both vaccines reduce total mortality without including also the non-randomized studies in their deliberations. In contrast, for the DTP vaccine they dismissed the non-randomized studies. This is inconsistent and scientifically unacceptable, particularly considering that the results for the cohort studies for the BCG and the measles vaccines varied equally much as those for the DTP vaccine.
The most important principle in medical ethics is: First, do no harm. I believe that the DTP vaccine should not be used unless being one of the interventions in a large randomized trial.
Author Information:
Peter C. Gøtzsche, Professor, DrMedSci, MSc
Kløvervang 39, DK-2970 Hørsholm, Denmark
Phone: +45 53 64 20 66
E-mail: [email protected]
