In 2002 a study was published using data from the DMSS, which looked at the date in a number of different ways including at the rates of diagnoses both before and after being vaccinated across the entire military during the years 1998 to 2000.  It concluded:

   ‘Results of surveillance efforts to date suggest that Anthrax Vaccine Adsorbed has few if any significant adverse health effects'[20]

The word ‘significant' in this context is defined as there being a less than 5% probability that observed differences in the rates of diagnoses between the vaccinated and unvaccinated happened by chance.  Such confidence intervals as were provided in the study, of which there were very few, were wide enough to indicate that there was insufficient power in the data to draw any meaningful conclusions as to whether there were significant, in the plain English definition of the word, adverse health effects.  If for instance 0.1% of the vaccinated population suffered a particular diagnosis, the confidence intervals in the observed data would have to be sufficiently narrow as to make such an observation less than 5% likely by chance, and that was not the case in some of the data provided.  Given the number of diagnoses that could potentially be triggered by a vaccine, the data as presented would not rule out 1 - 2% of vaccine recipients having a serious adverse reaction.

There were also other weaknesses in the study, as it noted:

   ‘However, there were some unaccounted for (residual) differences between the cohorts that were undoubtedly confounding. For example, in general, servicemembers who are immunized are healthier than those who are not (e.g. due to medical exemptions); servicemembers deploy or are assigned overseas are healthier than their counterparts who are ineligible (often for medical reasons) for such assignments; and medical encounters in treatment facilities on permanent military installations are more completely ascertained than those onboard ships or in deployed clinics and hospitals. Finally, there are regional, local, and assignment-related differences in endemic disease and injury hazards, in access to and utilization practices regarding health care resources (e.g. inpatient versus outpatient care for similar conditions), and in the natures, durations, and intensities of military and off-duty activities.’

It went on to say:

   ‘Assessments of causality require information more detailed than that routinely collected for medical surveillance purposes and analysis methods that consider, for example, biological plausibility, specific temporal relationships, medical histories, comorbidities, behavioral and other illness and injury risk factors, concurrent vaccinations, and variations in health care access, usage, and reporting.’

Also in 2002, the CDC established the Vaccine Analytic Unit (VAU) in collaboration with the DoD, which had access to DMSS with the intention that it would conduct detailed research.  The following year the IOM gave its final recommendations on research, which recommended data mining of the DMSS, and said of the VAU:

   The committee is concerned that insufficient priority is being given to the hypotheses testing work that is to be done through the Analytical Unit being established at AMSA.  The proposed funding level reflected in the version of the CDC-DOD Memorandum of Understanding provided to the committee (a total of $950000 over 3 years) does not appear adequate to support the kind of analysis of DMSS data that will be necessary to investigate hypotheses that have already been generated by AMSA’s routine screening of DMSS data and by the work of the IOM committee that reviewed the saferty and efficacy of the anthrax vaccine for DoD (IOM, 2002).[21]

AMSA stands for Army Medical Surveillance Activity.

By 2006 the DoD had got as far as creating a list of conditions which needed to be investigated:

   Peripheral neuropathy, Paresthesia, Mood cognition disorders, Atrial fibrillation, Stevens–Johnson syndrome/Toxic epidermal necrolysis/, Pemphigus, Unintentional injury, Guillain–Barre Syndrome, Optic neuritis, Diabetes mellitus, Syndromic illness*, Arthritis/arthralgia. [22]

Some studies were performed on rates of optic neuritis and diabetes mellitus as individual diagnoses in the vaccinated.  Other studies looked at the rates of all diagnoses between vaccinated and unvaccinated groups.  However the causes of chronic illness are not as simple as a single exposure leading to a single diagnosis, and nothing was ever done to investigate in detail the broad spectrum of illnesses being reported.

As late as 2008 Congress still felt it necessary to make the recommendation:

   ‘Evaluate the association of anthrax vaccine adsorbed (AVA) with chronic symptoms, Gulf War illness, and diagnosed diseases in personnel known to have received the anthrax vaccine during the Gulf War. These health outcomes should also be assessed at least five years after vaccination in deployment and era subgroups of personnel in the Millenium Cohort study as well as other groups vaccinated in association with the military's anthrax vaccine immunization program and federal anthrax vaccine trials.' [3]

The DoD lists a number of studies in the ‘Detailed Review of Anthrax Vaccine Adsorbed' published in 2015 [23] the most recent publication of its kind, which it claims support the safety of the vaccine.

​

Click on the link to see the full document.
 

The two largest and most detailed studies ever performed on anthrax vaccine are the Sulsky studies [24,25] which were published in 2011 and 2012.  The first looked at rates of all disabilities granted by the Army and found that 3.8% of those vaccinated had been granted disability against 6.3% of the unvaccinated, which was attributed:

   ‘It is likely that deployed personnel have greater levels of baseline physical fitness than nondeployed persons, and may be less injury prone or better able to recover from injuries [25]. We suspect this "Healthy Warrior Effect"[26] explains the lower risks of disability experienced by vaccinated compared to unvaccinated personnel for those soldiers who enlisted prior to 2000. The available data could not be used to limit the study population to those eligible for vaccination, which would have been the most valid subset of the population for these analyses.’[24]

There was an attempt to address the problem using Hostile Fire Pay (HFP) as a marker for deployment but it could not isolate the variables of interest, in part because the study population included members who had deployed and received HFP before anthrax vaccine started to be administered in 1998, while still further distortion was caused by servicemen who deployed under a vaccine waiver due to a preexisting medical condition.  Also, the entire precept of the study was that the Army would be equally likely to grant disability to those suffering anthrax vaccine reactions as for other forms of disability, which is inconsistent with many experiences described on veteran web sites.  The conclusions went on to say:

   ‘While it is possible that AVA may be causally associated with some disability separations from the army, this must be a rare situation, or such events would have been detected by the comprehensive analysis reported here.’[24]

There was no attempt to put a figure on ‘rare' despite it being the only number that mattered.

The same lead authoress followed up the next year with a study which looked at rates of both VA disability and Army disability, and found that 13.4% of those vaccinated had received a form of disability rating against 19.0% of the unvaccinated; [25] by which figures even if an extra one out of every eighteen anthrax vaccine recipients had suffered lifelong disability the resultant rate of disability in the vaccinated group would still have been lower than the unvaccinated group.

Any schoolchild who has completed the opening modules of statistics would readily be able to see every one of the studies promoted by the DoD to be worthless, mostly for similar reasons to the above.  Looking at rates of a single diagnosis generates a minimum of signal, while looking at rates of all disabilities including physical injuries and other disabilities clearly unrelated to the vaccine generates a maximum of background noise.  Each study was self evidently designed to reach a negative conclusion which could be promulgated as evidence of vaccine safety, and possibly appear as such to those who lack a schoolchild's understanding of statistics.  They form the entirety of the evidence presented to demonstrate the purported safety of the vaccine.

A genuine attempt to investigate adverse reactions to the vaccine would have started with looking at the patients presenting at military clinics with illnesses following the vaccine, from whom data could have been gathered for further study.  An epidemiological study is only as valid as its understanding of what to look for, which was one of the purposes the VHC were established to achieve but military medicine never used them.

As things stand, there is nothing to preclude the figure of 1 - 2% of recipients having a serious adverse reaction, as opposed to a serious adverse event, the difference being whether the event is indeed caused by the vaccine.  In spite of the long history of doubt about the safety of the vaccine, the only information provided to either providers or recipients are the package insert for Biothrax (anthrax vaccine)[26] and propaganda from the Defense Health Agency [27], both of which are extremely light on negative information about the vaccine as would be expected from such conflicted sources.

The definition of ‘safety' used by FDA is:

   ‘The relative freedom from harmful effect to persons affected, directly or indirectly, by a product when prudently administered, taking into consideration the character of the product in relation to the condition of the recipient at the time.'[12]

What the freedom from harm is ‘relative' to is not specified.  However clearly in this context the risk of the vaccine must be compared to the risk posed by anthrax as a weapon, which is a calculation that the Pentagon has never even made an attempt at.  One of the harder aspects of being a military commander is assessing the risk of casualties that is acceptable for a given operation.  For the Pentagon to say any such figures out loud would be political plutonium, but no such thought process can have taken place at all for policy to have been formed as it has.  Some risk from the vaccine has to be accepted when fighting an enemy which may use biological weapons, but a 1% risk of a serious adverse reaction would clearly be orders of magnitude too high even for the most vulnerable of those deployed at the height of the operations in 1991 and 2003.  A rate of serious reactions in 0.000001% of vaccine recipients would be significant to whether the vaccine should be administered to the crew of a warship anchored off the coast of Timor, as has happened since the mandate extended to the crews of the warships of Seventh Fleet stationed in Japan whose regular deployments remained in south east Asia.

The tools were available for a proper investigation, as noted by the IOM in 2002:

‘Only in the context of an extremely organized health care system, such as that administered by DoD, can exposures be truly quantified and outcomes more completely reported.'[12]

Despite the Pentagon having the resources, they were left unused.