Cochrane Reports: Vaccines for measles, mumps and rubella in children (Review)

The Cochrane collaboration has put out an updated report on the MMR vaccine.

Online discussions (for example, here) are (a) focused on the abstract only so far and (b) centered on two lines from the abstract:

Exposure to the MMR vaccine was unlikely to be associated with autism, asthma, leukaemia, hay fever, type 1 diabetes, gait disturbance, Crohn’s disease, demyelinating diseases, bacterial or viral infections

and

The design and reporting of safety outcomes in MMR vaccine studies, both pre- and post-marketing, are largely inadequate

Here is the full abstract from the paper:

Background
Mumps, measles and rubella (MMR) are serious diseases that can lead to potentially fatal illness, disability and death. However, public
debate over the safety of the trivalent MMR vaccine and the resultant drop in vaccination coverage in several countries persists, despite its almost universal use and accepted effectiveness.

Objectives
To assess the effectiveness and adverse effects associated with the MMR vaccine in children up to 15 years of age.

Search methods
For this update we searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2011, Issue 2),
which includes the Cochrane Acute Respiratory Infections Group’s Specialised Register, PubMed (July 2004 to May week 2, 2011)
and Embase.com (July 2004 to May 2011).

Selection criteria
We used comparative prospective or retrospective trials assessing the effects of the MMR vaccine compared to placebo, do nothing or a combination of measles, mumps and rubella antigens on healthy individuals up to 15 years of age.

Data collection and analysis
Two review authors independently extracted data and assessed methodological quality of the included studies. One review author
arbitrated in case of disagreement.

Main results
We included five randomised controlled trials (RCTs), one controlled clinical trial (CCT), 27 cohort studies, 17 case-control studies, five time-series trials, one case cross-over trial, two ecological studies, six self controlled case series studies involving in all about 14,700,000 children and assessing effectiveness and safety of MMR vaccine. Based on the available evidence, one MMR vaccine dose is at least 95% effective in preventing clinical measles and 92% effective in preventing secondary cases among household contacts.
Effectiveness of at least one dose of MMR in preventing clinical mumps in children is estimated to be between 69% and 81% for the
vaccine prepared with Jeryl Lynn mumps strain and between 70% and 75% for the vaccine containing the Urabe strain. Vaccination
with MMR containing the Urabe strain has demonstrated to be 73% effective in preventing secondary mumps cases. Effectiveness of
Jeryl Lynn containing MMR in preventing laboratory-confirmed mumps cases in children and adolescents was estimated to be between
64% to 66% for one dose and 83% to 88% for two vaccine doses. We did not identify any studies assessing the effectiveness of MMR in preventing rubella.

The highest risk of association with aseptic meningitis was observed within the third week after immunisation with Urabe-containing
MMR (risk ratio (RR) 14.28; 95% confidence interval (CI) from 7.93 to 25.71) and within the third (RR 22.5; 95% CI 11.8 to
42.9) or fifth (RR 15.6; 95% CI 10.3 to 24.2) weeks after immunisation with the vaccine prepared with the Leningrad-Zagreb strain.
A significant risk of association with febrile seizures and MMR exposure during the two previous weeks (RR 1.10; 95% CI 1.05 to
1.15) was assessed in one large person-time cohort study involving 537,171 children aged between three months and five year of age. Increased risk of febrile seizure has also been observed in children aged between 12 to 23 months (relative incidence (RI) 4.09; 95% CI 3.1 to 5.33) and children aged 12 to 35 months (RI 5.68; 95% CI 2.31 to 13.97) within six to 11 days after exposure to MMR vaccine. An increased risk of thrombocytopenic purpura within six weeks after MMR immunisation in children aged 12 to 23 months was assessed in one case-control study (RR 6.3; 95% CI 1.3 to 30.1) and in one small self controlled case series (incidence rate ratio (IRR) 5.38; 95% CI 2.72 to 10.62). Increased risk of thrombocytopenic purpura within six weeks after MMR exposure was also assessed in one other case-control study involving 2311 children and adolescents between one month and 18 years (odds ratio (OR) 2.4; 95% CI 1.2 to 4.7). Exposure to the MMR vaccine was unlikely to be associated with autism, asthma, leukaemia, hay fever, type 1 diabetes, gait disturbance, Crohn’s disease, demyelinating diseases, bacterial or viral infections.

Authors’ conclusions
The design and reporting of safety outcomes in MMR vaccine studies, both pre- and post-marketing, are largely inadequate. The
evidence of adverse events following immunisation with the MMR vaccine cannot be separated from its role in preventing the target
diseases.

The body of the paper summarizes the accepted papers and judges papers on criteria such as potential for bias and generalizability.

While much discussion online is given to the short abstract available in pubmed, consider the last three sections of the paper.

First, “Agreements and disagreements with other studies or reviews”

Agreements and disagreements with other studies or reviews
Currently, this is the only review covering both effectiveness and safety issues ofMMR vaccines. In agreement with results from other studies and reviews a significant association between autism and MMR exposurewas not found.The study of Wakefield (Wakefield 1998), linking MMR vaccination with autism, has been recently fully retracted (The Editors of The Lancet 2010) as Dr.Wakefield has been found guilty of ethical,medical and scientific misconduct in the publication of the paper; many other authors have moreover demonstrated that his data were fraudulent (Flaherty 2011). A formal retraction of the interpretation that there was a causal link between MMR vaccine and autism has already been issued in year 2004 by 10 out of the 12 original co-authors (Murch 2004). At that time (1998) an excessive and unjustified media coverage of this small study had disastrous consequences (Flaherty 2011; Hilton 2007; Offit 2003; Smith 2008), such as distrust of public health vaccination programmes, suspicion about vaccine safety, with a consequential significant decrease in MMR-vaccine coverage and re-emergence of measles in the UK.

Let’s look at the “Authors Conclusions” section at the end the paper:

AUTHORS’ CONCLUSIONS

This is in two sections:

first:

Implications for practice
Existing evidence on the safety and effectiveness of MMR vaccine supports current policies of mass immunisation aimed at global measles eradication and in order to reduce morbidity and mortality associated with mumps and rubella.

Repeat for emphasis: Existing evidence supports the current policies of mass immunisation.

This is followed by “implications for research”:

Implications for research
The design and reporting of safety outcomes in MMR vaccine studies, both pre and post-marketing, need to be improved and standardised definitions of adverse events should be adopted. More evidence assessing whether the protective effect of MMR could wane with the time since immunisation should be addressed.

At this point I’ll point out what I see as a false dichotomy. The discussion is not between “should vaccination continue” and “are the studies adequate” as some have framed it. Clearly Cochrane has called for both continued immunization *and* more research. This includes research into “assessing whether the protective effect of MMR could wane with the time since immunisation”, something which could point to the need for booster shots.

How about the MMR/autism question?

First off, let’s note that four of Mr. Wakefield’s papers were considered for the review. All four were rejected.

Wakefield 1998 {published data only}
Wakefield AJ, Murch SH, Anthony A, Linnell J, Casson DM, Malik M, et al.Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet 1998;351(9103):637–41.

Wakefield 1999a {published data only}
Wakefield AJ, Montgomery SM. Autism, viral infection and measles-mumps-rubella vaccination. Israel Medical
Association Journal 1999;1(3):183–7.

Wakefield 1999b {published data only}
Wakefield AJ. MMR vaccination and autism. Lancet 1999; 354(9182):949–50.

Wakefield 2000 {published data only}
Wakefield AJ, Montgomery SM. Measles, mumps, rubella vaccine: through a glass, darkly. Adverse Drug Reactions and
Toxicological Reviews 2000;19(4):265-83; discussion 284- 92.

Oddly, while the body of the Cochrane review notes that Wakefield 1998 has been retracted, the citation does not. The paper should either not be cited (as it has been effectively erased from the public record) or the citation should include the term “retracted”. The retraction is not the given reason for rejecting the study from the review. It was rejected due to being a case series. Wakefield 1999a and was rejected for lack of comparative data while Wakefield 1999b because it has no data.

As already quoted above, the authors of the Cochrane review *do* take notice of the GMC hearing and the Lancet retraction:

“The study of Wakefield (Wakefield 1998), linking MMR vaccination with autism, has been recently fully retracted (The Editors of The Lancet 2010) as Dr. Wakefield has been found guilty of ethical,medical and scientific misconduct in the publication of the paper; many other authors have moreover demonstrated that his data were fraudulent (Flaherty 2011).”

I note the statement that “many other authors have moreover demonstrated that his data were fraudulent ” First, the Cochrane Report is reporting that the data have been “demonstrated” to be fraudelent. Further, not “The BMJ” have claimed that the data were fraudulent, but “many other authors”.

In what I see as a strange move, the Hornig study (Lack of association between measles virus vaccine and autism with enteropathy: a case-control study.) was not considered by Cochrane with the given reason “Subjects affected by gastrointestinal disturbance”. Hornig et al. was the most on-target attempt to reproduce multiple critical aspects of Mr. Wakefield’s work.

What studies *were* included? Here’s the list, together with how Cochrane rated them for “risk of bias” and “generalisability”
Madsen 2002 moderate/high
Fombonne 2001 high/low
Uchiyama 2007 high/low
Smeeth 2004 moderate/medium
DeStefano 2004 moderate/medium
Mrozek-Budzyn 2010 moderate/medium
Fombonne 2006 moderate/medium
Honda 2005 moderate/medium
Makela 2002 moderate/medium
Taylor 1999 moderate/medium

To close, I’ll repeat the “implications for practice”:

Implications for practice
Existing evidence on the safety and effectiveness of MMR vaccine supports current policies of mass immunisation aimed at global measles eradication and in order to reduce morbidity and mortality associated with mumps and rubella.