Hazards to appraise: selection, misclassification, or information bias. Temporal relationships are hard to evaluate, and the research is largely dependent on accurate record-keeping of others.
International Statistical Classification of Diseases
- 4 alternative risk windows: 0-179, 180-364, 365-729, and more than 729 days after vaccination
- stratified age groups (10-29, 30-44 years)
- stratified country baseline incidence of demyelinating disease
- 2 additional analyses to address "if becoming a case contraindicates or delays vaccination in a significant number of participants". A) excluded a 30-day prevaccination period, B) including only vaccinated cases
1. Based on individual and complete records of vaccination and outcome diagnoses ascertained prospectively and independently in well-defined geographical areas
2. Statistical power to exclude meaningful increases in risk.
3. High degree of generalizability of our results. (Scheller et al., 2015)
1. Information on ethnicity, socioeconomic status, lifestyle factors, and family history was not available
2. Case definition relied on ICD-10 codes registered in national patient registries.
3. Previous vaccine safety studies have described a so-called unmasking phenomenon.
4. Study used date of diagnosis instead of date of disease onset causing differential misclassification
5. If becoming a case contraindicates or delays vaccination in a significant number of participants, the self-controlled case-series results can be biased. Supplementary self-controlled case-series analyses that addressed this concern by either excluding prevaccination time or including only vaccinated cases supported the primary self-controlled case-series analysis.
6. The vaccinated group was younger on average than the unvaccinated group, and since the background incidence of demyelinating disease peaks later among the included age groups, the crude rates were higher in the unvaccinated groups compared with the vaccinated group.
7. Although the study found no significant associations, the precision of the estimates did not allow the exclusion of < 3-fold increases in the risk of MS, and optic neuritis, or < 2-fold increased risk of other demyelinating diseases.
Conflict Of Interest Disclosures
- Arnheim-Dahlstrom et al is funded by grants from Sanofi Pateur, Merck, and GSK
- Fink et al has travel funds given to him by Biogen Idec
- Hvidd et al is funded by grants from Novo Nordisk Foundation
The study's limitation regarding system onset verses documented diagnosis in the time frame (2 years) could be problematic as stated by Yehuda Shoenfeld, MD, FRCP, who describes autoimmune disorders displaying up to 8 years after vaccination (Vaccine Safety, 2011). The authors based the 2-year time frame on a subsequent study (HepB & MS), and a single MS study that showed that 50% of patients who received a correct MS diagnosis within 1 year of symptom onset. Neither of the referenced studies embody the full scope that this articles tries to address. A 2-year time-frame is inadequate when trying to pick up a minority sub-population that is responding atypically from HPV vaccination in 9 separate ICD codes, only one being an MS diagnosis.
Less Then Two Doses Administered
Even though the authors conducted an additional analysis in an attempt to counter bias when individuals delay or drop out of the 3-course HPV injection, that analysis would only be valid if this occurred in a significant number of individuals. Further, this analysis was inspired by 5 peer-reviewed case studies describing HPV vaccines as being associated with central nervous system diseases (Sutton et al., 2008; Chang et al., 2011; Wildemann et al., 2009; DiMario et al., 2010; Menge et al., 2012). The majority of those case studies you can easily read that the individuals only received one HPV injection prior to their central nervous system disease diagnosis. 2 of the 5 studies you can not be certain without purchase of the article, and reading through the full description of occurrence (I may do this later). The Scheller et al study would then exclude the case studied individuals that they were trying to identify in the first place.
HPV Vaccine Science
Chang et al. (2011). Demyelinating disease and polyvalent human papilloma virus vaccination. J Neuro Neurosurg Psychiatry. Retrieved from http://jnnp.bmj.com/content/82/11/1296.extract
DiMario et al., (2010). A 16-year old girl with bilateral visual loss and left hemiparesis following an immunization against human papilloma virus. J Child Neurol. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/20189933
Marrie et al. (2005). Changes in the ascertainment of multiple sclerosis. Neurology. Retrieved from http://www.neurology.org/content/65/7/1066.short
Menge et al. (2012). Neuromyelitis optica following human papillomavirus vaccination. Neurology. Retrieved from http://www.neurology.org/content/early/2012/06/20/WNL.0b013e31825fdead
Scheller et al. (2015). Quadrivalent HPV vaccination and risk of multiple sclerosis and other demyelinating diseases of the central nervous system. JAMA. Retrieved from http://jama.jamanetwork.com/article.aspx?articleID=2088853
Sutton et al. (2008). CNS demyelination and quadrivalent HPV vaccination. Multiple Sclerosis Journal. Retrieved from http://msj.sagepub.com/content/15/1/116.abstract
Wildemann et al., (2009). Acute disseminated encephalomyelitis following vaccination against human papilloma virus. Neurology. Retrieved from http://www.neurology.org/content/72/24/2132.extract