Epi in the News

1. "The cumulative incidence of cervical cancer was 47 cases per 100,000 persons among women who had been vaccinated and 94 cases per 100,000 persons among those who had not been vaccinated." This was then adjusted for age at follow-up and covariates (county of residence, and parental characteristics, including education, household income, mother’s country of birth, and maternal disease history). Resulting in: "After adjustment for all covariates, the incidence rate ratio was 0.12 (95% CI, 0.00 to 0.34) among women who had been vaccinated before the age of 17 years and 0.47 (95% CI, 0.27 to 0.75) among women who had been vaccinated at the age of 17 to 30 years."

2. No mention that the study only follows girls until the age of 31? For example, if vaccination just delays onset of cervical cancer then we could see more cases in vaccinated populations in the future

3. This study was using the endpoint out cervical cancer incidence in individuals vs. previous studies looking at prevention of infection with the virus. Few thoughts for future studies:
a) Continuing to follow the same women over a longer time period would be valuable
b) Showing rates per person-year/decade etc. instead of risks to account for time of enrolment into the study
c) Conducting a similar study in another country to understand if the findings (e.g., preferred age window for vaccination) hold there (or if vaccination should be younger due to potentially earlier exposure to HPV).

Welcome to feedback - not sure on 2 and 3!

1. The report of a 88% lower risk and and 53% lower risk in women vaccinated before the age of 17 and from 17-30 respectively correspond to a relative risk of 0.12 and 0.47.

2.I didn't go ahead to read the actual study on this but looking at the principles of the epidemiology which have to do with time, place and person, I would go ahead to say there is no mention of the time period of the study or more specifically, the time period or age the participants were followed up.

3. I had to read portions of the main study for this question. While previous studies had mostly looked at the prevention of HPV infections and subsequently the progression to warts and certain cancers, this focused on the prevention of cervical CAs. This added finding implies that even if a woman a persistent HPV infection gets vaccinated for HPV her risk of developing invasive cervical CA is still lowerthan a woman with persistent HPV infection who does not received the vaccine. Both women may still however develop warts etc.
With respect to outcome, another study proposal could be the effect of the vaccine in slowing down progression of confirmed CA at various stages

P.S- not too sure if the answer for 3 has to do with getting a new idea for a different study or actually trying to fix the shortcomings of the study under review

Question; Does this study establish association or causation?

My feeling is that they established association as this is an observational open population cohort study which showed that there was an association between quadrivalent HPV vaccination and reduced risk of invasive cervical cancer with a larger effect noted among women who were vaccinated before the age of 17. Secondly, they only followed women till they turned 31.

However, their findings were reportedly consistent with 'limited trial data and with data from a previous ecologic study". Also, in terms of Bradford Hill criteria; it does meet the criteria of temporality and biological plausibility given that we know cervical cancer is HPV associated and we have comprehensive evidence that HPV vaccination prevents high-grade cervical lesions. The coverage in the media has seemed to attribute causality (I'm aware that you can't always trust the media's take on science/health but there was a quote included from a cervical cancer screening expert that said that this study shows HPV vaccination prevents cervical cancer).

I would have thought that the findings are extremely promising and there most likely is a causal association, but that further studies will be needed to follow women till they are older to ensure that cervical cancer is not just postponed and definitively establish a causal relationship. Or in conjunction with the evidence base available thus far, is this enough to establish a causal association?

Cant make my mind up on this so would appreciate any input!