H1N1 influenza vaccination during pregnancy not associated with problems in exposed offspring

In the full article the authors themselves state, " Although we observed an increased association between maternal 2009 pH1N1 influenza vaccination and pediatric asthma, the small magnitude of the association, which was no longer statistically significant after we accounted for multiplicity, along with evidence of a small correlation with our negative control outcome, implies that these results should be interpreted cautiously"

When I look at this paper I think of a couple reasons why I wouldn't put a lot of weight behind that hazard ratio.
1) To start with, it is a retrospective study so we can't say there are any any causal relationships.

2) A HR of 1.05 (1.02-1.09) can be interpreted as the criteria for "asthma" being met 1.02-1.09 times as often in the vaccination group as in the unvaccinated group. To me, that's pretty small to begin with. As we know, if 1.0 was included in that confidence interval it wouldn't have been considered significant.

3) Asthma wasn't "diagnosed" so much as labelled as I see it. "Diagnoses of asthma were ascertained from the Ontario Asthma dataset, which uses a validated algorithm (sensitivity 89%, specificity 72%) to identify cases of asthma from health administrative databases." That's a pretty low specificity. So a combined specificity that misses approximately 30% of cases and such a small increase in likelihood of asthma makes me question the validity of that likelihood right away. I didn't double check the original paper for the algorithm but I also wonder if the specificity is lower for children under the age of 5.

4) "Follow-up began on the date of birth and continued until the child either became ineligible for healthcare in Ontario (owing to emigration or death) or reached 5 years of age. However, for time-to-event outcomes (asthma, neoplasms, sensory disorders, mortality), the end of follow-up was the event date for those experiencing the outcome" Hence why we have a hazard ratio rather than a risk ratio - the kids weren't followed until the end. They could have been 'diagnosed' as early as 6 months and then not followed further as per the paper. Even a little more questionable for me.

5) The age of the children. Diagnosing asthma in children under 5 isn't necessarily easy. You can't really use spirometry. The Can. Ped. Soc. and the Can. Thoracic. Soc. have a guidelines that discusses the diagnostic challenges (https://www.cps.ca/en/documents/position/asthma-in-preschoolers).

6) Bias? Is there any population level bias that could result in the immunized population accessing healthcare for their children more often/easily or that they might trust the health system more easily? The authors didn't think so. "Although we believe that any misclassification of asthma status would likely have been non-differential according to exposure, vaccinated women may have been more likely to access healthcare for their infants, and this could have biased our results away from the null. Nevertheless, our sensitivity analyses designed to account for maternal propensity to access healthcare did not change any of the results."

6) More bias/confounding? Any geographic patterns? Were children labelled with asthma more likely to be living in areas with higher pollution, for example, and how would this compare to immunization in those areas. They did included rurality and public health unit in their standardization equation but I'm not sure how and if that would correlate with air quality. They also wouldn't have been able to account for household smoking - that wasn't the mother actively smoking during pregnancy (assuming someone even felt comfortable telling their healthcare provider if they were actually smoking)

Just some other things I think about
- There were statistically significant differences in vaccination rates by income. The lowest income group: vaccinated = 17.8%, unvacc = 24%. Highest income group: Vaccinated = 20.4%, unvacc = 14.2%. They standardized for this in the analysis. On it's own, it's food for thought on why there is a difference to begin with
- It was a monovalent pandemic vaccine, not a seasonal vaccine. I'm not sure I'd extrapolate any of the findings to other influenza vaccines.

I'm not sure the methods in this paper convince me that the children labelled as having asthma truly did, and that the severity was significant enough that it outweighed the risk of putting a pregnant mother, her baby/family/and contacts at risk for one of the worse influenzas we've seen in a while. Not my decision - it's hers - but I'd feel uneasy making a blanket statement that there is an increased risk of asthma.