Childhood vaccination rates have plateaued or fallen in most EU countries since the start of the COVID-19 pandemic

Is the study of good quality? Are the conclusions backed up by solid data?

"The study is of good quality, but its objective is debatable. It assesses trends, but does not assign specific values to those trends; instead, it considers whether they achieve statistical significance, which may be of little relevance, although the authors themselves explain this in the discussion. For example, a country with 70 per cent coverage that has reached 75 per cent in the latest observed trend will look good on paper, even if the reality is not exactly favourable. In other words, the study is well-conducted for its intended purpose, but I believe the assessment should have taken into account not only the trends but also the absolute figures for coverage.”

How does this study fit in with the existing evidence?

“There are few studies that assess such long-term trends; the end of the study coincides with the pandemic and the post-pandemic period, periods in which, in general, a slight decline in coverage has been observed, partly due to the prioritisation of COVID-19 vaccination and, subsequently in some cases, a possible increase in vaccine hesitancy.”

Have the authors taken confounding factors into account? Are there any significant limitations that need to be considered?

“There are several confounding factors that have not been taken into account, and only one of them is mentioned in the discussion. The period analysed—45 years—is very long; most of the vaccines assessed have been added to the vaccination schedule during this time. This means that, initially, when a vaccine is introduced into the schedule, a gradual increase in vaccination coverage is usually recorded (in most cases), so some of the initial increases in coverage may be due to the introduction of the vaccine into the schedule – something that is not ruled out; where this is the case, if there is subsequently a decline (even a small one) in coverage, it may give a negative impression that is not actually accurate when based on absolute coverage figures (rather than trends).

Furthermore, when analysing coverage figures spanning 45 years, these estimates may have changed due to alterations in the calculation method and may not actually reflect a real variation. For example, decades ago it was not uncommon for vaccination coverage to be calculated by dividing the total number of doses distributed by the number of children in the population; vaccines discarded due to expiry or any other possible wastage were not taken into account. Nowadays, improvements in registration systems ensure that coverage figures are accurate, as coverage is estimated by measuring the number of children actually vaccinated relative to the total population; in other words, improvements in coverage estimation methods work against the data and may lead to estimates of lower coverage than in other periods when calculations were carried out differently (lower but more accurate). This point is not discussed at any stage in the article.

Another potential source of bias is possible changes to the vaccination schedule. In this case, this is only mentioned in relation to measles and rubella, but is not taken into account in the study. However, the study measures vaccination coverage against a third dose of vaccines such as DTP, polio, etc., and in at least three European countries (Spain, France and Belgium) the vaccination schedule has changed, with that third dose now being administered at 11 months rather than at six months. It is well known that vaccination coverage is lower among older infants, but this fact is neither mentioned nor taken into account in the study.

In the case of Spain, the differing trends for the DTP and polio vaccines (Table 3) are striking, given that both vaccines have been administered in combination (in the same injection) for decades; therefore, any difference is either not real or not significant.”

What are the implications for the real world?

“The trends are interesting, but they must always be assessed alongside the actual coverage for each of these vaccines, rather than just the trend itself. The most recent downward trends may be due to various reasons, some of which may have no real impact; however, in any case, this serves as a wake-up call to continue monitoring both coverage levels and trends—something that those responsible for vaccination programmes routinely do.

[In Spain] the study shows, regardless of the trends, good vaccination coverage, although for the figures of 94 and 92 per cent, it would be desirable to achieve rates above 95 per cent.”