How Good are Government Data on Vaccinations?

  • 09/23/2021

How does a vaccination rate of 98% fit together with almost 50,000 cases of measles? Both figures come from Mongolia. Since such a high number of cases is hard to explain with almost 100% vaccination, this data is evidence in support of the theory of poor data quality for vaccination lrates. We from dwh are conducting research, commissioned by the Ministry of Health, on how precise figures can be found despite incomplete data. The goal is to support decision makers to make better decisions in the future.

Contradictory data

The World Health Organization (WHO) is collecting data on vaccination rates and measles cases worldwide. In the following scatter-plot each dot represents one country. The size corresponds to its population. The nation’s, respectivey their dots, are sorted by their measles cases per 1 million inhabitants (x-axis) and vaccination rates (y-axis). This means that the dot of a country with no reported case will be on the very left side of the graphics, while a higher number of cases per one million inhabitants will lead to a position more to the right of the figure. The altitude of a dot is dictated by the country’s vaccination rate. Nations with low vaccination rates are towards the bottom of the figure (e.g. South Sudan with 22%), while those found on the very top of it are having a vaccination rate of 99% (e.g. China and Thailand). Or better: shoud have.

 

Hover the mouse over a dot to get more information. Use the magnifier (bottom left) to zoom in, the cursor to navigate when zoomed in and the house to return to the full-size view.

Mongolia’s position within the graph’s coordinate system (in the top right corner) hints at a contradiction: since the vaccination against measles is a very effective one, it is extremely unlikely that a vaccination rate of 98% would allow 50,000 cases of measles. One could argue that Mongolia is a developing country and far away from Europe. But even at the heart of Europe one can find countries with doubtful data. And that is not only Bosnia and Herzegovina, but also EU-member Romania, as our graphic shows.

Of interest is the vertical band of nations on the very left edge. These are almost all archipelagic states and/or countries with extremely small populations. Apparently their isolated location is offering a certrain degree of protection from measles epidemics. But this protection from isoliation is only an illusion, as the latest events show. The currently ongoing (Dezember 2019) measles-epidemy in Samoa already killed more than 60 people. You can find Samoa in our graph (data from 2013-2017) on the very left, second from bottom. With more than 4300 reported cases (and a population of 200,000) it would now have a rate of 21,500 cases per 1m and hence be in the right third of the graph.

With sound data towards a measles free world

Measles are an extremely contagious and potentially lethal disease. Subsequently the goal is to eradicate this disease. The World Health Organisation (WHO) is keeping track of the progress towards the disease’s extinction. It collects government data on vaccination rates as well as cases of the disease. Since every case of measles is a reportable event this data can be expected to be fairly precise. Vaccination rates are a different story. Hardly any country is maintaining a compulsory record of vaccinations and hence these figures have to be estimated or extrapolated in a complex manner. The countrie’s governments are reporting the vaccination rates for 1-2 year old children to the WHO. The WHO aims at a 95% vaccination rate for this cohort, as this will leads to a protection of the whole population.

Austria, highlighted orange in the graphics, does also not posses a compulsory vaccination record. To ensure a high quality for figure of vaccination rates despite this lacking, dwh is conduction research, commissioned by the Ministry of Health and in cooperation with the Vienna University of Technology and Dexhelpp. The research results have been published by the Ministry of Health.

Data Sources:

Data used is from the period 2013-2017, Source for vaccination rates is the WHO, it represents the share of 1-year-olds that have recieved at least one dose of MCV (Measles Containing Vaccine). Data on recorded cases of measles is also taken from the WHO, population data was taken from the worldbank.