In the previous post, it was noted that Busby's claims about a deviating sex ratio in Fallujah (first article here, second article here) may not be such a significant finding as he makes it sound, and that Busby is well aware of it but doesn't change his approach about it. There are weaknesses in the study, both methodological and due to the difficult circumstances in performing the study. So, in lack of other data, the results of the study may be of interest, and if properly designed the survey may give much better results than other kinds of surveys. But with the weaknesses in mind it would be reasonable expect a more humble approach from Busby and co-authors about the conclusions, if they are serious about it, that is.
One of Busby's most significant findings, according to himself, is the deviating sex ratio for the children born in the years after 2004, the year of the battle of Fallujah. The first study shows a decrease in the number of boys with respect to the number of girls, 18% below the normal level (860 boys to 1000 girls instead of the expected 1055 boys to 1000 girls). According to Busby this must be due to mutagenic stress induced by radioactivity from uranium. To support this theory he cites studies about lower sex ratios when the parents have been exposed to uranium in mines, medical radition treatment, and the Hiroshima bomb. So if we ignore the weaknesses of the study, we may agree with Busby that an 18% reduction in the number of boys born is interesting.
The problem is that he consistently ignores all other possibilities. The wikipedia page on human sex ratio gives a number of environmental and sociological reasons for deviations in the sex ratio. Busby does not mention a single one of them. Considering the heavy fighting in the city, there may also be further reasons for deviations, including stress and the simple fact that maybe people are not putting priority on making babies when their homes and a good fraction of the city (and the country) have been smashed into rubbles. The issue about if uranium based weapons were used at all in Fallujah is an open question, there are opposing views on this issue (it has surely been used in other parts of Iraq). If we assume that uranium based weapons were used, then we would expect Busby to at least mention the known chemical toxicity of uranium. Instead he puts all emphasis on the radioactivity from uranium, his theories about radioactivity is the only thing that matters for this self-proclaimed international expert on radiation.
Considering how many attempts Busby has made with epidemiological studies (and failed badly with some of them) it is quite remarkable that he still has not learnt to be cautious with the most important parameter: low statistics. Furthermore, with all the possible reasons that he excludes as potential causes, he never asks the question: Is a deviation in sex ratio always due to mutagenic stress from radioactivity? It is always due to mutagenic stress at all? Let's find out.
As noted in an earlier blog post, Busby is quite upset with the Swedish Health Authorities (Socialstyrelsen) for not letting him use their cancer statistics data base, apart from the data that are publically available and that he already mistreated last year. I have some good news for him, he can play with another data base, the one from Statistics Sweden (Statistiska Centralbyrån, SCB), which has a lot of interesting data on the Swedish population. Let us use this data base in order to check the sex ratio for a few cases. Let us start with checking the sex ratio for the entire population in Sweden, i.e. the number of born boys every year divided with the number of born girls every year. As in Busby's article we normalize to 1000 born girls and expect the sex ratio to be around 1055.
We see that the sex ratio is indeed very close to 1055, the average rate is 1058. And it fluctuates very little, it is always well within the span 1040-1070, with minor statistical deviations. But we have lots of statistics when we use the number of born children in entire Sweden. So let us look at the same situation for a medium sized city in Sweden, for instance Avesta, the city where I was born. We use a blue line for the sex ratio for each year, and a red line for the 5-year average:
Interesting fluctuations indeed! Busby has made a lot of fuss about the level 860 boys to 1000 girls. This in a total cohort of 4843 persons. The population of Avesta is more than 21 000, so we should have more than enough statistics in order to make a fair comparison and even err on the side of caution. We see that the sex ratio (blue line) fluctuates year by year around the expected value of 1055 boys to 1000 girls, though in some years the sex ratio is down to 800 boys to 1000 girls. But in the Fallujah study the data were shown as cohorts of 5-year averages. The red line shows the data for 5-year averages for Avesta. This curve does not fluctuate so drastically as the blue line, the extreme values cancel out. In spite of this we see a drastic decrease in the sex ratio for the last five years, going down to slightly less than 900. This is not as low as the value 860 in Fallujah. But it is based on better statistics, and from a trustworthy source that probably has the numbers correct down to each individual child. And to my knowledge Avesta has not been bombarded with uranium based weapons recently.
Now, can we find any city in Sweden where the 5-year average of the sex ratio at some time in the period 1972-2010 is lower than 860? In order to be fair we should set a constraint that the city should not be too small. The population pyramid in Sweden is very different from the one in Iraq, so in order to have enough children born for a fair comparison we set, arbitrarily, that we want to find a city with a population of at least 10 000, where the 5-year average of the sex ratio at some time has been below 860. Well, look:
The figure is a bit messy, but the main point is to look at the extreme values for each city. We find eight cities that fulfill the requirement of a 5-year average sex ratio that at some time is below 860. The cities are: Trosa (pop 11 492), Åtvidaberg (11 474), Mörbylånga (14 152), Burlöv (16 825), Strömstad (11 965), Filipstad (10 506), Nora (10 462) and Hedemora (15 141). The data from Fallujah are shown as a broad grey line.
Wait a minute, you may say. These cities are probably from the same region, and share some common environmental effect. Hardly, the map below show their locations in Sweden. Furthermore, the fluctuations for the different cities do not agree in time with each other. It is therefore very unlikely to find a common cause. Except low statistics, just as in the sample from Fallujah. If we include Swedish cities with populations below 10 000 then we will find more than 30 with sex ratios below 860 in the 5-year average, some of them well below 750. The reason is, again, low statistics.
But what about the drastic decrease in the number of born children in Fallujah, the number of boys born went down with 50%? This must be due to an environmental effect, right? Well, not necessarily. There can be many different reasons for why there is a decrease in the number of born children, not the least after an intense battle occurring in the city. But let's take a look at the 8 Swedish cities again, now we look at the number of born boys:
Once again we see that the dramatic variations in Fallujah are not extreme when comparing with some of the Swedish cities. Each Swedish city has its own behaviour, mostly depending on local variations in the population. But most of them share a common drop from a peak value around the year 1992. For Hedemora the number of born boys is reduced to almost 50% over a ten year period, with 35% decrease as the most dramatic drop over a 5-year period. But why is there such a decrease in most of the studied cities? Well, let's look at the number of born children for all of Sweden:
We see that there was a peak around 1992 followed by a quite drastic decrease, in 1997 the 5-year average number of born boys was 26% lower than in 1992. Is it due to the bad economy of Sweden at the time? Or was it less "popular" to have children for a couple of years? There are surely studies available regarding likely causes. Whatever the reason, we can easily exclude uranium based weapons. This does not disprove any hypothesis about uranium being the cause in Fallujah. Furthermore, I have only looked at the sex ratio while the article deals with a number of health effects, but whoever argues for uranium being the cause has a lot to explain before talking about significant findings the way Chris Busby does.
A few conclusions:
- All these cities have drastic variations in the sex ratio, even for the 5-year averages, and reach values lower than 860 boys to 1000 girls.
- For several of the cities the 5-year averages varies dramatically, similar to in Fallujah.
- The cities are distributed in different parts of Sweden with different geographical/environmental conditions.
- The periods of low sex ratio for the 8 cities occur at different times, no common cause can be seen.
- None of the 8 cities have suffered from war during the last 200 years, and during the last 40 years Sweden has been among the top ranked countries in the world when it comes to health status of the population.
- The variations are as large, or larger, than in Fallujah, based on much more reliable data, and equal or better statistics.
- Chris Busby should give up all attempts of epidemiology. This is not the first time he fails in this discipline, he just can't do it right.
- We do not learn anything about the causes of the health effects in Fallujah by listening to self-proclaimed experts like Chris Busby.
There are indeed more things to say about Busby's studies on Fallujah. When time permits they will be brought up on this blog.
Mattias Lantz - member of the independent network Nuclear Power Yes Please
Related blog posts: