Smoking, pregnant women and their criminal offspring
Tuesday, November 16, 2010 I've been sitting on a new study (published today in the Journal of Epidemiology and Community Health) and I can reveal (cue drum roll) that "Mums who smoke heavily while pregnant run the risk of having kids who grow up to become repeat criminal offenders, research suggests".
I'm not making this up. According to the press release:
The findings showed that children whose mothers had smoked heavily during the pregnancy were the most likely to have a criminal record as an adult.
They had a 30% increased chance of having been arrested, and this applied to women just as much as it did to men.
The findings held true, even after taking account of a comprehensive range of family and social factors, such as mental ill health and deprivation, which are likely to influence behaviours, the research showed.
“While we cannot definitively conclude that maternal smoking during pregnancy (particularly heavy smoking) is a causal risk factor for adult criminal offending, the current findings do support a modest causal relationship,” conclude the authors.
I'm speechless, so I'll quote my source: "It must be true: it's peer reviewed."
See: Study claims heavy smoking during pregnancy can lead to a life of criminality (Liverpool Daily Post). Click HERE to view the paper in full.
Reader Comments (18)
Ha, ha, ha! This one's better than the Beano. Not even the Great British, stupid, sheeple will take this one seriously. I hope no Public money was involved.
Talk about scraping the bottom of the barrel.
FFS - *holding head in hands*
Sadly Frank, many of the stupid British sheeple will!
I heard about this on the news this morning and although only half awake, it seemed the study group was very small, around 3000, if I heard correctly.
I was also trying to equate this 'revelation' with the fact that a great many mothers smoked, and often heavily, during pregnancy in the 1950's and 60's, so how come most of us who were born then have not turned out to be criminals in any sense of the word (except nowadays for being a smoker or creeping a mile or 2 over a speed limit) never mind repeat offenders?
The research SUGGESTS. I suggest that researchers get proof before coming out with such ridulous statements.
I, genuinely, don't think so, Lyn. They are far from convinced about anything to do with ETS and most are aware the ban was a complete con, even the non smokers. The problem is 'what can we do about it' so they shrug shoulders and accept it. I refer to them as 'sheeple' because of the shrugging shoulders.
Even those that try have all avenues closed off to them. ATM, they're putting up with small transgressions like leaning on the door frame while smoking or lighting up in hospital grounds and shopping centres. In most cases they're now getting away with it so it will grow until one day it just falls apart.
I'm still tending to localism, though.
Fascinating piece of 'research' - and people shouldn't be so cynical !
It ties in rather neatly with similar findings of my own regarding mums who a) regularly feed their offspring Big Macs (and 'fries'), b) watch 'Hollyoaks', and c) collect Celine Dion CDs. There’s also a marked tendency NOT to listen to Radio 3, holiday in Cap Ferrat, or watch polo.
Of course, there COULD be something else going on here. More funding is urgently needed..................
BTW -
I think it highly irresponsible of Simon to publish that photograph above. It could encourage the more impressionable among us (eg children) to steal crow-bars - instead of saving up and buying one's own.
He's obviously a succesfull criminal or a beginner because he has a brand new crowbar.
Also he must be a non smoker because according to the antis he would not be fit enough to prise open a door.
The report ?
Elementary my dear Watson.
Complete fiction as well.
I would like to put to the researchers that crime in the UK has risen 10 fold since 1950 to 2009 as you can see from the quote below. Smoking reached its peak in 1950 in the UK when 66% of the adult population smoked and is now 22.5% So smoking has reduced by 66% and crime has increased 1,000%, so following a logical conclusion if we smoked more there would be less crime.
The key quote in the paper is: "CONCLUSION While we cannot definitively conclude that MSP (particularly heavy MSP) is a causal risk factor for adult criminal offending.."
"Between 1950 and 1992 recorded crime increased by 12 times from 461,000 to 5,592,000. By 1998/99 it had fallen back to 4,482,000, still nearly ten times the rate in 1950. The 2008/09 figure of 4,704,000 is still around ten times the 1950 figure."
http://www.civitas.org.uk/pdf/CrimeBriefingFeb2010.pdf
A brilliant example of a deliberate misrepresenting of correlation as causation. Or fraud, as some would call it.
Dave Atherton
To my reading, the key quote is this:
"Although most covariates had missing values for only a small number of subjects <10%), the cumulative effect of missing data was substantial. A complete case approach would include only 37.3% of the eligible participants, substantially decreasing the precision of the estimated effect of MSP. To reduce the number of omitted observations and obtain an indicator of parental ASB for the entire sample, multiple imputation was used."
So what they are saying is that they didn't have sufficient complete data records to enable the calculation of statistics that could possibly meet the 'p<0.05' significance test - so they replaced all of the missing data values with made-up ones.
They then proceded to perform a whole series of adjustments and multivariate weighting calculations on this made up data, and, not surprisingly, arrived at a conclusion that bears no relationship to the data with which they started.
An odds ratio of 1.31 (which is equivalent to a Relative Risk of about 1.2) is laughingly small anyway, but to claim that the Confidence Interval is as narrow as 1.06 to 1.62 is just hysterical. This CI is calculated only on the values that were stuffed into the 'logistic regression models', but don't forget that there had already been two rounds of data adjustments (making up missing values and confounder score weightings), so there has to be further statistical error to be incorporated into the final CI, and which would undoubtedly increase its width quite massively.
This is, sadly, an all too familiar and recurrent methodological failure in epidemiology - especially 'small-effects epidemiology' like this and most smoking-related studies. Too many mathematically illiterate people are let loose with complex statistical software tools that they really don't understand, use them inappropriately and then make great claims for the inevitably erroneous outcomes of their calculations.
This is not just a case of 'torturing data until it confesses', this data has been taken to Guantanamo Bay, been locked up for years, kept in dark and cramped conditions and water-boarded to breaking point. Now it seems like the researchers are expecting governments to hand over £millions in compensation - as if any government would do such a thing!
Hate Crime! Arrest these thugs! My beautiful, gorgeous, intelligent, hard working, talented children have been slandered because whoever paid for this study wants to get the wider public to hate their mother and foster mistrust of them as my children.
Pat
In the comments, the antismoker lobby are now busy offending the nonsmoking children of smoking mothers, who are not taking the accusation of criminality or the insults to their mothers at all well.
Smoking during pregnancy 'increases risk of children turning to crime'
http://www.dailymail.co.uk/health/article-1330127/Smoking-pregnancy-increases-risk-children-turning-crime.html?ITO=1490
Brian, great post. Note also that the authors admit they have not accounted for genetic influences or drink and drug taking (earlier in the paper giving their rarity as the reason). However, if repeat offending is also rare, this could have a very large effect. To be fair to the authors, their claims are fairly reasonable ie. not claiming much. They are cranked up for the "What we have added" statement and then probably cranked up again for the press release. After the journalists get their hands on it, it's "Send 3s 4d, we're going to a dance."
a 30% increase is not statistically significant
They don't take into account the fact that today there are no or very few deterrents re: crime (compared with in 1950), hence the rise in crime. As for saying that parents who smoke breed 'criminal kids' - well, if I was to believe that, I would have fallen out of a cuckoo clock - and I haven't! :)
@Brian
Thats a terrific review of the paper, thanks.
Sorry for the delay in replying Frank, been working too many hours to get time on the PC!
I agree that many of the sheeple are far from convinced regarding SHS, however I do believe they use it as a convenient excuse, otherwise there would be more backing the idea of getting smoking back into pubs and clubs, at least. Many also 'believe' it because it means they are not inconvenienced by the smell on their clothes and in their hair.
From the outset of the ban I have made a point of smoking in NHS car parks and anywhere else that is outside, as well as through the walkway in our town that runs alongside our Town Hall and which has notices stating that it has been designated as a No Smoking area. As far as I understand it there is no law against smoking in the open air (yet), so it would be impossible for them to make a case of it. So far I have not been challenged, but as I pay my taxes then I have a stake in public property and the right to smoke in the open air, regardless of their petty signs.