Newer SSRIs are more dangerous because research hasn't been done. Many of the disablities that valproate is alleged to cause are not discovered until a child is 2 or 3 and some children are only recognised as being affected in their early teens.
There were always warnings on the PILs for valproate but they were about congenital malformations at birth and that is what the research is usually based on. So when you look at the stats for valproate that include learning disabilites the risks are sky high (some report up to 90% reduced IQ!) compared with the 2 or 3 percent major malformations.
The doctor in the Telegraph article probably comes up against people with questions about this every day and she probably has no answer to give them. Drugs companies aren't doing research and the Drugs Agency MHRA doesn't seem to want to put two and two together.
In addition to this we have a slight conflict of interest going on with the medical departments. The neurologists don't ever get to meet the offspring of their valproate clients, the obstetricians are trained to look at MCMs and obvious health problems in utero, and the poor old paediatricians end up dealing with the result of disjointed thinking.
A court case last year ended up with 250 families losing funding for their high court battle against the makers of valproate. There are hundreds more out there, probably thousands of children undiagnosed.
Most of this would be preventable if government manned up to these drugs by funding cases and suing them for every penny that these hundreds of families are now costing the taxpayer. Only a beefed up legal system will prevent these problems.
Oh dear that was a bit of a rant. But I think what the doctor is saying in the article is - until decent tests are made we really don't know what the risks are with these drugs. Valproate is particularly difficult because of the mechanism by which it is teratogenic however it is a very old drug and there was plenty of time for it to be tested thoroughly.