The JayMan blogger is skeptical, and reports:
A new meta-analysis has dropped that sought to answer the question of whether childhood maltreatment leads to later mental health problems. This study is different from typical in that it tried to address the ever-pervasive problem of genetic confounding. ...That is, a child could inherit a gene for mental illness, become a huge behavior and discipline problem for the parents and caregivers, get mistreated, and grow up with a mental illness. But the illness was not caused by the mistreatment, but by the gene.
They were also diligent to check for publication bias and all those other things that are known to screw-up meta analyses. ...
I read through the paper and the supplements and it seems that there is one more confound that they didn’t address: child-to-parent effects. Troubled adults are quite often difficult, troubled children. Such children are more likely to elicit harsher treatment from caregivers. ...
Hence, the small residual seemingly causal association between childhood maltreatment and eventual adult mental health could be due to this. Now, the authors did also look at studies looking at institutional maltreatment (e.g., the infamous Romanian orphanages) and found a similar magnitude effect. Although child-to-caregiver effects are still a concern, the genetic confound also rears it head there, since there is a racial component to orphaned children there–more likely to be ethnic Romani, for example (a group with lower average IQ than ethnic Romanians).
For millennia kids grew up under conditions that we would consider harsh today. The harsh conditions did not appear to cause any mental illness.
JayMan is not suggesting that anyone mistreat kids. Just trying to understand the facts.