Background
Low birth weight is associated with mental health problems in child- and adulthood. However, it remains unclear if birth weight influences general versus specific mental health problems, and if such associations remain after controlling unmeasured confounds shared within sibling pairs.
Aims
1) Examine whether birth weight predicts general versus specific mental health problems.
2) Examine whether these associations remain after controlling unmeasured confounds shared within sibling pairs.
Method
We identified the oldest two siblings (differing by five years or less) in each family in Sweden born between 1958 and 1999 (N = 1,460,699 sibling pairs). Birth weight was retrieved from the Medical Birth Register. Psychiatric diagnoses (depression, anxiety, PTSD, OCD, eating disorder, alcohol abuse, drug use, autism, ADHD, bipolar, schizophrenia, and schizoaffective disorder) were retrieved from the National Patient Register (assigned by the attending physician in accord with ICD-8, 9, or 10).
Results
A four-factor EFA fit the data well (RMSEA = .002; CFI = .992), which we rotated toward one general and four specific factors (internalizing, externalizing, neurodevelopmental, and psychotic problems). We then regressed the EFA on birth weight using regular and fixed effects regression.
Birth weight was significantly associated with general (β = -.02, p < .001), externalizing (β = -.04, p < .001), and neurodevelopmental problems (β = -.03, p < .001). These associations remained significant after controlling unmeasured confounds shared within sibling pairs.
Discussion
In this population-based, within-sibling comparison study, low birth weight predicted general, externalizing, and neurodevelopmental problems even after controlling unmeasured confounds shared within sibling pairs. Restricted fetal growth indicates that insufficient energy supply for organ development contributes to a wide range of mental health problems in adulthood.
