"Findings demonstrate a significant causal link between skipping breakfast and an increased risk of ADHD" pubmed.ncbi.nlm.nih.gov/38566068/ Has Mendelian Randomization gone too far? Is this a valid application of the method @MarcusMunafo 🤔
@Neuro_Skeptic @MarcusMunafo "Further examination reveals that breakfast was skipped because people completely forgot about it."
@Neuro_Skeptic @MarcusMunafo The “breakfast skipping” variants used in this study are for a very crude breakfast-related phenotype: cereal intake. UKB never assessed meal skipping or food timing. The variants haven’t been validated yet so the instrument is weak; findings should be cautiously interpreted.
@Neuro_Skeptic Well, first of all it's (more accurately) evidence of a causal link between *genetic liability* to skipping breakfast and whatever. Also, for complex phenotypes you really need negative control outcomes and (ideally) triangulation of evidence from other methods as well.
@Neuro_Skeptic @MarcusMunafo could it be that: breakfast sleeping is correlated with sleeping in is correlated with sleeping late is correlated with being on devices too much at night is correlated with adhd
@Neuro_Skeptic @MarcusMunafo Is this a database of children, since ADHD is a childhood neurodevelopmental disorder, even if the diagnosis occurs after childhood?
@Neuro_Skeptic @MarcusMunafo so many wrong things in that abstract. non-sig results taken as no effect. confusing type i correction with "rigour" to name a few.