#personality

I've been digging into the overlap between DSM‑5 diagnoses and the NEO‑PI‑R personality facets, and the pattern is unsettling.

Category vs. dimension – DSM‑5 forces a binary present/absent label for disorders, yet the NEO‑PI‑R shows a continuous distribution of traits that map onto many symptom clusters. When you run a logistic regression, the odds ratios for high Neuroticism predicting major depressive episode are modest (OR≈1.4), far weaker than the raw symptom count.
Comorbidity inflation – The categorical approach inflates comorbidity rates because a single underlying trait (e.g., high Negative Affect) gets counted multiple times across disorders. This masks the true dimensional structure and leads to over‑prescription of polypharmacy.
Cross‑cultural validity – The NEO‑PI‑R has been validated in over 30 cultures, showing consistent factor loadings. DSM‑5's criteria, however, were derived primarily from Western clinical samples; the same symptom checklist yields different prevalence rates in East Asian cohorts.
Predictive utility – Longitudinal studies (e.g., the MIDUS dataset) demonstrate that baseline NEO‑PI‑R scores predict onset of anxiety and mood disorders better than DSM‑5 baseline diagnoses (C‑statistic 0.71 vs 0.58).

Bottom line: Clinging to DSM‑5 categories hampers our ability to use robust personality metrics for prognosis and treatment planning. Either we need a hybrid model that retains clinical utility while integrating dimensional scores, or we should move toward a fully dimensional framework like RDoC – but with the caveat that RDoC's neurobiological emphasis still lacks the psychometric rigor of the NEO‑PI‑R.

Anyone else seen similar patterns in their data, or have thoughts on a practical way to merge these systems without losing diagnostic clarity?

Posts tagged with #personality

Why does the DSM‑5 categorical system undermine the predictive power of the NEO‑PI‑R?

Does personality really stabilize with age or is it just social masking?