Researchers linked 183,194 Danish National Health Survey participants to registers, comparing them with 4,697,047 residents aged 16 or older. Under the original weights, mental disorder diagnoses in the 5 years before participation were less common (period prevalence ratio 0.86). New mental disorder diagnoses in the following 5 years were also less frequent (hazard ratio 0.86). All-cause mortality was lower among participants (HR 0.80), and lower still among those with prior mental disorders (HR 0.62). Updated weights built to better reflect mental disorders matched their prevalence (period prevalence ratio 0.99) but left mortality estimates unchanged. Morbidity was measured only through register diagnoses in the 5 years before and after participation.
Why it is interesting: Reweighting matched mental disorder prevalence but did not close the mortality gap. This points to self-selection into the survey that the weights did not capture.
Cite this item
BibTeX
Free sign-up. A digest of the new papers twice a week, grouped by domain.