Neighborhood-level predictors of age-at-first-diagnosis of psychotic disorders

Why did we do this research?

The relationship between neighborhood-level factors and the incidence of psychotic disorders is well established. However, it is unclear whether neighborhood characteristics are also associated with age-at-first-diagnosis of these disorders.

What did we do?

We used linked Swedish register data to identify a cohort of persons first diagnosed with an ICD-10 non-affective or affective psychotic disorder (F20-33) between 1997 and 2016.

Using multilevel mixed-effect linear modelling, we investigated whether neighborhood deprivation and population density at birth were associated with age-at-first diagnosis of a psychotic disorder. Our final cohort included 13,440 individuals, with a median age-at-first-diagnosis of 21.8 years for women (interquartile range [IQR]: 19.0-25.5) and 22.9 years for men (IQR: 20.1-26.1; p<.0001).

What did we find?

A one standard deviation increase in deprivation at birth resulted in a 47 day delay (95% CI: 18-77) in first diagnosis.

Interestingly, this association was first maksed by family income (in the unadjusted models), but once we accounted for income, delays in diagonsis related to neighbourhood deprivation became apparent (adjusted models).

We found no evidence this differed by sex or non-affective vs. affective diagnosis, and no evidence of an association with population density; patterns were similar when exposure during upbringing was used.

What does this mean?

Later age-at-first-diagnosis for individuals born in more deprived neighborhoods suggests structural barriers to accessing equitable psychiatric care.

There may be barriers to accessing timely care for psychosis amongst those born and growing up in more deprived communities, providing an opportunity for tailored, targeted early intervention.

Read full-text article.

This research was led by Spyros, one of the MSc students in the Division of Psychiatry and is part of an ongoing collaboration with researchers at Karolinska Institutet (Dr Anna-Clara Hollander & Prof Christina Dalman.