The impact of ethnic density on dispensing of antipsychotic and antidepressant medication among immigrants in the Netherlands

Purpose: A higher own-group ethnic density in the area of residence is often associated with a lower risk for psychotic disorder. For common mental disorders the evidence is less convincing. This study explores whether these findings are mirrored in data on dispensing of antipsychotics and antidepressants. Methods: Health insurance data on dispensed medication among all adults living in the four largest Dutch cities were linked to demographic data from Statistics Netherlands. Dispensing of antipsychotics and antidepressants in 2013 was analyzed in relation to the proportion of the own ethnic group in the neighborhood. Results: Higher own-group ethnic density was associated with lower dispensing of antipsychotics among the Moroccan-Dutch (N = 115,455), after adjusting for age, gender, and SES of the neighborhood (ORadj for the highest vs. the lowest density quintile = 0.72 [0.66-0.79]). However, this association vanished after adjustment for household composition (ORadj = 0.93 [0.85-1.03]). Similar results were found for the Turkish-Dutch (N = 105,460) (ORadj = 0.86 [0.76-0.96] and 1.05 [0.94-1.18]). For those of Surinamese (N = 147,123) and Antillean origin (N = 41,430), in contrast, the association between ethnic density and lower risk remained after each adjustment (P < 0.001). For antidepressants, a negative association with own-group ethnic density was consistently found for those of Antillean origin (ORadj = 0.62 [0.52-0.74]) only. Conclusion: These data on dispensing of psychomedication confirm the ethnic density hypothesis for psychosis alongside earlier equivocal findings for other mental disorders. The negative association between own-group ethnic density and dispensing of antipsychotics among the Moroccan- and Turkish-Dutch may be explained, at least in part, by a favourable household composition (i.e., living in a family) in high-density neighborhoods.