Authors' reply

We thank Professor Caan for his interesting correspondence on our paper and for providing some further insight into how ethnic density may impact on the risk of postnatal depression in our sample. The protective effect of ethnic density is thought to operate predominately through either (a) reducing exposure to racism, or (b) improved social support and social networks, ^1,2 and indeed, the latter is highly important in this context, given that poor social support is a strong predictor of postnatal depression. ³

We agree that family and household effects can have profound consequences on mental health, ⁴ and that the proximity and ease of access to close friends and relatives is highly beneficial, especially for new mothers. Although our article refers to the importance of social support, and how this may be driving the protective effects of high ethnic density in our sample, we failed to dissect fully the mechanisms underpinning the associations between ethnic density and postnatal depression. This area would certainly benefit from further research, by enquiring about the locality of close family friends and relatives, so we may begin to understand more fully how high ethnic density may be exerting its protective effects.