Previous studies associating neighbourhood context with mental health service use typically included limited sets of confounders. A data set including patients registered in a Case Register and population controls was subjected to multilevel analyses, including neighbourhood exposures and individual level confounders. In addition, days of care consumption of patients was addressed. The association between socioeconomic deprivation and social capital on the one hand and mental health service use rates on the other could be attributed to individual level differences. However, number of days of service consumption was higher in neighbourhoods with more informal social control. In residentially stable neighbourhoods only, socioeconomic deprivation was associated with lower levels of service consumption. Higher levels of social control may induce patients to remain in contact with mental health services. Furthermore, higher levels of deprivation in neighbourhoods with little population mobility may result in reduced expectations of recovery and/or increased tolerance.