Neighborhood social structure, criminal convictions, and substance use disorders in Sweden: a population-based...

We investigated how multidimensional neighborhood types, capturing the complexity of neighborhood social structure, related to criminal convictions and substance use disorders (SUDs) in Sweden while accounting for residential mobility and shared familial influences.

Using nationwide Swedish register data (1991-2020), we followed 1.82 million individuals born 1985-2002 across five neighborhood types previously identified using latent class analysis. These were defined by socioeconomic and demographic indicators, with the most common type (rural low-diversity neighborhoods) serving as the reference category. Associations between neighborhood types, violent and nonviolent convictions, and SUDs were estimated using Cox proportional hazards models with hazard ratios (HRs) and 95% confidence intervals (CIs). Neighborhood type was modeled as a time-varying exposure, capturing residential changes over time. Sibling comparison models were applied to account for familial confounding.

Risks of violent convictions, nonviolent convictions, and SUDs were highest in resource-limited neighborhoods (i.e., socioeconomically deprived areas with high residential turnover; violent: adjusted HR (aHR) = 4.70, 95% CI 4.26-5.19; nonviolent: aHR = 2.07, 95% CI 1.92-2.22; SUDs: aHR = 3.36, 95% CI 3.09-3.65). Moderate increases were observed in urban high-diversity neighborhoods (i.e., characterized by ethnic heterogeneity), whereas risks were lowest in socioeconomically advantaged urban areas. Associations were attenuated but remained statistically significant in sibling comparisons. Sex differences were modest and followed similar patterns across outcomes.

Neighborhoods marked by socioeconomic disadvantage and residential instability, rather than ethnic diversity, were linked to higher risks of criminal convictions and SUDs. Remaining associations after familial adjustment suggest modest but meaningful contextual effects.