Neighbourhood environment and socioeconomic inequalities in cancer admissions: a prospective study using UK Biobank and linked administrative records

Abstract Purpose: Neighbourhood environments may influence cancer risk. Average population effect estimates might mask differential effects by socioeconomic position. Improving neighbourhood environments could inadvertently widen health inequalities if important differences are overlooked. Methods: Using linked records of hospital admissions in UK Biobank, we assessed associations between admission with a primary diagnosis of cancer (any/breast/colorectal), and exposure to neighbourhood greenspace, physical activity facilities, and takeaway food stores, and whether household income and area deprivation modify these associations. We used adjusted Cox proportional hazards models, and estimated relative excess risks due to interaction (RERI) to assess effect modification. Results: Associations between neighbourhood exposures and cancer-related hospitalisations were weak to null overall, but with some evidence of effect modification. Most notably, more greenspace near home was associated with 16% lower hazard of cancer-related hospital admission in deprived areas (95%CI: 2%-29%). This was further pronounced for people in low-income households in deprived areas, and for breast cancer. Conclusions: In deprived neighbourhoods, increasing the amount of greenspace may help reduce cancer-related hospitalisations. Examining effect modification by multiple socioeconomic indicators can yield greater insight into how social and environmental factors interact to influence cancer incidence. This may help avoid perpetuating cancer inequalities when designing neighbourhood environment interventions.