PurposePeople with severe mental illnesses (SMI) have an increased risk of cardiovascular disease (CVD). Research in the general population suggests that social support may protect against increased CVD morbidity and mortality; however, this may not apply to those with SMI. We aimed to explore the association between perceived social support and attendance at primary care nurse CVD risk reduction clinic appointments and CVD risk-reducing behaviours in an SMI population with elevated CVD risk factors.MethodsWe used longitudinal and cross-sectional data from a randomised controlled trial on 326 adults with SMI recruited via 76 general practices in England. Multilevel regression analysis estimated the effect of perceived social support on attendance at CVD risk reduction clinic appointments over 6 months, and adherence to CVD medication, physical activity, diet, smoking and alcohol use at baseline, adjusted by age, sex, ethnicity, deprivation, psychiatric diagnosis and employment.ResultsPerceived social support predicted greater appointment attendance in unadjusted (IRR = 1.005; 1.000–1.010; p = 0.05) but not adjusted analysis (IRR = 1.003; 0.998–1.009; p = 0.25). Perceived social support was associated with greater adherence to medication; for each 1% increase in social support, there was a 4.2% increase in medication adherence (OR = 1.042; 1.015–1.070; p = 0.002). No association was found between greater perceived social support and greater physical activity, lower sedentary behaviour, healthier diet, lower alcohol use or being a non-smoker.ConclusionsSocial support may be an important facilitator for CVD medication adherence and is potentially important for primary care appointment attendance; however, alternative strategies might be needed to help people with SMI engage in physical activity, healthier diets and to reduce their smoking and alcohol use.