%0 Journal Article %A Carolina Ibarra-Castillo %A Marina Guisado-Clavero %A Concepció Violan-Fors %A Mariona Pons-Vigués %A Tomàs López-Jiménez %A Albert Roso-Llorach %A , %T Survival in relation to multimorbidity patterns in older adults in primary care in Barcelona, Spain (2010-2014): a longitudinal study based on electronic health records %D 2018 %R 10.1136/jech-2017-209984 %J Journal of Epidemiology and Community Health %P jech-2017-209984 %X Background Several studies have analysed the characteristics of multimorbidity patterns but none have evaluated the relationship with survival. The purpose of this study was to compare survival across older adults with different chronic multimorbidity patterns (CMPs).Methods Prospective longitudinal observational study using electronic health records for 190 108 people aged ≥65 years in Barcelona, Spain (2009–2014). CMPs were identified by cluster analysis. Mortality rates were estimated using the Catalan population structure and individual time at risk. Survival according to CMP (Cox regression) was analysed using hazard ratios (HRs) and 95% confidence intervals (CIs) with stratification by sex and age group (65–79, 80–94) and adjustment for age at onset, deprivation index, number of chronic conditions and invoiced drugs.Results The highest mortality rates were observed in men, adults aged 80–94 years, socially disadvantaged quintiles and people prescribed more drugs and with fewer conditions. Using the musculoskeletal pattern as the reference category, men with the digestive-respiratory pattern had a higher risk of death, with adjusted HRs of 6.16 (95% CI 5.37 to 7.06) in the 65–79 age group and 2.62 (95% CI 2.31 to 2.97) in the 80–94 age group. In women, the cardiovascular pattern was associated with the highest risk, with adjusted HRs of 6.34 (95% CI 5.28 to 7.61) in the 65–79 age group and 3.05 (95% CI 2.73 to 3.41) in the 80–94 age group. These patterns were also associated with the highest mortality rates.Conclusions Mortality and survival vary according to CMPs in older adults stratified by sex and age. Our findings are useful for guiding the design and implementation of clinical management strategies. %U https://jech.bmj.com/content/jech/early/2018/01/12/jech-2017-209984.full.pdf