An evaluation of the performance of SCORE Sweden 2015 in estimating cardiovascular risk

• Published in: European Journal of Preventive Cardiology
• Main Authors: Karjalainen, Tina, Adiels, Martin, Björck, Lena, Cooney, Marie-Therèse, Graham, Ian, Perk, Joep, Rosengren, Annika, Söderberg, Stefan, Eliasson, Mats
• Format: Article in Journal/Newspaper
• Language: English
• Published: Oxford University Press (OUP) 2016
• Subjects: Northern Sweden
• Online Access: http://dx.doi.org/10.1177/2047487316673142; http://journals.sagepub.com/doi/pdf/10.1177/2047487316673142; http://journals.sagepub.com/doi/full-xml/10.1177/2047487316673142

Background Risk prediction models for cardiovascular death are important for providing advice on lifestyle and in decision-making regarding primary preventive drug treatment. The latest Swedish version of the Systematic COronary Risk Evaluation (SCORE 2015) has yet not been tested in the population. Objective The objective of this study was to estimate the prevalence of high and very high risk of fatal cardiovascular disease (CVD) of the current population according to 2015 SCORE Sweden and to evaluate the predictive accuracy of the 2003 Swedish version of SCORE (2003 SCORE Sweden) and 2015 SCORE Sweden in a population with declining CVD mortality. Methods We estimated the high and very high risk group for cardiovascular death for individuals 40–65 years of age in the 2014 Northern Sweden MONICA population survey excluding subjects with known diabetes or previous CVD ( n = 813). Using the 1999 MONICA survey ( n = 3347) followed up for 10 years for CVD mortality, we assessed the calibration of both 2003 and 2015 SCORE Sweden. Results In 2014 2.6% of the population was considered at high or very high risk for fatal CVD, 95% were men and 76% were in the age group 60–65 years. Including subjects with a single markedly elevated risk factor, known diabetes or CVD, 12% of the population was at high or very high risk. During 10 years of follow-up of the 1999 cohort, 34 CVD deaths (24 men and 10 women) occurred. The 2003 SCORE overestimated the risk of death from CVD (ratio predicted/observed 2.3, P < 0.001) whereas the 2015 SCORE slightly overestimated the number of deaths (predicted/observed 1.3, P = 0.12). The 2015 SCORE predicted more accurately than the 2003 SCORE the number of deaths in the different risk and age categories. Conclusion The 2015 SCORE Sweden more adequately than 2003 SCORE Sweden predicts the number of deaths. In 2014, the proportion of high-risk individuals is small in northern Sweden. The main use of 2015 SCORE Sweden would therefore be as an educational tool between the physician and people ...