Prognostic validation of a non-laboratory and a laboratory based cardiovascular disease risk score in multiple regions of the world

16 April 2018

O bjective t o evaluate the performance of the non- laboratory i nterheart risk score ( n l - i hrs ) to predict incident cardiovascular disease ( c V D) across seven major geographic regions of the world. t h e secondary objective was to evaluate the performance of the fasting cholesterol- based i hrs (F c - i hrs ) . Methods U sing measures of discrimination and calibration, we tested the performance of the n l - i hrs (n=100 4 75) and F c - i hrs (n=107 8 63) for predicting incident c V D in a community-based, prospective study across seven geographic regions: s o uth a s ia, c h ina, s o utheast a s ia, Middle e a st, e u rope/ n o rth a m erica, s o uth a m erica and a f rica. c V D was defined as the composite of cardiovascular death, myocardial infarction, stroke, heart failure or coronary revascularisation. r e sults M ean age of the study population was 50.53 ( s D 9.79) years and mean follow-up was 4.89 ( s D 2.24) years. t h e n l - i hrs had moderate to good discrimination for incident c V D across geographic regions (concordance statistic ( c - statistic) ranging from 0.64 to 0.74), although recalibration was necessary in all regions, which improved its performance in the overall cohort (increase in c - statistic from 0.69 to 0.72, p<0.001). r e gional recalibration was also necessary for the F c - i hrs , which also improved its overall discrimination (increase in c - statistic from 0.71 to 0.74, p<0.001). i n 85 0 78 participants with complete data for both scores, discrimination was only modestly better with the F c - i hrs compared with the n l - i hrs (0.74 vs 0.73, p<0.001). Conclusions e x ternal validations of the n l - i hrs and F c - i hrs suggest that regionally recalibrated versions of both can be useful for estimating c V D risk across a diverse range of community-based populations. c V D prediction using a non-laboratory score can provide similar accuracy to laboratory-based methods