Congruity between WHO ISH laboratory and non-laboratory-based charts in an urban population in India: A cross-sectional community-based analytical study.

Abstract

Cardiovascular diseases (CVDs) are responsible for nearly one-third of all global deaths. WHO ISH non-laboratory-based risk prediction charts are useful for predicting the individual risk of CVD, especially in places without laboratory facilities. However, their results must be comparable to the laboratory-based charts. This cross-sectional study aimed to measure the congruity between the WHO ISH laboratory and non-laboratory-based charts in adults ≥40 years in an urban population in India. Data were collected using a semi-structured questionnaire, including details about behavioral and environmental risk factors, anthropometry, and biochemical measurements. Pearson's correlation and scatter plots were applied to measure the association between the risk scores in the two charts. Cohen's Kappa coefficient was applied to test the congruity between the risk categories given by the two charts. Receiver operator curve (ROC) plot for non-laboratory-based chart was plotted taking the laboratory-based chart as the gold standard. The mean CVD risk scores of the nonlaboratory and the laboratory-based charts are 6.62 ± 4.62% (IQR 3-9) and 7.99 ± 6.05% (IQR 3-11.75), respectively. The mean CVD risk score in males came to be more than that of females. Scatter plots showed a very strong association with a correlation coefficient of 0.861 (<i>P</i> < 0.001). The level of agreement between the two charts using Kappa statistics was 75.89%. The ROC curve showed a high area under the curve of 0.964. The non-laboratory-based chart can be applied to the population in low-and-middle-income countries like India, with reasonable accuracy.