The treatment of patients for coronary heart disease risk requires knowledge of the plasma lipid levels. Low density lipoprotein cholesterol (LDL) levels make a strong basis for therapeutic decisions. Although there are incongruities among values of LDL from different methods of determining LDL, the clinician is not routinely informed of the method used. The purpose of this study was to compare LDL levels determined by the Friedewald equation with those assayed by the Kyowa Madox method. The lipid results previously measured by Kyowa Madox method among Makerere University fasting students and reported earlier were retrieved. The measured values of total cholesterol (TC), High Density Lipoprotein cholesterol (HDL) and triacylglycerols (TG) were used to calculate LDL using Friedewald equation in which LDL= TC-HDL-TG/2.2 mmol/L. The values obtained were compared non parametrically with the assayed values previously reported. Our results showed a high value of correlation between measured and calculated LDL so that in general, the two methods can be used interchangeably in this population. However, in cases of dyslipidaemia, the calculated values tend to be lower than the assayed values. It is therefore recommended that clinical laboratories should report the LDL values along with the determination method used, the alert values, the reference ranges, the desirable ranges and the therapeutic targets.