Hypercholesterolemia and hypocholesterolemia are associated with mortality which warrants routine lipid profile testing. This financially burdens the already overwhelmed health sector especially in developing countries. Additionally, lipid profile test reagent stock-out or failure to afford all tests affects result interpretation. In 1972, James Friedewald published a statistical model to calculate low density lipo-protein. The study aim was to determine the percentage error of the James Friedewald equation in calculating all lipid profile test parameters. A retrospective study from 2018 was performed at Mildmay Uganda involving lipid profile results of 103 persons (48 HIV-positive and 55 HIV-negative) 50 years and older enrolled in a previous cross-sectional study. The Friedewald equation was used to calculate total cholesterol, high density lipoprotein, triglycerides and low density lipo-protein. The percentage error of calculated values in reference to measured values was ascertained. Pearson correlation between measured and calculated results was determined among all persons and classified by HIV status. The total error of calculated analytes was 7% (low density lipo protein), 17% (high density lipo-protein), 39% (triglycerides) and 4% (total cholesterol). Pearson correlations were 0.98 (all persons), 0.98 (HIV-negative) and 0.98 (HIV-positive) for low density lipo-protein, 0.89 (all persons), 0.90 (HIV positive) and 0.88 (HIV-negative) for high density lipo-protein, 0.75 (all persons), 0.76 (HIV-negative) and 0.77 (HIV-positive) for triglycerides, 0.99 (all persons), 0.98 (HIV-negative) and 0.99 (HIV-positive) for total cholesterol. In conclusion, Friedewald equation reliably calculated low density lipoprotein, total cholesterol (most accurate) and high density lipo-protein while triglycerides calculation was erroneous among persons aged ≥ 50 years.