Abstract: The Morinda officinalis sample (0.3 g) was taken, and 7 mL of nitric acid was added. The mixture was soaked overnight, dissolved by microwave digestion at 195 °C for 45 min, and heated to 1−2 mL at 100 ℃ in order to drive the acid. The solution was transferred into a 50 mL-volumetric flask with 5%(volume fraction, the same below) nitric acid solution, and 0.2 mL of 1 mg·L⁻¹ Au standard solution was added. The solution was made its volume up to 50 mL with 5% nitric acid solution, and the total element content test solution was obtained. The various elements in Morinda officinalis were extracted by water decoction, ethanol and bionic extraction solution (artificial gastric juice and artificial intestinal juice), 20 μg·L⁻¹ of the mixed internal standard solution of Sc, Ge, In and Bi was introduced online, and 22 elements (Be, B, Al, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Sr, Mo, Cd, Sn, Sb, Ba, Hg, Tl, and Pb) in the above solutions were determined by inductively coupled plasma mass spectrometry (ICP-MS). The extraction rates of the 22 elements under three extraction methods were analyzed, and their potential health risks to humans were assessed using the maximum theoretical limit values. As shown by the results, linear relationships between the response intensities after internal standard correction and mass concentrations of 22 elements were kept in definite ranges, with detection limits (3s/k) in the range of 0.001−0.528 mg·kg⁻¹. Test for recovery was made by the standard addition method, giving results in the range of 84.3%−103%, with RSDs of the determined values (n=6) less than 7.0%. The determined values of the reference material GBW10020 for citrus leaf component analysis were within the uncertainty ranges of the certified values. The extraction rates of water decoction extraction, ethanol extraction and bionic extraction for elements were 2.35%−75.9%, 0.205%−30.6% and 4.18%−93.1%, respectively. This method was used to analyze 14 batches of samples. Although the contents of elements in the samples varied, the changing trends of the contents among the elements were basically the same. The detected amounts of elements in the samples were less than the maximum theoretical limit values of the corresponding elements, indicating that the risk levels of the 22 elements were relatively low.