Objective To compare the effectiveness between operative and non-operative treatment for 3- and 4-part proximal humeral fractures in elderly patients. Methods Between January 2009 and January 2011, 35 patients with 3- or 4-part proximal humeral fractures were treated with open reduction and locking plate internal fixation (n=20, operative group) and with closed reduction and splint or cast fixation (n=15, non-operative group). There was no significant difference in gender, age, etiology, fracture type, and disease duration between 2 groups (P gt; 0.05). The postoperative rehabilitation protocol was performed in 2 groups. Results All patients of the operative group achieved healing of incision by first intention. All patients were followed up 16 months on average (range, 12-20 months). The X-ray films showed that the other fractures healed except 1 case (5.0%) nonunion in operative group. Varus malunion was found in 1 case (6.7%) of non-operative group and 2 cases (10.0%) of operative group. Humeral head necrosis was found in 1 case respectively in 2 groups (5.0% and 6.7%). There was no significant difference in complication incidence between 2 groups (P gt; 0.05). The Constant-Murley scores of non-operative group and operative group were 64.7 ± 9.9 and 66.8 ± 11.8 at last follow-up respectively, showing no significant difference (t=0.59, P=0.47). Conclusion Operative treatment has similar effectiveness to non-operative treatment for 3- and 4-part proximal humeral fractures. In elderly patients, non-operative treatment should be chosen.