目的评价采用曲线估计结合趋势季节模型预测医疗机构出院人次的精确性。方法采用曲线估计结合趋势季节模型拟合分析三六三医院2009~2015年各季度出院人次,预测2016年各季度出院人次,并计算预测值与实际值的相对误差。结果构建最优二次方回归方程为Y_t=3 006.050 1+202.350 8×t–3.544 4×t~2(决定系数R~2=0.927,P<0.001),结合趋势季节模型预测该院2016年出院人次为23 462人次,与真实值的相对误差为1.79%。结论曲线估计结合趋势季节模型是一种方便、直观的统计方法,在医疗机构出院人次或门急诊人次预测上具有较高的精确性,可以较好地为医院运营管理决策提供参考依据。 Objective To evaluate the accuracy of predicting the number of discharges by medical institutions by using the curve estimation combined with the seasonal season model. Methods The curve estimation and trend season model were used to fit and analyze the discharge data of hospitals in each of the 363 hospitals from 2009 to 2015, and predict the number of discharges in each quarter of 2016. The relative errors between the predicted and the actual values ​​were calculated. Results The optimal quadratic regression equation was Y_t = 3 006.050 1 + 202.350 8 × t-3.544 4 × t ~ 2 (coefficient of determination R 2 = 0.927, P 0.001) The number of discharged patients was 23 462, with a relative error of 1.79% from the true value. Conclusions The curve estimation combined with the trend season model is a convenient and intuitive statistical method. It has a higher accuracy in the number of hospital discharge or outpatient and emergency department visits and can provide a reference for hospital operation and management decision-making.