症状疗效量表的数学模型研究

目的：以多酶组合胶囊治疗消化酶不足引起消化不良的一临床研究为实例，建立一个症状疗效量表研究的数学模型。方法：症状评分与其发生率（频率）、重要性（专家评定）及严重性（轻中重）有关，由此建立含有以上3种因素（因子）的症状评分数学模型，形成消化不良症状疗效的评分量表，通过信度、效度、反应度和计算机模拟对此进行评价，确认其可行性和合理性。结果：6个症状（上腹胀痛、早饱、食欲下降、嗳气、反酸和恶心）作为疗效评分症状群，根据其发生率、重要性和严重性确定了各自权重因子和量表等级分值。信度结果可靠，一致性良好，反应灵敏；模拟发现，在不同的病例数（100～2000），让各症状的重要性判断和发生率在10％的范围波动，结果稳定。结论：本法建立的消化不良症状疗效评分量表，结果可靠，可为同类研究提供参考。

Study of mathematical models of scales for evaluating efficacy in symptomatology

AIM:To develop a symptom scale for evaluating efficacy of the multi-enzyme combination capsule in treatment of patients with dyspepsia symptoms be caused digestive enzyme insufficient. METHODS: The symptom score is always related to the incidence rate (frequency), the importance (evaluated by experts) and the serious degree (light, moderate and serious) in a measuring scale, and a mathematical model was proposed to calculate the weight factor and its scores for each symptom in the scale. In an example of the multi-enzyme combination capsule in treatment of patients with chemical dyspepsia, the efficacy was analyzed based on the symptom scales with this approach. The scale was evaluated with the reliability, the validity, the responsibility to change, and the simulation. RESULTS: Six symptoms (epigastric distention, easy satiety, the degression of appetite, belches, counter-sour and nausea) were selected as indices, and their weight factors and grades were calculated in score scales. Some evaluations demonstrated that the approach showed a better feasibility, uniformity, and responsibility to change. The sample size (100-2 000) and the flutter (10%) of the importance and the frequency yielded a little effect on total scores in simulation. CONCLUSION: This is a reliable approach in developing a scale of symptomatology for evaluating efficacy of drugs.