基于效标法与分布法的头颈癌生命质量量表最小临床重要性差值制定与评估

Establishing and evaluating minimal clinical important differences for the quality of life instrument in patients with head and neck cancer based on anchor and distribution method

目的使用效标法和分布法计算癌症患者生命质量测定量表体系之头颈癌量表[QLICP-HN(V2.0)]得分的最小临床重要性差值(MCID),为临床评价头颈癌治疗效果提供依据。方法依据头颈癌纳入和排除标准,使用QLICP-HN(V2.0)量表对云南省肿瘤医院2015-11-01-2016-12-31收治的100例头颈癌患者进行问卷调查,量表包括生理功能、心理功能、社会功能、共性症状与副作用和特异模块5个领域。依据量表计分规则,计算各领域得分及量表总分,采用配对Z检验进行治疗前后统计学检验;采用效标法(标准A和B)和分布法[效应值(ES)、标准测量误差(SEM)和可靠变化指数(RCI)]计算头颈癌MCID。结果效标法标准A结果显示,躯体功能、心理功能、社会功能、共性症状和副作用、特异模块及总量表的MCID分别为6.76、14.23、4.40、6.85、6.92和8.10;标准B结果显示,躯体功能、心理功能、社会功能、共性症状和副作用、特异模块及总量表的MCID分别为10.66、17.59、6.53、12.23、9.46和11.50。分布法结果显示,头颈癌MCID为1.74~10.03,其中ES法在上述各领域得分及量表总分的MCID分别为8.49、10.03、6.27、9.42、6.43和6.40,均高于SEM和RCI法,且与效标法计算的MCID相差不大。结论临床评价头颈癌治疗效果一般优先采用效标法计算结果,当效标法所选用锚点与测定量表相关系数较小或者符合效标定义的例数较少时,采用分布法中ES的计算结果。

Objective To calculate the minimal clinical importance difference(MCID) of the quality of life instruments for head and neck cancer patients(QLICP-HN) using the standard method and the distribution method, and provide a basis for clinical evaluation of the treatment effect of head and neck cancer. Methods According to the inclusion and exclusion criteria of head and neck cancer, a questionnaire survey of 100 patients with head and neck cancer admitted to Yunnan Cancer Hospital from November 1, 2015 to December 31, 2016 was conducted using the QLICP-HN(V2.0) scale. The scale includes physiological function, psychological function, social function, common symptoms and side effects, and 5 areas of specific modules. According to the scale scoring rules, the scores in each field and the total score of the scale are calculated, and the paired Z test is used for statistical testing before and after treatment;the effect standard method(standard A and B) and the distribution method [effect size(ES), standard error of measurement(SEM), reliabilitychange index(RCI)]are used to calculate the MCID of head and neck cancer.Results The results of standard A of the standard method showed that the MCIDs of physical function,mental function,social function,common symptoms and side effects,specific modules and total scale were 6.76,14.23,4.40,6.85,6.92and 8.10respectively.The results of standard B showed that the body function,psychological function,social function,common symptoms and side effects,specific modules and total scale were 10.66,17.59,6.53,12.23,9.46and 11.50,respectively.The results of the distribution method showed that the MCID of head and neck cancer was 1.74-10.03,and the MCID of the ES method in the above-mentioned fields and the total score of the scale were 8.49,10.03,6.27,9.42,6.43,and 6.40,which were higher than those of the SEM and RCI methods.And it is not much different from the MCID calculated by the effective standard method.Conclusions In clinical evaluation of head and neck cancer,the results of the treatment effect of head and neck cancer are generally calculated by the standard method.When the correlation coefficient between the anchor point selected by the standard method and the measurement scale is small or the number of cases conforming to the definition of the standard is small,the ES calculation in the distribution method is adopted to calculate the result.