| ObjectiveEstablish a chronic kidney disease(CKD)patient’ s PRO scale of traditional Chinese medicine(TCM)to response the patient’ s reality feeling of the clinical treatment,which is one of the evaluation of curative effect.It can provide guidance for clinical treatment and medication,expanse the idea of TCM efficacy evaluation,provide the basis of the search of TCM、the development of new drug and the extension of TCM treatment.According to the results of pilot study using item response theory to review CKDQOL-ND entries,which make the scale further perfection and lay a foundation for the further promotion of scale.By the way,compare the CKD1-3 patient with CKD4-5 patient who has non dialysis,found the CKD4-5 patient’ s special performance in quality of life and needs of PRO scale,who has non dialysis,and provide the basis of improve the quality of medical care in the future.Methods1.The Survey MethodInvestigation of Guangdong hospital of TCM,The First Affiliated Hospital of Traditional Chinese Medicine University Of Guangzhou,the Dongzhimen Hospital of Chinese Medicine,The Guanganmen Hospital of Chinese Academy of Medical Sciences,Liuan Hospital of TCM,The Hosptial of economic technical development area.This survey included 70 cases CKD1-3 patients and 263 cases CKD4-5 non dialysis patients.The patients must have a certain ability to read and write.The investigators should simple explain the scale to patients as a doctor.2.Item Response TheoryThe Semejim level reaction model was adopted to analysis each item of CKDQOL-ND unidimensional item response theory,according to the statistical dependency between projects,check list and local independence assumption,statistics of discrimination parameter and difficulty parameter values,use maximum likelihood estimation and the goodness of fit model fitting method,draw out the project item characteristic curve and information function curve.3.Variance AnalysisTo Evaluate the diference between CKD1-3 and CKD4-5 non dialysis population in various fields and the total score,matching by Independent-samples T test.Using single-factor analysis of variances method to study different demographic characteristics in the field of physiological,psychological,social,treatment,the self-assessment scale scores and patients of this a few different.4.The Statistical MethodsUsing SPSS 22.0 to complete the analysis of exploratory factor and reaction degree,Independent-samples T test and single-factor analysis of variances method.LISREL8.80 achieve confirmatory factor analysis.Using Multilog7.03 software and ConQust2.0 to analysis the item response theory.Results1.The Survey DataThe samples of this study from July 2014 to March 2015 of the inpatient and outpatient.Totolly Collected of 339 cases.After handling the missing value,there were 333 cases,including CKD1-3 patient 70 cases ane CKD4-5 non dialysis patient 263 cases.In those cases,the 210 cases achieved by Guangdong hospital of TCM is the main part,accounting for 63.1%.The First Affiliated Hospital of Traditional Chinese Medicine University Of Guangzhou have 30 cases,accounting for 9.0%.The The Hosptial of economic technical development area have 36 cases,accounting for 10.8%.A total of 15 regarding Liuan Hospital of TCM,accounted for 4.5%.The Dongzhimen Hospital of Chinese Medicine,a total of 19,accounted for 5.7%.The Guanganmen Hospital of Chinese Academy of Medical Sciences,a total of 23,6.9%.2.Unidimensional IRT(1)The Hypothetical TestUsing exploratory factor analysis(EFA)and confirmatory factor analysis(CFA)method to analysis,EFA analysis for total table first factor and the second factor eigenvalues ratio 3.112,physiological neighborhood first factor and the second factor eigenvalues ratio 4.76,psychological neighborhood first factor and the second factor eigenvalues ratio 3.002,the first factor and the second factor eigenvalues neighborhood society ratio 3.486,neighborhood first factor and the second factor eigenvalues ratio 1.58;In a single factor model descending the CFA examination,x 2/df is 4.511,NNFI is 0.78,NFI is 0.75,CFI is 0.78,RMSEA is 0.116,the original scale four factor model downward CFA analysis,x 2/df is 2.778,NFI is 0.85、CFI is 0.90、RMSEA is 0.082.Those results in line with the indicate unidimensional and localized hypothesis test.(2)The Difficulty and Discrimination ParameterEvaluated by the MML of GRM model,get the degree of discrimination parameter a and difficulty parameter b.Physiological neighborhood:a were greater than 0.3,the fluctuations in 0.72 to 2.26;Most project b between 2.95～2.95,and monotone increasing,the b4 value>2.95 in Ph9,Ph10,Ph15,Ph17,Ph18,Ph19,Ph20,Ph23,Ph24,Ph25,Ph26,Ph29,Ph32,Ph33,that entry is too difficult,Ph24 "sleep" project of b1 values<2.95.Psychological neighborhood:a value is between 0.91 to 4.92;The Ps3,Ps6,Ps8 correspond have to worry about discrimination,fun project,mood swings,b values were 3.01,-3.06,3.68,beyond the scope of ideal,the other project b value is between-1.81 to 2.70.Social neighborhood:a between 0.86 to 2.58,So5"friends" and So6 "sex" b is greater than 2.95.Treatment of dimensions:Th3,Th4a value is lower,divided into 0.32,0.23,the other project have a value between 0.67 to 3.17;"Th3",Th4,Th5,Th6 corresponding drug side effects,kidney disease risk factors,treatment of traditional Chinese medicine and diet control project,b4 value divided into 9.18,11.18,3.65,11.18,and the b1 value is lower in the project of Th3,Th4 and Th6,4.82,10.60 and 10.60,respectively.(3)The Item Characteristic Curve(ICC)and Information Function Curve(IFC).Graphical display,physiological neighborhood Phl～23,Ph25～33,psychological neighborhood Ps1～5,Ps8,social neighborhood So1～5,treatment of neighborhood Th1～2 of these projects,the ICC curve administrative levels feels strong,large range,which means the large amount of information and the project is ideal.Physical neighborhood Ph24,psychological field Ps6,Ps7 and social neighborhood So6,So7,treatment of neighborhood Th3,Th4 and Th6 curve probability value is small,the curve is low,a small value,IIC peak,and is generally flat which means these project needs further research to improvet.3.Multidimensional IRT(1)The Difficulty ParameterAll dimensions the patient’ s level of ability and scale project difficulty levels nomally distributed on the whole,the patient’ s level of ability distribution below the scale of the project difficulty levels,Ph15,Ph25,Ps3,Ps4,Ps5,So4,So5 and Th2 project difficulty levels higher than the patient’ s level of ability.(2)The MNSQ under Different Dimensions IRT Analysis Outfit MNSQ in the MIRT and UIRT analysis is greater than 1.3 projectshave Ph19,Ph24,Ph27,Ph29,Ps7 and Ps8,So6,So7 Ps4,Ps5 is less than 0.7.Under the MIRT and UIRT analysis of Infit MNSQ,So6,So7is greater than 1.3,Ps4,Ps5 are less than 0.7.The rest of the project’ s MNSQ values were between 0.7～1.3.(3)The Reliability Coefficient under Different Dimensions IRT AnalysisUnder the MIRT analysis,reliability coefficients were between 0.606～0.944;Under the UIRT analysis,reliability coefficients were between 0.534～0.915;the reliability coefficient of treatment field is less than 0.7,MIRT in all areas of the reliability coefficients is larger than UIRT’ s.(4)The Correlation Coefficient Under Different Dimensions IRT AnalysisThe correlation coefficient in the four dimensions were between 0.102~0.703,the psychological and social adaptation ability areas of correlation among the strongest,r=0.703;Field of physiological symptoms and treatment of body feel the worst domain correlation,r=0.102.4.Variance Analysis(1)The Different Between CKD1-3 and CKD4-5 Non DialysisDifferent stages in patients with chronic kidney disease(CKD)in the body physiological symptoms,mental health,social adaptation ability and self evaluation scores were significantly different(P<0.05),the score was no statistical difference in the field of treatment.Detailed analysis of two groups,we found that Ph1 and Ph2,Ph3,Ph4,Ph9,Ph13,Ph14,Ph15,Ph22,Ph31,Psl,Ps7,Sol,So4,So7,Th1,Th4,Th5 were differences,t statistic reached significant level,significant probability P<0.05,according to the eta 2 value,belongs were light intensity.(2)The Comparison of Different Demographic Characteristics of CKD4-5 Non DialysisDifferent age,hospitals,cultural level,medical expenses,disease progression in CKD4-5 non dialysis period which body physiological symptoms domain score difference was statistically significant(P<0.05).Different hospitals,occupation,living environment,medical expenses,grouping kidney disease progression in CKD4-5 non dialysis patients which mental health scores were significantly difference(P<0.05).Different gender,hospital,medical costs,grouping in CKD4-5 non dialysis patients’ social adaptation ability was statistically significant(P<0.05).Different age,occupation,hospital group in CKD4-5 non dialysis whose treatment feelings were difference,which was statistically significant(P<0.05).Grouping different hospitals,cultural degree,medical expenses in CKD4-5 non dialysis patients’ self assessment score were difference,which was statistically significant(P<0.05).Different age,hospitals,cultural level,medical expenses,kidney disease duration grouped in CKD4-5 non dialysis scale scores have significant differences,with statistical significance(P<0.05).Conclusion:1.CKDQOL-ND is multidimensional scale.Using MIRT analysis show that the scale of the project on the overall difficulty levels were higher than the patient’s level of ability,which should be overall difficulty levels to reduce the scale of the project.Within the project,we need to find out the inner structural reletions in the scale.Except treatment project,the test has good reliability,the items of treatment dimension needs to be improved.2.The different between CKD4-5 stage and CKD1-3 patients on the body physiological function,mental health,social ability had different performance.The scale is suitable for using in CKD4-5 stage of non dialysis patients,which can distinguish with CKD1-3 patients. |
PDF Full Text Request