Measurement Characteristics Evaluation Of SF-HSTCM And EQ-5D-3L Among Patients With Stroke

ObjectiveTo evaluate the measurement characteristics of the short Form-Health scale of traditional Chinese medicine(SF-HSTCM) and EQ-5D-3L and to have further understanding of the applicability of SF-HSTCM among patients with stroke.MethodThe study used a cross-sectional design. We investigated patients with stroke on the recovery stage in the Outpatient Chronic Disease Department, Outpatient and Inpatient Neurology Department of Guangdong Provincial Hospital of Chinese Medicine, and in the Acupuncture and Rehabilitation Department of the Guangdong Second Traditional Chinese Medicine Hospital. Questionnaires applied in the study included SF-HSTCM, EQ-5D-3L, mRS and BI The SF-HSTCM and EQ-5D-3L was randomly delivered to participants. Descriptive analysis was used to describe the general demographic characteristics and other factors. We analyzed the reliability of SF-HSTCM by the Cronbach’s alpha coefficient and the group intercorrelation coefficient (ICC);we adopted the confirmatory factor analysis to evaluated the validity of SF-HSTCM. The SF-HSTCM test retest reliability and the consistency reliability between the patient and the agent was assessed by group inter correlation coefficient (ICC). The frequency distribution method describes the test retest reliability of EQ-5D-3L and the consistency between the patient and the agent. The test retest reliability of SF-HSTCM, EQ-5D-3L, mRS and BI were analyzed by group intercorrelation coefficient. The correlation coefficient between SF-HSTCM and EQ-5D-3L was used to evaluate the convergent validity. We applied non parametric tests to assess the discriminant validity of SF-HSTCM, by using severity of the disease, course of disease, the location of survey, gender and age as the grouping factors. The multiple linear regression method was used to explain the influence of sociological factors on the score of mRS and BI.ResultsFrom April 2015 to January 2016, a total of 205 people were investigated, of which 135 were male and 70 were female. The recovery rate and response rate were high. The internal consistency of Cronbach’s a among patients with stroke in the recovery phase was 0.895, test-Retest reliability (ICC) was 0.991. The correlation coefficient (ICC) between the patient and the agent was-0.03. The fitting index of SF-HSTCM confirmatory factor analysis show that X2/df was 2.51, RMSEA was 0.086, NNFI was 0.92, CFI was 0.93, and GFI was 0.79. The test-retest reliability result of EQ-5D-3L showed that, the number of questionnaires of two surveys which was exactly the same was 27, accounting for 90%of the total number, and the number of inconsistent questionnaires were 4, accounting for 10%. The test retest reliability (ICC) of mRS and BI were 0.993 and 0.997, respectively. The spearman correlation coefficients of SF-HSTCM score and mRS score were 0.425 and-0.449, respectively. The spearman correlation coefficients of EQ-5D-3L score with BI score and mRS score were 0.828 and-0.449, respectively. The spearman correlation coefficient of EQ VAS and BI score was 0.59, and EQ VAS score and mRS score of Spearman correlation coefficient is 0.60. The distinguish ability of SF-HSTCM among the patients with stroke in the recovery phase was good for all the grouping factor, including severity of disease, course of disease location of survey and age, except the gender factor. Multiple linear regression analysis showed that the factors, including age, usually have no exercise, smoking history and absence of diagnosed diseases might impact the BI scores.ConclusionSF-HSTCM has good internal consistency and test-retest reliability when it is applied for health evaluation among patients with stroke in the recovery phase. The measurement purpose of SF-HSTCM and EQ-5D-3L, EQ VAS, Bi and mRS are not entirely consistent, they have moderate to good homogeneity. SF-HSTCM has good distinguish ability for disease severity and age among patients with stroke in the recovery phase. EQ-5D-3L has good reliability, and it has a certain degree of homogeneity with mRS and BI. SF-HSTCM and EQ-5D-3L can be used to evaluate the health status of the patients with stroke in the recovery phase.