| Object:To evaluate the implementation feasibility,reliability and validity of the EQ-5D-5L questionnaire in patients with cervical precancerous lesions,to test the responsiveness of the questionnaire in patients newly diagnosed and underwent surgical treatment,and to develop minimal clinically important difference(MCID),to assess the validity of the established MCID simultaneously.Through these analyses to provide convenient and effective measurement tool for cervical precancerous patients,and then used for health economic evaluation,as well as to provide a minimum threshold of clinical significance for the application of the questionnaire in judgment of clinical efficacy.Methods:Taking gynecology department of a grade-A tertiary hospital in Xinjiang as the investigation location,and the EQ-5D-5L questionnaire as the target measurement tool.Patients with cervical precancerous lesions diagnosed by cervical histopathology were the survey subjects.The outpatient department survey used the cross-sectional study for implementation feasibility,reliability and validity evaluation,and inpatient department survey used the longitudinal study for responsiveness assessment and MCID formulation.The outpatient department and the inpatient department conducted follow-up investigations 6to 8 days and 30 to 37 days after the first survey,respectively.Global rating of change questionnaire(GRCQ)was external anchor.Data collection was completed from November 2018 to August 2019.The acceptance rate,completion rate and completion time of questionnaire were used to evaluate the implementation feasibility of the EQ-5D-5L questionnaire.The test-retest reliability coefficient,intra-class correlation coefficient(ICC)and Bland-Altman diagram were used for evaluating the retest reliability.The split-half reliability and Cronbach's ? coefficient were for internal reliability evaluation.The Pearson correlation analysis of utility index(UI)and the scores of each dimension of the descriptive system and EQ-VAS were conducted for content validity and convergent validity.The discriminating validity was assessed by the results of Mann-Whitney U Test of UI and EQ-VAS in patients with different course of disease groups.The Wilcoxon Signed Rank Test,effect size(ES)and standardized response mean(SRM)were used to evaluated the responsiveness of the EQ-5D-5L questionnaire.MCID was triangulated using the distribution-based method,anchor-based method and instrument-defined method,and the validity of MCIDs were evaluated at the individual level and group level with 95%,90% and 80% confidence intervals(CIs)respectively.Results:1.248 outpatients and 50 inpatients were included in the analysis,and the average age were 45.86±9.24 years and 44.76±8.72 years,respectively.The outpatient and inpatient patients were mostly 40 to 50 years old,married,junior school education level,participate in medical insurance for urban workers,normal range body mass index(BMI),and annual household income was 30000 to 80000 yuan.The proportion of retired or housewives in the outpatient patients was the highest(29.00%),and the patients in the inpatientaccounted for the most was farmers(26.00%).Regarding disease-related characteristics,most patients in outpatient and inpatient departments seek a doctor due to problems in physical examination or screening,and patients with a disease duration of 0 account for 50.00% and 56.00%,and patients with high-grade squamous intraepithelial lesion account for 72.10% and 94.00%,respectively.49.60% of outpatients were untreated at the first survey,while all inpatients used surgery as the main treatment during hospitalization.2.The acceptance rate and completion rate of the EQ-5D-5L questionnaire in outpatients were both100%,and the average time to complete the questionnaire was 1.27±0.49 minutes,which indicated that the questionnaire has good implementation feasibility.The test-retest reliability coefficients of UI and EQ-VAS were 0.644 and 0.599,and ICCs were both ?0.75.The Bland-Altman diagram also showed good agreement between the two times of measurements.The split-half reliability and standardized total Cronbach's ?coefficient were 0.655 and 0.626,suggesting that the questionnaire has general internal reliability.The correlation coefficients between the five-dimensional scores and UI of the EQ-5D-5L questionnaire were-0.548,-0.613,-0.646,-0.565,and-0.682,which showed strong correlation;EQ-VAS was also moderately correlated with UI and anxiety/depression dimension,with correlation coefficients of 0.355 and-0.485;There were significant differences in UI and EQ-VAS between different course groups(all P<0.05),indicating good reliability.3.Inpatients who underwent surgical treatment,the UI and EQ-VAS were increased by 0.025±0.060 and6.92±14.01 after treatment(all P<0.05),and the effect size was small to moderate,which suggests that the questionnaire has good responsiveness.4.The range of the MCID formulated for UI and EQ-VAS by the distribution-based method were 0.025 to 0.058 and 6.99 to 11.93 respectively;The MCID developed by the anchor-based method for UI and EQ-VAS were 0.041 and 3.12;The MCID of UI obtained based on the instrument-defined method was0.064.5.Among MCIDs developed in this study,the MCID of UI and EQ-VAS that could correctly judge whether 95% of group level patients' health is improved are at least 0.025 and 3.12.The MCID of at least0.058 for UI with 80% CI to suggest whether the health of individual level patients is improved,but the MCID established for EQ-VAS can not be used for individual level patients.Conclusion:The EQ-5D-5L questionnaire has good implementation feasibility,reliability,validity and responsiveness in cervical precancerous patients,which can be used as a health measurement instrument and health economic evaluation tool for cervical cancer screening programs.The MCID developed for the UI can be used for judgment of clinical efficacy at the individual level and group level,while the MCID for EQ-VAS can only be used for group level patients.We recommend that future studies could base on large sample sizes to further develop MCID for EQ-VAS that can be used at the individual level.And MCID can also be developed in other clinical settings such as different interventions,different follow-up intervals and different directions of health change. |
PDF Full Text Request