| Patient-reported outcomes and listening to the true feelings of the patient are the hot spot in cancer research both in China and abroad recently, especially the pediatric patients reported outcomes. Self-report is the standard to evaluate children’s health status, while because of the age differences and different cognitive level, so parents need to report their health related information on behalf of their children in some special circumstances(i.e., serious disease, the presence of cognitive dysfunction). Compared to children’s self-reported outcomes, parent-reported outcomes show high correlations with clinical indicators, and it directly affects the medical decision-making and clinical resource utilization. Therefore, both the child self-report and parent-report outcomes have its own value. Based on the development of Chinese Version of Pediatric Patient-Reported Outcomes Measurement Information System(PROMIS)(PROMIS)-Child Version(C-C-Ped-PROMIS),which is one part of a large international standardization scale system developed by National Institutes of Health(NIH), cooperated with tutor Dr. Pamela S Hinds’ research team, we aimed to introduce the Chinese Version of Pediatric PROMIS-parent Version(P-C-Ped- PROMIS), thus to finish the measurement analysis(i.e., reliability, validity, responsiveness and minimal important differences) of two versions, and try to conduct the analysis of measurement invariance of C-C-Ped-PROMISacross age, gender and culture. The introduction and measurement test of full version of PROMIS, provides a strong evidence for the credibility, validity, sensitivity and comparability of this system, and guarantee its effective application in the field of childhood cancer. It has a great influence in improving the level of cancer care for Chinese children, promote the development of the field of health outcomes of children with cancer and clinical communication and research collaboration internationally. ObjectivesThe objectives of this study are: 1. to develop the Chinese version of parent proxy reported PROMIS measures and examine its reliability and validity in children with cancer in China; 2. to evaluate the responsiveness and clinical minimal importance difference of Chinese version of pediatric and Parent Proxy reported PROMIS in the children with cancer who are experiencing chemotherapy; 3. to examine the measurement invariance of C-C-Ped-PROMIS eight measures across age and gender; 4. to complete cross-cultural measurement invariance test of seven measures of C-C-Ped-PROMIS in cooperation with American research team; 5. to compare the reported health outcomes between children-report and parents-report PROMIS in chemotherapy stage, and to explore their consistency and differences of health outcomes at different time points. MethodsThe study is divided into three parts:1. Measurement Examination of C-C-Ped-PROMIS.(1) Responsiveness and minimum important differences(MID): C-C-Ped-PROMIS eight measureswere used to investigate children’s health status with cancer from 8 to 17 years at the stage of chemotherapy(pre-chemotherapy T1, chemotherapy T2, post-chemotherapy T3, three points), a paired t-test was used to analyze responsiveness of C-C-Ped-PROMIS at T1-T2 and T2-T3; The criterion method and distribution method were used to calculate the two-stage C-C-Ped-PROMIS minimum important difference.(2) Measurement invariance across age and gender: multiple indicators multiple causes(MIMIC) model was used to evaluate 232 children’s health outcomes of C-C-Ped-PROMIS eight scales, and analyzing differential item functional(DIF) of each item was analyzed;(3) Cross-cultural Measurement Invariance: Cooperated with Pamela S Hinds team, multi-group categorical confirmatory factor analysis(multi-group CCFA) was used to analyzehealth outcomes of children from 8 to 17 years old between China and the US, and the cross-language measurement invariance between Chinese and English version Pediatric PROMIS seven measures was analyzed;2. Development and measurement examination of P-C-Ped-PROMIS.(1) Chinese version development and reliability and validity: The functional assessment of chronic illness treatment translation methodology(FACIT) was used to complete the translation ofP-C-Ped-PROMIS eight measures, and conducted cultural validation through parents’ cognitive interviews; cross-sectional study design was used to investigate 209 parents and to analyze construct validity, internal consistency reliability, scale model estimation reliability of P-C-Ped-PROMIS eight measures.(2) The responsiveness and the minimum important difference: P-C-Ped-PROMIS eight measures were used to conduct a longitudinal survey of 115 parents whose children were 5 to 17 years old and at the chemotherapy stage(T1, T2, T3), and the statistical analysis methods were same to the test of children’s version;3. Comparative study of health outcomes between C-C-Ped-PROMIS and P-C-Ped-PROMIS. Based on the paired longitudinal data of 75 parents whose children were 8 to 17 years old,(1) The status and changing trend of health outcomes between C-C-Ped-PROMIS and P-C-Ped-PROMIS were described and analyzed;(2) to analyze health outcomes differences were analyzed by using a paired t-test or Wilcoxon signed rank test at each time points;(3) The outcome consistency between C-C-Ped-PROMIS and P-C-Ped-PROMIS was analyzed by using the intraclass correlation coefficient(ICC) at three time points(T1, T2, T3) in different age,gender and diseases. Results1. Measurement examination result of C-C-Ped-PROMIS:(1) Inthe stage T1-T2, five measures of C-C-Ped-PROMIS had good responsiveness(effect size(ES): 0.19 ~ 0.51, standardized response mean(SRM): 0.27 ~ 0.66), including depression, fatigue, pain interference, physical function-mobility and physical function-upper extremity, the MID with patient self-rated subjective anchor were 3.91 ~ 7.53; MID value based on distribution were between 9.36 ~ 24.59; In T2-T3 stage, except peer relationship, other measures had good responsiveness(ES: 0.18 ~ 0.40, SRM : 0.26~ 0.62), MID evaluated based on subjective anchor were 5.57 ~ 11.15, and MID were 8.55~20.46 based on distribution.(2) The DIF across gender was not existed in C-C-Ped-PROMIS eight measures items, except 6 items had DIF across the age, DIF of other items was not statistically significant(P> 0.05);(3) Except physical functions-upper extremity, the symptoms of depression, anxiety, fatigue, pain interference, peer relationships, physical function-mobility of C-C-Ped-PROMIS was achieved cross-cultural measurement invariance hypothesis after testing of invariance parameters at different levels;2. Development and measurement examination of P-C-Ped-PROMIS(1)The translation of P-C-Ped-PROMIS eight measureswas completed by a team of Chinese and American experts, there was no need to modify the items after the cognitive interviews; the reliability and validity results showed that correlation coefficients of eight measures item were more than 0.6, each single factor model fit good, Cronbach’α values were between 0.770 ~ 0.939, model estimated scale reliability were between 0.835 ~ 0.940;(2) eight measures responsiveness test of P-C-Ped-PROMIS results were equivalent to the CC-Ped-PROMIS, in the T1-T2 stage, symptoms of depression and other four measures had good responsiveness(ES values, 0.18 ~ 0.52, SRM: 0.22 ~ 0.67); MID based on parentssubjective anchor were 3.14 ~ 5.33, with MID calculated by distribution were 9.15-23.53; In the T2-T3 stage, except peer relationship, other seven measures had good responsiveness(ES: 0.21 ~ 0.42, SRM: 0.29 ~ 0.47); The MID based on the subjective anchor value were 2.56 ~ 5.23; MID based on distribution method were 11.00~22.75.3. Comparative study of health outcomes between C-C-Ped-PROMIS and P-C-Ped-PROMIS.A total of 75 parentswhose children were 8 to 17 years old were investigated;(1) Choosing health outcomes of United States children population as a reference, the average score of peer relationships, mobility, and upper extremity function average level of C-C-Ped-PROMIS and P-C-Ped-PROMIS were lower than norm(P<0.05); Expect fatigue during chemotherapy(P<0.05),children-reported anger,anxiety, pain interference were lower than the reference(P<0.05);parents-reported anxiety, fatigue and pain interference were higher than the normal(P<0.05), except peer relationship measure, health outcomes at different time point of chemotherapy showed significant difference between C-C-Ped-PROMIS and P-C-Ped-PROMIS;(2) Except peer relationship measure, the difference of seven health outcomes(depression, anger, anxiety, fatigue, pain affect, physical function-mobility and physical function-upper extremity function) of C-C-Ped-PROMIS and P-C-Ped-PROMIS were statistically significant(P<0.05);(3) Except peer relationship of a poor correlation(ICC<0.4) before and during chemotherapy stage, the other health outcomes of C-C-Ped-PROMIS and P-C-Ped-PROMIS had a moderate or strong correlation(ICC> 0.4).Except small correlations,most the C-C-Ped-PROMIS and P-C-Ped-PROMIS had a moderate or strong correlation with different values in patients withdifferent age,gender and disease group, Conclusions1. The P-C-Ped-PROMIS had good reliability and validity, and it can be used to measure health-related outcomes assessment in children with 5 to 17 years old in China;2.Except peer relationship, depression, anxiety, anger, fatigue, pain interference, physical function-mobility and physical function-upper extremity measures of C-C-Ped-PROMIS and P-C-Ped-PROMIS have good responsiveness and MID; This result can provide evidence for PROMS measure clinical application;3.Except the invariance of physical function-upper extremity across age, other seven measures of C-C-Ped-PROMIS can be used for comparative effectiveness of different health outcomes of cancer children across age and gender;4. Except physical function-upper extremity and anger measure, other six measures of C-C-Ped-PROMIS have cross-cultural measurement invariance, and it can be used for compare the effectiveness of different health outcomes of cancer children between China and America;5. Chemotherapy has a great impact on the children’s physical, psychological and social health. The significant difference of physical and psychological status,fatigue, pain and social health in different time point are existed between C-Ped-PROMIS and P-C-Ped-PROMIS and with the moderate and high degree of consistency at different time point; Therefore, children’s self-report should be the primary reference on the prewise of children is able to report themselves, otherwise, the parents-reported health outcomes can also provide valuable information for clinical practice and scientific research. |
PDF Full Text Request