A Comparative Study On The Health Utility Value Measurement Of Breast Cancer Patients

BackgroundBreast cancer is the most frequent diagnosed cancer in women. The health burden of breast cancer is increasing in China, cases in China account for 12.2% of all newly diagnosed breast cancers and 9.6% of all deaths from breast cancer worldwide. Breast cancer is becoming a major health threat in developing countries, it not only brings serious burden of disease, but also caused great social loss and economic burden.Health economic evaluation is an important method and tool in health economics, which can provide support and information for decision-makers. As one of the main methods of health economics evaluation, cost-utility analysis has been recognized and widely used. In health sector, utility refers to the ability of health service programs to meet people s health need or desire, which represents the preference of a social or personal values or the weight of one health status, also known as health utility value. The measurement of health status can be classified into non-preference-based and preference-based. Because of the certain limitations about non-preference-based method, the health utility values are the most widely preferred preference-based concept in this study.Health utility value is a composite index which reflects health status of the individual, and refers to a will or preference of some kind of health status, the value is in the range of 0 (death) ～ 1 (full health). Internationally, the utility value measurement methods mainly include two types:direct measurement and indirect measurement.Direct measurement is based on the patient’s own health preferences, while the indirect measurement is based on the general population preferences.The health utility value measurement research in a variety of chronic diseases are increasing now, many studies both used the direct and indirect measurement methods, and compare the two types, but it is still relatively rare in breast cancer. The domestic studies have focused on the quality of life about breast cancer patients by using breast cancer-specific measurement scales and general quality of life measurement scales (such as SF-36, EQ-5D scale, etc.) and few studies involved in health utility values.The measurement research about health utility value of breast cancer patients is still in the blank in our country. Therefore, it is necessary to carry out the utility value measurement research of breast cancer patients in our country. It provides reference for clinical doctors and decision-making basis for policy makers, but also provides basic data support for the future cost-utility analysis.ObjectiveThe general objective of this study was to use direct measurement (TTO) and indirect measurement (SF-6D% EQ-5D-5L) to measure the health utility value of breast cancer patients in our country in order to compare the agreement between different methods, to explore the best measurement for china women and provide reference for the evaluation of health economics research in the future. Specific goals include:1. Using the TTO and SF-6D、EQ-5D-5L to measure health utility values of breast cancer patients;2. Measuring the health utility value of breast cancer patients by using three kinds of methods and exploring influencing factors;3. Comparing and analyzing the consistency of the three methods and to find the best measurement method for Chinese breast cancer women.MethodsThis study derived from two aspects:One is the collection of existing literature data; the second is the questionnaire survey to the breast cancer outpatients and inpatients in the Breast Surgery of Qingdao Municipal Hospital from November 2014 to February 2015. We interviewed 621 breast cancer patients to collect the personal basic information and clinical information. And we use TTO and SF-6D、EQ-5D to measure the health utility values.This study used the current international TNM breast cancer staging, SF-6D and EQ-5D-5L have been authorized by the R & D institutions, and we formed the final TTO on the basis of seeking for existing international literature design and the characteristics of breast cancer in our country. After obtained informed consent, we completed the survey with the help of the trained investigators.Analysis method mainly includes:1) the descriptive statistics, including mean,95% confidence interval and the median statistical indicators; 2) the single factor analysis, including Wilcoxon rank test and Kruskal Wallis test;3) the multivariate factor analysis, binary logistic regression analysis; 4) We used the Bland-Altman Plot method and ICC to check the consistency of three methods.ResultsThe basic information of respondents:We investigated breast cancer patients in total 621 people, including 13 people were excluded in the final analysis results because they could not be completed the questionnaire. Overall, the mean age was 48.0 (SD=9.6) and the mean duration of illness was 38.2 (SD=40.9); nearly one-third of the respondents have no qualifications or primary education certificate; in the occupational structure, enterprise workers (workers) account for more than one-third and the farmers/migrant workers account for about one quarter; A large number of the subjects (88.7%) were married; The average annual household income between 30000-80000 yuan account for nearly half of the respondents; The most respondents participated in health insurance for urban workers and NCMS, accounted for 47.2% and 43.3% health respectively; the average duration of respondents was 38.2±40.9 months; About one third of respondents with stage Ⅲ, the proportion of respondents with stage IV was the lowest; 54.1% respondents had ER/PR positive and 26.0% were ER/PR negative tumors. Menopause account for more than half of the respondents (54.9%),40.3% respondents with second and following years after primary breast cancer recurrence, the proportion of respondents with first year after recurrence just was 10.7%.The utility values of EQ-5D-5L:The mean health utility value of breast cancer patients using EQ-5D-5L is 0.83±0.16 (95%CI 0.82-0.85,the median is 0.87). The analysis results of influence factors using the EQ-5D-5L demonstrated that there are statistically significant in the subgroups such as residence, marital status, occupation, education level, health insurance, household income, and duration of disease, clinical stage, treatment options and disease status(P<0.05). According to the multivariate analysis, education level, home income and treatment options could influence the health utility value of breast cancer patients.The utility values of SF-6D: The mean health utility value of breast cancer patients using SF-6D is 0.65±0.13 (95%CI 0.64～0.66, the median is 0.62). The analysis results of influence factors using the SF-6D demonstrated that there are statistically significant in the subgroups such as marital status, occupation, education level, health insurance, household income, and treatment options and disease status(P<0.05). The multivariate analysis showed that residence, occupation, education level, health insurance, household income, and treatment options could influence the health utility value of breast cancer patients.The utility values of time trade-off:The mean health utility value of breast cancer patients using SF-6D is 0.80±0.25 (95%CI 0.77～0.82,the median is 0.90). The analysis results of influence factors using time trade-off demonstrated that there are statistically significant in the subgroups such as marital status, education level, household income, and treatment options and disease status(P<0.05). According to the multivariate analysis, education level, household income, and pathological diagnosis information could influence the health utility value of breast cancer patients.The comparison of EQ-5D-5L scale、SF-6D scale and time trade-off: There is a higher ceiling effect (28.62%) about EQ-5D-5L scale, while there is a floor effect (4.44%) about time trade-off. The overall intraclass correlation coefficient between the three methods was 0.46, which is moderate; There was a perfect agreement between EQ-5D-5Lscale and SF-6D scale, while fair agreement between EQ-5D-5L scale and time trade-off, and weak agreement or inconsistency between the SF-6D scale and time trade-off. According to the Bland-Altman Plot, the three methods are both inconsistency.ConclusionThe health utility values was different between three methods in 608 respondents, and there was moderate agreement between EQ-5D-5L scale、SF-6D scale and time trade-off. The study demonstrated that health utility values are affected by many factors, including education level, disease status and so on. In addition, the study suggests that SF-6D scale more suitable in Chinese breast cancer patients.