Disease burden of schizophrenia patients visiting a Chinese regional mental health centre

This study constructed a decision-analytic model to estimate the disease burden of schizophrenia by comparing the simulated health outcomes and direct medical costs associated with a created clinical cohort of Chinese patients with schizophrenia from a Chinese regional mental health center and the age and gender-matched Chinese general population. The Chinese healthcare system for mental diseases usually consists one regional mental health center in provincial capital city to provide tertiary care to schizophrenia patients at varied disease stages and community mental health rehabilitation institutes in the lower level cities and towns in the province for patients with relapsed disease. The created clinical cohort with schizophrenia from the regional mental health center in Hunan province was used to conduct a retrospective cohort study, which was used to develop the disease analytic model for the disease burden of schizophrenia. This regional mental health center provides tertiary care for schizophrenia patients with varied disease stages across Hunan province, which has over 60 million residents. A literature review and telephone interview of the relevant experts were conducted to estimate the model variables that could not be derived from the retrospective cohort study. The constructed decision-analytic model was used to perform base-case analysis and sensitivity analyses for the point estimations and uncertainty of the disease burden of schizophrenia, which was estimated using the differences in the simulated life years, quality-adjusted life years (QALY) and direct medical costs over lifetime between the created clinical cohort of schizophrenia patients and the age and gender-matched Chinese general population. The cost perspective in this study was healthcare payers. All costs in this study were converted to US dollars according to the currency exchange rate on 31 December 2017 (¥1 = US$0.15). This study was reviewed and approved by the ethics board of the Second Xiangya Hospital where the retrospective cohort study was conducted.

This study first reported the patient characteristics, treatment pattern, health resources utilization and direct medical costs from the included schizophrenia in the retrospective cohort study. The results of the base-case analysis, one-way sensitivity analyses and PSA in the developed schizophrenia disease burden model were reported subsequently.

To our knowledge, this study was the first research to quantify the disease burden of schizophrenia in China by simulating a real-world clinical cohort of schizophrenia patients over the entire lifetime in a decision-analytic model. Through comparisons with the age and gender-matched model control cohort with general Chinese population, the Chinese patients with schizophrenia could expect to lose about 18 years survival time and triple the lifetime direct medical costs under current treatment pattern, which mainly included oral antipsychotic medications. Thus, this study confirmed that schizophrenia has become a major health problem with a heavy disease burden that requires attention and actions to address the unmet medical needs of Chinese schizophrenic patients.

As the provincial mental healthcare center, our hospital received referred schizophrenia patients across the Hunan province with a population of over 68 million. The schizophrenia patients managed in our hospital had highly comparable demographics (young age and evenly distributed gender) and socioeconomic status (high unemployment rate and more patients living in the county) as the Chinese schizophrenia patients included in previous observational studies [22,23]. Thus, the schizophrenia cohort in this study should well represent the Chinese schizophrenia patients and the estimated disease burden of schizophrenia from the simulation of the created schizophrenia cohort should be comparable as the disease burden of the schizophrenia patients living in other Chinese provinces.

The estimated disease burden of Chinese schizophrenia was extremely heavy as the reduced survival time was as much as 20.6 years. After the review of the patient characteristics, treatment pattern and disease natural history, the heavy disease burden of schizophrenia in Chinese patients was likely to strongly correlated with the following factors. First, the Chinese schizophrenia patients were relatively young with a mean age of 31.1 years. Since schizophrenia is not curable, the longer duration of schizophrenia increases mortality. Second, the longer duration of schizophrenia increases the risk of treatment relapse, which reduces the quality of life and increases hospitalizations, the two main contributors to the disease burden of schizophrenia. Third, the current treatment pattern was predominated by oral antipsychotic medications that were associated with poor treatment compliance [24]. Thus, the high risk of treatment relapse due to poor treatment compliance could further worsen the disease burden of schizophrenia by increasing both mortality and direct medical costs. Because the demographics of schizophrenia patients could not be changed, how to reduce the risk of treatment relapse in schizophrenia patients could be the main direction to reduce the disease burden of schizophrenia in China.

Changing the current antipsychotic treatment pattern could be an effective solution to address the heavy disease burden of schizophrenia. Even though LAI treatment was introduced in China in 2012, LAI was only prescribed in 5% of the included patients in this study. Thus, the introduction of LAI treatment did not substantially change the treatment landscape of schizophrenia in China. Based on the synthesized evidence on the treatment effects of identified main oral antipsychotic medications and paliperidone palmitate injection, which was the only LAI identified in the included patients, the hazard ratio of treatment relapse associated with LAI relative to oral antipsychotic medications was 0.367. The superiority of LAI over oral antipsychotic medications for the reduced risk of treatment relapse was expected to substantially reduce the disease burden of schizophrenia. However, the reduction of the disease burden associated with the increased treatment rate of LAI was modest in the one-way sensitivity analysis likely due to the short treatment duration of LAI. The treatment duration of LAI in our real-world patient cohort was only 8.1 months because of the lack of reimbursement support. None of LAIs were listed for reimbursement in China during the study observation time. The schizophrenia patients and their families in China usually have limited income and they were unlikely to afford the high drug acquisition costs of the approved LAI. Thus, the improving the patient access to LAI through reimbursement could be an effective solution to improve health outcomes in Chinese schizophrenia patients.

The estimated lost survival time associated with our schizophrenic patients in the base-case analysis was 20.6 years, which was in the range of the reported lost life years associated with schizophrenia in the literature (14.5–25 years) [25,26]. Different from other chronic diseases, the impact of schizophrenia on survival was not well recognized and previous disease burden studies for schizophrenia have mainly focused on disability and costs. Our study estimated the mortality risk associated with schizophrenia in Chinese patients irrespective of residence location. In a real-world setting, the schizophrenia patients living in the rural area could be associated with much higher mortality risk than those living in the urban cities because of higher suicide rates [27], higher prevalence of other mental diseases and poorer management [28]. Given the fact that over half of our patients could live in a rural area, the lost life years associated with our schizophrenic patients could be even higher than our estimates.

Similar to previous economic burden studies, our study confirmed the substantial economic burden of schizophrenia from the perspective of healthcare payers. Our retrospective cohort study reported that the medical costs associated with one hospital episode were ¥13,769, which was 55% above the average costs of one hospital episode in China in 2017 [20]. According to the one-way sensitivity analyses, the economic burden of schizophrenia in China was mainly driven by the direct medical costs associated with the local mental health rehabilitation institutes, which have been well established across China to control the violence and threaten of mental diseases to the society [29]. The telephone interviews with these institutes found that LAI treatment was not used in the community at all due to the lack of access. Thus, the lack of LAI treatment in the community care for schizophrenia patients could further increase the risk of disease relapse and cause patients to heavily use local mental health rehabilitation institutes.

Even though this study leveraged real-world patient characteristics, treatment pattern and direct medical costs associated with over 3000 schizophrenia patients from a Chinese tertiary care hospital to maximize the robustness of the data sources for the development of schizophrenia disease burden model, the challenges and limitations associated with model development could still damage the reliability and validity of the estimated disease burden of schizophrenia. First, selection bias could be associated with the created study cohort, which could include more patients with newly diagnosed schizophrenia and patients living in the provincial city where the hospital located. The younger age associated with newly diagnosed schizophrenia patients could overestimate the disease burden because they would have a longer disease duration. Additionally, the higher medical costs associated with mental health rehabilitation institutes in the provincial city could overestimate the economic burden of schizophrenia. Second, reimbursement coverage had a substantial impact on treatment selection. LAI was not listed for reimbursement during the observation time defined in the retrospective cohort study. Recently, LAI has been listed for national reimbursement formula and the use of LAI for schizophrenia could be much improved. Thus, the measured disease burden of schizophrenia in our study should be interpreted under the treatment pattern without reimbursement coverage of LAI. Third, this study did not conduct chart review to extract more patient clinical characteristics, such as disease history and previous treatments, which could impact our data analysis regarding the differences in the measured outcomes associated with LAI and oral treatment. Third, the model assumptions in the decision analytic model could have a strong impact on the measured disease burden. For example, our study assumed that patients with more than two treatment relapses were managed in local mental health rehabilitation institutes. This assumption could underestimate the direct medical costs associated with tertiary care as some included patients had more than two hospital episodes in the retrospective cohort study. The management pathways associated with schizophrenia patients in real-world settings should be investigated in the future to further improve the robustness of our decision-analytic model. Fourth, our study still used a large amount of non-Chinese evidence to estimate the model variables that could drive the disease burden of schizophrenia. For example, the treatment efficacies of antipsychotic treatments were mainly based on studies from high-income countries. The quality of life associated with Chinese schizophrenia patients was based on a utility formula developed from a French study. Even though this study used conversion methods to improve the validity of these estimations in Chinese patients, future studies are still needed to validate these conversions and further confirm these estimations. Finally, the medical costs associated with local mental health rehabilitation institutes were based on the rough estimation from the telephone interview. These cost estimations might not take into account the additional medical costs associated with managing other mental disorders, which were prevalent in our study cohort as well. Thus, the economic burden of our schizophrenia patients could be underestimated. Future studies are still needed to obtain real-world cost data from those institutes to improve the accuracy of the estimated economic burden of schizophrenia.

In summary, this study used a real-world clinical cohort of Chinese patients with schizophrenia and a decision-analytic model to demonstrate the disease burden of schizophrenia under treatments, mainly oral antipsychotic medications. Based on the model simulation of these real-world patients, schizophrenia was associated with substantial disease burden by reducing survival time over 20 years and tripling the lifetime direct medical costs when compared with age and gender-matched general Chinese population. Because the mortality and high medical costs associated with relapsed schizophrenia mainly drove up to the disease burden of schizophrenia, improving patient access and compliance to treatment with the least risk of treatment relapse, such as LAI, could be an effective method to reduce the disease burden of schizophrenia in Chinese patients.

Q Zhong, W Chen, J Chen and H Tang formulated the research idea and developed the study protocol. Q Zhong, Y Tana, H Huang, J Huang, S Li, Z Teng, M Shen, C Wu, L Wang, S Cao, H Wu and H Tang conducted patient identification and data extraction at the study site. H Huang conducted the data analysis and developed the disease burden model of schizophrenia under the supervision of W Chen, J Chen and H Tang. Q Zhong, W Chen, J Chen and H Tang developed the manuscript. The submitted manuscript has been critically reviewed and approved by all authors.

This study was sponsored by Xian Janssen Pharmaceutical Ltd. W Chen owns a consulting firm receiving industry funds to conduct health economics and outcomes research. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

This study was reviewed and approved by the ethics board of the Second Xiangya Hospital where the retrospective cohort study was conducted.