Background As the main cause of cancer death, lung cancer imposes seriously health and economic burdens on individuals, families, and the health system. In China, there is no national study analyzing the hospitalization expenditures of different payment methods by lung cancer inpatients. Based on the 2010–2016 database of insured urban resident lung cancer inpatients from the China Medical Insurance Research Association (CHIRA), this paper aims to investigate the characteristics and cost of hospitalized lung cancer patient, to examine the differences in hospital expenses and patient out-of-pocket (OOP) expenses under four medical insurance payment methods: fee-for-service (FFS), per-diem payments, capitation payments (CAP) and case-based payments, and to explore the medical insurance payment method that can be conducive to controlling the cost of lung cancer. Method This is a 2010–2016, 7-year cross-sectional study. CHIRA data are not available to researchers after 2016. The Medical Insurance Database of CHIRA was screened using the international disease classification system to yield 28,200 inpatients diagnosed with lung cancer (ICD-10: C34, C34.0, C34.1, C34.2, C34.3, C34.8, C34.9). The study includes descriptive analysis and regression analysis based on generalized linear models (GLM). Results The average patient age was 63.4 years and the average length of hospital stay (ALOS) was 14.2 day; 60.7% of patients were from tertiary hospitals; and 45% were insured by FFS. The per-diem payment had the lowest hospital expenses (RMB7496.00/US$1176.87), while CAP had the lowest OOP expenses (RMB1328.18/US$208.52). Compared with FFS hospital expenses, per-diem was 21.3% lower (95% CI = -0.265, -0.215) and case-based payment was 8.4% lower (95% CI = -0.151, -0.024). Compared with the FFS, OOP expenses, per-diem payments were 9.2% lower (95% CI = -0.130, -0.063) and CAP was 15.1% lower (95% CI = -0.151, -0.024). Conclusion For lung cancer patients, per-diem payment generated the lowest hospital expenses, while CAP meant patients bore the lowest OOP costs. Policy makers are suggested to give priority to case-based payments to achieve a tripartite balance among medical insurers, hospitals, and insured members. We also recommend future studies comparing the disparities of various diseases for the cause of different medical insurance schemes.

In this work we study user controlled table-totext generation where users explore the content in a table by selecting cells and reading a natural language description thereof automatically produce by a natural language generator. Such generation models usually learn from carefully selected cell combinations (clean cell selections); however, in practice users may select unexpected, redundant, or incoherent cell combinations (noisy cell selections). In experiments, we find that models perform well on test sets coming from the same distribution as the train data but their performance drops when evaluated on realistic noisy user inputs. We propose a fine-tuning regime with additional user-simulated noisy cell selections. Models fine-tuned with the proposed regime gain 4.85 BLEU points on user noisy test cases and 1.4 on clean test cases; and achieve comparable state-of-the-art performance on the ToTTo dataset. 1

No abstract

No abstract