Propensity score (PS) is increasingly used to establish causal relationships in observational data. PS is a valuable tool because health economics research such as economic evaluation of health guidance for patients with metabolic syndrome must rely on observational data such as health insurance claims. PS is the probability that a case will be assigned to treatment. If cases from treated and untreated groups of the same PS are compared, then the comparison becomes a quasi-randomized trial and hence establishes a causal relationship. Covariates must be those preceding treatment and outcome. Health check data are appropriate as covariates for PS calculation because health checks precede health guidance and health insurance claims. However, PS requires a certain condition of "strongly ignorable" treatment assignment for it to be valid. Selecting covariates that best fulfill the strongly ignorable assignment constitutes the most difficult task.

The Hosmer-Lemeshow goodness-of-fit test result, as well as the discriminatory power measured by the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, serve as the two main indicators for evaluating the appropriateness of PS calculation. Both indicators should be close to 1 but if AUC is too close to 1, there may not be enough overlapping of PS to secure a good sample size. After PS is properly calculated, analyses will be conducted by 1) matching, 2) subclassification and 3) covariance analysis. One-to-one matching is the first choice but it may reduce the sample size because not all cases can be matched. Subclassification can use all the cases but interpretation may be difficult when the results are not consistent across subclasses. Covariance analysis simply uses PS as an additional explanatory variable in the regression model. A Tobit regression model should be employed when the outcome is health care cost, which will be a zero-truncated value. Although PS programs are readily available, the most important task, selecting covariates, is left to the researcher's personal skills and knowledge. Therefore, researchers must develop such skills and knowledge to use PS appropriately. In observational studies, selection of control groups is crucial. Results may be manipulated when control groups are arbitrarily selected. Hence, PS requires high ethical standards and morals among researchers. To ensure research integrity, it would be better to have different researchers perform the PS calculation and outcome measurement tasks. It is also recommended that multiple researchers analyze the same datasets and combine them using systematic review.

This paper investigates the distribution patterns of medical care and long-term care expenditures in Japan by using the complete set of claim bill data provided by the public insurers of Fukui Prefecture from fiscal years 2003-07. Our research confirmed the well-known fact that a major portion of medical care expenditures are concentrated on a few heavy users. The expenditures on long-term care services, on the other hand, are more equally distributed among users.

We also found that medical care and long-term care expenditures have a weak, but negative relationship, because heavy users of inhospital services and those of institutional long-term care were mutually exclusive. Excluding these heavy users turned the association null, or a weakly positive one. These associations were confirmed by use of seeming unrelated regression analysis.

Finally, we analyzed the longitudinal change in the distribution of expenditures by utilizing the sample of five­year survivors in the dataset. Results showed that the expenditures for the top 10% of high-cost medical care users tend to wane considerably within a relatively short period, while that of high-cost long-term care users remains at a high level for a longer period.

One of the major goals of establishing the Long-term Care Insurance System (LTC) was to resolve the problem of social hospitalization, unnecessary inpatient service use to compensate for limited access to nursing care. This paper investigates whether LTC has alleviated social hospitalization by analyzing the claim bill data of elderly beneficiaries（≤70 years) enrolled in Fukui Prefecture's National Health Insurance Program for fiscal year 2007.

A commonly used definition of social hospitalization often relies on the length of hospitalization, though it is susceptible to upward bias in the estimation of medical expenditure because it can include those patients with a need of chronic medical care. Instead, we follow Fukawa (1995) and identify cases of social hospitalization as those with per-diem expenditure less than 1.1 times of basic inhospital service expenditure.

In sum, we find the problem of social hospitalization does remain, but at the magnitude of a half to two-thirds compared to Fukawa's previous estimation in 1993. Specifically, social hospitalization accounts for 7.5% to 18.4% of total inpatient cases, 1.9% to 4.6% of the total number of health care insured patients, 6.9% to 23.5% of overall inpatient payments, and 3.2% to 10.9% of total health care expenditures. Hence, we conclude that the introduc­tion of the LCT system alleviated the problem of social hospitalization to some extent, though it still remains prevailing in Japan.

Using the long-term care insurance claim data of Fukui Prefecture, this study investigates how newly provided long-term care services affect the long-term care expenditures of residents.

Until March 2007, 4 towns in Fukui Prefecture, Eiheiji, Ikeda, Mihama, and Ohi, did not have outpatient rehabilitation providers for long-term care. In April 2007, an existing provider of a long-term care health facility in Ohi expanded its services to include outpatient rehabilitation for long-term care.

We regarded samples from Eiheiji, Ikeda, and Mihama as the control group, and those from Ohi as the treatment group. Our analysis compared both groups to determine the respective averages of long-term care expenditure in 2006 and 2007. To measure the effect of service expansion on long-term care expenditures, we estimated the ordinary least squares with a difference-in-differences (DID) specification.

Two findings are evident from the analysis. First, the average expenditure of outpatient rehabilitation for long­-term care increased, while that of in-home long-term care services decreased in Ohi from 2006 to 2007, but not in other municipalities. The econometric analysis using DID showed that the introduction of outpatient rehabilitation for long-term care increased rehabilitation service expenditures by 1200 yen per month, and decreased in-home long-term care service expenditures by 5500 yen per month. Second, the expenditures on some in-home long-term care services increased while other decreased or unchanged, suggesting the introduction of outpatient rehabilitation affected the choice of service mix.

The result implies a possibility that a wider set of service options available enabled the users to efficiently choose service mix and to save long-term care expenditures.

We employ the original Internet survey data to examine the impacts of the recently largest cigarette price increase in October 2010 and of the unexpected supply shock to the tobacco industry caused by the 2011 off the Pacific Coast of Tohoku Earthquake (the Great East Japan Earthquake) on smoking behavior. We find that these shocks have statistically significantly negative impacts on smoking behavior. Our estimates suggest that the increase in cigarette price in October 2011 significantly reduces smoking probability by 0.5 percent, the number of cigarettes smoked per day by 7.0-7.6 percent, and the nicotine intake per day by 0.70-0.74 mg (by 0.38-0.42 mg for a continuing smoker). Likewise, the unexpected supply shock from March to July 2011 significantly reduced smoking probability by 0.3 percent, the number of cigarettes smoked per day by 2.7 percent, and the nicotine intake per day by 0.33-0.34 mg (by 0.18 mg for a continuing smoker).

By using panel data from Japanese young women (age, 20s to 40s) from the Japanese Panel Survey of Consumers 2003-2008 (JPSC), this paper analyses how social status affects self-reported health status. Because women in the 20s-40s age range play multiple roles such as workers, wives, mothers and caregivers, this paper takes such dynamic changes in work and health status into consideration by employing dynamic panel probit models to obtain a consistent estimator following Woolridge. The results proved that health status in a previous period, years of education, being homemakers, and being employed were significant predictors of current health status.