In order to decrease disparity in health status, the development of social capital through improved mutual cooperation would be more effective than the expansion of health services. This is the premise for the “new public health movement” aimed at creating a society which supports healthy activities. However, the effectiveness of such social engineering has not been adequately evaluated. Moreover, some have proposed that improving the educational system for young children and their parents would be more effective in decreasing disparity.

Disparity in health care is a policy issue distinct from disparity in health status. A lack of policy intervention to constrain disparity in health care threatens the effectiveness and safety of health services, the prevention of personal bankruptcy from health care costs, and the containment of public expenditures in health care. The reason lies on the supply side and the demand side.

On the supply side, health care is provided in unique and unrepeatable situations, so it is very difficult to evaluate whether the services provided had been “appropriate” or “inappropriate”. Thus, services tend to be provided, particularly in a life or death situation, even if its effectiveness is doubtful and its safety has not been adequately evaluated. Moreover, once a technology comes to be used, its use will expand from the appropriate area for which it was originally intended.

On the demand side, an individual usually does not know when the demand would arise or how much it would cost Therefore, to avoid risk, the demand for health care translates into a demand for health insurance. In the insurance market, the most attractive product is the one that offers the best benefits for the lowest premium. There are two ways to do it First, the insurer can act as an agent for the consumer and contract with providers who can deliver high quality services efficiently. Second, the insurer can adjust premiums and benefits to correspond to the risk profile of the consumer. The second way is far easier, and so market principles in the health insurance market lead to more disparity, and increases in personal bankruptcies for the uninsured and those with inadequate coverage. Moreover, since what is considered as “appropriate” would inherently expand to meet the criteria of the most generous plan, health care expenditures in total would increase.

Although Japan's health care system is basically equitable and has relatively low costs, it has not necessary been positively evaluated by the general public for the following underlying reasons. The adjustments for age and income among plans are unclear and not refined, the measures to alleviate patients' out-of-pocket expenses are inadequate, the revision process that sets fees and prices lacks transparency, and quality assurance is inadequate. These issues should be resolved. However, the major effort should be placed on refocusing the current policy discussion about the inter-generational disparity of cost burden to the regional disparities of service benefits. After the national government has adjusted for the disparity of cost burden arising from differences in age and income among the prefectures through subsidies, the relationship between premiums and benefits in each prefecture would become clear, providing an incentive for prefectural governments to evaluate, redesign and monitor the health delivery system for efficiency and to reflect the preferences of their residents.

The purpose of this paper is to investigate the relationship between the health care costs of employees on one hand, and their self-reported health and stress levels, on the other. For our analysis, we have used a combined dataset of personal information obtained from our questionnaire on life-style and panel data of health care costs of four thousand volunteering employees in a particular Japanese firm. For the sake of simplicity, we have limited our sample to lifetime non-smokers in this paper.

First, we have analyzed the factors that influence the subjective health rating. Our findings suggest that, while (1) sex and age do not affect their ratings, (2) many so-called “life-style diseases”, such as hypertension, stomach cancer, stroke, alcohol-dependency, substantially lower them. Since most of the employees in our sample are quite healthy and free from these diseases, however, in explaining the variation in their subjective health ratings, (3) such common complaints as stiff-shoulders, lower-back pains, pains in joints of arms and legs, numbness in arms and legs, and overall-fatigue, play far more important roles than these serious diseases, and (4) differences in their self-reported stress levels seem to be the single most important factor.

Secondly, we have analyzed the effect of subjective health rating on the health care costs by estimating the health care cost equation and subjective health rating equation simultaneously. We have relied heavily on the rich personal information supplied by our questionnaire for our instruments. According to our result, (4) a one-level improvement of subjective health rating reduces the health care costs by more than 30 percent, and (5) a one-level increase in the self-reported stress reduces the subjective health level by 0.3. Thus we can conclude that a one-level increase in self-reported stress increases the health care costs by almost 10%.

In general, it seems fair to say that Japanese public health insurance system has been primarily designed to deal with physiological changes of patients. But for stress-related changes, psychological approaches or therapies may turn out to be far more efficient. Some firms are reported to have already started to provide psychologist's services to their employees, and, if our analysis is correct, we can expect some concrete results in controlling health care costs. Moreover, in a recent study on the subjective health of the elderly population, depression is found to be an extremely important determinant. If that is the case, we should start investigating whether or not we should allocate more of our health care resources to psychological therapies, particularly in the care of our elderly.

In this paper, the length of stay at the hospital was analyzed using the data from patients hospitalized for cataract ((IR-DRG(International Redefined Diagnosis Related Groups) 2041) and had lens operation performed. The discrete type proportional hazard model was used in the analysis. The factors which might affect the length of stay were analyzed. We found that the child dummy, place to go back to after hospitalization and types of operations were important factors affecting the length of stay. There were surprisingly big differences in the lengths of stay by hospitals even after considering the characteristics of the patients. The factors, which might affect the lengths of stay by hospitals, were also analyzed. The findings were i) higher profit rates made the lengths of stay longer and ii) the lengths of stay were longer in the cold region (Hokaido and Tohoku region) than in other regions.