Objective: “Local drugs” (or “country drugs”) are medicines which are approved and distributed only in Japan. “Drug lag” and “drug loss” indicates medicines approved in other countries (the US or Europe) which receive delayed or no approval in Japan, respectively. Specifically, the former problem indicates the “overdevelopment” of medicines, while the latter indicates “under-development.” Notably, drug lag remains a controversial issue in Japan from 2000s. This study quantitively clarifies the current situations of these problems in Japan.
　Methods: For local drugs, I used the list of medicines approved by Japan’s Pharmaceuticals and Medical Devices Agency (PMDA) from 2017 to 2023. These data were combined with the approval status of the US Food and Drug Administration (FDA) and European Medicines Agency (EMA). Next, the new medical entities (NMEs) approved by the FDA from 2017 to 2023 were examined to quantify the problems of drug lag and drug loss in Japan. Besides Japan, the launch days of the following countries were obtained from the Multinational Integrated Data Analysis System (MIDAS) database by IQVIA: Europe (Germany), the UK, Switzerland, Canada, Australia, New Zealand, Korea, Taiwan, and Singapore. These days were compared with the approval days by the FDA.
　Results: For new medicines without new ingredients, the percentage of local drugs in Japan is approximately 40%. This number reaches 60% if the manufacturers are limited to Japanese companies. The main developers of local drugs are relatively small-size Japanese companies. Next, 20% of the new medicines with new ingredients were not approved either by the FDA or EMA. This number rises to 40% of medicines which were manufactured by Japanese companies. Meanwhile, the percentage was even higher for regenerative medical products. Further, 70% (80% for Japanese companies) of the medicines approved first in Japan were not approved by the FDA or EMA. Conversely, regarding the drug lag and drug loss, the proportion of the launched medicines in Japan to the FDA-approved medicines was inferior to Europe (Germany) and the UK. However, it was the comparable with Canada and Switzerland, and higher than that in both the Oceanian (Australia and New Zealand) and Asian countries (Korea, Taiwan, and Singapore). The results remain unchanged during the subgroup analysis for anti-cancer drugs, orphan drugs, and drugs developed by new companies. Finally, the time to launch in Japan from the FDA approval day is similar to that in Canada and Switzerland. It is much faster than the three Asian countries. This is especially true for the launch speed of anti-cancer drugs, where it is the highest except that in Germany and the UK. However, the launch speed of orphan drugs tends to be lower.
　Discussion: Clearly, the problem and percentage of local drugs are not negligible. In terms of drug lag and drug loss, most medicines are available in Japan except that in Germany and the UK. As Germany is one of the best countries among European countries in terms of the access to new medicines, the Japanese situation may be better than several European countries. Considering these results, further investigation is needed regarding how much investments should be made to improve access to medicines. Finally, the policy’s cost-effectiveness about drug lag or drug loss should be discussed.

View full abstract

　This paper examines the effects of advertising levels on sales, business profits, and profit margins per sale in the Japanese pharmaceutical market by using the data from the financial reports of 52 pharmaceutical firms. Using a fixed-effects model, I show that increasing the level of advertising has significantly positive effects on firms’ sales and business profits, but that advertising elasticities are small. To consider the advertising characteristics, I use the subsamples of firms producing prescription and over-the-counter drugs, and estimate the advertising effects separately. My results indicate that advertising stocks have significantly positive effects on sales and business profits for prescription drug manufacturers, but that they have insignificant effects on all dependent variables for over-the-counter drug manufacturers. I find that additional 10 thousand yen of advertising increases sales by 92.6 thousand yen and business profits by 39.2 thousand yen in the overall pharmaceutical market.

View full abstract

Purpose: It is well known that workplace social support is positively related to workers’ mental health. Most of the previous studies used cross-sectional data to investigate the association, therefore, unobserved individual-specific confounders were not controlled for. In this study, we examined whether workplace social support is associated with mental health as well as fatigue recovery even after controlling for unobserved individual-specific confounders. In addition, we compared the magnitude of estimated coefficients between controlling and noncontrolling for unobserved individual-specific confounders.
　Method: 10,000 workers were recruited through an online research company in November 2021, and a followup survey was carried out in the following year. 2,410 regular workers were used for analysis. By using a twoyear longitudinal data, we applied a fixed-effect model to control for unobserved time invariant individualspecific confounders. We also ran previous cross-sectional research methods and compared coefficients.
　Results: Even after controlling for unobserved individual-specific confounders, workplace social support is significantly and positively associated with depression tendency, subjective well-being, work engagement, and fatigue recovery. Regarding the absolute value of coefficients, the coefficients became significantly smaller when unobserved time invariant individual-specific confounders are controlled. The same positive association and smaller value of the coefficient were also observed when the results were analyzed separately by gender (excluding the results for some female sub-samples).
Conclusions: This study confirmed that workplace social support is positively associated with mental health outcomes even after controlling for unobserved time invariant individual-specific confounders. However, the magnitude of association became smaller than the coefficients from cross sectional analysis. The results of the cross-sectional study method included bias derived from individual characteristics, so it was suggested that the strength of the relationship should be considered with some reservation.

View full abstract