In this article, we provide a brief summary of study design of case-control study within a cohort and an introduction of two case-control studies within a cohort conducted in Japan recently using antihypertensive drug database based on the post-marketing surveillance data of pharmaceutical companies. In the case-control study within a cohort, it is possible to avoid bias caused in a case-control study and conduct more efficiently than cohort study. Therefore the case-control study within a cohort widely has been used in pharmacoepidemiological studies with a database. (Jpn J Pharmacoepidemiol 2013; 18(2): 77-83)

In this article, we provide a brief summary of a case-cohort study design and an introduction of Japan Statin Study (JSS) as an example for case-cohort study conducted in Japan. In the case-cohort study, the control as a sub-cohort is randomly selected from a research cohort at the beginning of the study. As it is not necessary to be sampled a control by interesting event in the research, the study design as well as cohort study can examine some events in the same time. Therefore the study design can be also applied to pharmaco-vigilance survey. The case-cohort study design has wide application range. (Jpn J Pharmacoepidemiol 2013; 18(2): 84-89)

Case-crossover study is research using only data from cases, patient with interested event. Getting control from own data, the case-crossover study is classified into self-controlled study. It has a lot of condition to be valid study. When study satisfies the conditions, the case-crossover has a lot of advantage in which it is not necessary to collect information of control group patients, and the control is to be matched genetic and other background. The paper is summarized the case-crossover study design. (Jpn J Pharmacoepidemiol 2013; 18(2): 90-94)

In pharmacoepidemiology, cohort studies and case-control studies have been commonly used as research methods to examine causal relationship between exposures to medicines and occurrences of advance events. For both study designs, we could assume a common population at risk, in which cases are developed. A cohort study defines a research cohort within the population at risk and tries to investigate the research cohort directly, and a case-control study tries to investigate the research cohort partly by the control sampling. Assuming an underling research cohort, it becomes possible to understand cohort studies and case-control studies within an unified framework. We revisited several sampling methods to select controls in case-control studies and effect measures implied by the sampling methods. (Jpn J Pharmacoepidemiol 2013; 18(2): 95-111)

Objective: Medical information databases provide useful Real World Evidence (RWE) and a comprehensive view of medical activities. However, since each database has limited coverage and cannot be self-sufficient, combining information from multiple databases is a useful research technique. In this study, we examined methods of estimating patient numbers by combining information from multiple databases in order to assess the respective databases and identify the respective characteristics, biases and idiosyncrasies. This process also allowed us to propose improvements in the grand design of medical information databases in Japan.
Design: Retrospective observational cohort study
Methods: We attempted to estimate the numbers of patients treated for certain diseases and the numbers prescribed a drug by three methods: i) We estimated patient numbers for seven diseases using an insurance claims database, adjusting the proportion of elderly patients according to a hospital medical records database; ii) Sales information for drug X was combined with the prescribed volume per person estimated from pharmacy claims databases to estimate the number of patients administered X; this number was divided by the prescription rate obtained from a medical claims database to calculate patient numbers for the associated disease; and iii) We examined two surveys of the National Institute of Infectious Diseases (NIID) for timely estimation of patient numbers for influenza, referring to estimates from an insurance claims database.
Results: In Method i)-iii), we proved that it is possible to estimate patient numbers for many diseases and administered drugs by effectively combining multiple medical information databases. Validation could be claimed when multiple methods lead to similar results.
Conclusion: These databases provided by government agencies and private corporations are separately managed, and there is no grand plan to integrate them into one platform. It is crucial that databases, rather than being designed to stand alone, are standardized according to widely used systems under a solid master data management strategy. This will make it easier to combine information from multiple databases and to maximize their values. Mutual use of these databases by academic researchers for epidemiological and clinical studies and by government policy makers and data scientists of pharmaceutical companies may improve the usefulness of these databases and expand their application in research.

As for spontaneous case reports of suspected adverse drug reactions, the following problems are known, a part of phenomenon actually recognized to be side effects in clinical site is reported, moreover, the thing which report bias which fluctuate produces by a factor with the various rates exists, since an appearance ratio is incalculable, it cannot be adapted in the analysis technique for common side-effects data. Therefore, the search algorithm analyzing method of signal detection has been established. Japanese Adverse Drug Event Report database (abbreviation; JADER) is exhibited from Pharmaceuticals and Medical Devices Agency (abbreviation; PMDA) to April 2012, and can use anyone now without restrictions. It is expected that the quality of research of pharmacoepidemiology will improve by practical use of JADER. However, as for the present condition, there are few reports of the practical use example in a pharmaceutical company as a subject which tends to attract attention from “signal detection” as the practical use method, consequently regulatory agency should cope with. Although the pharmacoepidemiology group of the author Japanese Society for Biopharmaceutical Statistics is not comprehensive before public presentation of JADER, the list obtained by “case report line list search with which side effects are suspected” which PMDA offers was collected and put in a database, and practical use in a pharmaceutical company has been tried by referring to the examination method of pharmacoepidemiology. This paper explains the processing method of the JADER data for enabling signal detection. Using this method, application to the subject of much pharmacoepidemiology is performed actively, and it expects to contribute to improvement in the quality of research of pharmacoepidemiology.

Japanese Adverse Drug Event Report database (Abbreviation; JADER) was disclosed April 2012 and will be expected to have a major role in the establishment of the adequate information for the proper usage of drugs. In this paper, we were compared using the shape parameters that were estimated by fitting the Weibull distribution, whether the adverse drug reaction of suicide- or diabetes-related onset time profiles vary depending on the type of interferon formulations. The data were used JADER of August 2013. The combined number of adverse drug reaction and drug that duplicates removed was 702,925. In diabetes-related side effects, the distribution of adverse drug reaction time was different among formulations. The shape parameters of the Weibull distribution are 1.49 (1.09-1.94), 0.84 (0.66-1.05) and 1.07 (0.92-1.23)(point estimates and two-sided 95% confidence interval) in interferon-α,-β and peginterferon, respectively. Interferon-α is categorized as the wear-out failure type adverse drug reaction onset time profile from which its 95% confidence interval exceeds 1. Peginterferon is classified as the random failures type from which its point estimates is almost 1. Since the upper 95% confidence interval is near 1, the time profile of interferon-β is close to the early failures type. We conclude that the combination of the shape parameter of the Weibull distribution and the visual inspection, like histogram and box plot, is useful in the monitoring of adverse drug reaction onset time profile.

From April 2012, Japanese Adverse Drug Event Report database (JADER) has become downloadable for utilization in the public, under the specified acceptable use policy. Given the situation, we focused on the severe eruptions which cases are increased in the public Relief System for Sufferers from Adverse Drug Reactions, for the purpose to analyze the characteristics of typical severe eruptions and a trend or a commonality in the corresponding reported drugs, by utilizing JADER. Disproportionate reporting obtained with ROR (Reporting Odds Ratio) and distribution parameter estimations obtained with Weibull distribution fit for the onset time of drug adverse reactions, were applied for the analysis in addition to the summary of frequency. We obtained 10,171 cases of severe eruptions from JADER, after exclusion of duplicated reports. In the Drug Induced Hypersensitivity Syndrome (DIHS), which has characteristics in clinical time course and causal drugs, we confirmed that typical causal drugs such as anti-epilepsy are frequently reported in JADER. On the other hand, drugs other than typical causal drugs also showed high ROR signal values. In the estimation of Weibull distribution shape parameter fit for drug adverse reaction onset time, DIHS gave estimation apparently different from other severe eruptions. Coincide with the estimation, histogram of onset time for DIHS showed the peak at around 20 days after drug administration, which is later than other severe eruptions. We conclude that analytical approach to obtaining information from multiple aspects of JADER data should be a useful effort for the persons who are engaged in preventive action for drug adverse reactions.