Prediction of Milk Transfer of Drugs Using Machine Learning Methods / 医薬品情報学

Objective: Currently, limited information is available on the milk transfer properties of drugs when consumed by lactating women. Therefore, we aim to construct a prediction model of milk transfer of drugs using machine learning methods.Methods: We obtained data from Hale’s Medications & Mothers’ Milk (MMM) and SciFinder®, and then constructed the datasets. The physicochemical and pharmacokinetic data were used as feature variables with M/P ratio ≥ 1 and M/P ratio < 1 as the objective variables, classified into two groups as the classification of milk transferability. In this study, analyses were conducted using machine learning methods: logistic regression, linear support vector machine (linear SVM), kernel method support vector machine (kernel SVM), random forest, and k-nearest neighbor classification. The results were compared to those obtained with the linear regression equation of Yamauchi et al. from a previous study. The analysis was performed using scikit-learn (version 0.24.2) with python (version 3.8.10).Results: Model construction and validation were performed on the training data comprising 159 drugs. The results revealed that the random forest had the highest accuracy, area under the receiver operating characteristic curve (AUC), and F value. Additionally, the results with test data A and B (n = 36, 31), which were not used for training, showed that both F value and accuracy for the random forest and the kernel method SVM exceeded those with the linear regression equation of Yamauchi et al. Conclusion: We were able to construct a predictive model of milk transferability with relatively high performance using a machine learning method capable of nonlinear separation. The predictive model in this study can be applied to drugs with unknown M/P ratios for providing a new source of information on milk transfer.

Japanese Medical Academic Organizations Surveyed on the Current Situation of Medical Information Providedto the Public on the Internet / 医薬品情報学

Objective: The internet isflooded with drug information; however, some of it isinappropriate and thisinadequate information could expose the public to health hazards. Therefore, we conducted research on the idea of transmitting drug information to the public via the internet and the current state of the information currently provided by each academic society relevant to the field of Medical Informatics.Methods: A questionnaire was mailed to the website managers for the website of 129 specialist medical societies, all members of the Japanese Association of Medical Sciences. We conducted our research between October to November 2018. We investigated each website administrator’s opinion about offering drug information for consumers via the internet and what information each academic society is currently providing.Results: The effective response rate was 43.4% (56/129 groups). Most respondents thought that drug information overflowed in the current Internet society. Moreover, more than half of the respondents thought that the quality of drug information transmitted to the public wasinadequate. Currently, 30 of the academic groupssurveyed are providing information to the public. When providing information, they did not refer to the “Drug Guide for Patients” much. More than 80% of respondents said they would cooperate with linking to the information provided to the public. However, each academic society felt there would be many problems with doing so such as the need for a system to check the contents of the information provided and a system to perform maintenance.Conclusion: The results showed that the website administrators recognized that there is a need to improve the quality of and system for providing drug information to the public. We believe that an integrated information system can be constructed by aggregating the drug information held by each academic society. However, this cannot be realized without first solving many problems.

2．A Checklist of Important Points in Research Using JADER / 薬剤疫学

Spontaneous reporting is an important source of information in pharmacovigilance. In Japan, the Japanese Adverse Drug Event Report database (JADER) was released in 2012, and this has led to numerous conference presentations and academic research papers that have reported the detection of signals based on the use of data mining methods. However, spontaneous reporting generally has certain limitations, including under-reporting, a lack of denominator information, and the effects of reporting bias, and these problems apply equally to JADER. The system of collecting spontaneous reports also influences the results obtained based on JADER analysis, as JADER in principle comprises serious adverse drug event reports and includes solicited reports. The detection of signals showing statistical significance does not necessarily imply a causal relationship between a particular drug and adverse events, and consequently, the cause of the signals detected requires careful interpretation. However, it has been pointed out that findings are sometimes accepted without considering the limitations.For pharmaceutical companies, the Guidance on good pharmacovigilance practices Module Ⅸ and Guidance for Industry - Good Pharmacovigilance Practices and Pharmacoepidemiologic Assessment are available in the European Union and United States, respectively, for the purpose of signal management in pharmacovigilance activities. In contrast, there are limited resources to which researchers can refer when they publish scientific findings obtained using spontaneous reporting databases. To rectify this deficiency, we created a “checklist of important points to be noted during research that uses the data mining method in JADER (mainly signal detection by disproportionality analysis)” for the benefit of researchers using JADER. That study was supported by a Grant for Research Projects of the Japanese Society of Drug Informatics in 2017. In this article, we provide an overview of the checklist, with reference to the “Report of CIOMS Working Group Ⅷ： practical aspects of signal detection in pharmacovigilance,” which was used as a source when creating the checklist.

Creating a Checklist and a Survey on Research That Used the Japanese Adverse Drug Event Report Database / 医薬品情報学

Objective: The purpose of this study was to create a checklist that summarizes checkpoints that should be noted when using the Japanese Adverse Drug Event Report database (JADER). After we created the checklist, we then used it to survey published academic papers that used JADER.Method: First, we created a draft checklist for research that uses JADER by citing the report of CIOMS working group VIII “Practical Aspects of Signal Detection in Pharmacovigilance”. Then, we conducted a pilot test and revised the draft checklist. Finally, the checklist was completed after the review by a pharmacoepidemiology expert. The checklist was applied to published academic papers that used JADER, and the fulfill rate of each checkpoints was calculated.Results: A “checklist of important points to be noted during research that uses the data mining method in JADER (mainly signal detection by disproportionality analysis)” was created. We also revealed problems with published academic papers that used JADER. For example, some researchers were thought to be inappropriately using JADER as a source of their research while others used an inappropriate version of MedDRA.Conclusion: The checklist created in this study summarizes key points that could be noted in research that uses JADER and is thought to contribute to an improvement in quality of research that uses JADER. Additionally, in our investigation of published academic papers that used JADER, we found the possibility that both the role of signal detection and the impact on analysis of JADER using the updated MedDRA version are not well understood.

Characteristics of Adverse Events in Pregnant Women / 医薬品情報学

<b>Purpose: </b>The purpose of this study is to elucidate the characteristics of adverse events in pregnant women, the offending drugs, and patient backgrounds from reports of adverse events.  We performed a case series study.<br><b>Methods: </b>We used CARPIS, a database of adverse events and toxication reported in Japan spanning from 1987 to 2014 and created by the Drug Information Center, Meijo University.  We extracted cases of adverse events in pregnant women, their fetuses, and newborns and investigated the age, primary disease, and history of allergies of the women and the intended use of/offending drugs, therapeutic category, and names for adverse events.<br><b>Result: </b>We collected 434 cases of adverse events in pregnant women, and 251 pediatric cases with adverse events.  The most frequent offending drug in both groups was ritodrine hydrochloride.  The most frequent adverse event in pregnant women was pulmonary oedema due to the administration of ritodrine hydrochloride.  The most frequently reported adverse events in pediatric cases were transient hypothyroidism and withdrawal symptoms in newborns and birth abnormalities in fetuses and newborns, all of which were caused by drugs given for the underlying diseases of their mothers.<br><b>Discussion: </b>We elucidated serious adverse events in pregnant women caused by the administration of ritodrine hydrochloride.  Frequent factors for adverse events were the onset of physiological factors in pregnant women and complicated factors of the mechanism of action of ritodrine hydrochloride.  We need to monitor both mothers and fetuses during the drug administration.  It is suggested that adverse events in pediatric cases are associated with drugs given for underlying diseases in mothers.  Thus, it is necessary to give appropriate information and communicate the risks of taking these drugs before pregnancy.  We believe the results could be helpful in the early detection of adverse events in the future.

The Influence of Drugs Taken during Breastfeeding on Infants: Report and Analysis of Adverse Events Using FAERS / 医薬品情報学

<b>Introduction</b>: The purpose of this study is to elucidate the influence of drugs on infants and to provide information about safe drug treatments during breastfeeding using the FDA’s Adverse Event Reporting System (FAERS).<br><b>Study Design</b>: Case series based on FAERS data.<br><b>Methods: </b>We used the cleaned FAERS data in JAPIC AERS, extracted cases of adverse events from the category “neonatal exposure through breastfeeding (2000189)” in Standardized MedDRA Queries, and collected “the cases of breast-feeding infants” by system organ class, the type of adverse event, therapeutic category of first suspected drugs, and the generic name of the drugs.  We aggregated the data of the most reported cases of suspected drugs from first to fifth by the name of the adverse event.  Additionally, we investigated the properties and disposition of each suspected drug and verified the transitivity of breast milk as well.<br><b>Results: </b>Only 551 cases of breast-feeding infants (0.01%) were extracted from the data of JAPIC AERS.  The aggregated data of adverse events in terms of system organ class showed high numbers of gastrointestinal disease (such as diarrhea and vomiting) and impairment of the nervous system (such as drowsiness), and drugs which acted on the nervous system were the most suspected drugs.  There were 26 cases of adverse events associated with lamotrigine, which was the most frequently reported.<br><b>Discussion: </b>We could observe the characteristics of adverse events and suspected drugs that were shown as the influence of drugs taken during breastfeeding which were exposed to infants.  Our study showed 2 conclusions: 1) the drugs that have had adverse events frequently reported have the characteristic of facilitating the drug’s migration into breast milk, and 2) the most frequently reported cases were those in which nursing mothers or medical experts could recognize the correlation between breast milk and the adverse event(s) immediately after breast milk was given to the infant.

Analysis of Erroneously Dispensed Named Generic Drugs / 医薬品情報学

<b>Objective: </b>The purpose of this study is to investigate incidents of erroneously dispensed drugs on the assumption that the incidents stem from the similar names of the drugs.  The investigation was also conducted to prevent such dispensing incidents in the future, i.e. to search for factors that can prevent future incidents, and finally to propose a prevention plan which takes each of these factors into account.<br><b>Methods: </b>We extracted incident cases related to generic drugs reported by pharmacies in Japan and from those cases examined those that were categorized as cases of erroneously dispensed medicine.  We used this data to categorize the difference in relationship between the drugs which were supposed to be prescribed and those which were erroneously dispensed, and to analyze the association between the “Flowchart for Avoiding Confusion Errors between Similarly Named Drugs” and the name similarity index based on this flowchart.<br><b>Results: </b>The types of incident cases due to name similarities of generic drugs were categorized into specification mistakes and brand mistakes.  The edit of the name similarity index were especially important factors for dispensing incidents between generic drugs.<br><b>Conclusion: </b>This study focusing on generic drugs revealed the factors that result in dispensing incidents due to name similarity.  Further empirical studies investigating the usefulness of interventions that alter the name similarity index is required.

Analysis of Actual Pharmacist Research Activity at Japan Pharmaceutical Association (JPA) Congress of Pharmacy and Pharmaceutical Science / 医薬品情報学

<b>Objective: </b>The purposes of this study were to verify the actual state and to identify the problems associated with research activities by pharmacists.<br><b>Methods: </b>The abstracts presented at the 42nd and the 37th Japan Pharmaceutical Association (JPA) Congress of Pharmacy and Pharmaceutical Science were investigated.  They were categorized into six research types: “Outcome research”; “Intervention research”; “Observational study”; “Status report”; “Case report”; and “Other”.  They were then ranked according to evidence level, with “Outcome research” ranked the highest.  In addition, “Outcome research”, “Intervention research” and “Observational studies” were checked for whether they had structured abstracts, and form scores were awarded.<br><b>Results: </b>We found that only 0.8% of abstracts were “Outcome research”.  but the level was higher at the 42nd Congress than the 37th (<i>p</i>=0.03).  Structured abstracts were also more common at the 42nd Congress than at the 37th (<i>p</i><0.01).  However, the form scores were not significantly different between the 42nd and 37th Congresses.<br><b>Conclusions: </b>Our findings suggest that pharmacist research activities are improving, but that there is still room for further development.  The referee system and the guidelines for research design are useful methods to encourage more activity.  The target of this study was community pharmacy, and further studies are necessary for hospital pharmacy.