Machine Learning-based Models for Outpatient Prescription of Kampo Formulations: An Analysis of a Health Insurance Claims Database.

BACKGROUND: Despite the widespread practice of Japanese traditional Kampo medicine, the characteristics of patients receiving various Kampo formulations have not been documented in detail. We applied a machine learning model to a health insurance claims database to identify the factors associated with the use of Kampo formulations. METHODS: A 10% sample of enrollees of the JMDC Claims Database in 2018 and 2019 was used to create the training and testing sets, respectively. Logistic regression analyses with lasso regularization were performed in the training set to construct models with prescriptions of 10 commonly used Kampo formulations in 1 year as the dependent variable and data of the preceding year as independent variables. Models were applied to the testing set to calculate the C-statistics. Additionally, the performance of simplified scores using 10 or 5 variables were evaluated. RESULTS: There were 338,924 and 399,174 enrollees in the training and testing sets, respectively. The commonly prescribed Kampo formulations included kakkonto, bakumondoto, and shoseityuto. Based on the lasso models, the C-statistics ranged from 0.643 (maoto) to 0.888 (tokishakuyakusan). The models identified both the common determinants of different Kampo formulations and the specific characteristics associated with particular Kampo formulations. The simplified scores were slightly inferior to full models. CONCLUSION: Lasso regression models showed good performance for explaining various Kampo prescriptions from claims data. The models identified the characteristics associated with Kampo formulation use.

Association between Maoto Use and Hospitalization for Seasonal Influenza in a Nonelderly Cohort in Japan.

Objective Maoto is a traditional Japanese Kampo formula used to treat influenza. However, clinical evidence for maoto has been limited to small-scale studies of its effect in alleviating symptoms. The present study evaluated whether or not the addition of maoto to a neuraminidase inhibitor was associated with a reduction in hospitalization following influenza. Methods Using the JMDC Claims Database, we identified outpatients <60 years old who were diagnosed with influenza by an antigen test from September 2013 to August 2018. One-to-five propensity score matching was conducted between patients who received maoto in addition to a neuraminidase inhibitor and those who received a neuraminidase inhibitor alone. Hospitalization within seven days of the influenza diagnosis was compared in the matched groups using the Mantel-Haenszel test. Results We identified 1.79 million cases of influenza from the database in the 5-year study period. Maoto was prescribed for 3.9% of the 1.67 million cases receiving a neuraminidase inhibitor. In the 64,613 propensity score-matched groups of patients, the 7-day hospitalization rate was 0.116% (n=75) for patients with maoto and 0.122% (n=394) for patients without maoto. The difference between these treatment groups was nonsignificant (common odds ratio, 0.95; 95% confidence interval, 0.74 to 1.22; p=0.695). Conclusion The addition of maoto to a neuraminidase inhibitor was not associated with a decrease in hospitalization among nonelderly patients with influenza. Further research is necessary to clarify the indication and efficacy of maoto.

Outpatient Prescriptions of Kampo Formulations in Japan.

Objective Kampo is a traditional Japanese medicine using formulae of natural agents. Although Kampo is widely practiced, information regarding the current prescriptions of Kampo formulations is lacking. The aim of the study was to describe the outpatient use of Kampo formulations in the current Japanese health insurance system. Methods From the JMDC Claims Database, we identified subscribers with outpatient prescriptions of Kampo extract formulations between April 2017 and March 2018. Prescription records were summarized at the individual level to describe the pattern of each formula's use, such as the frequency of prescription and the number of days within a year that were covered by the prescriptions. We also examined whether or not Kampo formulations were prescribed in combination with other drugs. Results Of the 4.5 million subscribers, 13.5% received prescriptions of Kampo extracts within 1 year, and 54% of Kampo users were women. The most commonly prescribed Kampo formulae included kakkonto, shoseiryuto, and maoto, which were used for the short term covering a median of 5 to 7 days. There were also several formulae that were prescribed for longer periods. The median numbers of days covered by kamishoyosan and keishibukuryogan were 60 and 56, respectively. Kampo formulations were used in combination with Western drugs in 85% of prescriptions. Conclusion Kampo formulations are commonly prescribed under the Japanese insurance system and are frequently used in combination with Western drugs. The pattern of prescriptions varied across different formulae.