A questionnaire survey was conducted on citizens taking medication for chronic diseases (type 2 diabetes, hypertension, asthma, atopic dermatitis). As a result, 52.7% had residual drugs. Patients who use GPs or pharmacies tend to self-stockpile drugs in preparation for a disaster. These patients may have previously been informed about the need of self-stockpiling drugs in case of emergency. Those who responded that they were at risk of a secondary disaster tended to self-stockpile drugs in preparation for long-term evacuation. Although the risk of self-stockpiling of drugs by patients with chronic illness may not be high, it is desirable for pharmacies to be more actively involved to ensure safe self-stockpiling.
To efficiently share clinical information, regional healthcare networks have been developed nationwide. As per a survey, only 1% of the population is registered in such networks. Unless residents in each region become aware of the service, they would not willingly register themselves to share their clinical data. Accordingly, this study administered an online questionnaire to measure their recognition on the networks and their attitude toward online-sharing of clinical information.
The surveillance showed that the recognition rate was 17.8% and the approval rate was 63.3%, as national average. The results indicated a positive correlation between the recognition and the approval rate, a weak correlation between the number of networks and the recognition in a prefecture, and no correlation between the number of networks and the approval rates. The results suggested that promotion campaigns of the networks might contribute to the online sharing of patient information.
The collection, organization, and analysis of data in the healthcare and welfare administration involves a huge amount of manual work. The authors believe that efficiency can be improved by applying machine learning technology.
In this paper, the authors used a logistic regression model as a machine learning model. As a result a logistic regression analysis model, the authors were able to create a model that reproduced the judgment processes of the care-need assessment committees based on the precedents of the secondary review of the initial computer assessment.
The authors believe that this method is effective for improving the efficiency of data collection, organization and analysis in health care and welfare administration and could be used to streamline other tasks.
The Administrative Reports on Hospital Bed Function (ARHBF) was started in 2014 to provide data for facilitating the Regional Medical Care Vision (RMCV) targeted to the year 2025. The RMCV aims at converting excessive acute care wards into less-costly chronic or rehabilitative wards. All general hospitals (excluding psychiatric hospitals) and clinics with inpatient beds are required by the Medical Care Act to report detailed data as of July each year and hospital-specific data are published on prefectural governments' website as a form of Excel files. The author compiled the five-year cumulative data into data warehouse and analyzed the progress of converting ward-specific clinical functions through ward-matching technique. Approximately 30% of acute care wards in 2014 have already converted as of July 2018, and the negative linear correlation between the actual converting rate and the realization rate of the self-proclaimed target suggested the unwillingness of most hospitals for conversion of their ward functions. ARHBF data can be used effectively by data-warehousing to facilitate RMCV.