Healthcare Data Science Research Unit

Background

EBPM, an extension concept of EBM, entails the utilization of scientific evidence as the foundation for policy development, implementation, and assessment. This concept stems from the recognition that previous policies lacked empirical evidence. Consequently, efforts have been made to produce empirical evidence. Despite the accumulation of evidence, its execution in real-world scenarios often deviates from expectations. This calls for a focus on implementation science. Our research concentrates on elucidating the practical implementation of preventive medical interventions, the efficacy of which is epidemiologically evident.
In clinical medicine, evidence implementation is partially achieved through regulation via clinical guidelines and health systems. In contract, the preventive health interventions we focus on, particularly in the community and occupational settings, present greater challenges to implementation due to a number of stakeholders and less regulated actors. Additionally, challenges arise in generating evidence as predominant research often explores risk factors rather than concrete measures to modify these factors. For instance, while it is established that excessive salt intake is known to contribute to hypertension and stroke, there is limited research on methods to reduce salt intake in daily life.
Moreover, in Japan, policy formulation is rarely evidence-based. In this context, an approach that assesses policy effectiveness post-implementation and subsequently considers de-implementation is imperative. This means adopting an action-research methodology that is closely linked to the target, verifying effects through a Plan-Do-Check-Act(PDCA) framework, and iteratively improving interventions.

Verification of implementation of preventive medical interventions

Against the backdrop outlined above, the primary focus of the Data Health Research Unit centers on preventive medical intervention measures. Our research emphasizes the prerequisites for effective policy implementation. Workplace-related studies have demonstrated a correlation between changes in individual health (metabolic syndrome) and the specific workplace environment (Kakinuma et al., 2019). Additionally, previous studies have illustrated a positive association between sustained use of personal health records tools and health outcomes (Nakao et al., 2020). We have tried to extract factors that are conducive to a higher implementation of preventive interventions. (Hamamatsu et al., 2021). The measures implemented by health insurers and their corresponding outcomes are necessary. Analyzing data solely at the individual level may overlook the collective influence of the group or organization to which an individual belongs. Therefore, it is crucial to conduct empirical research that integrates characteristics and environments at both the group and individual levels. To facilitate this, it is essential to establish a framework for continuous data collection from health insurers.

Establishment of infrastructure for measurement and evaluation

We are engaged in research that establishes the basis for validating the effectiveness of preventive interventions through data analysis. In the face of societal challenges, such as shortage of human resources and declining labor productivity in an aging society, companies have embraced the promotion of employee health as a national policy. However, the absence of easily measurable indicators to assess the impact and effectiveness of health investments. We proposed “Single-Item Presenteeism Question” (SPQ), to measure the loss of labor productivity due to employee health/physical conditions. The validity of the index has been demonstrated (Muramatsu et al., 2021) and is presently being incorporated into the government’s Health and Productivity Management Organization Certification System.
Medical claims are used to assess the effectiveness of preventive interventions. To gain insights into the allocation of medical costs across diseases, we developed a method for disease-specific allocation of medical costs (Hiramatsu et al., 2022). This method has practical applications, including evaluating the cost-effectiveness of therapeutic and preventive interventions and visualizing the allocation of medical costs, for example in end-of-life care.
The development of these indices and methodologies stands is a foundational achievement in validating the effectiveness of preventive interventions and will underpin future research.

Educational Initiatives

It is important to seamlessly integrate education and academic research results. Recognizing the importance of practitioner education as a cornerstone for empirical research advances, we have expanded our educational efforts. Since 2018, we have conducted training courses for health insurers. For National Health Insurance System, We has been offered “the Prefectural Leadership Program” since 2020. In addition, we have written guidebooks for the preparation of “data health plans” that each insurers is required to by law to prepare, textbooks to promote health management, and manuals on prevention of lifestyle-related diseases.
We have also initiated educational activities for children too. We are developing a program that will integrates the data health plan as a supplementary teaching material for elementary school classes, with a particular focus on lifestyle-related disease prevention education.
Our commitment to knowledge dissemination extends to undergraduate and graduate students at the University of Tokyo’s School of Medicine and Graduate School.