the authors applied machine learning techniques to the model construction to achieve
an even higher sensitivity and specificity (Yokota, Endo, & Ohe, 2017).
The authors stated that nursing researchers can generate knowledge through the
secondary analysis of the data stored in the EMR, create artificial intelligence by
automatically processing the knowledge on a computer, and further utilize artificial
intelligence to create higher-level nursing care. It has been argued that the development of artificial intelligence may impair the ability of medical professionals to
think, however, we strongly expect that next-generation nurses who can utilize
artificial intelligence as a tool will provide high-quality care based on realworld data.
3.2 Real-World Data-Based Effective Pressure Ulcer
Prediction and Management
3.2.1 Impact of Pressure Ulcers on the Healthcare System
The following examples are related to technological advances in pressure ulcer
management regarding assessment technologies that were achieved through collaboration with molecular biologists and engineers.
Pressure ulcers are major healthcare problems and are most commonly found in
elderly patients (Thomas, 2007). The ulcers are associated with prolonged hospital
stays (Allman, Goode, Burst, Bartolucci, & Thomas, 1999), decreased quality of life
(Gorecki et al., 2009), increased mortality (Khor et al., 2014; Manzano et al., 2014),
and increased medical costs (Brem et al., 2010).
We have demonstrated that having pressure ulcers affects not only those burdens,
but also the patient’s discharge destination (Nakagami et al., 2020). We analyzed
medical big-data using the Japanese Diagnosis Procedure Combination (DPC)
database, which is a case-mix patient classification system; it was launched in
2002 by Japan’s Ministry of Health, Labour, and Welfare, and it covers nationwide
inpatient administrative claims and discharge data for acute-care hospitals across the
country. Detailed patient data and administrative claims data are collected for all
inpatients discharged from participating hospitals. All 82 academic hospitals in
Japan are obliged to participate in the database; participation of community hospitals
is not mandatory. The database contains data on approximately 50% of all acute-care
inpatients. It includes the following: unique identifiers of hospitals; patient age and
sex; body mass index; smoking status (non-smoker or current/past smoker); the level
of consciousness at admission (classified according to the Japan Coma Scale); type
of admission (planned or urgent); use of ambulance service; use of home care before
admission; ADL scores on admission; primary diagnoses and comorbidities on
admission and adverse events after admission, recorded in accordance with the
International Classification of Diseases tenth Revision (ICD-10) codes; procedures
and surgeries recorded using original Japanese codes; pressure ulcer stage on
admission and discharge based on DESIGN-R classification (Sanada et al., 2011);
Real-World Data-Based Care Innovation: Lessons Learned from Nursing Science
87
an even higher sensitivity and specificity (Yokota, Endo, & Ohe, 2017).
The authors stated that nursing researchers can generate knowledge through the
secondary analysis of the data stored in the EMR, create artificial intelligence by
automatically processing the knowledge on a computer, and further utilize artificial
intelligence to create higher-level nursing care. It has been argued that the development of artificial intelligence may impair the ability of medical professionals to
think, however, we strongly expect that next-generation nurses who can utilize
artificial intelligence as a tool will provide high-quality care based on realworld data.
3.2 Real-World Data-Based Effective Pressure Ulcer
Prediction and Management
3.2.1 Impact of Pressure Ulcers on the Healthcare System
The following examples are related to technological advances in pressure ulcer
management regarding assessment technologies that were achieved through collaboration with molecular biologists and engineers.
Pressure ulcers are major healthcare problems and are most commonly found in
elderly patients (Thomas, 2007). The ulcers are associated with prolonged hospital
stays (Allman, Goode, Burst, Bartolucci, & Thomas, 1999), decreased quality of life
(Gorecki et al., 2009), increased mortality (Khor et al., 2014; Manzano et al., 2014),
and increased medical costs (Brem et al., 2010).
We have demonstrated that having pressure ulcers affects not only those burdens,
but also the patient’s discharge destination (Nakagami et al., 2020). We analyzed
medical big-data using the Japanese Diagnosis Procedure Combination (DPC)
database, which is a case-mix patient classification system; it was launched in
2002 by Japan’s Ministry of Health, Labour, and Welfare, and it covers nationwide
inpatient administrative claims and discharge data for acute-care hospitals across the
country. Detailed patient data and administrative claims data are collected for all
inpatients discharged from participating hospitals. All 82 academic hospitals in
Japan are obliged to participate in the database; participation of community hospitals
is not mandatory. The database contains data on approximately 50% of all acute-care
inpatients. It includes the following: unique identifiers of hospitals; patient age and
sex; body mass index; smoking status (non-smoker or current/past smoker); the level
of consciousness at admission (classified according to the Japan Coma Scale); type
of admission (planned or urgent); use of ambulance service; use of home care before
admission; ADL scores on admission; primary diagnoses and comorbidities on
admission and adverse events after admission, recorded in accordance with the
International Classification of Diseases tenth Revision (ICD-10) codes; procedures
and surgeries recorded using original Japanese codes; pressure ulcer stage on
admission and discharge based on DESIGN-R classification (Sanada et al., 2011);
Real-World Data-Based Care Innovation: Lessons Learned from Nursing Science
87