Background: Clinical information have been stored electronically as a hospital information
system(HIS). The database stores all the data related with medical actions, including accounting information,
laboratory examinations, and patient records described by medical staff and becomes the
indispensable infrastructure for clinical decision process. However, clinical environment is very complex,
and flexible and adaptive service improvement is crucial in maintaining quality of medical care.
Thus, incremental software development in hospital information system and its evaluation is important.
Methods: The following software development process is proposed. First, data extracted from hospital information system
is used to capture the peculiarities of the divisions in a university hospital. Then, the mining results are interpreted by
medical staff and the solutions are discussed. Based on the discussions, new interfaces are developed, and their performance
was evaluated using the service logs. The data used for hypothesis generation is the chronological change of the
number of clinical orders and waiting time. Analytical method of temporal data is based on multiscale matching method.
Experiments were conducted from fiscal year 2010 to 2012. In the ends of fiscal year 2010 and 2011, the new software
was embedded into hospital information system and service logs are collected. From the service log, the times when a patient
came to visit a reception and a laboratory division, when results of laboratory examinations were output, when a doctor
started to examine in his/her clinic were extracted, the time differences between events were calculated and these values
were used for the evaluation statistics. Fiscal year 2010, 2011 and 2012 were regarded as the baseline, the period
when the first improvement was implemented, and that when the second improvement implemented, respectively. Comparison
of statistics (median and mean) were used for evaluation and Kruskal-Wallis test was applied for checking the differences
among three years. For statistical analysis, R3-1-1 was used.
Results: Two divisions, hepatology and rheumatology were selected for comparison. The obtained results gave a hypothesis
that the workflow of rheumatology is different from that of hepatology, which reflected the ordinary workflow in the
outpatient clinics of the university hospital and caused the problem where experts forgot to issue the orders. The first step
was to implement an interface for double checking: if a clinician input the comment on laboratory examination in a reservation
sheet but he/she has not yet issued an order before they closed their windows for a patient, an alert will come up
from the screen. After one year trial, statistics showed a small improvement in waiting time was very small as shown in
the next section, we discussed with rheumatologists again, and in fiscal year 2012, we set up the management screen
where all the forgotten orders for patients who visited that day would be displayed, and the doctors could go back to issue
orders. The workflow and waiting time were improved after the second installation.
Conclusion: The process, which can be viewed as a variant of active mining process, will give a new framework for
quantitative evaluation of software development in hospital information system, which can be viewed as an application of
active mining process.