Page 177 - 臺大管理論叢第33卷第1期

NTU Management Review Vol. 33 No. 1 Apr. 2023

place from January 2001 to June 2002 and 39 from July 2002 to December 2003.

2.2 Data Sources
We collect data related to NHI reimbursements of hospital charges, the medical

records of patients, and the basic characteristics of surgeons from the hospital being
investigated. We select the research subjects according to aforementioned ICD procedural
codes. We also identify comorbidities using ICD 9 codes in any primary or secondary

diagnosis fields (Deyo, Cherkin, and Ciol, 1992). The Charlson Comorbidity Index (CCI)
is then computed for each patient. Additionally, we obtain the basic information regarding
each surgeon from the medical center’s personnel information archive, including age,
specialty, and years of service.

2.3 Clinical Pathways (CPs) Protocol
The CPs are operational tools for executing best practices based on local practice
and clinical guidelines that are shaped by interdisciplinary teams (De Bleser et al., 2006).

These pathways create a consistent workflow for care delivery. The main objectives of the
CPs are to improve the quality of medical care, maximize the effectiveness of available
medical resources, define consistent and standard medical care procedures, and control
medical costs (Rotter et al., 2012). Following the four essential components of CPs, i.e.,
a timeline, the categories of care or activities and their interventions, intermediate- and

long-term outcome criteria, and the variance record (De Bleser et al., 2006), we consider
the transition from hospitalization to discharge a standard procedure; however, within the
CPs’ framework. First, we develop the treatment procedures and observe the connection

between the variation in procedures and treatment results. Secondly, we continually
modify the procedures to optimize medical quality and utilize medical resources more
efficiently.
In the current study, the workflow of the CPs developed for a thyroidectomy is as
follows. First of all, we identify major diagnostic categories, types of diagnosis, and

treatment intervention before selecting the development procedure for the CPs. Secondly,
we examine the current procedure and clarify the related factors to develop the CPs. Third,
we assess whether the new CPs are applicable for the current study. Finally, we constantly

169