Page 187 - 臺大管理論叢第33卷第1期

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

Table 5 Difference-in-Differences Regression for LOS and Hospital Charges (N = 376)
Variables LOS Hospital Charges
Intercept 6.721*** 50,991***
Patient sex
Male vs. Female 0.423 1,209
Patient age -0.153** -770*
Patient age squared 0.002*** 9**
Major diagnostic category
Malignant neoplasm of the thyroid gland vs. Other 0.310 3,573
diagnoses
Benign neoplasm of the thyroid gland vs. Other -0.490 -4,313
diagnoses
Major intervention
Thyroid lobectomy vs. Total thyroidectomy 0.450 2,006
Partial thyroidectomy vs. Total thyroidectomy 0.844 -556
CCI scores 0.770*** 3,494***
Surgeon age -0.010 -91
Treatment vs. Control group -0.752 2,348
GS-division vs. ENT division
Time for implementing CPs 0.614 2,058
Interaction term for the time when implementing the -1.640** -7,478
CPs and Treatment groups
R = 0.270 R = 0.140
2
2
p < 0.000 p < 0.000
Note: CCI scores = Charlson comorbidity index scores; CPs = clinical pathways; ENT = ear, nose,
and throat; GS = general surgery; LOS = length of stay.
* p < 0.05; p < 0.01; p < 0.001.
***
**

division, the CPs for this type of surgery usually require one dose of antibiotics only (some
even require none), and no antibiotics are prescribed when patients are hospitalized and

after they have been discharged. Further, the reasons for the reduction in other fee items
such as laboratory, special medical supply, and X-ray fees may be attributed to the fact
that the CPs or medical operation models have been jointly developed by multidisciplinary

professionals, who have designed them to be the optimal, most cost-effective treatment
models. The treatment procedures, which are laid out in a day-by-day agenda, have
enabled the physicians involved to achieve a certain level of consistency and efficiency.

179