Cost Control of Treatment for Cerebrovascular Patients Using a Machine Learning Model in Western China

Table 2

Multiple linear stepwise regression results of factors influencing hospitalization expenditure in cerebrovascular disease patients.

Variables	Regression coefficient	Standard deviation	T-statistic	value
Gender (take the male as a reference)
Female	−229.84	44.53	−5.16	<0.000
LOS (take less than nine days as a reference)
9 days∼13 days	2360.58	57.49	40.12	<0.000
More than 13 days	6463.63	59.93	119.86	<0.000
Level of hospital (take grade B secondary hospital as a reference)
Grade A secondary hospital	1369.27	66.80	20.50	<0.000
Grade B tertiary hospital	3542.44	75.10	47.17	<0.000
Grade A tertiary hospital	6038.54	76.85	78.57	<0.000
Surgery (take having surgery as a reference)
No surgery	−1480.79	90.11	−16.43	<0.000
Status on discharge (take recovery as a reference)
Transfers	−669.11	265.52	−2.52	0.012
Death	3035.03	265.08	11.45	<0.000
Others	16.50	64.99	0.25	0.80
Midway check-out	−665.90	819.97	−0.81	0.416
Type of insurance (take urban as a reference)
Rural	−260.14	46.82	−5.56	<0.000
Main comorbidities complications (take cerebral arteries lack blood supply as a reference)
Lacunar infarction	−3814.00	67.45	−56.55	<0.000
Cerebral infarction	−3440.60	74.52	−46.17	<0.000
Chronic cerebral ischemia	−1622.16	86.57	−18.74	<0.000
Others	−3925.15	120.92	−32.46	<0.000
Age (take less than 45 as a reference)
45∼60	385.83	115.15	3.35	0.001
More than 60	512.51	108.66	4.72	<0.000