Statistical learning methods as a preprocessing step for survival analysis: evaluation of concept using lung cancer data

Table 1 Patient characteristics

Characteristic	Incident n	Incident mean/SD or %	Censored n	Censored mean/SD or %	p-value
Age	59	69.58/7.85	92	65.42/8.84	0.0038*
Grade	59	2.22/0.62	92	2.10/0.68	0.2651*
One	6	10.17%	17	18.48%	0.1656
Two	34	57.63%	49	53.26%	0.5988
Three	19	32.20%	26	28.26%	0.6053
Gender
Male	38	64.41%	34	36.96%	0.0010
Female	21	35.59%	58	63.04%	0.0010
Histology subtype
Adenocarcinoma	29	49.15%	58	63.04%	0.0919
Squamous	25	42.37%	20	21.74%	0.1510
Large Cell	3	5.08%	11	11.96%	0.1555
Adenosquamous	2	3.39%	3	3.26%	0.9655
Smoking status
Non-Smoker	12	20.34%	19	20.65%	0.9629
Smoker	47	79.66%	73	79.35%	0.9629

The patient characteristics are summarized in this table. The incident column refers to group-2 patients (unfavorable survival outcome) and the censored column refers to group-1 patients (favorable survival outcome). The mean and standard deviation (SD) are provided for the continuous variables and percentages (%) are provided for the other variables by group. The number of patients (n) for each variable is given for each group. The incident and censored group characteristics were compared with either the t-test (*) or binomial proportional test. The relevant p-values are provided in the last column (right).

ISSN: 1475-925X