The association between active participation in a sports club, physical activity and social network on the development of lung cancer in smokers: a case-control study

Table 1 The characteristics of the CoSmoS-study sample (n = 302)

Variable	Coding	Missing values	Lung cancer patients		Control patients
		n	n	%	n	%
Gender	male	0	104	65.8	89	61.8
	female		54	34.2	55	38.2
Age	35-44	2	3	1.9	23	16.1
	45-54		44	28.0	31	21.7
	55-64		61	38.9	51	35.7
	65-76		49	31.2	38	26.6
Education	no vocational/university education	3	34	21.8	15	10.5
	lower vocational education		113	72.4	102	71.3
	higher vocational/university education		9	5.8	26	18.2
Marital status	married	3	48	30.6	46	32.4
	not married		109	69.4	96	67.6
Residence	city	2	36	22.8	25	17.6
	country		122	77.2	117	82.4
Religion	religious	0	36	22.8	37	25.7
	not religious		122	77.2	107	74.3
Physical activity	physically active at least one hour per week	1	44	27.8	72	50.3
	not physically active during the week		114	72.2	71	49.7
Pack years	≤ 15 pack years	10	91	59.5	109	78.4
	> 15 pack years		62	40.5	30	21.6
Friends	many	0	72	46.6	72	50.0
	none/few		86	54.4	72	50.0
Relatives	many	0	72	45.6	79	54.9
	none/few		86	54.4	65	45.1
Participation in a sports club	yes	0	19	12.0	35	24.3
	no		139	88.0	109	75.7

The results of the stem-and-leaf plots revealed high outliers for both number of friends (mean = 6.2; maximum outlier = 150) and number of relatives (mean = 5.4; maximum outlier = 50). Two separate regression analyses were conducted, one with and one without the detected outliers (median split at 3) [30]. No significant differences were found without the outliers present. The outliers were excluded for the present regression analysis.

ISSN: 1756-0500