Table 2.

Association between lifestyle variables and diabetes or ever-participation in DSME, respectively, estimated by logistic regression analyses

		Participants with diabetes (ref. participants without diabetes)	DSME-ever participants (ref: DSME-never participants)
		OR* (95% CI)*	OR* (95% CI)*
Fruit intake	At least once daily (ref: less than once daily)	1.25 (1.08 to 1.45)†	1.15 (0.87 to 1.53)†
Vegetable intake	At least once daily (ref: less than once daily)	1.13 (0.97 to 1.30)†	0.97 (0.73 to 1.30)†
Physical activity	Walking ≥30 min/day (ref:<30 min/day)	1.01 (0.87 to 1.17)†	1.19 (0.88 to 1.30)†
	Cycling ≥1 day/week (ref:<1 day/week)	0.70 (0.59 to 0.84)†	1.62 (1.15 to 2.30)†
	Doing sports in general ≥1 day/week (ref:<1 day/week)	0.74 (0.63 to 0.86)†	1.16 (0.86 to 1.57)†
	Doing sports in general ≥2 day/week (ref:<2 day/week)	0.71 (0.62 to 0.82)†	1.25 (0.90 to 1.73)
	Endurance training ≥150 min/week & strength training twice/week (ref: no)	0.59 (0.49 to 0.72)†	1.41 (0.95 to 2.08)†
	≥150 min aerobic PA/week including walking (ref.: less)	0.69 (0.60 to 0.80)†	1.42 (1.03 to 1.94)†
	≥150 min aerobic PA/week without walking (ref: less)	0.65 (0.56 to 0.76)†	1.48 (1.08 to 2.03)†
Tobacco use	Smoking (ref. never-smoking)	1.00 (0.81 to 1.23)‡	1.32 (0.87 to 2.00)‡
	Ex-smokers (ref. never-smokers)	1.28 (1.09 to 1.50)‡	1.39 (1.03 to 1.88)‡
BMI	Overweight (ref. normal)	2.09 (1.67 to 2.60)‡	0.97 (0.65 to 1.45)‡
	Obesity (ref. normal)	4.92 (3.95 to 6.12)‡	1.17 (0.79 to 1.74)‡

This table shows results of logistic regression analyses, dependent variables were lifestyle variables (fruit and vegetable intake, physical activity variables, smoking, BMI); DSME participation and diabetes as independent variables.

*All results are adjusted for age, sex, socioeconomic status, living together in marriage/as an unmarried couple or alone, limitation due to health problems for at least 6 months, occupational status, AKSU Index of self-efficacy and attentiveness to own health. Age, socioeconomic status [3 to 21], self-efficacy [1 to 5] and how much participants pay attention to their health [1 to 5] were included as continuous variables in the model. Categorial variables included in the model were: Sex (male/female), limitation due health problems for at least 6 months (high limitation/moderate limitation/none), occupational status [employed/unemployed/ retired or unable to work], living situation (as married or unmarried couple/ living alone] and how much participants do generally pay attention to their health [not at all/not so much/moderate/much/very much).

†Results were calculated using weighted multinomial logistic regression analyses, BMI categorization was according to WHO categorization¹⁶ as normal (<25 kg/m²), overweight (25-<30 kg/m²) and obese (≥30 kg/m²).

‡Results were calculated using binary weighted-logistic regression analyses.

BMI, body mass index; diabetes, diabetes mellitus; DSME, diabetes self-management education; PA, physical activity; SES, socioeconomic status.