Machine learning for predicting diabetes risk in western China adults

Table 2 Multivariate logistic regression analysis in the development group

Category	Features	Multivariate logistic regression analysis
Category	Features	Beta	OR(95%CI)	P value
Questionnaire	Sex, n(%)
	Male
	Female	0.1095	1.116 (1.103 1.129)	< 0.001
	Age(year)	0.0299	1.03 (1.030 1.031)	< 0.001
	Ethnicity, n(%)
	Uyghur
	Han	0.384	1.469 (1.452 1.485)	< 0.001
	Kazak	0.549	0.578 (0.564 0.592)	< 0.001
	Hui	0.465	1.592 (1.56 1.624)	< 0.001
	Khalkha	0.732	0.481 (0.445 0.52)	< 0.001
	Mongol	0.524	0.592 (0.554 0.632)	< 0.001
	Tajik	1.09	0.335 (0.264 0.419)	< 0.001
	Other	0.142	1.153 (1.094 1.215)	< 0.001
	EH, n(%)
	Balanced diet
	Meat based	0.0762	1.079 (1.037 1.123)	< 0.001
	Vegetarian based	0.018	1.018 (0.985 1.052)	0.282349
	SS, n(%)
	Never smoked
	smoking	0.0645	1.067 (1.048 1.086)	< 0.001
	Quit smoking	0.171	1.186 (1.131 1.244)	< 0.001
	HTN, n(%)
	No
	Yes	1.212	3.361 (3.324 3.398)	< 0.001
	CAD, n(%)
	No
	Yes	0.471	1.603 (1.579 1.627)	< 0.001
	PDM, n(%)
	No
	Yes	1.310	3.706 (3.575 3.841)	< 0.001
Routine examination	WC (cm)	0.0121	1.012 (1.012 1.013)	< 0.001
Routine examination	BMI (kg/m2)	0.0147	1.015 (1.013 1.017)	< 0.001
Laboratory test	HGB, g/L	0.001	0.999 (0.999 0.999)	< 0.001
	WBC, × 10⁹/L	0.126	1.134 (1.13 1.138)	< 0.001
	PLT, × 10⁹/L	0.002	0.998 (0.998 0.998)	< 0.001
	FBG, mmol/L	0.946	2.574 (2.556 2.593)	< 0.001
	ECG, n(%)
	TC, mmol/L	0.005	1.005 (0.999 1.012)	0.090
	TG, mmol/L	0.255	1.29 (1.278 1.303)	< 0.001
	LDLC, mmol/L	0.019	0.981 (0.974 0.988)	< 0.001
	HDLC, mmol/L	0.196	0.822 (0.810 0.833)	< 0.001

EH eating habits, SS smoking status, HTN hypertension, CAD coronary heart disease, PDM parental diabetes mellitus, WC waist circumference, BMI body mass index, HGB hemoglobin, WBC white blood cell, PLT platelet, FBG fasting blood glucose, ECG electrocardiogram, TC total cholesterol, TG triglyceride, LDLC low-density lipoprotein cholesterol, HDLC high-density lipoprotein cholesterol

ISSN: 1758-5996