Skip to main content
. 2022 Apr 11;22:720. doi: 10.1186/s12889-022-13007-0

Table 2.

Associations between diabetes risk factors and undiagnosed pre-or type 2 diabetes (HbA1c ≥ 39 mmol/mol), Saharawi refugee camps Algeria, 2014; from logisitic regressions (n = 301 in adjusted analysis)

Bivariate model Adjusted modela Regression coefficient (B) Scoreb Regression coefficient
after shrinkage
(B)c
Scoreb
OR (95% CI) OR (95% CI)
Intercept -4.16 -3.76
Gender (female as reference) 0.49 (0.25, 0.95)
Age (vs.< 45 years)
 ≥ 45 and < 60 years 4.66 (2.09, 10.37) 3.62 (1.51, 8.68) 1.29 4 1.10 4
 ≥ 60 3.78 (1.64, 8.72) 3.60 (1.45, 8.91) 1.28 4 1.10 4
BMIc (vs. < 25 kg/m2)
 ≥ 25 and < 30 kg/m2 6.78 (2.65, 17.31) 4.08 (1.40, 11.85) 1.41 5 1.21 4
 ≥ 30 kg/m2 10.86 (4.25, 27.73) 3.74 (1.12, 12.55) 1.32 4 1.13 4
WCd (vs. F: < 80 cm/M: < 94 cm)
 F: ≥ 80 and < 88 cm/M: ≥ 94 and < 102 cm 5.27 (1.73, 16.04) 2.31 (0.65, 8.12) 0.84 3 0.72 2
 F: ≥ 88/M: ≥ 102 10.84 (4.29, 27.38) 3.57 (1.08, 11.77) 1.27 4 1.09 4
Physical activity (vs. none)
 At least 30 min per day 0.63 (0.32, 1.25)
Intake of fruits and vegetables (vs. no)
 At least once per day 0.83 (0.41, 1.66)
Blood pressure medication (vs. no) 0.99 (0.97, 1.02)
Diabetes in first-degree relative (vs. no) 1.76 (0.82, 3.76)

a Adjusted model created by backward elimination (starting with all variables tested in bivariate models), variables significantly associated with undiagnosed prediabetes or T2D were kept in the model.

b Based on Schneeweiss’s scoring system [27] (Schneeweiss, 2003 #258)

c Calculated from internal validation using Bootstrap analysis

d Body mass index

e Waist circumference