Skip to main content

Table 2 Factors related to anemia: results from two-level logistic regression in 2001 and 2005 a

From: Assessment of maternal anemia in rural Western China between 2001 and 2005: a two-level logistic regression approach

Characteristics

2001

2005

 

OR(95%CI)

P-value

OR(95%CI)

P-value

Null model

 

u 0j

0.7 ± 0.1

<0.001

0.8 ± 0.1

<0.001

VPC

16.5%

-

20.4%

-

Full model

 

Fixed part

 

Provinces b

 

Ningxia

0.58(0.40,0.83)

<0.01

0.63(0.33,1.22)

0.173

Xinjiang

0.63(0.34,1.14)

0.124

1.14(0.62,2.11)

0.663

Gansu

0.77(0.51,1.15)

0.201

0.31(0.13,0.74)

<0.01

Inner Mongolia

0.85(0.59,1.23)

0.386

0.49(0.26,0.93)

<0.05

Sichuan

1.04(0.72,1.51)

0.83

1.21(0.68,2.17)

0.515

Guizhou

-

-

1.56(0.82,2.96)

0.174

Chongqing

1.20(0.80,1.82)

0.378

2.00(1.03,3.85)

<0.05

Qinghai

1.46(1.01,2.11)

<0.05

1.86(1.00,3.48)

<0.05

Guangxi

2.32(1.54,3.50)

<0.001

1.31(0.71,2.44)

0.387

Individual characteristics

 

Ethnicity

 

Han

1.00

 

1.00

 

Tibet

1.29(0.92,1.79)

0.136

1.23(0.79,1.92)

0.355

Uighur

2.35(1.28,4.30)

<0.01

1.42(0.74,2.73)

0.288

Hui

1.36(0.99,1.87)

0.060

0.57(0.40,0.82)

<0.01

Zhuang

1.06(0.68,1.63)

0.806

2.09(1.34,3.26)

<0.01

Others

1.26(0.93,1.70)

0.133

1.42(1.05,1.93)

<0.05

Parity

 

1

1.00

 

1.00

 

2

0.91(0.79,1.05)

0.194

0.84(0.71,0.98)

<0.05

>2

0.76(0.57,0.99)

<0.05

0.83(0.62,1.12)

0.228

Wealth index

 

Poorest

1.00

 

1.00

 

Middle

0.93(0.81,1.06)

0.280

1.05(0.91,1.22)

0.496

Wealthiest

0.90(0.80,0.99)

<0.05

1.09(0.92,1.28)

0.313

Mothers’ age, years

1.00(0.99,1.03)

0.252

1.02(1.01,1.04)

<0.01

Breastfeeding duration

0.99(0.98,1.00)

<0.05

0.995(0.991,0.999)

<0.05

County characteristics

 

Altitude, m

 

<500

1.00

 

1.00

 

500 ~ 1500

1.03(0.70,1.53)

0.868

1.19(0.69,2.04)

0.530

>1500

2.47(1.58,3.84)

<0.001

1.86(1.05,3.64)

<0.05

Random part

 

u 0j

0.4 + 0.1

<0.001

−1.7 ± 0.3

<0.001

VPC

10.2%

-

16.2%

-

  1. Abbreviation: VPC variance partition coefficient, u 0j , random effect.
  2. a Values were given as mean ± SD or odds ratio (95%CI) unless otherwise stated.
  3. b Effect coding was used for provinces. The reference group is the average of all 10 provinces.