Skip to main content

Table 5 Three-level mixed-effects logistic regression modeling associations between individual, household, and community level factors and women’s modern contraceptive use

From: Women’s empowerment, intrahousehold influences, and health system design on modern contraceptive use in rural Mali: a multilevel analysis of cross-sectional survey data

Variables

n

%

Adjusted OR (95% CI)

p valueb

Individual level

    

 Age in years

   

 < 0.0001

  15–19

1839

13.1

1.0

Ref

  20–24

2377

16.9

2.37 (1.52, 3.68)

 < 0.001

  25–29

2746

19.6

1.89 (1.18, 3.04)

0.008

  30–34

2447

17.4

2.29 (1.39, 3.77)

0.001

  35–39

2998

14.2

1.97 (1.17, 3.32)

0.010

  40–44

1499

10.7

1.31 (0.75, 2.27)

0.342

  45–49

1126

8.0

0.80 (0.43, 1.49)

0.485

 Number of living children

   

0.0053

  None

2376

17.0

1.0

Ref

  1–2

3858

27.5

1.52 (1.05, 2.20)

0.028

  3–4

3674

26.2

1.42 (0.95, 2.13)

0.084

  5–6

2688

19.2

1.60 (1.04, 2.46)

0.031

  7 + 

1416

10.1

2.39 (1.49, 3.83)

 < 0.001

 Marital status

   

0.0501

  Not currently married

1341

9.6

1.0

Ref

  Polygynous

5409

38.6

1.61 (0.97, 2.67)

0.067

  Monogamous

7272

52.9

1.79 (1.09, 2.92)

0.020

 Education

    

  None

12,515

89.3

1.0

Ref

  Any

1499

10.7

3.28 (2.52, 4.27)

 < 0.001

 Participates in paid labor

    

  No

12,111

86.3

1.0

Ref

  Yes

1917

13.7

1.71 (1.35, 2.17)

 < 0.001

 Mobility

   

0.0028

  Been to no place

4767

34.1

1.0

Ref

  Been to any place but none alone

2498

17.9

0.58 (0.42, 0.79

0.001

  Been to any place alone

6728

48.1

0.82 (0.64, 1.03)

0.093

 Spousal violence

    

  Not tolerated

3853

28.1

1.0

Ref

  Tolerated

9872

71.9

1.00 (0.80, 1.25)

0.978

 Decision-making

    

  None

10,210

73.0

1.0

Ref

  Any

3769

27.0

1.29 (1.04, 1.60)

0.019

Household level

    

 Someone else in the household using modern contraception

    

  No

13,682

97.7

1.0

Ref

  Yes

324

2.3

3.04 (1.95, 4.73)

 < 0.001

 Distance to health centerc

   

 < 0.0001

  < 2 km

2859

20.4

1.0

Ref

  2–4.99 km

3586

25.6

0.50 (0.33, 0.75)

0.001

  5–6.99 km

3329

23.7

0.33 (0.20, 0.53)

 < 0.001

  7–9.99 km

2677

19.1

0.33 (0.19, 0.55)

 < 0.001

  ≥ 10

1581

11.3

0.70 (0.38, 1.31)

0.266

 Food insecurity

    

  None

11,931

85.1

1.0

Ref

  Any

2084

14.9

1.05 (0.78, 1.41)

0.745

 Wealth quintiled

   

0.4582

  Richest

2385

17.2

1.0

Ref

  Rich

2592

18.7

1.04 (0.79, 1.37)

0.788

  Middle

2689

19.4

0.91 (0.68, 1.23)

0.554

  Poor

2997

21.6

0.86 (0.64, 1.18)

0.354

  Poorest

3193

23.0

0.78 (0.55, 1.09)

0.140

Community level

    

 CHW services available

   

0.2920

  None

9649

68.8

1.0

 

  Satellite village

1908

13.6

0.81 (0.48, 1.39)

0.452

  Posted village

2475

17.6

1.27 (0.82, 1.97)

0.291

 Health catchment areae

   

 < 0.0001

  Dimbal

3145

22.4

1.0

Ref

  Lessagou

2195

15.6

1.90 (1.14, 3.17)

0.014

  Doundé

1756

12.5

1.75 (1.00, 3.07)

0.051

  Ende

728

5.2

3.28 (1.67, 6.44)

0.001

  Soubala

2407

17.2

0.56 (0.30, 1.03)

0.063

  Kanibozon

1448

10.3

4.05 (2.36, 6.97)

 < 0.001

  Koulongon

2353

16.8

1.86 (1.10, 3.14)

0.020

Random effects

    

 Village-cluster level (level three) variance (SD)

0.22 (0.09)

 Compound-within-cluster level (level two) variance (SD)

0.98 (0.29)

 Level three ICC (95% CI)

 

0.05 (0.02, 0.10)

 Level two ICC (95% CI)

 

0.27 (0.18, 0.38)

Log likelihood

 

− 2081.39

  1. aN = 13,376 complete cases, or 95% of all women in the analytic sample
  2. bValue provided in line with the categorical variable name is the result of the likelihood ratio test
  3. cVillage distance to health center substituted if household distance to health center was missing (N = 560)
  4. dHousehold wealth quintile in the regression models includes water, sanitation, and hygiene measures, as these are not explored separately
  5. eLargest health catchment area in terms of sample population is used as the reference category