Table 3 |
||||
|
Odds ratios and 95% confidence intervals for multilevel logistic regression models |
||||
|
Variables |
Model 0 (Empty model) |
Model 1 (Child-level variables) |
Model 2 (Mother-level variables) |
Model 3 (Community-level variables) |
|
|
||||
|
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
|
|
|
||||
|
Fixed effects |
||||
|
Individual characteristics |
||||
|
Sex of child |
||||
|
Female |
1.02 (0.85-1.22) |
1.01 (0.83-1.22) |
1.00 (0.80-1.26) |
|
|
Male |
1 |
1 |
1 |
|
|
Birth order/birth interval |
||||
|
First birth (order 1) |
1.03 (0.79-1.34) |
0.98 (0.72-1.33) |
0.99 (0.69-1.41) |
|
|
Order 2-4 & <24 months |
1.00 (0.73-1.38) |
1.03 (0.73-1.44) |
0.89 (0.62-1.31) |
|
|
Order 2-4 & 24-47 months |
1 |
1 |
1 |
|
|
Order 2-4 & 48+ months |
0.78 (0.52-1.17) |
0.66 (0.43-1.01) |
0.65 (0.39-1.07) |
|
|
Order 5+ & < 24 months |
0.51 (0.33-0.79)* |
0.54 (0.33-0.89) |
0.61 (0.34-1.09) |
|
|
Order 5+ & 24-47 months |
0.85 (0.65-1.11) |
0.84 (0.60-1.18) |
0.87 (0.59-1.30) |
|
|
Order 5+ & 48+ months |
0.98 (0.67-1.42) |
0.84 (0.54-1.32) |
0.70 (0.41-1.21) |
|
|
Mother's age |
||||
|
15-18 |
0.55 (0.27-1.11) |
0.45 (0.19-1.10) |
||
|
19-23 |
0.99 (0.73-1.36) |
0.89 (0.62-1.29) |
||
|
24-28 |
1 |
1 |
||
|
29-33 |
0.98 (0.72-1.32) |
0.81 (0.57-1.16) |
||
|
≥ 34 |
1.33 (0.95-1.85) |
1.33 (0.91-1.94) |
||
|
Marital status |
||||
|
Single |
1.18 (0.52-2.66) |
0.92 (0.35-2.40) |
||
|
Currently married |
0.89 (0.53-1.49) |
0.67 (0.37-1.19) |
||
|
Formerly married |
1 |
1 |
||
|
Ethnicity |
||||
|
Hausa/Fulani/Kanuri |
1 |
1 |
||
|
Igbo |
1.66 (1.15-2.41) |
2.47 (1.28-4.76) |
||
|
Yoruba |
1.90 (1.29-2.81) |
1.47 (0.78-2.80) |
||
|
Others |
1.68 (1.27-2.23) |
1.47 (0.96-2.25) |
||
|
Mothers' education |
||||
|
No education |
0.85 (0.62-1.16) |
0.85 (0.58-1.25) |
||
|
Primary |
1.05 (0.79-1.39) |
1.04 (0.74-1.48) |
||
|
Secondary or higher |
1 |
1 |
||
|
Mothers' occupation |
||||
|
Prof./Tech./Manag. |
1 |
1 |
||
|
Clerical, sales, services, skilled manual |
0.62 (0.40-0.96) |
0.56 (0.34-0.93) |
||
|
Agric. self., Agric. employee, household & domestic, unskilled manual |
0.81 (0.49-1.33) |
0.88 (0.49-1.60) |
||
|
Not working |
0.68 (0.43-1.08) |
0.68 (0.40-1.16) |
||
|
Wealth index |
||||
|
Poorest |
0.44 (0.29-0.65) |
0.35 (0.21-0.59) |
||
|
Poorer |
0.49 (0.34-0.71) |
0.47 (0.27-0.81) |
||
|
Middle |
0.66 (0.47-0.93) |
0.64 (0.42-0.97) |
||
|
Richer |
0.63 (0.46-0.87) |
0.57 (0.39-0.84) |
||
|
Richest |
1 |
1 |
||
|
Community characteristics |
||||
|
Community mother's education |
||||
|
Low |
0.76 (0.55-1.06) |
|||
|
Middle |
1 |
|||
|
High |
1.02 (0.73-1.41) |
|||
|
Community hospital delivery |
||||
|
Low |
0.62 (0.40-0.94) |
|||
|
Middle |
1 |
|||
|
High |
1.12 (0.75-1.68) |
|||
|
Community prenatal care by doctor |
||||
|
Low |
1.20 (0.79-1.84) |
|||
|
High |
1 |
|||
|
Region of residence |
||||
|
North Central |
1.15 (0.69-1.91) |
|||
|
North East |
0.90 (0.48-1.70) |
|||
|
North West |
0.83 (0.42-1.67) |
|||
|
South East |
0.36 (0.18-0.73) |
|||
|
South South |
0.48 (0.26-0.87) |
|||
|
South West |
1 |
|||
|
Random effects |
Empty |
Child-level |
Mother-level |
Community-level |
|
Community-level |
||||
|
Variance (SE) |
0.393 (0.119)** |
0.567 (0.189)** |
0.210 (0.112) |
0.272 (0.133)* |
|
VPC (%) |
9.7 |
13.6 |
4.9 |
13.9 |
|
Explained variation (PVC) (%) |
Reference |
44.3 |
63 |
-29.5 |
|
Mother-level |
||||
|
Variance (SE) |
0.382 (0.193)* |
0.296 (0.145)* |
0.790 (0.216)* |
0.576 (0.241)* |
|
VPC (%) |
9.4 |
7.1 |
18.4 |
13.9 |
|
Explained variation (PVC) (%) |
Reference |
22.5 |
-170 |
27 |
|
Model fit statistics |
||||
|
DIC |
3341 |
3337 |
3158 |
2269 |
|
|
||||
|
Note: Model 0 contained no variables; Model 1 included child-level characteristics; Model 2 adjusted for mother-level characteristics; Model 3 additionally adjusted for community-level characteristics. Abbreviations: VPC = Variance partition coefficient; DIC = Deviance information criterion; SE = Standard error: OR = Odds ratio; CI = Confidence Interval. Data source: 2003 Nigeria Demographic and Health Survey *p < .05; **p < .01; ***p < .001 |
||||
|
Antai BMC Infectious Diseases 2009 9:181 doi:10.1186/1471-2334-9-181 |
||||
