Table 2 |
||||
|
Multilevel logistic regression of the association between individual and community characteristics and unintended pregnancy in Orellanaa |
||||
|
Model 1b |
Model 2c |
Model 3d |
Model 4e |
|
|
|
||||
|
Individual variables |
OR (95% CI) |
OR (95% CI) |
OR (95% CI) |
|
|
|
||||
|
Age groups |
||||
|
|
||||
|
=>40 |
ref |
ref |
||
|
|
||||
|
30-39 |
1.11 (0.68-1.80) |
1.13 (0.70-1.84) |
||
|
|
||||
|
20-29 |
1.75 (1.04-2.93) |
1.78 (1.06-3.00) |
||
|
|
||||
|
15-19 |
2.31 (1.21-4.40) |
2.34 (1.22-4.47) |
||
|
|
||||
|
Level of education |
||||
|
|
||||
|
More than primary |
ref |
ref |
||
|
|
||||
|
Primary |
1.48 (1.09-2.0) |
1.53 (1.12-2.10) |
||
|
|
||||
|
No school |
1.82 (1.14-2.89) |
1.89 (1.18-3.03) |
||
|
|
||||
|
Marital status |
||||
|
|
||||
|
Married/union |
ref |
Ref |
||
|
|
||||
|
Single |
1.98 (1.27-3.09) |
2.00 (1.28-3.11) |
||
|
|
||||
|
Other |
1.01(0.51-1.98) |
0.98 (0.50-1.94) |
||
|
|
||||
|
Ethnic group |
||||
|
|
||||
|
Non-indigenous |
ref |
ref |
||
|
|
||||
|
Indigenous |
1.65 (1.21-2.24) |
2.01 (1.10-3.65) |
||
|
|
||||
|
No. of living children |
||||
|
|
||||
|
1-2 |
ref |
ref |
||
|
|
||||
|
3-4 |
2.55 (1.77-3.67) |
2.56 (1.77-3.70) |
||
|
|
||||
|
>4 |
5.29 (3.28-8.52) |
5.46 (3.38-8.48) |
||
|
|
||||
|
Community variables |
||||
|
|
||||
|
Ethnic community |
||||
|
|
||||
|
Neighbourhood urban |
ref |
ref |
||
|
|
||||
|
Neighbourhood rural |
1.31 (0.84-2.04) |
0.82 (0.53-1.27) |
||
|
|
||||
|
Non-indigenous rural |
0.95 (0.63-1.42) |
0.84 (0.55-1.27) |
||
|
|
||||
|
Indigenous rural |
2.02 (1.40-2.92) |
0.72 (0.37-1.40) |
||
|
|
||||
|
Random effects |
||||
|
|
||||
|
Community random variance (SE)f |
0.11 (0.06) |
0.00 (0.00) |
0.003 (0.034) |
0.00 (0.00) |
|
|
||||
|
VPC (%)f |
3.37 |
0.0 |
0.09 |
0.0 |
|
|
||||
|
PCV (%)f |
reference |
100 |
97.39 |
100 |
|
|
||||
|
MORf |
1.38 |
1.00 |
1.05 |
1.00 |
|
|
||||
|
a Variables that showed statistically significant association are in bold. b Model 1 presents a null model with no fixed-effect predictors and a random variance component for communities. c Model 2 presents individual predictor variables in the fixed part and a random variance at community level. d Model 3 presents community predictor variables in the fixed part and a random variance at community level. e Model 4 presents both individual and community predictor variables in the fixed part and a random variance at community level. f SE= Standard error; VPC= Variance partition coefficient; PCV=Percentage change in variance; MOR= Median odds ratio. |
||||
|
Goicolea and San Sebastian International Journal for Equity in Health 2010 9:14 doi:10.1186/1475-9276-9-14 |
||||
