Table 3. Multi-variable multilevel binary logistic regression analysis result of both community and individual level factors associated with utilization of deworming medication in pregnant mothers in Ethiopia, EDHS 2016.
Community and individual level factors | Models | |||
---|---|---|---|---|
Null model | Model I | Model II | Model III | |
(AOR 95%CI) | (AOR 95%CI) | (AOR 95%CI) | (AOR 95%CI) | |
Marital status | ||||
Married/ in union | 1 | 1 | ||
Not married | 1.24 (0.69–2.24) | 1.28 (0.71–2.30) | ||
Religion | ||||
Orthodox | 1 | 1 | ||
Muslim | 1.23 (0.77–1.98) | 1.20 (0.76–1.89) | ||
Protestant | 1.12 (0.69–1.84) | 1.10 (0.67–1.78) | ||
catholic/traditional/other | 1.83(0.52–6.51) | 1.77 (0.51–6.13) | ||
Working status of the respondent | ||||
Not working | 1 | 1 | ||
Working | 1.56 (1.09–2.24) | 1.57 (1.09–2.26)* | ||
Media exposure | ||||
Yes | 1.49 (1.01–2.03) | 1.6 (0.97–2.19) | ||
No | 1 | 1 | ||
ANC visit | ||||
ANC 1 | 1 | 1 | ||
ANC 2 to 3 | 1.99 (0.76–5.21) | 1.983 (0.75–6.51) | ||
≥ ANC 4 | 2.85 (1.05–7.74) | 2.83 (1.03–7.79) * | ||
IFA status | ||||
No | 1 | 1 | ||
Yes | 3.35 (2.11–5.33) | 3.42 (2.15–5.43)* | ||
Decision maker | ||||
Respondent | 1 | 1 | ||
Others | 0.67 (0.41–1.11) | 0.67 (0.41–1.83) | ||
Community education level | ||||
Low | 1 | 1 | ||
High | 0.91(0.65–1.28) | 0.76 (0.51–1.13) | ||
Distance to Health facility | ||||
Big problem | 1 | 1 | ||
Not Big problem | .89(0.64–1.24) | 0.72 (0.50–1.52) | ||
Community poverty level | ||||
High | 1 | 1 | ||
Low | 1.95 (1.35–2.84) | 1.84 (1.23–2.75)* | ||
Measure of random effect (variation) | ||||
Variance | .9913999 | 0.97654 | 0.8624397 | 0.9688124 |
ICC (%) | 23.2% | 23.1% | 20.8% | 22.8% |
PVC (%) | Reference | 1.5% | 13% | 2.3% |
MOR | 2.57 | 2.55 | 2.4 | 2.5 |
Model fitness | ||||
Log likelihood | -1226.4761 | -1159.74 | -1218.96 | -1154.1187 |
Deviance | 2452.95 | 2319.48 | 2437.92 | 2308 |
AIC | 2456.952 | 2343.489 | 2449.92 | 2338.237 |
1- Reference category, AOR-adjusted odds ratio, CI- confidence interval, * and old letter—statistically significant variables in the full model.