Table 2.
Multilevel logistics regression model for individual and household/community-level factors associated with long-acting reversible contraceptive use among sexcually active women of reproductive age in Nigeria
| Variables | Model 0 | Model I AOR [95%CL] |
Model II AOR [95%CL] |
Model III AOR [95%CL] |
|---|---|---|---|---|
| Age | ||||
| 15–24 | 1 | 1 | ||
| 25–34 | 0.85(0.61–1.18) | 0.84(0.61–1.18) | ||
| 35 and above | 0.87(0.59–1.27) | 0.86(0.59–1.26) | ||
| Working status | ||||
| Yes | 1.28(1.00-1.64) | 1.28(1.00-1.63) | ||
| No | 1 | 1 | ||
| Marital status | ||||
| Never married | 1 | 1 | ||
| Currently married | 1.08(0.69–1.68) | 0.98(0.63–1.52) | ||
| Pregnancy termination | ||||
| Yes | 0.63(0.49–0.81) ** | 0.62(0.48–0.80) ** | ||
| No | 1 | 1 | ||
| Educational level | ||||
| No formal education | 1 | 1 | ||
| Primary education | 1.37(0.99–1.90) | 1.37(0.99–1.90) | ||
| Secondary education & above | 1.59(1.16–2.18) ** | 1.63(1.17–2.26) ** | ||
| Husband education level | ||||
| No Education | 1 | 1 | ||
| Primary | 0.94(0.64–1.38) | 0.95(0.65–1.39) | ||
| Secondary education & above | 1.03(0.73–1.46) | 1.00(0.71–1.41) | ||
| Total children ever born | ||||
| No child | 1 | 1 | ||
| 1–2 children | 9.66(1.18–78.7) | 9.47(1.17–76.5) | ||
| 3–4 children | 13.2(1.61–108.6) | 12.7(1.56–103.8) | ||
| 5 & above children | 15.4(1.86–127.5) | 14.8(1.80-121.8) | ||
| visited health facility last 12 months | ||||
| no | 1 | 1 | ||
| yes | 1.12(0.93 1.33) | 1.08(0.90–1.29) | ||
| fertility preference | ||||
| want more | 1 | 1 | ||
| Undecided | 1.46(1.01–2.09) | 1.48(1.04–2.12) | ||
| want no more | 1.62(1.29–2.05) ** | 1.65(1.30–2.08) ** | ||
| Media | ||||
| No | 1 | 1 | ||
| Yes | 1.07(0.83–1.38) | 1.11(0.85–1.44) | ||
| Religion | ||||
| Christianity | 1 | 1 | ||
| Islam | 0.84(0.65–1.09) | 0.74(0.57–0.97) | ||
| Traditionalist & Others | 1.65(0.36–7.57) | 2.50(0.56–10.9) | ||
| Ethnicity | ||||
| Hausa | 1 | 1 | ||
| Yoruba | 0.67(0.46–0.97) | 1.14(0.70–1.87) | ||
| Igbo | 0.34(0.22–0.52) ** | 0.97(0.56–1.69) | ||
| Others | 0.79(0.56–1.11) | 1.27(0.86–1.89) | ||
| Household/Community level | ||||
| Residence | ||||
| Urban | 1 | 1 | ||
| Rural | 0.82(0.64–1.04) | 0.77(0.60–0.98) | ||
| Wealth index | ||||
| Poorest | 1 | 1 | ||
| Poorer | 0.97(0.65–1.45) | 0.89(0.59–1.34) | ||
| Average | 1.05(0.71–1.57) | 0.97(0.64–1.47) | ||
| Richer | 1.05(0.69–1.60) | 0.99(0.63–1.55) | ||
| Richest | 1.08(0.67–1.74) | 1.00(0.60–1.67) | ||
| Geopolitical Zone | ||||
| North-Central | 1 | 1 | ||
| North-East | 0.52(0.37–0.75) ** | 0.71(0.49–1.03) | ||
| North-West | 1.21(0.86–1.69) | 2.07(1.36–3.16) ** | ||
| South-East | 0.33(0.22–0.47) ** | 0.32(0.19–0.53) ** | ||
| South-South | 0.63(0.44–0.90) | 0.56(0.38–0.81) ** | ||
| South-West | 0.66(0.49–0.90) ** | 0.68(0.46-1.00) | ||
| Sex of household head | ||||
| Male | 1 | 1 | ||
| Female | 0.92(0.65–1.30) | 0.97(0.68–1.37) | ||
| Community literacy level | ||||
| Low | 1 | 1 | ||
| Medium | 1.12(0.86–1.47) | 0.99(0.75–1.31) | ||
| High | 1.21(0.90–1.64) | 1.05(0.76–1.44) | ||
| Community socioeconomic status | ||||
| Low | 1 | 1 | ||
| Medium | 0.86(0.63–1.16) | 0.84(0.62–1.14) | ||
| High | 0.64(0.44–0.94) | 0.66(0.45–0.97) * | ||
| PSU Variance (95% CI) | 0.96[0.70–1.32] | 0.84[0.59–1.19] | 0.71[0.49–1.04] | 0.67[0.45–0.99] |
| ICC | 0.23 | 0.20 | 0.18 | 0.17 |
| LR Test | χ2 = 108.17 p < 0.001 | χ2 = 80.24, p < 0.001 | χ2 = 62.96, p < 0.001 | χ2 = 53.19, p < 0.001 |
| Wald χ2 | Reference | 114.52*** | 62.79*** | 165.05*** |
| Model fitness | ||||
| Log-likelihood | -2075.31 | -2010.97 | -2044.33 | -1982.74 |
| AIC | 4154.62 | 4067.93 | 4122.66 | 4041.48 |
| Number of clusters | 1042 | 1042 | 1042 | 1042 |
Weighted NDHS, 2018
Exponentiated coefficients; 95% confidence intervals in brackets; AOR = adjusted Odds Ratios; CI = Confidence Interval; RC = Reference Category
*p < 0.05; **p < 0.01; ***p < 0.001
PSU = Primary Sampling Unit; ICC = Intra-Class Correlation; LR Test = Likelihood ratio Test; AIC = Akaike’s Information Criterion
Model 0 is the null model, a baseline model without any determinant variable
Model I include individual-level variables (Age of respondent, working status, marital status, educational level, husband educational level, total children ever born, visited health facility last 12 months, fertility preference, media, religious, ethnicity)
Model II include household/community level variables (Place of residence, wealth index, region, sex of household head, community literacy level, and community socioeconomic status)
Model III the final model included both individual and household/community level variables