Table 3.
Unadjusted and adjusted associations between sociodemographic and family characteristics, health seeking behaviour, family planning and unsafe abortion
| Women who had unsafe abortion* (n=58 266) |
Women who had safe abortion* (n=30 845) |
Unadjusted OR (95% CI)† missing indicator (n=89 111) |
P value | Adjusted OR‡ (95% CI)† missing indicator (n=89 111) |
P value§ | |
| Sociodemographic characteristics | ||||||
| Maternal age (years) | ||||||
| 15–19 | 2548 (4.2) | 1041 (3.4) | 1.22 (1.11 to 1.33) | <0.001 | 1.09 (1.00 to 1.18) | <0.001 |
| 20–24 | 17 852 (30.3) | 7889 (24.6) | 1.17 (1.12 to 1.23) | 1.13 (1.09 to 1.18) | ||
| 25–29 | 18 932 (32.3) | 10 354 (32.4) | 1.00 (ref) | 1.00 (ref) | ||
| 30–34 | 10 908 (18.9) | 6716 (22.4) | 0.86 (0.82 to 0.90) | 0.88 (0.84 to 0.92) | ||
| 35–39 | 5508 (9.7) | 3460 (12.0) | 0.87 (0.82 to 0.92) | 0.82 (0.78 to 0.87) | ||
| 40–44 | 1833 (3.4) | 1070 (3.9) | 0.94 (0.85 to 1.04) | 0.82 (0.75 to 0.89) | ||
| ≥45 | 685 (1.3) | 315 (1.3) | 1.00 (0.86 to 1.18) | 0.95 (0.83 to 1.09) | ||
| Missing | 0 (0) | 0 (0) | N/A | N/A | ||
| Marital status | ||||||
| Married | 58.012 (99.6) | 30 729 (99.6) | 0.91 (0.69 to 1.20) | 0.5125 | N/A¶ | |
| Single | 254 (0.4) | 116 (0.4) | 1.00 (ref) | N/A¶ | ||
| Missing | 0 (0) | 0 (0) | N/A | N/A¶ | ||
| Maternal education | ||||||
| Tertiary and above | 2970 (4.8) | 2242 (7.7) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| Secondary school | 17 641 (27.3) | 12 032 (34.2) | 1.27 (1.18 to 1.37) | 0.94 (0.88 to 1.00) | ||
| Primary school/below | 14 395 (23.8) | 7644 (24.3) | 1.56 (1.44 to 1.68) | 1.13 (1.05 to 1.20) | ||
| Illiterate | 23 165 (43.9) | 8879 (32.6) | 2.07 (1.92 to 2.23) | 1.48 (1.39 to 1.59) | ||
| Missing | 95 (0.2) | 48 (0.1) | 1.92 (1.21 to 3.05) | N/A** | ||
| Maternal employment status | ||||||
| In-paid employment | 740 (11.0) | 3716 (10.6) | 1.00 (ref) | 0.0606 | N/A¶ | |
| Not in-paid employment | 50 474 (88.3) | 26 927 (88.8) | 0.96 (0.91 to 1.01) | N/A¶ | ||
| Missing | 391 (0.7) | 202 (0.6) | N/A¶ | |||
| Place of residence | ||||||
| Urban | 9455 (18.4) | 7127 (27.6) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| Rural | 48 811 (81.6) | 23 718 (72.4) | 1.69 (1.62 to 1.76) | 1.26 (1.21 to 1.32) | ||
| Missing | 0 (0) | 0 (0) | N/A | |||
| Religion | ||||||
| Hindu | 47 322 (80.3) | 25 948 (82.4) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| Muslim | 9209 (17.8) | 4319 (16.2) | 1.13 (1.08 to 1.18) | 1.16 (1.12 to 1.22) | ||
| Christian | 625 (0.9) | 253 (0.7) | 1.33 (1.11 to 1.60) | 1.39 (1.20 to 1.62) | ||
| Others | 1016 (0.9) | 277 (0.6) | 1.48 (1.23 to 1.76) | 2.05 (1.78 to 2.36) | ||
| Missing | 94 (0.2) | 48 (0.1) | 1.23 (0.78 to 1.95) | N/A†† | ||
| Social group | ||||||
| Others | 41 708 (72.4) | 22 826 (74.8) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| Schedule caste | 11 351 (20.5) | 5290 (18.9) | 1.12 (1.07 to 1.17) | 1.08 (1.04 to 1.12) | ||
| Schedule tribe | 5113 (7.0) | 2681 (6.2) | 1.17 (1.10 to 1.25) | 0.86 (0.81 to 0.90) | ||
| Missing | 94 (0.2) | 48 (0.1) | 1.24 (0.78 to 1.97) | N/A** | ||
| Wealth/asset index | ||||||
| 5: Highest | 11 977 (20.6) | 8684 (30,2) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| 4 | 11 496 (19.8) | 6726 (21.8) | 1.29 (1.23 to 1.36) | 1.08 (1.03 to 1.13) | ||
| 3 | 11 012 (19.8) | 5421 (17.5) | 1.57 (1.49 to 1.66) | 1.18 (1.13 to 1.24) | ||
| 2 | 10 113 (16.8) | 4521 (13.7) | 1.67 (1.58 to 1.76) | 1.24 (1.18 to 1.31) | ||
| 1: Lowest | 9443 (15.7) | 3348 (9.8) | 2.02 (1.90 to 2.14) | 1.45 (1.38 to 1.53) | ||
| Missing | 4225 (7.2) | 2145 (7.1) | 1.46 (1.36 to 1.57) | 1.33 (1.25 to 1.42) | ||
| Health seeking behaviour | ||||||
| Antenatal care use | ||||||
| No | 46 392 (79.7) | 22 656 (71.2) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| Yes | 11 874 (20.3) | 8189 (28.8) | 0.63 (0.60 to 0.66) | 0.69 (0.67 to 0.72) | ||
| Missing | (0) | (0) | N/A | N/A | ||
| Family characteristics | ||||||
| Number of surviving children at the point of interview | ||||||
| 0 | 2203 (3.9) | 703 (1.9) | 1.85 (1.66 to 2.07) | <0.001 | 1.30 (1.16 to 1.46) | <0.001 |
| 1–3 | 35 751 (59.8) | 20 165 (61.5) | 1.00 (ref) | 1.00 (ref) | ||
| 4 or more | 8991 (17.2) | 4924 (19.8) | 0.94 (0.90 to 0.99) | 0.96 (0.91 to 1.00) | ||
| Missing | 11 321 (19.2) | 5053 (16.8) | 1.06 (1.01 to 1.11) | 0.79 (0.68 to 0.93) | ||
| Proportion of surviving female children among total surviving children | ||||||
| 0% | 12 596 (21.0) | 7206 (21.8) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| 10%–30% | 6148 (11.6) | 3554 (12.4) | 0.97 (0.91 to 1.03) | 0.97 (0.92 to 1.03) | ||
| 40%–60% | 11 704 (20.4) | 6481 (21.2) | 0.99 (0.94 to 1.05) | 1.05 (1.00 to 1.10) | ||
| 70%–90% | 4736 (8.3) | 2749 (9.0) | 0.96 (0.90 to 1.03) | 1.00 (0.94 to 1.06) | ||
| 100% | 10 343 (17.0) | 5379 (16.1) | 1.10 (1.04 to 1.15) | 1.12 (1.07 to 1.17) | ||
| Missing | 12 739 (21.6) | 5476 (19.6) | 1.14 (1.08 to 1.20) | 1.43 (1.23 to 1.66) | ||
| Family planning | ||||||
| Use of family planning methods at the point of interview | ||||||
| No | 24 762 (42.2) | 10 325 (34.7) | 1.00 (ref) | <0.001 | 1.00 (ref) | <0.001 |
| Yes | 22 823 (39.1) | 16 101 (49.4) | 0.65 (0.63 to 0.67) | 0.69 (0.66 to 0.71) | ||
| Missing | 10 681 (18.8) | 4419 (15.9) | 0.97 (0.92 to 1.02) | 1.02 (0.97 to 1.06) | ||
| Others | ||||||
| Self-reported mental illness | ||||||
| No | 57 752 (99.0) | 30 589 (99.1) | 1.0 (ref) | 0.5773 | N/A¶ | |
| Yes | 420 (0.8) | 207 (0.7) | 1.07 (0.88 to 1.31) | N/A¶ | ||
| Missing | 94 (0.2) | 49 (0.1) | 1.20 (0.76 to 1.90) | N/A¶ | ||
*Frequencies are unweighted (true counts). Rates/proportions are weighted for survey design and clustering effects.
†The 95% CIs were calculated using linearised standard errors p value for a t-test showing the significance level for the overall univariable regression model.
‡Multivariable logistic regression, adjusting for the other variables/potential risk factors in the final model.
§P value for a Wald test indicating whether the incremental adjustment for an independent variable significantly improves the fit of the model.
¶N/As are shown because these variables were not included in the final multivariable logistic model.
**N/As are shown because the OR associated with the missing category was extremely large, which can be explained by the small number of missing values in each group.
††N/A is shown because the OR was omitted due to collinearity between missing categories.
‡‡P value for a t-test showing the significance level for the overall univariable regression model.