Effect of the Mobile Phone Intervention on Primary and Secondary Outcomes at 2 Weeks and 4 Months Follow-Up
| Outcome | Intervention Arm n/N (%) | Control Arm n/N (%) | P Value (χ2) | Unadjusted OR (95% CI) | Adjusted OR (95% CI)a |
|---|---|---|---|---|---|
| Primary outcomes | |||||
| LARC use at 4 months | 48/389 (12%) | 59/383 (15%) | .22 | 0.77 (0.51, 1.17) | 1.06 (0.53, 2.13)b |
| LARC use with MI at 4 months (100 imputations) | 55/484 (11%) | 72/478 (15%) | .11 | 0.73 (0.49, 1.08) | 0.95 (0.49, 1.83)c |
| Secondary outcomes | |||||
| 4-month follow-up | |||||
| Effective modern method use (any method)d | 214/389 (55%) | 204/383 (53%) | .63 | 1.07 (0.81, 1.42) | 1.04 (0.75, 1.43)b |
| Subsequent pregnancy | 6/389 (2%) | 10/383 (3%) | .30 | 0.58 (0.21, 1.62) | 0.48 (0.16, 1.43)e |
| Subsequent MR or abortion | 2/389 (<1%) | 4/383 (1%) | .40 | 0.49 (0.09, 2.69) | 0.43 (0.08, 2.41)e |
| 2-week follow-up | |||||
| LARC use | 48/413 (12%) | 56/411 (14%) | .39 | 0.83 (0.55, 1.26) | 1.57 (0.68, 3.62)b |
| Effective modern method use | 223/413 (54%) | 221/411 (54%) | .95 | 1.01 (0.77, 1.33) | 1.07 (0.74, 1.54)b |
| 4-month contraceptive calendar (log rank test) | Discontinuations/time at risk (weeks) | Discontinuations/time at risk (weeks) | Log rank test | ||
| LARC discontinuation | 2/661 (<1%) | 2/844 (<1%) | .81 | ||
| Injectable discontinuation | 20/620 (3%) | 14/517 (<3%) | .56 | ||
| Pill discontinuation | 27/1693 (2%) | 32/1555 (2%) | .33 | ||
| Adverse outcomes at 4 months | |||||
| Physical IPV | 42/386 (11%) | 25/382 (7%) | .03 | 1.74 (1.04, 2.92) | 1.97 (1.12, 3.46)b |
| Physical IPV with MI (100 imputations)f | 57/484 (12%) | 31/478 (6%) | .03 | 1.87 (1.07, 3.27) | 2.16 (1.16, 4.02)c |
| Sexual IPV | 47/386 (12%) | 36/379 (10%) | .23 | 1.32 (0.83, 2.09) | 1.25 (0.78, 2.02)b |
| Physical violence from in-laws | 6/384 (2%) | 4/380 (1%) | .54 | 1.49 (0.42, 5.33) | 1.20 (0.31, 4.62)b |
| Anything happened to them as a result of being in the study | |||||
| Nothing | 335/385 (92%) | 366/381 (96%) | .06 | ||
| Something good | 28/385 (7%) | 13/381 (3%) | 2.22 (1.13, 4.36) | 2.25 (1.14, 4.44)g | |
| Something bad | 2/385 (<1%) | 2/381 (<1%) | 1.03 (0.14, 7.36) | 2.10 (0.19, 3.60)g |
Abbreviations: CI, confidence interval; IPV, intimate partner violence; IUD, intrauterine device; LARC, long-acting reversible contraception; MI, multiple imputation; MR, menstrual regulation; OR, odds ratio; PPI, Poverty Probability Index; SES, socioeconomic status.
↵a Total number varies due to missing data on baseline covariates included in the model (see Table 1).
↵b Adjusted for baseline use of these methods/experience of this outcome, age, and SES (PPI Score).
↵c Adjusted for baseline use/experience only.
↵d Pill, injection, implant, IUD, or sterilization. At 4 months, participants were asked about how regularly they take the pill and were only classified as pill users if they reported always or usually taking it on time.
↵e Adjusted for baseline LARC use, age, and SES (PPI score).
↵f Analysis not specified in study protocol.
↵g Adjusted for age and SES (PPI score).