Whitaker 2014.

Methods	Location and time frame: Chicago, IL (USA), from May 2007 to July 2010; Evanston, IL (USA), from April 2009 to January 2011 Design: randomized safety and efficacy study Sample size calculation and outcome of focus: assumed 60% of standard insertion would return, 20% of each group would discontinue, estimated 80% postplacental group and 48% standard group would use IUD in 12 months; 46 per group for 90% power to detect difference at alpha = 0.05; assumed 15% loss to follow‐up; planned 108 participants to have 92 participants complete trial; outcome of focus was use at 12 months
Participants	General with N: 42 women in 2 hospitals (20 postplacental and 22 standard group) Inclusion criteria: pregnant; ≥ 18 years; planning scheduled cesarean delivery; desires LNG‐IUS for contraception; speaks English Exclusion criteria: allergy to polyethylene or levonorgestrel or other contraindication to LNG‐IUS use; positive test for gonorrhea, chlamydia, or trichomoniasis during pregnancy without treatment and subsequent test of cure confirming negative result; leiomyomata distorting uterine cavity; uterine anomaly precluding IUS placement; current cervical cancer or carcinoma in situ; desires repeat pregnancy in 12 months; history of postabortal or postpartum sepsis; randomized to postplacental IUD placement and could not undergo insertion within 10 minutes because of hemorrhage or surgical complication Exclusion at delivery: clinical evidence of infection; prolonged rupture of membranes; or no longer desiring LNG‐IUS
Interventions	LNG‐IUS after cesarean delivery Treatment: immediate postplacental insertion (within 10 minutes) Comparison: standard insertion, 4 to 8 weeks after delivery
Outcomes	Use of LNG‐IUS at 12 months after delivery; proportion with IUS inserted per protocol, expulsion, satisfaction with IUD, complications Assessment times: 3 and 6 months after delivery by telephone and 12 months after delivery in clinic (by phone for those with IUD inserted off protocol expelled without replacement or removed)
Notes	Results also in ClinicalTrial.gov posting Slow recruitment, early termination, small N (42 women)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated blocked‐randomization scheme, stratified by site
Allocation concealment (selection bias)	Low risk	Sequentially numbered, sealed opaque envelopes
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Research staff members conducting follow‐up assessments were blinded to participant allocation. ClinicalTrials.gov listing states "single blind (investigator)"
Incomplete outcome data (attrition bias) All outcomes	High risk	Loss to follow‐up: 33% (30% immediate; 36% standard)