Prophylactic Use of Immediate Postpartum Sertraline to Prevent Postpartum Depression: A Double Blind Randomized Placebo Controlled Trial

Eligible women for our study would be identified antepartum from our three obstetrical groups delivering at Cooper University Hospital: Cooper Faculty Group, Women's Care Center, and CamCare. Potentially eligible women would be referred to the study coordinator or principal investigator to discuss the nature of the study. Those women who agreed to the trial would be further screened with a baseline structured psychiatric evaluation in the third trimester of the pregnancy to rule-out current depressive illness. If there was no evidence of depression at the antepartum evaluation, patients delivering a liveborn singleton fetus at 34 0/7 weeks or greater would be re-evaluated prior to discharge and, if scoring </= 12 on the Edinburgh Postpartum Depression scale, would then be enrolled in the trial.

Patients would be assigned to either sertraline 50 mg daily or identical appearing placebo for 12 weeks. Group allocation would be determined by restricted-randomization technique with variable block length, with the sequence generated by someone not associated with participant assignment. Assignment would be kept in sequentially numbered, opaque, sealed envelopes (SNOSE) in the pharmacy, which would dispense the medication (or placebo). Patients would be given a 30 day supply on the day of discharge, with refills provided by the study coordinator (through the pharmacy)for 12 weeks and a four-day supply of 25 mg Sertraline or placebo at the end of 12 weeks as a taper. Patients would also undergo follow-up blinded structured psychiatric evaluations at 4 weeks, 8 weeks, and 12 weeks to assess for adverse reaction to the assigned treatment agent, and for administration of questionnaires/evaluation to assess for development of depression. Any patient with recognized clinical depression would immediately be removed from further active participation in the study and referred to our Cooper Psychiatry department or outside psychiatrist for ongoing evaluation and treatment. Medication received (Sertraline or Placebo) would necessarily be revealed to Psych only for purposes of guiding appropriate further treatment.

All women randomized would be analyzed according to group assignment (intent-to-treat). Demographic information, including patient age, race/ethnicity, gravidity, parity, gestational age at delivery, infant birth weight, as well as infant weights from standard Pediatric visits (obtained verbally from the mother)would be recorded and compared using the Student t-test for normally distributed continuous data, Mann-Whitney U for non-normative continuous data, and the Chi Square test or Fisher Exact test for categorical data.

A sample size calculation was performed. Based on an anticipated rate of postpartum depression (PPD) of 30% in the placebo group, we would need 62 subjects in each group to detect a reduction in PPD to 10% in the sertraline group, with a power of .8 and a Type I error of .05. Based on the 2200 deliveries occurring annually at Cooper University Hospital, we anticipate that it would take 2-3 years to recruit 124 subjects into this study.