checkedThe Carlat Psychiatry Podcast2026-11-16last verified 2026-11-16

did 98% of women with postpartum psychosis reach full recovery in one study?

verdict: overstated
directionally right; magnitude or certainty inflated
“With that approach, 98 percent of women reached full recovery in one study.”

a host, a psychiatrist said on The Carlat Psychiatry Podcast, 2026-11-16, at 06:52. hear it in context → the speaker cited: one study.

the episode: Kraepelin on Clancy Part 3 (at 06:52). we link the show's own page; we do not re-host audio.

first, the part that holds

the number is real. in the 2015 american journal of psychiatry cohort of 64 women admitted for first-onset postpartum psychosis or mania, 98.4% reached complete remission under a stepped protocol of benzodiazepines, then antipsychotics, then lithium, and none needed ect.

what we found when we went and looked

the study exists and reports the figure: "nearly all patients (98.4%) achieved complete remission within the first three steps," which is 63 of 64 women in one dutch inpatient cohort. but "complete remission" in that paper had a narrow definition: no psychotic, manic or severe depressive symptoms for at least 1 week. by 9 months postpartum, sustained remission had fallen to 79.7%, so about one in five had relapsed. a later review by the same group says that only 20-50% of women have an isolated postpartum episode, and that the rest usually go on to have episodes outside the perinatal period. "98 percent reached full recovery" is a fair account of acute remission. as a statement about recovery, it overstates what the study measured.

assertion by assertion

assertionverdictwhat the record says
in one study, 98% of women with postpartum psychosis reached remissionconfirmed98.4% (63 of 64) achieved complete remission within the first three steps of the protocol, and none needed ect. this is one consecutive-admission cohort of first-onset postpartum psychosis or mania, not a trial. [1]
those women reached 'full recovery'overstatedremission was defined as no psychotic, manic or severe depressive symptoms for at least 1 week. sustained remission at 9 months postpartum was 79.7%. multiparity and nonaffective psychosis predicted relapse. a related review reports that 50-80% of women go on to have later episodes, usually in the bipolar spectrum. [1][2]
the 98% came from treatment with lithium, antipsychotics and as-needed benzodiazepinesoverstatedthe 98.4% covers the whole cohort under a sequential algorithm (benzodiazepines, then antipsychotics, then lithium, with ect as step 4). it does not count only women who received all three. the abstract says remission came 'within the first three steps', so some women remitted before lithium was added. the abstract we read does not specify as-needed benzodiazepine dosing, and we could not read the full text to confirm it. [1]
lithium is the part of the approach linked to staying wellconfirmedwomen on lithium maintenance had a significantly lower relapse rate than women on antipsychotic monotherapy. the 2016 review calls this n=64 study the largest evidence base and says it shows lithium is highly effective for both acute and maintenance treatment. this was observational: maintenance followed what each woman had needed to remit, so it was not randomized. [1][2]
the approach caused the 98% remission rateunverifiablethere was no control or comparison arm, so the study cannot show how many women would have remitted under a different regimen. the 2016 review notes that fewer than 30 publications address treatment of postpartum psychosis at all. [1][2]

both directions

for the speaker: the figure is accurate to the source: 98.4% complete remission in 64 consecutively admitted women, with no ect required. lithium maintenance was associated with significantly fewer relapses than antipsychotic monotherapy. a later review by the same research group calls this the largest study of postpartum psychosis treatment and describes lithium as highly effective for acute and maintenance treatment. as a description of short-term response to a structured protocol, the speaker's number holds up.

against: "full recovery" is not what the study measured. remission meant at least one week without psychotic, manic or severe depressive symptoms, and by 9 months only 79.7% were still in remission. the 98% applies to a stepped sequence in which some women remitted on benzodiazepines alone or with an antipsychotic, before lithium. it does not apply to a three-drug combination given to everyone. it is a single-center, uncontrolled cohort of first-onset cases, so it cannot show that the protocol caused the remission rate. the review literature reports that most women who have postpartum psychosis go on to have later mood episodes, often bipolar, outside the postpartum period.

what this cannot say

we read the pubmed abstracts of the 2015 cohort and the 2016 review. the full text of the 2015 paper was blocked to us. that means we did not verify the per-step remission counts, the exact relapse percentages by maintenance group, the inclusion and exclusion details, or whether benzodiazepines were given as needed. the study is 64 women at one center with no comparison arm, so it cannot say how other regimens would perform or how well the result carries over to women with a prior psychiatric history. we also did not hear the full podcast segment, so "that approach" may have been described in more detail than the quote shows.

the medication pages behind this check

sources, each one fetched and read

  1. treatment of psychosis and mania in the postpartum period. american journal of psychiatry, 2015;172(2):115-23. pmid 25640930, doi 10.1176/appi.ajp.2014.13121652 What we read there: 64 women consecutively admitted for first-onset postpartum psychosis or mania were treated with a 4-step algorithm (benzodiazepines, antipsychotics, lithium, ect). 'nearly all patients (98.4%) achieved complete remission within the first three steps', and none required ect. remission was defined as 'the absence of psychotic, manic, and severe depressive symptoms for at least 1 week'. sustained remission at 9 months postpartum was 79.7%. lithium-treated patients had a significantly lower relapse rate than those on antipsychotic monotherapy. Fetched 2026-11-16 (HTTP 203). https://pubmed.ncbi.nlm.nih.gov/25640930/
  2. postpartum psychosis: madness, mania, and melancholia in motherhood. american journal of psychiatry, 2016. pmid 27609245, doi 10.1176/appi.ajp.2016.16040454 What we read there: fewer than 30 publications have studied treatment of postpartum psychosis, and the largest (n=64) found lithium highly effective for acute and maintenance treatment. 20-50% of women have isolated postpartum psychosis, and the rest usually have later episodes outside the perinatal period, typically in the bipolar spectrum. relapse risk in a later pregnancy after isolated postpartum psychosis is 31% (95% ci 22-42). Fetched 2026-11-16 (HTTP 203). https://pubmed.ncbi.nlm.nih.gov/27609245/

speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam