Evidence and support

The evidence ledger

Findings with population, result, applicability and uncertainty attached.

Research preview · selected sources rechecked September 14, 2026 · independent clinical review pending. These editorial assessments are not formal GRADE ratings. No study here establishes an effect of Brighter Postpartum.

61 results
supported

Counseling can reduce perinatal depression risk

2025 draft pooled RR 0.83 (95% CI 0.72–0.95). Relative reductions were 31% in increased-risk and 5% in lower-risk/unselected groups. Historical 2019 pooled RR: 0.61 (0.47–0.78).

Read population and limitations ↗
supported

Proactive trained peer support has prevention evidence

At 12 weeks, 14% versus 25% scored above the EPDS threshold; reported NNT 8.8.

Read population and limitations ↗
supported

ROSE interpersonal prevention program

Depression at six months occurred in 16% versus 31%.

Read population and limitations ↗
promising

Task-shared prenatal anxiety intervention

Among 755 assessed postnatally, major depression was 12% versus 41%; adjusted odds ratio 0.19.

Read population and limitations ↗
supported

Supervised nonspecialist teletherapy can scale care

Trained nonspecialists were noninferior to specialists, and telemedicine to in-person therapy.

Read population and limitations ↗
promising

Technology-assisted peers can deliver an established therapy

Technology-assisted peer delivery was noninferior to the established Thinking Healthy Programme.

Read population and limitations ↗
mixed

Universal automated CBT prevention can fail

MDE occurred in 2.35% versus 2.91%; HR 0.85 (95% CI 0.61–1.20), not statistically significant.

Read population and limitations ↗
promising

Prenatal insomnia treatment has promising downstream effects

Probable depression at three months postpartum was 4% versus 18%, a secondary outcome.

Read population and limitations ↗
supported

Exercise can improve symptoms after appropriate recovery

Depressive symptoms: SMD −0.52 (95% CI −0.80 to −0.24). Four small trials found depression odds OR 0.55.

Read population and limitations ↗
mixed

Mindfulness is a reasonable optional adjunct

One review reported depressive symptoms SMD −0.53; another found no significant pooled EPDS difference.

Read population and limitations ↗
promising

Yoga evidence is encouraging but small

Symptom improvements favored yoga.

Read population and limitations ↗
mixed

Journaling is not a proven PPD prevention package

No significant differences in the main outcomes at one or six months.

Read population and limitations ↗
limited

Faith may offer meaning and community, not a cure guarantee

Organized participation was associated with fewer symptoms in this cohort.

Read population and limitations ↗
unsupported

NLP has insufficient evidence for health claims

Most included randomized studies did not show significant benefits; bias was substantial.

Read population and limitations ↗
limited

Vitamin D is a deficiency discussion, not a cure stack

EPDS mean difference −2.70 (95% CI −3.68 to −1.71), rated low certainty.

Read population and limitations ↗
mixed

Large DHA prevention trial was not positive

High EPDS symptoms: 9.67% versus 11.19%; RR 0.85 (95% CI 0.70–1.02), not significant.

Read population and limitations ↗
limited

One probiotic signal is not a class effect

Mean depression symptom score was 1.2 points lower with HN001.

Read population and limitations ↗
unsupported

Magnesium and zinc are not established PPD treatments

Neither mineral significantly improved depression or anxiety compared with placebo.

Read population and limitations ↗
limited

Amino-acid/blueberry product: negative primary, positive exploratory result

No primary mood-induction effect; exploratory baby-blues score favored supplementation (effect size 0.62).

Read population and limitations ↗
avoid

Placenta capsules have no established benefit and infection risk

No robust mood benefit in the pilot; contaminated capsules were implicated in recurrent infant infection.

Read population and limitations ↗
avoid

Theanine lactation data are insufficient

LactMed reports no breastfeeding-use data and advises avoiding supplemental theanine when nursing a newborn or preterm infant.

Read population and limitations ↗
promising

Bright-light therapy is not the same as sunlight advice

Study-defined remission favored bright light, 73% versus 27%.

Read population and limitations ↗
supported

Sertraline has randomized treatment evidence

Response 59% versus 26%; remission 53% versus 21%.

Read population and limitations ↗
promising

Preventive medication may be considered after prior PPD

Recurrence: 1/14 sertraline versus 4/8 placebo in a small pilot.

Read population and limitations ↗
supported

Zuranolone can improve PPD symptoms rapidly

Day-15 HAM-D change −17.8 versus −13.6; difference −4.2 points (95% CI −6.9 to −1.5).

Read population and limitations ↗
supported

Brexanolone listings need an availability correction

FDA withdrew Zulresso approval effective April 14, 2025 after the sponsor reported it was no longer marketed.

Read population and limitations ↗
supported

Psychosis needs urgent professional care

Usually begins abruptly in the first two weeks; hallucinations, delusions, mania or severe confusion warrant emergency assessment.

Read population and limitations ↗
supported

Unwanted intrusive thoughts are not equivalent to intent

Unwanted infant-harm thoughts/OCD were not associated with increased reported maternal aggression in this sample.

Read population and limitations ↗
promising

A prescription digital therapeutic already exists

At least four-point EPDS improvement: 86.3% versus 23.9% with sham app.

Read population and limitations ↗
supported

Guided digital CBT has diagnosis-based trial evidence

At follow-up, 15/19 assessed intervention participants versus 4/22 usual-care participants no longer met diagnostic criteria.

Read population and limitations ↗
mixed

First births can carry increased risk, but parity is not destiny

Primiparity pooled RR 1.76 (95% CI 1.59–1.96) in one evidence synthesis.

Read population and limitations ↗
promising

Hormone sensitivity differs across individuals

Significant mood symptoms occurred in 5/8 with PPD history and 0/8 controls.

Read population and limitations ↗
supported

Depressive symptoms can appear later in the first year

7.2% reported symptoms at 9–10 months; 57.4% of later-positive respondents had not reported symptoms at 2–6 months.

Read population and limitations ↗
supported

Prior symptoms, adversity and support are important risk contexts

History of depression/anxiety, low support, violence and stressful life events repeatedly correlate with PPD.

Read population and limitations ↗
supported

Thyroid symptoms can overlap with postpartum depression

Thyroid dysfunction can cause mood, energy and sleep symptoms; assessment may be appropriate based on history and presentation.

Read population and limitations ↗
supported

ECT remains a specialist option for severe illness

Guidance includes ECT in selected severe cases; 2026 review summarizes a limited predominantly nonrandomized literature.

Read population and limitations ↗
mixed

Anemia treatment is a clinical issue, not a universal iron mood stack

At six weeks, EPDS outcomes did not differ significantly between three active iron-treatment groups.

Read population and limitations ↗
limited

Accelerated TMS for PPD remains a research question in an ongoing trial

Mount Sinai lists the SAINT study as recruiting, with completion planned for 2029.

Read population and limitations ↗
promising

Clinician-delivered EMDR has a postpartum birth-trauma trial

Two clinician-delivered sessions between two and five weeks postpartum reduced PTSD symptoms and related distress compared with telephone contacts.

Read population and limitations ↗
limited

Massage may help immediate comfort without proving depression prevention

Day-two massage improved immediate discomfort and several mood measures relative to bed rest.

Read population and limitations ↗
mixed

Acupressure plus music had a short-lived screening signal

EPDS-based status favored the intervention at one week; no significant difference remained at six weeks.

Read population and limitations ↗
promising

Group singing has a qualified early postpartum trial signal

Overall faster improvement was not significant (P=.16); the moderate/severe symptom subgroup favored singing (P=.033).

Read population and limitations ↗
promising

A larger singing trial found later symptom differences

Between-group EPDS differences were not significant at weeks 6/10. Later post-hoc comparisons favored singing: week 20 difference -1.59; week 36 -1.56.

Read population and limitations ↗
limited

Singing-related communication findings need replication

The secondary analysis reported improvements in some maternal communication measures after singing.

Read population and limitations ↗
limited

Mothers described identity and achievement, not only symptom relief

Reported themes included shared creative experience, time for themselves, achievement/identity, perceived infant calming and bonding.

Read population and limitations ↗
promising

Relational savoring can support connection in an adjacent population

Greater immediate gratitude/pride and closeness; later sensitivity to toddler cues improved. Overall parenting wellness did not improve at three months.

Read population and limitations ↗
promising

Post-NICU savoring has an immediate wellbeing signal

Immediate increases in closeness, parenting satisfaction and emotional wellbeing favored savoring.

Read population and limitations ↗
limited

Positive-affect treatment is useful inspiration, not app validation

Composite clinical improvement favored positive-affect treatment, with a small between-group advantage. Separate positive-affect and interviewer-anhedonia outcomes did not establish superiority.

Read population and limitations ↗
mixed

Less distress does not automatically mean more positive affect

The program reduced negative affect but did not improve positive affect or life satisfaction in this report.

Read population and limitations ↗
promising

Self-guided skills can help selected parents, but dropout matters

Depression symptom change favored the program (-3.35 versus -1.48); anxiety symptoms also improved more.

Read population and limitations ↗
supported

Coparenting belongs in the intervention target

Benefits were reported across multiple coparenting, parenting and adjustment outcomes.

Read population and limitations ↗
limited

Compassion work is promising but not a one-line cure

Self-esteem, rumination and loneliness outcomes favored the intervention.

Read population and limitations ↗
limited

Peer storytelling is promising design territory, not treatment proof

Accounts suggest that stories can offer recognition, information and a sense of connection.

Read population and limitations ↗
limited

Singing together may affect how a moment feels

Immediate closeness, affect and anxiety outcomes favored the singing condition.

Read population and limitations ↗
promising

Distal digital interventions can modestly improve postpartum symptoms

Depression favored intervention after treatment (d −0.28, 95% CI −0.41 to −0.15) and at follow-up (d −0.27, 95% CI −0.52 to −0.02).

Read population and limitations ↗
supported

Pregnancy-to-postpartum hormones are a transition, not a personal mood forecast

Placental endocrine support ends after delivery; estrogen/progesterone withdrawal can matter in susceptible individuals.

Read population and limitations ↗
supported

Prolactin and oxytocin do different jobs in milk production and release

Prolactin supports production; oxytocin contributes to milk ejection. Effective milk removal also matters.

Read population and limitations ↗
promising

Brain remodeling is not evidence of lost intelligence

Structural brain trajectories during pregnancy and postpartum have been observed.

Read population and limitations ↗
mixed

Fathers have a transition too, but not the same placental hormone withdrawal

Some longitudinal findings link becoming a father with lower testosterone; different observational populations show different relationships.

Read population and limitations ↗
limited

Invisible planning is work, but correlation is not an app trial

Greater cognitive household labor was associated with poorer maternal wellbeing measures.

Read population and limitations ↗
supported

Postpartum care is an ongoing process

Contact in the first three weeks, ongoing care as needed and a comprehensive visit no later than twelve weeks are recommended.

Read population and limitations ↗