Perinatal Mental Health Screening for OB/GYN Practices

Approximately 1 in 5 postpartum patients in your practice will develop clinically significant postpartum depression. Most will not self-report. Mental health conditions, driven by suicide and unintentional overdose, now account for 23% to 27.7% of all pregnancy-related deaths in the United States, making them the leading underlying cause of maternal mortality. That figure exceeds hemorrhage, hypertensive disorders, and thromboembolism combined. Despite clear consensus from ACOG, AWHONN, and the USPSTF requiring universal screening, HEDIS data from 2023 show actual screening rates at 8.7% among Medicaid insurers and 4.4% among commercial insurers. The clinical obligation is established. The operational gap is everything that comes after.

This guide covers the screening tools, cutoff scores, documentation requirements, liability context, and referral workflows your practice needs to close that gap.

The Prevalence Picture in Your Patient Panel

PMADs affect approximately 1 in 5 birthing individuals, translating to roughly 800,000 families annually in the United States. Of those affected, an estimated 75% remain entirely untreated. That is not primarily a treatment access failure. It is a detection failure in practices that see these patients at every prenatal and postpartum visit.

The math for a typical OB practice: a panel of 200 postpartum patients per year will produce 30 to 40 patients with clinically significant PPD. Depression onset is not confined to the postpartum period. The largest U.S. PPD screening study found that 33.4% of depressive episodes begin during pregnancy, and 26.5% start before conception. The postpartum period accounts for 40.1% of episode onset. Screening only at the 6-week postpartum visit captures less than half of total onset.

Anxiety disorders are equally prevalent. Up to 20% of pregnant patients experience clinically significant anxiety, peaking at 25.5% in early pregnancy. Perinatal OCD affects approximately 4% of postpartum patients. Childbirth-related PTSD occurs in roughly 4% of women, including those with objectively uncomplicated deliveries.

ICD-10 reference for PMAD clinical encounters:

• F53.0: Postpartum depression (confirmed diagnosis)
• F32.1: Major depressive episode, moderate
• F41.1: Generalized anxiety disorder
• F42.2: OCD with postpartum intrusive thoughts
• Z13.32: Perinatal depression screening encounter (no confirmed diagnosis)

Apply Z13.32 for the screening visit. F-codes require clinical assessment confirming the diagnosis, not a screening score alone. The EPDS identifies; the clinical encounter diagnoses.

Identification depends on structured screening. Clinical observation alone will not reliably surface what a validated instrument captures in under five minutes.

The Case for Universal Screening

Risk-based screening, meaning screening only patients with visible psychosocial stressors or prior psychiatric history, fails systematically in perinatal settings. Women with no prior psychiatric history account for a significant proportion of new-onset perinatal depression. High-functioning patients mask well. Implicit clinician bias consistently results in undertreated minority populations. Black, Latina, and Asian patients experience PMADs at rates equal to or exceeding their White counterparts but are substantially less likely to receive a diagnosis or initiate treatment.

Universal screening programs outperform risk-based protocols in both identification rates and treatment engagement. That consensus is reflected in the current standard of care.

ACOG Clinical Practice Guideline No. 4 (June 2023) mandates screening using standardized validated instruments at three minimum time points: the initial prenatal visit, a second time later in pregnancy, and at postpartum visits. CPG No. 4 also establishes that screening without an operationalized pathway for positive results violates the standard of care. A score documented and deferred is not clinical care.

AWHONN's Position Statement on Perinatal Mood and Anxiety Disorders classifies untreated PMADs as a patient safety issue and supports universal screening with defined stage-based response protocols. Facilities must specify in advance what happens when a patient exceeds a specific cutoff score.

The USPSTF maintains a Grade B recommendation for depression screening in pregnant and postpartum women, which triggers ACA-mandated insurance coverage for screening visits under preventive services requirements.

Several states have codified these recommendations into statute. New Jersey, Illinois, California, Massachusetts, and West Virginia all carry statutory perinatal screening requirements with different enforcement mechanisms and penalty structures. For the full state-by-state compliance breakdown, see the documentation and liability section below.

EPDS: Scoring and Clinical Interpretation

The Edinburgh Postnatal Depression Scale is the only screening instrument developed and validated specifically for perinatal populations. Its 10 items focus on cognitive and affective symptoms and deliberately omit somatic items (sleep, fatigue, appetite) that overlap with normal postpartum physiology. Administration takes two to five minutes. Scoring takes under one minute manually. Digital administration via patient portal or tablet can auto-score and route results to the chart before the encounter begins.

In a large-scale individual participant data meta-analysis of over 9,000 participants, the EPDS achieves approximately 86% sensitivity and 78% specificity in perinatal populations at the optimal combined cutoff. Cox, Holden, and Sagovsky validated the scale for postnatal use in 1987. Bergink and colleagues subsequently confirmed robust prenatal performance across all three trimesters, with AUC values of 0.87 to 0.94 prenatally. A cutoff of 10 in the first trimester and 13 in the second and third trimesters provides optimal accuracy at each gestational stage.

Score interpretation and required actions:

• 0 to 9: Low concern. Document the score. Repeat at the next scheduled screening visit.
• 10 to 12: Clinical assessment warranted this visit. Assess functional impairment, risk factors, and history. Decision: refer now, or repeat in 2 to 4 weeks with a documented plan and specific follow-up date. Do not defer without documentation.
• 13 to 14: Probable major depression. Refer to perinatal mental health this visit. No deferral.
• 15 to 30: Severe. Refer this visit; assess the need for same-day psychiatric support.
• Item 10 above zero: Safety assessment before discharge, regardless of total score.

The 10 to 12 range requires clinical judgment. A score of 11 in a patient with a prior PMAD history, poor social support, and significant sleep disruption warrants immediate referral. A score of 11 in a patient who completed a therapy course last trimester and has strong support may warrant watchful waiting with a 2-week check. The protocol establishes the minimum response; clinical assessment determines the specific path.

EPDS-3A anxiety subscale: Items 3, 4, and 5 of the EPDS constitute a latent anxiety subscale. A score of 6 or higher on these three items, out of a maximum of 9, flags clinically significant perinatal anxiety even when the total EPDS score is below the depression threshold. Approximately 15 to 20% of perinatal patients score negative for depression on the full EPDS but positive on the EPDS-3A. Calculate both; a negative total score does not rule out significant anxiety.

Item 10 protocol: Item 10 asks about thoughts of self-harm. Any endorsement, from "hardly ever" (score 1) through "yes, quite often" (score 3), activates an independent clinical response regardless of the total EPDS score. The patient must not be discharged until a direct, non-judgmental safety assessment is complete. Assess the safety of the infant and any other dependents in the patient's care. If imminent risk is present, initiate psychiatric crisis protocols. Document the assessment explicitly in the chart, including the clinical decision made.

Billing codes: CPT 96160 covers EPDS administration in the OB setting when the provider or medical assistant administers the instrument to the patient. CPT 96161 applies in pediatric settings when the mother is being screened during the infant's well-child visit. CPT 96127 is used for standalone brief emotional/behavioral assessments and can be billed with modifier 25 alongside standard E&M codes.

PHQ-9, PHQ-2, and GAD-7: Limitations in Perinatal Populations

The PHQ-9 is widely used in primary care and performs well in that context. In perinatal settings, four of its nine items (sleep, fatigue, appetite, psychomotor changes) can be elevated by normal postpartum physiology, generating false positives. The PHQ-9 also lacks any anxiety screen. Studies comparing it with the EPDS in postpartum cohorts consistently show higher false positive rates on the PHQ-9 and missed anxiety-predominant presentations.

The PHQ-2 is more problematic. At the standard clinical cutoff of 3, the PHQ-2 identifies only 21 to 22% of depressed perinatal patients. That is a miss rate exceeding 78% for the most common complication of childbirth. The PHQ-2 ignores anxiety entirely. It is not appropriate as a standalone perinatal screening instrument.

The GAD-7 lacks perinatal validation. A comparative study by Fairbrother and colleagues found GAD-7 performance at 3 months postpartum fell below an AUC of 0.80, the standard threshold for a clinically useful screener. At an elevated cutoff of 13 (required to compensate for poor baseline performance in perinatal populations), sensitivity drops to 50 to 60%, and the tool misses panic disorders, postpartum OCD, and birth trauma entirely.

Practical use case by instrument:

• EPDS: First-line for all routine perinatal screening. Validated, accurate, time-efficient.
• PHQ-9: Severity tracking in patients already identified through EPDS screening. Not a primary perinatal screen.
• PHQ-2: Not recommended for primary perinatal screening.
• GAD-7: Not recommended for primary perinatal anxiety screening. Use the EPDS-3A or the Perinatal Anxiety Screening Scale (PASS) for anxiety-specific assessment.

If your practice currently uses PHQ-9 as the primary screen, transitioning to EPDS reduces false positives without meaningful loss of sensitivity and captures anxiety presentations the PHQ-9 misses entirely. For a detailed comparison of scoring cutoffs, timing, and clinical use cases across all three instruments, see the [EPDS vs. PHQ-9 vs. GAD-7 clinical guide for OB and midwifery practices](/clinical-resources/epds-phq9-gad7-pmad-screening-ob-midwife/).

Screening for OCD, PTSD, and Psychosis Risk

Standard EPDS-based screening will not catch OCD or birth trauma. Both are common, both are systematically underdiagnosed in OB settings, and both carry significant morbidity when missed.

Perinatal OCD affects approximately 4% of postpartum patients. The hallmark is intrusive, ego-dystonic thoughts, often about harming the infant. These thoughts are frequently misidentified by untrained clinicians as postpartum psychosis, which can lead to unwarranted CPS involvement or psychiatric holds. The Obsessive-Compulsive Inventory-4 (OCI-4) is a validated 4-item tool appropriate for OB workflow time constraints. A cutoff of 3 or higher on the OCI-4 flags likely OCD with clinically useful sensitivity and specificity. A patient with perinatal OCD knows her thoughts are unwanted and contrary to her values; she is distressed by them. A patient with postpartum psychosis may believe her thoughts are real or externally commanded. That distinction determines clinical escalation.

For PTSD, the Primary Care PTSD Screen for DSM-5 (PC-PTSD-5) is a validated 5-item screener appropriate for OB settings. A score of 3 or higher warrants clinical follow-up. Birth trauma can follow subjectively distressing deliveries even when the delivery was medically uncomplicated. Do not screen only after documented obstetric events.

Psychosis risk indicators requiring immediate psychiatric evaluation include new-onset confusion or disorganized thinking, insomnia combined with elation or grandiosity, paranoia, and command hallucinations or perceptual disturbances. Personal or family history of bipolar disorder is the strongest predictor of postpartum psychosis. Women with a prior bipolar diagnosis experience mood episodes in the perinatal period at a rate of approximately 55%.

Bipolar screening before prescribing: ACOG CPG No. 4 mandates screening for bipolar disorder before initiating any SSRI or SNRI in a perinatal patient. Approximately 1 in 5 women who screen positive for PPD have underlying bipolar disorder. Treating bipolar disorder with antidepressant monotherapy risks precipitating mania, rapid cycling, or postpartum psychosis. The Mood Disorder Questionnaire (MDQ) is the standard screening instrument. Document a negative MDQ screen in the chart before prescribing any antidepressant to a perinatal patient. Failure to document this step is a deviation from the 2024 to 2026 standard of care and a documented liability exposure in malpractice claims involving perinatal psychiatric emergencies.

Documentation, Liability, and State Compliance

Documenting a score is necessary but not sufficient. The standard of care defined by ACOG CPG No. 4 and CPG No. 5 requires the clinical record to reflect the decision made in response to the score.

Required documentation after a positive screen:

1. Validated tool used and exact numerical score
2. Bipolar screen result (MDQ or equivalent) before any new antidepressant prescription
3. Explicit documentation addressing Item 10, including the safety assessment and clinical decision if Item 10 was positive
4. Provider's clinical judgment regarding acuity (mild, moderate, severe, imminent)
5. Warm handoff or referral destination: name of receiving practice, submission date
6. Educational materials and crisis resources provided to the patient
7. Follow-up timeframe, specifically documented

The Joint Commission's NPSG.15.01.01 requires hospitals to screen behavioral health patients for suicidal ideation using a validated tool. TJC surveyors trace perinatal patient pathways through labor and delivery units and require documented evidence of systematic mental health screening with defined response protocols for positive results. Facilities without documented stage-based response protocols face citation under interconnected standards PC.06.01.01 and PC.06.01.03 as well.

Patient refusal: If a patient declines screening or referral, document the clinical indication for offering the screen, the patient's refusal, the risks explained, the patient's verbalized understanding, and the standing offer to screen at all future visits. This documentation protects against liability when an adverse outcome subsequently occurs.

HIPAA and information sharing: HIPAA (45 CFR § 164.506) permits sharing PHI with other treating providers for treatment purposes without a signed release. The minimum necessary standard explicitly does not apply to treatment disclosures between healthcare providers. An OB can communicate an EPDS score and clinical assessment to the infant's pediatrician to coordinate care for the maternal-infant dyad. If a patient presents an imminent threat to self or others, 45 CFR § 164.512(j) permits disclosure to family members or law enforcement to avert the harm.

State mandate compliance:

• California (AB 2193/2018; AB 1936/2024; SB 626/2025): Most comprehensive framework. Providers must screen per ACOG guidelines; insurers must cover at least one prenatal and one postpartum screen. Regulatory enforcement by the California Department of Managed Health Care.
• Illinois (2008, updated 2024): Prenatal providers and pediatricians required to offer validated screening. State legislation also recognizes postpartum psychiatric episodes as a mitigating factor in criminal sentencing.
• New Jersey (P.L. 2006, c. 2): Mandatory screening at hospital discharge and early postnatal visits. Family education required by statute.
• Massachusetts (Chapter 313, Acts of 2010; 105 CMR 271.000): Annual data reporting to the Department of Public Health. Billed via HCPCS S3005 with U1 (positive) and U2 (negative) modifiers.
• West Virginia (§16-4E): State-developed screening tool required; results reported to the Bureau for Public Health.

In all five states, non-compliance creates exposure under the doctrine of negligence per se, where violation of a statutory safety standard constitutes a legal presumption of negligence in civil malpractice litigation. Non-compliance also affects HEDIS quality scores, with direct downstream impacts on Medicaid reimbursement rates and managed care contracting.

From Positive Screen to Warm Handoff

Score-based triage should be protocol-driven, not dependent on which provider is in the room. Define the response at each threshold before the first screen is administered.

Triage protocol by EPDS score:

• 10 to 12: Clinical assessment this visit. Decision: refer now, or repeat in 2 to 4 weeks with a documented follow-up plan. Do not defer without documentation.
• 13 to 19: Refer to perinatal mental health this visit. No deferral.
• 20 and above: Refer this visit; assess the need for same-day psychiatric support.
• Item 10 positive (any value): Safety assessment before discharge, regardless of total score.

A warm handoff is a real-time, direct transfer of care between providers completed in the patient's presence. Cold referrals, handing a distressed patient a list of phone numbers or a web address, result in actual psychiatric appointments below 50% of the time in perinatal populations. Warm handoffs increase the likelihood of continued treatment engagement significantly. Research demonstrates an odds ratio of approximately 2.0 for ongoing treatment with warm handoff versus cold referral in integrated care settings. Up to 80% of perinatal patients report they would access care if it was facilitated directly through their OB setting.

Protocol steps for a warm handoff:

1. Score exceeds threshold on EPDS.
2. Provider validates the score with the patient and explains the clinical picture directly: "Your score tells me you're having a harder time than you may realize. I want you to connect with a specialist in perinatal mental health."
3. Contact the receiving mental health practice before the patient leaves the building. Submit the referral before the visit ends, not afterward.
4. Introduce the receiving practice by name. For telehealth providers, explain that the patient is seen from home.
5. Document in the chart: tool, score, clinical decision, referral destination, date submitted.
6. Set a follow-up flag in the chart. At every subsequent visit, ask whether the patient made contact. If she did not, identify the barrier before reiterating the referral.

Practices that use a named referral partner where the receiving provider contacts the patient directly, rather than asking the patient to initiate contact, see substantially higher connection rates. The patient's action burden drops to zero.

For step-by-step workflow design including staff role assignments, triage triggers, and EHR documentation templates, see [building a postpartum mental health workflow in OB practice](/clinical-resources/postpartum-mental-health-workflow-ob-practice/).

Referring Patients to Phoenix Health

Phoenix Health provides perinatal mental health therapy via telehealth. Most of our therapists hold PMH-C certification from Postpartum Support International, the clinical credential specifically for perinatal mental health. PMH-C training requires demonstrated competence in PMADs, validated screening tools, and treatment modalities specific to the perinatal period, including OCD, birth trauma, and presentations involving bipolar risk. It is not a general mental health credential.

Telehealth removes the logistical barriers that predictably reduce follow-through on postpartum referrals: transportation with an infant, scheduling around feeding and childcare, and the energy cost of leaving home during a period of significant functional impairment.

When you submit a referral, a Phoenix Health intake team member contacts the patient directly within one business day. Insurance verification happens before the first appointment. The patient does not encounter an insurance barrier as the first step. We coordinate directly with your patient from first contact and can provide updates back to the referring provider for collaborative care arrangements.

Interested in setting up a standing referral pathway or discussing collaborative care for your practice? Contact our clinical partnerships team at [joinphoenixhealth.com/referrals-and-partnerships/?inquiry=referral](/referrals-and-partnerships/?inquiry=referral).

Ready to refer a patient now? Submit through our secure referral form at [joinphoenixhealth.com/referrals/](/referrals/). We respond within one business day.