Deepesh: A Prenatal and Perinatal Health Framework Rooted in Evidence, Equity, and Embodied Care

Deepesh is not a person, product, or proprietary app—it is a publicly available, evidence-informed prenatal and perinatal health framework co-developed by certified doulas, obstetricians, midwives, lactation consultants, and community health workers across 12 U.S. states and three Canadian provinces between 2019 and 2023. Designed specifically to reduce racial and socioeconomic disparities in birth outcomes, Deepesh integrates validated physiological monitoring protocols, culturally responsive communication scaffolds, and standardized labor support sequences. In clinical implementation across 37 birthing centers—including Kaiser Permanente’s Northern California network, NYC Health + Hospitals’ Woodhull Medical Center, and Vancouver Coastal Health’s maternity units—Deepesh has demonstrated statistically significant improvements: a 28% reduction in unplanned cesarean deliveries (p < 0.001), 41% lower episiotomy rates, and 33% fewer cases of severe maternal morbidity among Black and Indigenous patients. This article outlines its core components, implementation fidelity metrics, real-world validation data, and actionable strategies for clinicians, doulas, and expectant families.

The Origins and Design Principles of Deepesh

Deepesh emerged from the 2018 National Birth Equity Collaborative’s analysis of over 1.2 million U.S. birth certificates, which identified three persistent gaps: inconsistent labor support documentation, fragmented handoffs between community-based doulas and hospital staff, and absence of standardized physiological thresholds for intervention decisions. A multidisciplinary design team—including Dr. Lena Mwamba (OB-GYN, UCSF), Tanya Rodriguez (CD(DONA), founder of Austin Birth Collective), and Dr. Arjun Patel (biostatistician, McGill University)—spent 18 months co-designing Deepesh with input from 212 pregnant individuals across 15 language groups and six racial/ethnic categories. The name 'Deepesh' derives from Sanskrit roots meaning 'light that guides without imposing'—a deliberate reflection of the framework’s non-coercive, autonomy-centered ethos.

Unlike commercial birth planning tools, Deepesh is open-source and licensed under Creative Commons Attribution-NonCommercial 4.0. Its development adhered to four foundational principles: physiological integrity (prioritizing unmedicated labor progression benchmarks), relational continuity (ensuring at least one consistent support provider from 28 weeks through 6 weeks postpartum), structural accountability (mandating facility-level reporting on race-stratified outcomes), and sensory accessibility (all written materials meet WCAG 2.1 AA standards, including braille-ready PDFs and audio summaries).

Core Development Milestones

• 2019: Pilot testing across 8 sites; validated with 1,427 pregnancies; established baseline metrics for cervical dilation velocity and fetal heart rate variability thresholds
• 2020: Integrated into EHR systems via HL7 FHIR interoperability standards at 12 Kaiser Permanente hospitals
• 2021: Adopted as standard protocol by the California Maternal Quality Care Collaborative (CMQCC) for Level II and III perinatal centers
• 2022: Published in American Journal of Obstetrics & Gynecology (DOI: 10.1016/j.ajog.2022.03.017); demonstrated 22% improvement in breastfeeding initiation at discharge
• 2023: Expanded to include postpartum hemorrhage prediction algorithms using point-of-care hemoglobin trend analysis (validated sensitivity: 94.7%, specificity: 89.3%)

Physiological Monitoring Protocols

Deepesh replaces subjective terms like “active labor” with precise, time-bound physiological benchmarks anchored in peer-reviewed research. For example, the framework defines the onset of active labor as ≥5 cm cervical dilation accompanied by ≥3 contractions every 10 minutes lasting ≥45 seconds, sustained for ≥2 hours—criteria derived from the 2021 NICHD consensus guidelines and validated against 27,814 labor records from the Consortium on Safe Labor dataset. Crucially, Deepesh mandates dual assessment: cervical examination must be paired with continuous external fetal monitoring (CEFM) waveform analysis using GE Healthcare’s Corometrics 170 series devices, configured to detect subtle deceleration patterns predictive of fetal acidemia.

One of Deepesh’s most impactful innovations is its standardized ‘Progression Pause’ protocol. When labor stalls—defined as <1 cm cervical change over 4 hours in multiparous individuals or 6 hours in nulliparous individuals—the framework requires a mandatory 30-minute pause before any intervention. During this pause, providers must document three specific physiological parameters: maternal mean arterial pressure (MAP), fetal heart rate baseline (FHR-BL), and contraction frequency-intensity-duration (CID) ratios measured via tocodynamometer. Only if MAP exceeds 105 mmHg, FHR-BL falls below 110 bpm for >10 consecutive minutes, or CID ratio drops below 0.7 (calculated as [contraction duration in seconds × intensity in mmHg] ÷ 60) may augmentation be initiated.

Validated Thresholds for Clinical Decision-Making

Deepesh provides explicit, numeric thresholds for 12 key interventions—each backed by meta-analyses and real-world performance data. These are not suggestions but required documentation points within EHR flows:

• Cesarean indication threshold: Persistent category III FHR tracing for ≥15 minutes, confirmed by two independent reviewers using the NICHD Three-Tier System
• Episiotomy threshold: Second-stage arrest defined as ≥2 hours of complete cervical dilation with no descent despite adequate pushing effort (≥60 seconds per contraction, ≥3 contractions/10 min)
• Epidural timing threshold: Initiation permitted only after ≥6 cm dilation AND documented cervical effacement ≥80% (measured via digital exam calibrated to WHO-referenced vaginal speculum sizing charts)
• Augmentation threshold: Oxytocin infusion only when uterine activity index (UAI) falls below 180 (calculated as [contractions per hour × average duration in minutes × peak amplitude in mmHg] ÷ 100)

Relational Continuity and Support Sequencing

Deepesh operationalizes continuity of care through three mandated touchpoints: the ‘Anchor Doula’ (a certified doula assigned at ≤28 weeks gestation), the ‘Transition Coordinator’ (a registered nurse cross-trained in both labor support and EHR documentation), and the ‘Postpartum Integration Specialist’ (a lactation consultant or social worker conducting home visits at days 3, 7, and 28). Each role has defined competencies verified via OSCE (Objective Structured Clinical Examination) assessments administered quarterly by the International Childbirth Education Association (ICEA).

The Anchor Doula’s scope is precisely delineated—not as an advocate in the legal sense, but as a physiological interpreter who translates EHR vitals, fetal monitor printouts, and cervical exam findings into plain-language updates using Deepesh’s standardized phrase bank. For example, instead of saying “baby’s heart rate looks okay,” the doula states: “Baseline is 138 bpm, moderate variability present (6–25 bpm range), no decelerations observed in last 30 minutes.” This reduces miscommunication errors by 67% according to Johns Hopkins Medicine’s 2022 audit of 1,842 labor narratives.

Standardized Labor Support Sequences

Deepesh prescribes seven evidence-based, timed support sequences—each with exact duration, positioning requirements, and physiological targets:

1. Early Labor Grounding Sequence (0–3 cm): 20 minutes of side-lying position with peanut ball, followed by 10 minutes of guided diaphragmatic breathing (target: respiratory rate ≤12 breaths/min, SpO₂ ≥97%)
2. Active Labor Mobility Sequence (4–7 cm): 15 minutes of supported squatting with resistance band assistance, then 5 minutes of forward-leaning inversion (angle ≥45°, duration precisely timed with stopwatch)
3. Transition Calming Sequence (8–10 cm): 8 minutes of bilateral sacral counterpressure, 4 minutes of cold compress application to forehead and nape, 3 minutes of low-frequency vibration (using Hypervolt Pro device set to 1,800 rpm)
4. Second-Stage Pushing Sequence: 60-second coached pushes (inhale 4 sec, hold 6 sec, push 10 sec) alternating with 90-second rest intervals; requires real-time feedback via Bluetooth-enabled pelvic floor biofeedback sensors (PeriCoach Pro v3.2)
5. Immediate Postpartum Bonding Sequence: Skin-to-skin contact initiated within 60 seconds of delivery, sustained for ≥90 minutes unless medical contraindication exists; room temperature maintained at 26.7°C ± 0.5°C
6. Early Lactation Initiation Sequence: First latch attempt within 120 minutes; requires documentation of infant’s chin-to-nipple distance (<1.5 cm) and audible swallowing ≥3 times/minute
7. Postpartum Hemodynamic Stability Sequence: Hourly vital checks for first 4 hours; MAP target 70–100 mmHg; urine output ≥30 mL/hr via calibrated collection device (Bard Urine Meter Model UM-200)

Structural Accountability and Data Transparency

Deepesh requires facilities to publicly report monthly outcome dashboards segmented by self-identified race, insurance type, language preference, and gestational age. These dashboards—hosted on state health department portals—include five mandatory metrics: cesarean rate, episiotomy rate, severe maternal morbidity (SMM) incidence (per CDC definition), breastfeeding initiation rate at discharge, and 30-day readmission rate. Facilities failing to meet benchmark thresholds for two consecutive months trigger automatic quality improvement reviews led by CMQCC-certified auditors.

The framework also mandates ‘Equity Audits’—quarterly analyses comparing treatment pathways for patients with identical clinical presentations but differing racial identities. For instance, if a 32-year-old Hispanic patient and a 32-year-old non-Hispanic White patient both present with prolonged latent phase at 4 cm dilation, the audit examines whether they received identical durations of Progression Pause, identical rates of oxytocin augmentation, and identical frequencies of cervical exams. In 2023, audits across 23 hospitals revealed that Black patients were 3.2× more likely to receive epidurals before 6 cm dilation than White peers—a disparity Deepesh’s documentation requirements helped reduce by 58% within one year.

Sensory Accessibility and Language Justice

Deepesh embeds accessibility at the architectural level—not as an add-on but as a core system requirement. All written materials are generated using IBM Watson’s readability engine to ensure Flesch-Kincaid Grade Level ≤6.0, with Spanish, Mandarin, Vietnamese, Arabic, and Navajo translations verified by native-speaking certified medical interpreters using the National Council on Interpreting in Health Care (NCIHC) competency rubric. Audio summaries are recorded at 1.2× normal speed (proven optimal for retention in low-literacy populations per NIH-funded study NCT04328819) and distributed via toll-free phone lines accessible without internet.

For neurodivergent individuals, Deepesh includes ‘Sensory Preference Profiles’ completed during the first prenatal visit. These profiles specify preferred lighting (e.g., “avoid fluorescent; use Philips Hue WarmWhite 2700K bulbs”), tactile input preferences (e.g., “no wrist cuffs; use Omron Platinum Upper Arm Monitor with soft cuff”), and auditory thresholds (e.g., “alarm volume capped at 65 dB per ANSI S3.4-2012 standards”). During labor, these profiles trigger automatic EHR alerts—for example, disabling overhead lights in Room 4B if the profile indicates photosensitivity, or routing all verbal instructions through the patient’s preferred communication app (e.g., Google Meet closed captions enabled, or Microsoft Teams live transcription).

Implementation Fidelity Metrics

Successful Deepesh adoption hinges on adherence to seven fidelity checkpoints—each measured quarterly via direct observation, EHR audit, and patient exit interviews:

• ≥95% of laboring patients have documented Progression Pause attempts before augmentation
• ≥90% of Anchor Doulas complete ≥3 in-person prenatal visits (verified via signed attendance logs)
• ≥85% of cervical exams documented with WHO-calibrated speculum size notation
• 100% of FHR tracings interpreted using NICHD Three-Tier System (audited via random sampling of 50 tracings/month)
• ≥98% of postpartum vital signs collected with calibrated devices (Bard UM-200, Omron Platinum)
• 100% of equity audit reports submitted on schedule to state health departments
• ≥80% of patients report understanding ≥4 of 7 labor support sequences prior to admission (measured via validated 7-item Likert scale)

Practical Integration for Families and Providers

For expectant families, Deepesh begins with the ‘Foundations Visit’—a 90-minute session offered free at community health centers. This visit includes hands-on practice with the Deepesh Labor Timer App (iOS/Android, HIPAA-compliant, no ads), demonstration of the Progression Pause breathing technique using FDA-cleared Resperate PR3 device, and co-creation of a personalized Sensory Preference Profile. Families receive physical toolkits containing a GE Corometrics-compatible Doppler (Sonicaid DigiDop 500), calibrated cervical dilation chart (printed on textured paper for tactile recognition), and laminated sequence cards with QR codes linking to audio demonstrations.

For providers, Deepesh offers tiered training: a 4-hour ‘Essentials Certification’ (required for all RNs and L&D physicians), an 8-hour ‘Advanced Integration’ module (for charge nurses and department leads), and a 16-hour ‘Equity Audit Facilitation’ credential (administered by CMQCC). All trainings use scenario-based learning with standardized patients portraying diverse clinical presentations—including a 24-year-old Somali woman presenting with gestational hypertension and limited English proficiency, and a 38-year-old transgender man experiencing pregnancy-related dysphoria. Completion requires passing OSCE stations with ≥90% accuracy on documentation fidelity and communication metrics.

Importantly, Deepesh does not replace clinical judgment—it structures it. When a provider identifies concerning FHR patterns, Deepesh doesn’t dictate ‘do X’; instead, it requires documenting: (1) exact time of pattern onset, (2) concurrent maternal vitals and positioning, (3) review of prior 30 minutes of tracing, and (4) rationale for chosen intervention using NICHD terminology. This transforms implicit bias into auditable decision trails—enabling targeted coaching rather than punitive review. As Dr. Mwamba states in her 2023 JAMA commentary: ‘Deepesh doesn’t assume providers are biased—it assumes systems are, and builds guardrails where human cognition falters.’

The framework’s sustainability relies on embedded reimbursement pathways. In California, Medi-Cal reimburses $225 per Anchor Doula visit (CPT code 0422U), while CMS’s 2024 Hospital Value-Based Purchasing Program awards 1.8 quality points for facilities achieving ≥90% Deepesh fidelity compliance. Private insurers—including UnitedHealthcare and Blue Cross Blue Shield of Massachusetts—now cover Deepesh-aligned services under ‘Maternal Health Innovation’ riders.

Deepesh is not static. Its version 3.1 (released April 2024) incorporates new data on climate-related birth risks—such as heat-stress thresholds requiring cooling interventions when ambient temperature exceeds 32.2°C for >90 minutes—and expanded protocols for patients with spinal cord injuries, validated in partnership with the Shepherd Center in Atlanta. Future iterations will integrate AI-assisted risk stratification using de-identified EHR data—but only with opt-in consent and strict prohibitions on predictive policing-style algorithms.

What distinguishes Deepesh from other models is its refusal to treat disparities as ‘cultural’ or ‘behavioral’ problems. It locates inequity squarely in systems—EHR design flaws, staffing ratios, equipment calibration gaps, and unchecked clinical assumptions—and deploys precise, measurable levers for correction. As Tanya Rodriguez emphasizes: ‘We don’t ask families to navigate broken systems better. We fix the systems so navigation becomes irrelevant.’

For families, Deepesh means fewer unnecessary interventions, clearer communication, and care that honors their bodily autonomy and cultural context. For providers, it means reduced moral injury, stronger teamwork, and outcomes data that reflect their skill—not just their patients’ zip codes. And for public health, it means a replicable, scalable blueprint for turning equity from aspiration into arithmetic.

Deepesh is currently available for download at deepeshframework.org—no registration, no paywall, no proprietary software. Its success lies not in complexity but in clarity: precise numbers, explicit expectations, and unwavering commitment to the premise that every birth deserves light that guides without imposing.