Skip to main content
SMFM 2026 Pregnancy Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Additional registration fee required
Additional registration fee required
Faculty have requested this content not be shared outside of the session
Faculty have requested this content not be shared outside of the session
CME Credit Offered
CME Credit Offered
Abstract Award
Abstract Award
Recording available 2/16-5/2
Recording available 2/16-5/2
AIUM Credit
AIUM Credit
Foundation Awardee
Foundation Awardee
Poster Icons
Abstract Award
Abstract Award
Foundation Awardee
Foundation Awardee

Poster Session 1

Category: Digital Health Technologies (DHT)

Poster Session 1

(141) Streaming Labor: What 28,436 YouTube Videos Reveal About Childbirth Content

Wednesday, February 11, 2026
10:30 AM - 12:00 PM  

    Submitting Author and Presenting Author(s)

  • Ayala Hirsch, MD, N/A (she/her/hers)

    Resident
    Shaare Zedek medical hospital
    Jerusalem, Yerushalayim, Israel

    • Coauthor(s)

  • CM

    Chen Mordehai

    Ben-Gurion University of the Negev
    Beer Sheva, HaDarom, Israel

  • MR

    Mika Rabinovich

    Ben-Gurion University of the Negev
    Beer Sheva, HaDarom, Israel

  • AW

    Adi Y. Weintraub, MD

    Professor of Obstetrics & Gynecology; Urogynecology & Pelvic Floor Medicine Specialist
    Soroka University Medical Center, Faculty of Health Sciences, Ben‑Gurion University of the Negev
    beer sheba, HaDarom, Israel

  • Reut Rotem, MD, MPH

    Resident
    Cork University Maternity Hospital
    Cork, Cork, Ireland

Objective:
This study aimed to evaluate childbirth-related YouTube videos at scale using a Big Data approach, analyzing their emotional tone, classifying them according to content type (e.g., medical recommendations, personal experiences), and thematic content.

Study Design:
Using 446 unique childbirth-related search queries, we scraped 28,436 publicly available YouTube videos. We downloaded the audio track for each video, then transcribed it automatically. Metadata, transcripts, and thumbnails were analyzed by integrating multiple natural language processing (NLP) classifiers, topic modeling approaches, and image-captioning pipelines. This framework enabled sentiment analysis (TextBlob), content classification and medical recommendation assessment (T5 model), topic modeling (LDA), thumbnail image analysis (gpt2), and medication extraction (Med7). Descriptive statistics (frequencies, percentages, medians, interquartile ranges) were used to summarize video characteristics, sentiment distribution, and classification outputs.

Results:
Short-form videos (< 10 minutes) made up 63.3% of the dataset. Of 5,402 videos containing medical recommendations, the vast majority (96%) were classified as balanced. Sentiment analysis revealed that 88% of transcripts were positive, 7.7% negative, and 4.3% neutral. Thumbnails showed a shift toward neutral-to-negative sentiment. The most frequently mentioned medications were opioids (e.g., fentanyl, morphine), oxytocin, and antibiotics. “Can you refuse a C-section” was the most viewed hashtag, reflecting high public interest in maternal autonomy.

Conclusion:
Childbirth-related YouTube content serves both emotional and educational roles. While much clinical content is balanced in presenting both positive and negative aspects, its framing is shaped by storytelling and platform aesthetics. Engaging with digital platforms offers healthcare providers an opportunity to improve online health literacy and ensure access to trustworthy, relatable information.