Skip to content

Hero image: A woman speaking confidentially with a healthcare professional in a private room

All insights

Intimacy & Identity

Intimate Partner Violence in a Connected World: Clinical Care and Digital Safety

10 min readEvidence synthesis
Read the evidence

The question in focus

An expert-grade guide to intimate partner violence, survivor-centred first-line care, reproductive coercion, technology-facilitated abuse and safe digital design.

Evidence at a glance

32%

women worldwide exposed to physical or sexual violence

WHO's 2023 estimates found that nearly one in three women had experienced physical and/or sexual intimate-partner or non-partner sexual violence during their lifetime.

Violence Against Women

Violence is a health issue, not a private failure

WHO estimates that nearly one in three women worldwide, 32%, have experienced physical and/or sexual violence by an intimate partner or non-partner sexual violence [1]. Intimate partner violence also includes psychological abuse, coercive control, stalking and economic abuse. Health effects can include injury, chronic pain, gastrointestinal and sexual health problems, unintended pregnancy, sexually transmitted infection, anxiety, depression and post-traumatic stress.

Reproductive coercion can involve contraceptive sabotage, pressure to become or remain pregnant, or interference with pregnancy decisions. Technology can extend control through location sharing, account access, connected-home devices, message monitoring, image-based abuse and repeated digital contact. These behaviours can occur without physical assault and can escalate around separation, pregnancy or disclosure. A health consultation may be one of the few private contacts available.

  • Ask about fear, control and safety, not only physical injury.
  • Include reproductive and technology-facilitated coercion in assessment.
  • Recognise that leaving may increase immediate risk and is the survivor's decision.

First-line support starts with privacy and choice

WHO recommends immediate, women-centred first-line support for anyone who discloses violence. The LIVES approach is to listen, inquire about needs and concerns, validate, enhance safety and support connection to services [2][3]. The consultation must be private, confidential within clearly explained legal limits, non-judgmental and responsive to the person's priorities. Pressuring disclosure or directing someone to leave can undermine safety.

Professional interpreters need careful safety consideration. A partner, child or family member should not interpret a violence assessment. Clinicians should know local mandatory reporting rules before asking and explain them. Documentation should be factual, use the patient's words where possible and avoid information that could create danger if a partner can access a portal, printed summary or insurance correspondence.

  • Confirm whether the person is alone before asking.
  • Explain confidentiality and its limits before disclosure.
  • Offer options and respect the survivor's decisions.

Digital safety is part of clinical safety

ACOG identifies monitoring phone use, social-network stalking, non-consensual intimate-image sharing and digital humiliation as forms of abuse [4]. A safe service should ask which device, phone number, email, portal and time are safe for contact. Changing passwords or disabling location services may alert an abusive partner, so generic advice can increase risk. Individual safety planning with a trained advocate is preferable.

The Safety Net Project recommends survivor-centred policies for text, chat, video, email and remote advocacy [5]. Healthcare organisations should suppress sensitive notifications when requested, avoid shared-account leakage, review proxy access, minimise lock-screen content and offer a quick-exit function on public information pages. A quick-exit button cannot erase browser history or network logs, so the limitation must be stated.

  • Record a safe contact method without exposing it in a shared view.
  • Do not send detailed violence-related text without explicit consent.
  • Build a process for confidential portal segmentation and proxy-access review.

Healthcare response needs more than screening

ACOG recommends periodic IPV screening, including during obstetric care, in a private setting [4]. Screening only helps when a service can provide first-line support and connection to advocacy, medical, mental health, sexual health, legal and emergency resources. Metrics should therefore include safe disclosure response, completed warm referrals and patient-reported helpfulness, not only the percentage screened.

Clinical care may involve injury treatment, sexual-assault care, emergency contraception, STI testing, pregnancy care and trauma-informed mental health support according to need and timing. A negative screen does not prove safety. People may choose not to disclose because the setting, timing or risk is not right. Universal education about healthy relationships and confidential resources can provide help without requiring disclosure.

  • Make referral pathways available before screening begins.
  • Offer resource information discreetly to all patients when safe.
  • Revisit safety at future contacts without punishment for non-disclosure.

Designing for the person with the least privacy

A digital health product should assume that some users share devices, passwords, phone plans, transport and finances. Survivor-centred digital-safety practice advises assessing technology use and reducing avoidable exposure [5]. Threat modelling should therefore include an abusive household member with legitimate account access, not only an outside hacker. Sensitive data should be minimised, segmented and visible only to those who need it. Audit logs, access revocation and safe alternatives to digital contact are clinical requirements.

Automated risk scoring is especially hazardous if its logic is opaque or it triggers contact that the user did not request. No algorithm can know whether a notification is safe at a particular moment. Human review, explicit consent and local emergency knowledge remain essential. A survivor should be able to decline digital tools without losing access to care.

  • Test the service with survivor advocates before launch.
  • Offer phone, in-person and paperless alternatives.
  • Never expose a violence flag in a shared family dashboard.

What the evidence cannot yet answer

  • Global prevalence estimates vary by survey method, definition, setting and willingness or safety to disclose.
  • Evidence for screening alone is not the same as evidence for a complete survivor-centred intervention.
  • Digital safety advice is context-specific and can increase danger if actions alert an abusive person.

Questions worth taking into care

  1. Is it safe to speak now, and what are the limits of confidentiality here?
  2. Is anyone monitoring your device, accounts, location, medicines or contraception?
  3. Which contact method and time are safe, if any?
  4. What support would be most useful today without increasing risk?
  5. How does this service prevent sensitive information appearing in proxy or shared access?

Source record

Evidence used in this review

Sources were selected for clinical authority, methodological relevance and traceability. Links open the original guidance, public-health record or research publication.

  1. [1]
    Violence Against Women

    World Health Organization · 2026

  2. [2]
  3. [3]
  4. [4]
    Intimate Partner Violence

    American College of Obstetricians and Gynecologists · 2012

  5. [5]
    Best Practices for Digital Services

    National Network to End Domestic Violence Safety Net Project · 2025

  6. [6]
    About Intimate Partner Violence

    Centers for Disease Control and Prevention · 2024

Editorial standard

This evidence synthesis is for general information. It does not diagnose a condition or replace care from a qualified health professional. Treatment choices depend on individual history, examination, local guidance and informed preference. Emergency or rapidly worsening symptoms need urgent local medical assessment.