Domestic and Elder Abuse

Intimate partner violence and elder abuse are among the few conditions in orthopaedic practice where the diagnosis depends almost entirely on whether the clinician thinks of it. The injuries are ordinary - a distal radius fracture, a facial laceration, a bruised forearm. What is not ordinary is the cause, and the cause is only found by asking.

Orthopaedics is unusually exposed to these patients. In Bhandari and colleagues' study of the musculoskeletal manifestations of physical abuse after intimate partner violence, head and neck injuries were the most prevalent (40%) and musculoskeletal injuries the second most common. Domestic violence has been described as the leading cause of non-fatal injury to women, and musculoskeletal injury is its second commonest physical manifestation - which places the fracture clinic, the emergency department and the trauma list squarely in the path of these patients.

Abuse of an adult differs from abuse of a child in one fundamental respect: a capacitous adult may decline your help, and managing that refusal well is the central clinical skill. In a child, suspicion compels referral. In a competent adult, it compels a conversation — and if that conversation ends in refusal, your task becomes documentation, safety planning and leaving the door open, not escalation over the patient's head. For the paediatric fracture patterns, the skeletal survey and the safeguarding pathway that follows them, work through non-accidental injury alongside this page.

Patterns of Injury in Intimate Partner Violence

• Head, face and neck injuries are the commonest - reported as 40% in Bhandari's series. Facial fractures, periorbital and nasal injury, dental injury, tympanic membrane perforation, and scalp haematoma or traction alopecia.
• Musculoskeletal injuries are the second commonest. The pattern that should raise suspicion is the defensive injury: fractures of the ulnar shaft and forearm sustained raising the arm to block a blow, distal radius and hand and metacarpal injuries, and bruising to the extensor surfaces of the forearms.
• Injuries in a covered distribution - the trunk, upper arms, medial thighs, breasts and abdomen - areas concealed by clothing.
• Bilateral, symmetrical or grip-pattern injuries: fingertip bruising on the upper arms, linear or parallel marks from a slap, loop-shaped marks from a cord or belt, ligature marks, and bite marks.
• Injuries of clearly differing ages, on the skin or on imaging.
• Non-fatal strangulation - see the safety alert below. External signs may be minimal or absent.
• Injury during pregnancy, particularly abdominal.
• Sexual assault, which requires specialist forensic pathways and must not be examined opportunistically.

Patterns in Elder Abuse

• Bruising in an inflicted distribution - see the discriminators below.
• Fractures with a mechanism that does not fit, particularly in a non-ambulant person, where any long bone fracture demands explanation.
• Neglect: pressure injury in unexpected sites or of unexpected severity, dehydration, malnutrition and unexplained weight loss, poor hygiene, overgrown nails, soiled clothing, an untreated or late-presenting fracture, and withheld analgesia.
• Medication problems: over-sedation, missing medication, or hoarded medication.
• Financial signs: unexplained inability to pay for care or medication despite adequate income, a companion who controls all money, sudden changes to legal arrangements.
• Behavioural signs: fearfulness, flat affect, deference to the companion, and reluctance to speak.
• A carer who is hostile, dismissive of the patient, refuses to leave, or gives a history that conflicts with the patient's.

View larger

Other Conditions to Consider Before Concluding Abuse

What Good Practice Achieves

• Identification is the intervention. Most of the benefit of an orthopaedic contact comes from asking, believing, documenting and referring - not from anything technical.
• Because abuse is recurrent and escalating, a single well-handled contact can change the trajectory even where the patient declines referral on that occasion, by establishing that disclosure is safe and that help is available.
• Documented contemporaneous records are frequently the decisive evidence in later protective or legal proceedings, long after the clinical episode is forgotten.
• PRAISE's finding that only 14% had ever been asked means the achievable improvement is enormous and requires no new technology, funding or equipment - only the question.

Follow-Up

Related Topic

For suspected abuse in children - the characteristic fracture patterns, the skeletal survey protocol, the differential diagnosis of metabolic bone disease, and the paediatric safeguarding pathway - see the separate OrthoVellum topic Non-Accidental Injury at /topics/non-accidental-injury. That material is deliberately not duplicated here. The overlap that matters clinically is this: where an adult discloses abuse, a child in the same household is at risk and a separate duty is engaged, and Bhandari's series found child protective services were concomitantly involved in half of the women living in abusive relationships.

How to Ask

• Get the patient alone. Use a legitimate clinical reason - accompanying them to imaging, a dressing change, a chaperoned examination - and use a room, not a curtained bay.
• Use a professional interpreter where needed, never a family member or the accompanying person.
• Normalise the enquiry. For example: "Because injuries like this are sometimes caused by someone else, I ask everyone about this. Has anyone hurt you, frightened you or made you feel unsafe at home?" Framing it as routine removes the implication of accusation.
• Ask about the range of abuse, not just physical violence - fear, control, humiliation, isolation, money, and being prevented from getting care.
• Ask specifically about strangulation given its lethality significance.
• Establish who is in the household, especially children and dependent adults, because that changes your duty.
• Do not press for forensic detail you do not need. Repeated questioning can be distressing and can compromise a later formal account.
• Believe the patient, and say so. A supportive first response substantially influences whether they seek help again.
• If they decline to discuss it, leave the door open: document that you asked, give information in a safe form, and say clearly that they can return at any time.

Validated Screening Tools

No tool substitutes for a private conversation, but knowing the named instruments is examinable and they structure the enquiry.

Examination

• Examine the whole patient, with consent and a chaperone, including areas covered by clothing. Injuries in a concealed distribution are only found if you look.
• Examine the neck, eyes, ears and mouth specifically, for strangulation signs, tympanic perforation and dental injury.
• Document every lesion on a body map: site, size in centimetres, shape, colour, whether tender, and whether patterned.
• Assess for neglect: skin integrity and pressure areas, hydration, nutrition, hygiene, continence care.
• Assess cognition and capacity, and mental state including risk of self-harm.
• Image appropriately for the clinical injury. There is no adult equivalent of the paediatric skeletal survey as a routine screen, but a low threshold for imaging areas of unexplained tenderness is reasonable, and old healed fractures found incidentally are relevant evidence.
• Where sexual assault is disclosed or suspected, do not examine opportunistically - refer to the specialist forensic service, because examination and sampling must be done correctly and once.

View larger

Photography

• Photography is valuable and is frequently the evidence that matters, but it requires care:
  • Specific, documented, informed consent for photography, including what the images will be used for, who may see them, and that they may be disclosed in legal proceedings. Consent to photography is separate from consent to examination.
  • Include an identifier and the date, and a scale marker alongside each lesion.
  • Take a sequence from wide to close - orienting view, regional view, then close-up - so the anatomical location of a close-up is unambiguous.
  • Use the hospital's own equipment and secure storage within the clinical record system. Never store clinical images on a personal device.
  • Never alter, crop, filter or enhance images; maintain an auditable chain of custody.
  • The patient may decline photography and that must be respected - in which case the body map and written description carry the evidential weight.

Consent, Confidentiality and Information Sharing

• Seek consent to share information wherever possible; explain who you would share it with and why.
• Confidentiality is not absolute. Sharing without consent may be justified where there is a risk of serious harm to the patient or to others, where the patient lacks capacity, or where a child or dependent adult is at risk. Where you share without consent, document the justification and, wherever possible, tell the patient you are doing so unless that would increase risk.
• Take advice. Discuss with your safeguarding lead or named professional before acting on a difficult decision, and record that discussion. Rarely is an individual clinician expected to make these judgements alone.

Global Framing

• Both intimate partner violence and elder abuse are global phenomena. PRAISE deliberately sampled across five countries on three continents and found intimate partner violence in orthopaedic fracture clinics everywhere it looked, with some between-country variation in past-year risk. Yon's meta-analysis drew on 28 countries and found elder abuse prevalence associated with country income classification but not with gender.
• Robust prevalence data are sparse in low-income and middle-income countries, which Yon and colleagues explicitly identify as a gap and a reason elder abuse remains a neglected global public health priority. Absence of local data should never be read as absence of local abuse.
• Statutory frameworks differ fundamentally between countries and this topic deliberately describes pathways in generic terms - "adult safeguarding or adult protection service", "safeguarding lead or named professional" - rather than any one nation's legislation. The clinical obligations are universal even where the legal machinery is not.

Side-by-Side Guidance

Areas of Genuine Uncertainty

• Universal screening versus targeted clinical enquiry. Whether all patients should be routinely screened, or only those with indicators, remains contested. The concern is that screening without accessible support services may not improve outcomes and could increase risk. What is not contested is that clinical enquiry where abuse is suspected is mandatory.
• Whether identification improves outcomes. Prevalence is well established; high-quality evidence that healthcare-based identification alone reduces subsequent violence is much weaker, and interventions that work appear to depend on the availability of advocacy and support.
• The optimal screening instrument in an orthopaedic setting, and whether self-completed questionnaires outperform direct questioning. PRAISE used both direct questions and two instruments.
• Reliability of injury pattern as evidence. The patterns described are statistical associations from selected populations, not diagnostic signs, and their positive predictive value in an unselected clinic population is unknown.
• Elder abuse prevalence in low-income and middle-income countries, and in institutional as opposed to community settings, is under-studied.
• How to respond to coercive control in a formally capacitous patient - where the decision is legally the patient's but is substantially constrained by fear - remains ethically unresolved.
• Detection in cognitive impairment, where no validated tool exists and the patient cannot advocate for themselves, is a recognised gap.

High- versus Limited-Resource Practice

• Well-resourced settings: hospital safeguarding teams and named professionals, specialist domestic abuse advocates embedded in services, forensic examination services, formal lethality risk assessment, refuge provision, professional interpreting, dedicated photography services with secure image storage, and multi-agency risk conferencing.
• Limited-resource settings: the highest-value interventions cost essentially nothing. A private room and a direct question. A believing response. A body map drawn by hand with measurements. A written record in the patient's own words. Establishing where the patient is going and with whom before discharge. Knowing one telephone number for the local protection service. None of that requires funding, and PRAISE's finding that only 14% had ever been asked shows that the deficiency in every setting is the question, not the resources.
• Everywhere: know your local pathway before you need it, including out of hours; use professional interpreters rather than family members; never discharge a patient into the care of a suspected perpetrator; and take advice rather than carrying these decisions alone.

Related pages. The paediatric parallel: Non-Accidental Injury is the same reasoning applied to a child, and it is worth reading alongside this page because the two share a household - the Bhandari data show child protective services were involved in half of the abusive relationships studied, so identifying one form of abuse should prompt enquiry about the other.