Category Archives: Child protection

Adverse Childhood Experience

Associated with range of negative outcomes.

Later lifestyle

Adverse childhood experiences increase probability of smoking in adulthood. Physical harm in Eastern European countries increases the probability of heavy drinking by about 3.4%, but not in other regions. Exposure to child neglect (little understanding) increases the probability of alcohol abuse by 2.7% in Nordic Countries but not other macro-regions. The experience of a poor relationship with parents is a strong predictor of alcohol abuse for the female subsample in Latin countries (2% higher).

While ACE does not appear to have a substantial effect on excess weight in any macro-region, childhood trauma (physical harm) appears to have a major impact on the likelihood of being obese later in life. Obesity has a more obvious impact on chronic illness than poverty, smoking or alcohol.

Contraception and sexual health

All methods with exception of condoms more than 99% effective  – if you use it as directed, of course! Combined – Rigevidon has v safe progesterone.  Evra is a patch (replace each week for 3 weeks then week free).  Nuvaring is monthly ring, less effected by GI problems but more expensive.

Contraindications for any combined product – migraine with aura, first 6 weeks breast feeding.  DVT risk related to which progesterone is in combination – risk triples with levonorgestrel (Rigevidon), norethisterone, norgestimate (Cilest) but quadruples for others.  But cf risk in pregnancy, more than 10x higher. UKMEC has risk table for family history etc.

Move towards only 4 pill free days – to avoid risk of ovulation if you miss day 1.  Ultimately going to 63-84 days continuously (3-4 packs) but potentially confusing as need to stop and start on different days of the week.

Progesterone only pills were just barrier methods, due to effect on mucus. Cerazette (desogestrel) different, inhibits ovulation without other oestrogen effects. Bleeding is quite common in early days.  Good for controlling cycle related problems eg menorrhagia, catamenial migraine. Good for young people because continuous. Depot good as lead in for implant (else weight gain as side effect).

Nexplanon is implant, under local, lasts 3 years.  But side effects include irregular bleeding. 

Enzyme inducers – cbz, phenytoin, topimarate! And st john’s wort! Rifampicin.  Lamotrigine is not an inducer, but interacts with COCP/POP so avoid unless no other option, in which case needs dose adjustment and must be continuous method.

Consent to sexual activity often confused! Under 13 cannot consent (so different from medical treatment consent).  Consent must be “free agreement” so sleep, coercion, under influence not allowed.  Disclosure of child protection concerns is tricky due to fear of disengagement with service.

Screen time

Sedentary time spent in front of screens programs metabolism and brain neurochem.  Similar to addiction.

But it’s complicated! Partly because there are so many different kinds of activity that can now be done with mobile phones and tablets, and because it’s difficult to control or randomize.

How people interact with screens is changing too. In the mid-2000s in the US, children over 8 spent an average of 6.43 hours a day on electronic media, but this was mainly watching TV.  Evidence of stress response (reduced cortisol increase) on waking after using screens for average 3 hours a day.

But now mobile devices are the centre piece of young people’s social lives.  Boys tend to spend more time gaming, girls more time on social networking.

Effect of high levels of screen time does not seem to be attenuated by equivalent exercise.

High levels of screen time (over 4 hours daily) is associated with poorer school performance. Social skills are poorer. These children tend to form cliques with shared interest, that create further social isolation.

Violent games and media are associated with aggression in children as young as pre-school. Aggression in children can be manifest physically, or verbally, or relationally (ignoring, excluding, spreading rumours). There is also significantly more hostile attribution bias, where you interpret behaviour (such as not being invited to a party) as hostile, even when it is not, or at least ambiguous.

Parental involvement matters – how frequently parents watch TV with their children, discuss content with them, and set limits on time spent playing video games.

Exposure to media violence must also be seen within a risks/resilience approach. [Gentile and Coyne, Aggressive Behaviour 2010] ]

Some evidence that some “social” games themed around cooperation and construction eg Animal crossing have benefits. Similarly, links between media and relational violence pretty weak, but that could be because relational violence content isn’t really examined! Actual content probably more important than time spent playing. Note that in that study of Animal Crossing, background mental health problems seemed to reduce benefit.

Parenting and permanence orders

Permanence order is mechanism for local authority to apply for parental rights and responsibilites to be removed from parents.

Not specifically detailed in law (2007) but “threshold test” must be satisfied:

  1. living with parent poses threat of serious detriment to welfare of child
  2. the need to safeguard and promote welfare of child is paramount consideration
  3. that it is better for the child that the order be made, than that the order not be made

Supreme Court decision In the matter of EV (A Child) (No 2) (Scotland) 2017 –

Not duty of parents to prove parenting ability, but for social work to prove lack of ability with full assessment (or adequate records and sworn evidence of non-engagement).

Also, although allegations of harm may be sufficient reason to place child in care, not sufficient for seeking “permanence order”.  Given that this may mean waiting on criminal proceedings to be completed, children may be stuck in hearing system for longer than before.



Harmful parent child interactions

Emotional abuse and neglect – There is an obligation on a Local Authority to prove their case. This is important both for allegations made against a parent and for assessing the capacity of a parent to look after the child. In some cases the allegations against a parent are unsubstantiated or not proven in any other context when that decision is taken by Social Work.

The Supreme Court decision In the matter of EV (A Child) (No 2) (Scotland) puts the emphasis on meeting the threshold test- a risk of serious detriment is not sufficient in itself. The Local Authority must prove its case. Crucially the Local Authority must address the three following issues through explanation and evidence:

  1. What is the detriment to the child in staying in the care of his or her parents?
  2. Why is this detriment considered serious?
  3. Why is this detriment considered likely?

Potentially more delays in system while evidence is gathered.

The message from the Supreme Court is clear that the onus is not on a parent to show they have the necessary parenting skills to parent the child, but rather for the Local Authority to assess and prove they do not have the capacity to parent the child.

Fabricated and induced illness

Five possible causes for discrepancy between reported and observed symptoms/signs suggested in the RCPCH guidelines:

  1. Exaggeration due to anxiety, poor understanding, lack of knowledge
  2. Carer misperception of child’s illness leading to genuine belief that child is ill
  3. Carer actively promotes sick role by non-treatment or fabrication/falsification or induction of illness (‘true’ FII)
  4. Carer suffers from psychiatric condition which leads them to believe the child is ill
  5. True medical condition

It is important to keep an open mind and to carefully plan appropriate assessments for both medical causes and evidence of maltreatment, without putting the child at further risk of harm.

RCPCH National guidance supports the clinician and team in withholding concerns about FII from the parents at early stage of investiagion. This highlights the main difference in dealing with suspected FII compared to other forms of abuse. Documentation and information sharing need to be handled carefully, as alerting the parents to your concerns may put the child at greater risk of harm – concerns should not be recorded in case notes, parents should not be informed.  Important that team is united.

Gather information about the parents’ background and any known health problems, including some assessment of parenting capacity and risk factors such as domestic violence, mental health issues or drug abuse. This is essential not optional.

Chronology – given multiple attendances for multiple children with different services, helps see overall picture for a family.  Should also include significant events eg moves, bereavements etc.

It is important to feed back your findings to the parents that there appear to be no medical problems and that this is good news.  How family responds to initial assessments and management plans is key to making the diagnosis.

In-patient: clarify nursing ability to supervise 24hrs a day.  Can child leave ward with/without nurse escort?  Who gives medication/food/drink?  Where should notes be kept?

If parents demand a new consultant, you can agree to involve another consultant for a specific medical issue eg asthma/epilepsy.  You should definitely discuss with named doctor.  CAMHS could also be useful for discussing case (and supporting staff, esp if conflicting views)

Disclose to parents – if decision is made to disclose concerns, keep it positive (health of child, etc).  Bring in dad, gran etc if potentially useful.  Don’t confront or challenge, acknowledge how parents and professionals can have different perceptions and responses to a child’s problems.  Present united front, and unambiguous plan.


Sudden Unexpected Death in Infancy (SUDI)

Or Cot death?  Or SIDS (Sudden infant death syndrome)?

It is well recognised that some babies go to sleep apparently healthy, and then don’t wake up in the morning.  Even after a full post mortem (PM) investigation, no cause is found.  This unexplained phenomenon however has some very well recognised features eg age 2-6 months, prematurity, maternal smoking, poor socio-economic conditions, prone sleeping.

SUDI was originally defined by CESDI (Confidential Enquiry into Stillbirths and Deaths in Infancy) as death between 7 and 365 days where unexpected and unexplained at autopsy, during an acute illness that was not recognised as life-threatening, due to an acute illness of less than 24 h duration in a previously healthy infant (or death after this if life had only been prolonged by intensive medical care); definition also includes deaths from a pre-existing occult condition, and deaths from any form of accident, trauma or poisoning.

I find SUDI most useful for describing the initial situation one may find oneself in, particularly from the point of view of bereavement, need for medical and police investigation.  Interestingly, many of the same risk factors pertain to both deaths unexplained (ie SIDS, or strict SUDI) and to accidental deaths (with the exception of prone sleeping).

SIDS is the ICD recognized term, so is what is generally put on a death certificate.  However pathologists vary in their use of the terminology, some will use “Unascertained” to mean SIDS, others will use SIDS but reserve Unascertained for cases where there are additional factors that somehow cast doubt on the diagnosis.

Similarly, overlying (smothering) as a cause of SUDI is often inferred from the history, but may be specified on the death certificate to differentiate from SIDS.

PM finds a cause in about a 1/3 of cases) eg

  • Infection
  • Cardiomyopathy, anomalies of coronaries
  • Ion channelopathies
  • Metabolic disorders eg MCAD

See also Prevention, and Sudden unexpected postnatal collapse.

Head injury

In minor head injury (definition?!), statistically significant correlation between intracranial haemorrhage and:

  • skull fracture
  • focal neurology
  • history of loss of consciousness
  • GCS abnormality (difficult to gauge in preverbal children…)

Headache and vomiting were not found to be predictive and there was great variability in the predictive ability of seizures. (meta-analysis, ArchDisChild 2004;89)

SIGN 110 suggests immediate CT for:

  • GCS less than 14
  • high speed mechanism
  • witnessed loss of consciousness for more than 5 minutes
  • Suspicion of open or depressed skull fracture
  • Any sign of basal skull fracture
  • Tense fontanelle
  • Focal neurological deficit

Otherwise, early (ie within 8 hours) CT should be considered if:

  • bruise/swelling/laceration >5cm on head
  • post-traumatic seizure without epilepsy (and not reflex anoxic)
  • amnesia (antero- or retrograde) >5 minutes
  • suspicion of NAI
  • Significant fall
  • 3+ discrete episodes of vomiting
  • abnormal drowsiness
  • GCS other than 15 in under 1yr old, assessed by experienced provider

If suspicion of NAI, extra rule applies – CT should be done “as soon as child is stable” (and ideally within 24 hrs) if under 1 yr, or neuro signs (incl haemorrhagic retinopathy).

Any loss of consciousness should be assessed, but interestingly retrograde amnesia has to be for >30 minutes to warrant assessment, whereas NICE would do immediate CT! Otherwise 2+ vomits, severe and persistent headache, coagulopathy, difficulties with assessment or social situation, or any other indication for CT.

Admit if any indications for CT, although it also says discharge can be considered if social situation suitable!

NICE head injury (2017) guidelines

similar criteria, but suggests immediate CT for more. Change in practice from admit and watch (Royal College of Surgeons guidelines) to diagnose and decide. Leads to far fewer skull XRs, a lot more CTs and maybe half as many admissions. Some centres have seen cost savings due to earlier discharge.

CT within 1 hour for:

  • age over 1 year, GCS<14 on initial assessment;
  • age under 1 yr; GCS<15 on initial assessment.
  • GCS<15 at 2 hours after injury.
  • age under 1yr plus bruise, swelling or 5cm laceration.
  • Suspicion of NAI.
  • Loss of consciousness >5min (witnessed).
  • Post-traumatic seizure without epilepsy.
  • Abnormal drowsiness.
  • Suspected open or depressed skull fracture, or tense fontanelle.
  • Any sign of basal skull fracture – haemotympanum, panda eyes, CSF leak from ears/nose, Battle’s sign.
  • Focal deficit.

Plus CT within 1 hour if MORE than 1 of the following:

  • Witnessed loss of consciousness more than 5 minutes
  • Abnormal drowsiness
  • 3+ discrete episodes of vomiting
  • Dangerous mechanism eg high speed road traffic accident, fall >3m, high speed projectile
  • Amnesia (retro or antegrade) >5 min

If only 1 of the above, then observe minimum 4 hours – go to CT if during that time:

  • GCS <15
  • Further vomiting
  • Episode of abnormal drowsiness

In children under 10yr, CT for spine should be avoided (risk to thyroid) unless severe head injury (eg GCS<=8), strong suspicion despite plain films, or inadequate plain films. Over 10yr, CT is investigation of choice if:

  • GCS<13 (so 1 point less than for head).
  • intubated.
  • inadequate plain films.
  • Continued suspicion.
  • Needing multi-region scan anyway!

Neuroscience centres are expected to be able to perform initial management of multiple injuries in children. Local guidelines for transfer should be drawn up – there are benefits for being in a neurosurgical centre even if surgery is not required.

Kids with a fracture are not as prone to intracranial lesions as adults, at the same time they are more likely to have intracranial lesion without a fracture!

Note increased risk of malignancy with CT.  So observe for 4 hours if persistent vomiting, review by senior clinician to decide further observation rather than CT. Involve parents in decision [BMJ 2019;365:l1875]


No good RCTs! Avoid secondary brain injury – 1 episode hypotension post head injury triples mortality. Cerebral blood flow is low in first 24hr, peaks at 48hr. Depends on temperature, seizures, pain/anxiety.

Glasgow Coma Score (GCS) 9-12 is moderate, <=8 is severe (equivalent to P or U in AVPU score) and is indication for ventilation to protect airway as reflexes potentially unreliable.

Diffuse axonal injury progresses over 24+ hrs, difficult to see on scan.

Consider external drain/ventriculostomy for intracranial haemorrhage. ?Remove contused brain ?Decompressive craniectomy

Neuroprotective strategy:

  • Head up 30deg, straight
  • Maintain pCO2 at 35-40mmHg
  • Cool if febrile (awaiting data on role of hypothermia). Paralyse to avoid shivering. Paralysis will make seizures difficult to recognise: role for prophylactic anti-epileptics?
  • Analgesia
  • (steroids not helpful)
  • CVP&arterial BP monitoring, ensure adequate perfusion pressure
  • ICP monitoring if neuro signs, GCS <9, post decompression. Bolt gives data but does not allow CSF drainage. ICP takes 7-10 days to settle

For RICP, 3% NaCl 3-5ml/kg bolus – Keep osmo <310mmol/l.

For induction, thiopentone is traditionally used. Ketamine theoretically increases ICP but no real evidence. Adding fentanyl smooths cardiovascular response to procedure.

CT@72h is prognostic.

Shaken Baby: lethargy, vomiting, apnoeas, seizures (40-80%), opisthotonus, irritability. See NAI.


Typical areas of accidental bruising:

Pattern of accidental bruising
Maguire, 2010

Typical pattern of abusive bruising:

Pattern of abusive bruising
Maguire, ADC Ed & Pract 2010

From prospective longitudinal study of children (<6 years):

  • 6.7% of premobile children had at least one bruise (2.2% of babies who could not roll over and 9.8% in those who could)
  • Most common site affected in all groups was below the knees, followed by ‘facial T’ and head in premobile and early mobile.
  • The ears, neck, buttocks, genitalia and hands were rarely bruised (<1%).
  • Gender, season or the level of social deprivation not associated with bruising patterns, although having a sibling increased the mean number of bruises.
  • There was considerable variation in the number of bruises recorded between different children, which increased with developmental stage, and was greater than the variation between numbers of bruises in collections from the same child over time – so some kids do just bruise more than others?

[Arch Dis Child 2014]


Are you sure it isn’t a Mongolian blue spot? Or capillary haemangioma? Or erythema nodosum?

Cupping in Chinese culture! Dermatitis artefacta?


Child Sexual Abuse

Any sexual activity under 13yr classifed as assault, according to RCPCH. Not the same rules as capacity and consent for medical treatment.  Sexual offences Scotland act 2009.  Consent must be “free agreement” so sleep, coercion, under influence not allowed.  Disclosure of child protection concerns is tricky due to fear of disengagement with service.

Police (family protection unit) refer for forensic medical viz joint police surgeon and paediatrician. Video colposcopy and swabs. Consent incl data protection important. Bloods only if high risk.  Rapid HBV immunisation.

Social work can consent if guardian is perpetrator.

Disclosure of child protection concerns is tricky due to fear of disengagement with service.

Difficult to exclude internal injuries since speculum/rectal exams not acceptable for pre-pubescents: relies on details of history.

ULIPRISTAL (ellaOne) new post intercourse contraception, window of 5 days. IU device not suitable for poorly oestrogenised.

Support services: advice leaflets, SW follow up, GUM clinic at 1/52, CAMHS.