The Myth of Invisible Men: fathers who kill or injure their babies

15 September 2021

“You are the first person who has ever wanted to know what actually happened, what I went through. No one has ever asked me before….no one has ever bothered with me – you are the first person that has ever sat down and asked me my story.”

Quote from a man convicted of killing a baby, interviewed as part of the Child Safeguarding Practice Review Panel’s review of non-accidental injury of infants by male caregivers, ‘The Myth of Invisible Men’


The UK’s Child Safeguarding Practice Review Panel’s in-depth and hard-hitting report, The Myth of Invisible Men, published today (16 September 2021) shines a light on the rare but tragic phenomenon of fathers and father-figures who kill or injure babies (not by accident). It makes three recommendations:

  • The engagement of fathers must be embedded in prospective and current programmes, including Family Hubs, the Troubled Families Programme and the follow up work stemming from the Leadsom Review into ‘Best Start for Life: A vision for the first 1001 critical days’.
  • A pilot project should be funded to holistically work with expecting fathers who meet the risk factors outlined in this review, providing them with perinatal health provision, local mental health and substance misuse services, and local initiatives to tackle domestic abuse, in a collective and integrated service response.
  • There should be further research into the backgrounds, characteristics and trigger factors of male perpetrators of serious harm, with a view to understanding how practitioners across agencies can more effectively engage with those who might present a potential risk to babies in their care.

The report draws on three key sources: a detailed analysis of 23 serious child safeguarding case reviews where babies were killed or injured by their father, stepfather or another adult male; a series of interviews with eight men convicted of such crimes and currently serving time in prison for their offences; and an evidence review, conducted by us at the Fatherhood Institute.

Our evidence review comprises, in fact, two separate but complementary evidence reviews. One explored what recent research (since 2010) tells us about the prevalence of, and risk factors behind, such cases: in short, how many men kill or injure babies, what do we know about them, and why they do it? The other looked at safeguarding services’ role in preventing them: in short, what are we doing about this problem, and how might we do it better?

This was a big piece of work – a job of national importance, that presented serious methodological challenges, but needed to be done well. It required us to refine areas of knowledge that underpin much of our ‘bread and butter’ work training professionals in father-inclusive practice: narrowing in on the risks some fathers may pose, and examining how services whose role is to protect children, engage with men in families, in the UK and in other comparable countries. We rigorously searched research literature databases, the FI’s own library and a wide range of organisational websites to uncover less well known research studies and interventions, and read and analysed more than 500 studies.

The Panel’s main report, which draws and builds on our own review, makes for dark but essential reading for managers and practitioners at all levels, and in every geographical corner, of the UK’s safeguarding services – including universal services such as maternal and infant health. We wholeheartedly commend it to you.

Below we summarise some key findings of our evidence review, the full version of which runs to more than 80 pages. If you want to delve deeper into the Panel’s report – and our evidence review more specifically – follow the links at the bottom of this article. One of them takes you to a recording of a FREE lunchtime webinar we held on 22 September, where we summarised the findings of our evidence review.

About the dads

To give a sense of the scale of the problem: there are an average of eight cases per year in which fathers/ stepfathers/ mothers’ partners are convicted of killing an infant, in England and Wales. Roughly a quarter of such killings involve death by shaking.

Fathers or mother’s partners (gender not specified) are identified perpetrators in around half of serious case reviews involving non-fatal severe physical assault – in around two fifths of such cases they acted jointly with the mother. The median child age was three months (range 0-17.5 years, with 75% aged under one year).

We found mixed[1] evidence from the UK and internationally on whether fathers (including ‘stepfathers’) outnumber mothers as perpetrators of infanticide, and whether biological fathers outnumber biological mothers. In the only UK analysis we found, covering convicted infanticides in England and Wales over the period 1997-2006, infants were roughly twice as likely to be killed by a father (as main perpetrator), as by a mother.

Parental neonaticides (homicides within 24 hours of birth) were almost exclusively perpetrated by biological mothers. But fathers outnumbered mothers as perpetrators of identified abusive head trauma (AHT), both for AHT deaths and across all identified AHT cases, in samples predominantly of infants. This is a consistent finding in international data.

Biological fathers outnumbered ‘stepfathers’ (broadly defined) as father-perpetrators of identified physical abuse, in studies exclusively or predominantly of infant cases – again, this is a consistent finding across nearly all UK and international studies of infanticide, non-fatal physical abuse and AHT. A recent UK study found biological fathers outnumbering ‘stepfathers’ as convicted principal perpetrator, by a ratio of 10 to 1, and 15 to 1 for shaking-caused deaths.

The ratio of biological fathers vs ‘stepfathers’ as convicted perpetrators evens out or reverses for father-perpetrated homicide of older babies and pre-school children aged 1 to 5 years. So whilst ‘stepfathers’ are a minority of father-perpetrators for infanticide (9%) and shaking-caused infant deaths (6%) in England and Wales – as one might expect, given the rarity of stepfathers in infants’ lives – they are over-represented compared to their prevalence (around 2% in the UK).

Confounding factors mean we do not know from this data whether there is a causal effect: it may be that other risk factors for paternal infanticide and AHT deaths are more prevalent among stepfamilies[2].

Why do they do it?

What do we know about why fathers and father-figures injure or kill their babies? Analyses of serious case reviews often cite parental risk factors such as poor mental health, young parental age, misuse of alcohol and drugs, past criminal convictions, acrimonious parental relationships and separations, partner violence, previous involvement with public authorities (social services, police, criminal justice), poverty and homelessness, alongside prior concerns by practitioners about abuse and neglect in the family.

Large-scale evidence about how these relate to fathers’ use of violence is unclear. But detail from serious care reviews and qualitative research – including the eight interviews drawn on in the Panel’s report – suggest a complex mix of co-occurring factors are likely to be at play in these troubled men’s lives.

How can we prevent such deaths?

The will and capacity to ‘see’, and where necessary seek out, fathers who are struggling – in order to then provide individualised support to the small number who may present a risk to babies – are likely to be key to prevention efforts.

But our review found that as things stand, the services best-placed to develop, assess – and, where necessary, challenge – men’s parenting, do not engage systematically or confidently with them. We agree wholeheartedly with the Panel that this needs to change, and our review found lot of possible approaches that might help.

Article written by Dr Jeremy Davies


[1] Which men are included as ‘stepfathers’ and therefore as ‘fathers’ will influence the differing ratios of father-perpetrators to mother-perpetrators, with few studies or administrative data sources giving explicit definitions.

[2] These might include mother and/or father age, father education, father mental health, family size, family deprivation and whether the mother was cohabiting with the infant’s biological father at birth.

Tags: , , , , , ,