Bereaved by drug and alcohol deaths: the children behind the statistics

Hundreds of thousands of children have had a parent die from drugs, and almost no country counts them. The best estimate, for the United States, is that 321,566 children had a parent die of an overdose between 2011 and 2021^[1]. Scotland is one of the very few places that records these children routinely. For deaths caused by alcohol, the children are almost entirely missing from the statistics.

When a parent dies from drugs or alcohol, a child often grieves a person they were already, in some ways, missing. Months or years of a parent's illness, absence or unpredictability may come first. Then comes a sudden death, often an investigation into it, and a cause of death that adults may find hard to name. Their grief is the ordinary grief of a child whose parent has died. What makes it different is the silence and stigma that can surround it, and how rarely services know the child exists. This study gathers what is known about these children worldwide. A sibling study covers adults bereaved by drug and alcohol deaths.

In the US, overdose has become a major cause of parental death in childhood

The largest body of evidence comes from the United States, where overdose deaths rose for two decades. Two research teams have used national death records to estimate how many children were left behind. Neither counted the children directly; both modelled them from death records combined with survey or census data, so both figures are modelled estimates.

A team from federal health agencies combined death records with a national household survey on drug use to estimate how many of the adults who died were parents. Between 2011 and 2021, 649,599 adults aged 18 to 64 died from a drug overdose. The team estimated that 321,566 children living in the community (95% confidence interval 276,592 to 366,662) were bereaved of a parent in those deaths^[1]. The rate rose from 27.0 per 100,000 children in 2011 to 63.1 per 100,000 in 2021. The parent who died was more often a father (an estimated 192,459 children) than a mother (129,107)^[1].

A second team used a demographic projection model over a longer period. It estimated that 759,000 young people under 18 (95% CI 722,000 to 800,000) had a parent die from drug poisoning between 1999 and 2020^[4]. Over those years, parental deaths due to drugs rose 345% in the model, against 24% for all causes other than drugs and firearms. In 2020, drugs and firearms together accounted for 23% of all parental deaths, double the 12% of 1999^[4]. The two studies cover different years and use different methods, so their totals should not be compared directly.

Children's family losses go beyond parents. Using kinship modelling, a third team estimated that more than 1.4 million US children had had a family member die of an overdose, “often losing central figures like parents or grandparents”^[5]. That figure covers other family members, not only parents.

These deaths do not fall equally. In 2021, children of American Indian and Alaska Native parents were bereaved by overdose at an estimated 187.1 per 100,000, more than double the rate for children of non-Hispanic White (76.5) or non-Hispanic Black (73.2) parents^[1].

Almost nowhere counts these children, and Scotland shows what counting looks like

We found no country that publishes a national count of children bereaved by drug or alcohol deaths as part of its official death statistics. Scotland comes closest. Its National Drug-Related Deaths Database, run by Public Health Scotland, gathers case records on each drug-related death, including whether the person was a parent.

For deaths registered in 2022, where this was known, 28% of the people who died (243 people) were a parent or parental figure to at least one child under 16^[2]. At least 387 children had a parent or parental figure die that year. Of those children, 17% (64) were living in the same household as the parent when they died^[2]. The 387 is a minimum: parental status was unknown for 13.8% of the deaths reviewed^[2]. These deaths are concentrated where families have least: over half (52%) of the people who died lived in Scotland's 20% most deprived neighbourhoods^[2].

The other direct count comes from one US state. Researchers in Michigan linked death certificates to the birth certificates of the children of the people who died. Between 2000 and 2023 they identified 115,558 children whose parent had died. For 38,429 of them, the parent died from what the authors call a stigmatised cause: drug overdose, an alcohol- or drug-induced death, suicide or homicide^[3]. That share rose from 28.5% of parental deaths in 2008 to 41.8% in 2023, or “2 in every 5”. Across counties it ranged from 21.3% to 47.2%^[3]. Because only biological parents named on birth certificates could be found, and fathers are named less often, the authors describe their figures as conservative^[3].

Children bereaved by alcohol deaths are the least visible of all

Alcohol kills far more people than drugs worldwide. The World Health Organization estimates that 2.6 million deaths a year were attributable to alcohol in 2019, 4.7% of all deaths, against 0.6 million attributable to drug use^[6]. Of these, 2 million alcohol deaths and 0.4 million drug deaths were among men. WHO also reports that the highest proportion of alcohol-attributable deaths in 2019, 13%, was among people aged 20 to 39^[6], ages at which many people are raising young children. These are modelled global estimates of deaths attributable to alcohol, not counts from death certificates.

Yet we could not find a single study, in any country, that estimates how many children have had a parent die from an alcohol-related cause. The Michigan study includes alcohol-induced deaths, but only within its wider group of stigmatised causes^[3]. Scotland's database covers drug deaths only^[2]. In the Norwegian register study described below, the causes of parental death reported are drug-related, suicide or self-harm, injury, cancer and other diseases; no separate figure is given for alcohol^[7].

There are reasons the gap persists. Most alcohol deaths are deaths from illness: WHO estimates that 1.6 million of the alcohol-attributable deaths in 2019 were from noncommunicable diseases, including 474,000 from cardiovascular disease and 401,000 from cancer^[6]. A child whose parent died of liver disease after years of heavy drinking may be counted, if at all, as bereaved by illness. Counting these children would mean linking parenthood to the full range of alcohol-attributable deaths, which no national system we found currently does.

Outcomes are often harder after a drug death, and family circumstances may explain part of the gap

Few studies follow children bereaved by drug or alcohol deaths separately, but the ones that do point the same way. Norway's population registers offer the clearest worked case. Researchers followed 574,229 people born in Norway in 1977–87, of whom 18,339 (3.2%) had a parent die before they turned 18; for 1,001, the parent's death was drug-related^[7]. The study measured years spent between ages 22 and 29 not in education, employment or training.

Every cause of parental death was linked to more such years than among people whose parents were alive. The link was weakest when a parent died of cancer (incidence rate ratio 1.19) and strongest when the death was drug-related (2.36, 95% CI 2.17 to 2.56)^[7]. The figures vary sharply with parental education. In families without a university-educated parent, the predicted number of such years after a drug-related death was 2.28, against 1.03 where a parent had a university education^[7]. The authors suggest the stronger associations in families with less education “may largely reflect confounding factors”; the models adjust for birth year and migration background, not for everything that differs between families.

A US survey of 36,309 adults, asked about their childhoods, separated the death of a parent from a parent's substance problems. People who had experienced both before age 18 reported an average of 5.34 adverse childhood experiences, more than any other group. They had between 2.06 and 3.59 times the odds of a suicide attempt and of six substance-use and mental health disorders in adulthood, compared with bereaved people whose parents had no substance problems^[8]. Most differences remained after adjusting for other adversity. The parents' deaths were not necessarily caused by drugs or alcohol, and these are associations in recalled histories, not proof of cause. Higher odds describe groups, not what will happen to any one child.

In Michigan, researchers also looked at deaths among the children themselves. Of 21,368 children whose parent died from overdose or drug-related causes in 2009–2023, 79 died before age 18. That is a crude rate of 36.97 per 10,000, against 5.22 per 10,000 for all Michigan children^[9]. These rates are not adjusted for age, poverty or anything else, and the study cannot say why. The authors present the finding as evidence of the need for preventive support for bereaved children. Our study of mental health after a parent's death sets these findings in the wider evidence.

Children's grief often starts before the death, and support rarely reaches them

For many children, a parent's death from drugs or alcohol comes after a long period of what researchers call ambiguous loss. A US nursing paper applied this idea to children of parents with opioid use disorder. It describes children living with a parent's “physical absence and psychological presence”, when the parent is away in treatment, in prison or seeking drugs, and with their “psychological absence but physical presence”, when the parent is at home but intoxicated or in withdrawal^[10]. It is a conceptual paper, not a study with data, but it helps explain why a child's grief may be mixed with relief, anger or long-held worry. The authors conclude: “Interventions that are child-focused are lacking.”^[10].

The little evidence on support suggests these children are often missed. In a Norwegian survey of 255 family members and close friends bereaved by a drug-related death, about half had received help themselves. Few of those with children in the family said the children had received help, and fewer than a third of those were satisfied with it^[11]. It was a convenience sample, not representative of all bereaved families, but its conclusion was plain: “younger age groups and children need particular recognition, and a family perspective from services is essential”^[11].

The Michigan researchers reached a similar view from the counts. They argue that children bereaved by these deaths may face “a more complex bereavement process”, and that county-level figures can help direct bereavement services to the places with most need^[3]. For how children fare after a parent dies by suicide, which shares much of this stigma, see our study of children bereaved by suicide.

What the evidence asks of us

The evidence is thin, but it points in a consistent direction.

• Count the children. Scotland shows that recording whether a person who died was a parent, and how many children they had, is possible within routine reviews of drug deaths^[2]. Michigan shows that linking death and birth records can do the same for every cause^[3]. Countries and regions that review drug or alcohol deaths could add the same questions. Without them, services cannot plan for children they cannot see.
• Include alcohol. No system we found counts children bereaved by alcohol-related deaths, though WHO links over four times as many deaths to alcohol as to drugs^[6]. Any count should cover alcohol deaths recorded as illness, not only those coded as poisoning.
• Reach children who did not live with the parent. In Scotland, most children whose parent died a drug-related death were not living with them^[2]. Support offered only to the household where the death happened will miss most of them.
• Take a family perspective. The Norwegian survey found that children were rarely offered help even when adults were^[11]. Researchers in Michigan and Norway both call for bereavement services to plan for these children specifically^[3][11].
• Recognise what came before. Where a parent's substance problems preceded the death, children may carry other adversity too. The authors of the US survey suggest this “may warrant additional trauma-informed mental health care in bereavement services”^[8].
• Put support where deaths concentrate. Drug deaths cluster in deprived areas and some communities^[1][2]. Local figures, like Michigan's county estimates, can guide where funders and services place childhood bereavement support^[3].

Our review of what helps bereaved children covers the evidence on support itself, and the research library on bereaved children brings the wider evidence together.

Language: this piece uses plain words such as “died” rather than euphemisms, because they are clearer. It uses “stigmatised causes” only where that is the researchers’ own term. Where it discusses suicide, it follows the Samaritans’ media guidelines.

Why we're publishing this

Children whose parent dies from drugs or alcohol are some of the least visible bereaved children. Their parent's death may be surrounded by silence, and the systems that record these deaths rarely ask whether the person who died had children. The charities and services that support bereaved children, and the families themselves, have long said these children need recognition. We want their case to be easy to make, with the best figures available, from every country that has them, set out plainly.

How to read this data

The US figures for how many children were bereaved of a parent by overdose are modelled estimates, built from death records and survey or census data, not counts of children. The two main US studies cover different years and methods and should not be compared directly^[1][4]. Scotland's figure is a count from case records and is a minimum, because parental status was not known for every death^[2]. Michigan's figures are counts for one US state, covering biological parents only^[3]. WHO's global alcohol and drug death figures are modelled^[6]. The outcome studies show associations. The Norwegian study adjusted only for birth year and migration background, and its authors suggest the stronger associations in less-educated families may largely reflect confounding. The Michigan child mortality rates are crude. The US survey relied on adults' recall of their childhoods. Almost all the evidence comes from the US, Scotland and Norway, and we found none that looks at children bereaved by alcohol deaths on their own.

Use this data

Free to cite with attribution to Understanding Bereavement. Charities, schools, services and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This study summarises published research about groups of people for general information. It is not medical advice or bereavement support, and it cannot describe or predict anything about an individual child or family. If you are worried about a child or about yourself, please speak to a GP or another qualified professional.

Sources

1. Jones CM, Zhang K, Han B, et al. Estimated number of children who lost a parent to drug overdose in the US from 2011 to 2021. JAMA Psychiatry, 2024. europepmc.org/article/MED/38717781
2. Public Health Scotland. The National Drug-Related Deaths Database (Scotland) report: analysis of deaths registered in 2021 and 2022. Public Health Scotland, 2026. publichealthscotland.scot/media/37992/2026_03_17_ndrdd_report-formatted.pdf
3. McCabe SE, Hulsey E, Kcomt L, Evans-Polce RJ, Radford G, Tennant SD, McCabe VV. Children’s experiences of parental deaths due to suicide, homicide, overdose, alcohol, or drug use. JAMA Network Open, 2025. europepmc.org/article/MED/40928783
4. Schlüter BS, Alburez-Gutierrez D, Bibbins-Domingo K, Alexander MJ, Kiang MV. Youth experiencing parental death due to drug poisoning and firearm violence in the US, 1999–2020. JAMA, 2024. europepmc.org/article/MED/38703404
5. Verdery AM, Ryan-Claytor C, Smith-Greenaway E, Sarkar N, Livings M. More than 1.4 million US children have lost a family member to drug overdose. American Journal of Public Health, 2024. europepmc.org/article/MED/39326002
6. World Health Organization. Over 3 million annual deaths due to alcohol and drug use, majority among men. WHO, 2024. www.who.int/news/item/25-06-2024-over-3-million-annual-deaths-due-to-alcohol-and-drug-use-majority-among-men
7. Delalic L, Wörn J, Reme BA. Early parental death and being not in education, employment, or training (NEET-status) in Norway: a population-wide study on the moderating role of parental education. European Journal of Public Health, 2025. europepmc.org/article/MED/40628386
8. Giang CW, Evans-Polce RJ, Kcomt L, et al. Parental death and parental substance use problems during childhood and the risk of DSM-5 substance use, suicidality and mental health disorders. Social Psychiatry and Psychiatric Epidemiology, 2026. europepmc.org/article/MED/41824018
9. McCabe SE, Kcomt L, Evans-Polce RJ, Radford G, Tennant SD, Hulsey E, McCabe VV. Childhood mortality by parental cause of death. JAMA Network Open, 2026. europepmc.org/article/MED/41870432
10. Mechling BM, Ahern NR, Palumbo R. Applying ambiguous loss theory to children of parents with an opioid use disorder. Journal of Child and Adolescent Psychiatric Nursing, 2018. europepmc.org/article/MED/30298538
11. Kalsås ØR, Titlestad KB, Dyregrov K, Fadnes LT. Needs for help and received help for those bereaved by a drug-related death: a cross-sectional study. Nordisk Alkohol- & Narkotikatidskrift, 2023. europepmc.org/article/MED/37969900