A Growing Number of U.S. Children Live in Counties Affected by the Drug Epidemic

The number of U.S. children living in counties with high drug overdose death rates has increased sharply since 2000, according to a Population Reference Bureau (PRB) analysis of county-level data from the U.S. Centers for Disease Control and Prevention (CDC) and the U.S. Census Bureau.

While drug overdoses and drug-related deaths are much more common among adults than children, it is likely that millions of children have been affected by the epidemic, either directly—through parental addiction or death—or indirectly through the experiences of friends, classmates, or neighbors.

Between 2000 and 2016, the number of children under age 20 living in counties with high drug overdose death rates—defined as 20 or more deaths per 100,000 population—rose from less than 250,000 to nearly 22.6 million, nearly a 100-fold increase (see Figure 1). The share of children living in high-rate counties also increased sharply during this period, from less than 1 percent of children in 2000 to 28 percent in 2016.



The rapid increase in drug overdose deaths in the United States is cause for alarm. The opioid epidemic in the United States has gained widespread public attention, but this attention has focused mainly on its effect on adults and those who have died of overdoses. Less consideration has been given to the potential effects of the opioid crisis on children. However, our results suggest that a growing number of children—regardless of racial/ethnic background, socioeconomic status, or area of residence—may be affected.

The Racial/Ethnic Composition of Children at Risk Has Shifted

As the opioid crisis spread, the demographic composition of children in affected counties shifted. In 2000, more than half of children (under age 20) living in high-rate counties (58 percent) were African American (see Figure 2). Baltimore, Maryland, accounted for most of the children at risk. By 2016, the majority of children in high-rate counties (57 percent) were non-Hispanic white.



The share of children in high-rate counties who are Latino has also increased sharply over time, from 6 percent in 2000 to 20 percent in 2016, but remains lower than the national share of Latinos in the population under age 20 (25 percent).

While whites made up the largest share of children in high-rate counties, American Indian/Alaska Native children were the most likely to live in high-rate counties, at 41 percent (see Figure 3). In contrast, 18 percent of Asian/Pacific Islander children and 23 percent of Latino children lived in counties with high drug overdose death rates.



The poverty status of children living in counties with high drug overdose death rates has also shifted over time. Between 2000 and 2016, the poverty rate among children under age 18 living in high-rate counties fell from 28 percent to 21 percent, while the child poverty rate in counties with lower overdose death rates increased, from 16 percent to 19 percent (see Figure 4). These divergent trends reflect the rising number of drug-related deaths in higher-income communities, including many large suburban areas.1


Children in the Southwest Are Among the Most Vulnerable

In recent decades, the risk of drug-related deaths has been highest in economically-distressed rural (nonmetropolitan) counties with high rates of poverty, unemployment, and disability.2 However, the share of children in high-rate counties classified as nonmetropolitan decreased from 2000 to 2016, from 23 percent to 14 percent. Again, these trends point to growing concentrations of drug-related deaths across the country—not only in economically distressed rural communities.

Although states in Appalachia have been hit hard by the opioid crisis, the drug epidemic is also widespread in parts of the Southwest, putting a large number of children in that region at risk. In four states, more than eight in 10 children live in counties with high drug overdose death rates. All of these states are in the Southwest (Arizona, Nevada, New Mexico, and Utah) (see Figure 5). In Nevada and Utah, nearly all children (92 and 95 percent, respectively) live in counties with high overdose death rates.



At the other end of the spectrum, five states (Hawaii, Iowa, Nebraska, New York, and North Dakota) have no counties with high drug overdose death rates, and thus no children living in counties with high rates. In another five states (Kansas, Minnesota, South Dakota, Texas, and Vermont), less than 5 percent of children live in high-rate counties.

We can also rank states by the number of children living in counties with high drug overdose death rates. Ten states (Florida, Arizona, Pennsylvania, Ohio, California, Massachusetts, Michigan, Utah, Kentucky, Tennessee) account for more than half (56 percent) of all children at risk, despite accounting for only 37 percent of the nation’s children (see Figure 6). Florida alone accounts for 13 percent of all U.S. children at risk, although it is home to only 6 percent of all U.S. children. The large number of unregulated pain clinics in Florida has prompted policymakers in that state to crack down on so-called pill mills that have been flooding communities with controlled prescription drugs.



California is the most populous state in the nation (home to 12 percent of U.S. children), but accounts for only 5 percent of children in high-rate counties. Texas, the nation’s second-most populous state (10 percent of U.S. children), has such a small share of children in high-rate counties that it does not appear in the top 10 states for number of children at risk. One possible explanation for Texas’s relative success may be that it is one of the few states that had systems in place for decades to track prescriptions of Schedule II drugs—those with a high potential for abuse.3 Doctors in Texas were prescribing fewer addictive painkillers, resulting in fewer drug-related deaths and fewer children at risk, relative to most other states.

How Drug Use in Families and Communities Affects Children

The rapid increase in drug overdose deaths in the United States is cause for alarm and has important implications for children’s well-being. Data from the National Survey on Drug Use and Health indicate that in the period from 2009 to 2014, more than 2 million children were living with parents who were dependent on or who abused illicit drugs.4 Children growing up with parents who are addicted to opioids face a high risk of becoming addicted to drugs themselves. A recent review of the literature showed that between 47 percent and 59 percent of children with opioid-dependent parents exhibit “substance misuse behavior.”5(Misuse behavior includes any use of a substance other than as directed, such as taking medication prescribed for another person.)

Our analysis shows that millions of other children are potentially exposed to drug addiction and drug-related deaths in their communities. Little research is available on the potential neighborhood-level effects of drug use on children’s well-being. A recent study by the Department of Health and Human Services, however, shows that counties with higher overdose death and hospitalization rates also have higher child welfare caseload rates.6 Another study shows that parents living in areas with a greater availability of drugs report higher levels of child abuse.7 Neighborhood quality—including signs of drug use on the street—has also been linked to problem behavior among children and adolescents.8

Drug-related deaths among young adults are rising. Since 2014, among young adults ages 20 to 24, the number of accidental poisonings—mostly from drug overdoses—has exceeded the number of young-adult deaths due to homicides.9 Death rates from drug overdoses and other forms of accidental poisoning among young adults increased nearly fivefold between 1999 and 2016, rising from 3.8 deaths per 100,000 to 18.9 deaths per 100,000. More than 4,200 young adults died as a result of accidental poisoning in 2016, accounting for nearly one in five (19 percent) of all young adult deaths.

Some states are taking action, implementing forward-looking policies and programs to help address the impact of the opioid epidemic on children and families. In Kentucky—a state with one of the highest drug overdose death rates—the Sobriety Treatment and Recovery (START) program provides an intensive, family-centered approach to ensure that children are reunited with their parents. Not only do women who participate in START have higher rates of sobriety, but their children are 50 percent less likely to be removed from their homes.10

More public policy attention is needed to reduce the potential impact of drug addiction and drug-related deaths on America’s children, and prevent substance abuse from being transmitted to the next generation.

Data Sources and Limitations

County-level data on the population ages 0 to 19—by race/ethnicity—are from the U.S. Census Bureau’s Population Estimates Program. Data from the Census Bureau’s Small Area Income and Poverty Estimates (SAIPE) are also used to investigate patterns of child poverty (ages 0 to 17) in counties with high/low drug overdose death rates.

Data on drug-related deaths are from the CDC’s modeled county-level database on drug overdose death rates. Rates are age-adjusted and represent the number of county-level deaths due to drug poisoning per 100,000 population. Because these data are modeled estimates, discrete values are not published; rather, data are reported for ranges of drug poisoning deaths per 100,000 population. In our analysis, counties are classified as having high drug overdose death rates if they had at least 20 deaths per 100,000 population (based on the lower limit of CDC’s range of values) in a given year.

Because available data are limited to the county level, this analysis does not account for neighborhood-level variations in death rates within counties. Within low-rate counties, there may be neighborhoods with high concentrations of drug-related deaths, and within high-rate counties there may be neighborhoods with relatively few deaths. Similarly, we lack data for a measure of within-family or within-household risk. We caution readers not to presume that all children in counties with high overdose death rates are at high risk (just as not all children in counties with low overdose death rates are at low risk).



[1]  Marjory Givens et al., 2017 County Health Rankings Key Findings Report (Madison, WI: University of Wisconsin Population Health Institute, 2017), accessed at, on Sept. 28, 2018.

[2]  Jessica Y. Ho, “The Contribution of Drug Overdose to Educational Gradients in Life Expectancy in the United States, 1992-2011,” Demography 54, no. 3 (2017): 1175-1202.

[3]  Mike Hixenbaugh, “Texas Doles Out Fewer Opioids Than Most States, but Addiction Crisis Remains,” Houston Chronicle, July 7, 2017, accessed at, on Oct. 23, 2018

[4]  Rachel N. Lipari and Struther L. Van Horn, “Children Living With Parents Who Have a Substance Use Disorder,” The CBHSQ Report, Aug. 24, 2017, accessed at, on Oct. 23, 2018.

[5]  Cory Morton and Melissa Wells, “Behavioral and Substance Use Outcomes for Older Youth Living With a Parental Opioid Misuse: A Literature Review to Inform Child Welfare Practice and Policy,” Journal of Public Child Welfare 11, no. 4-5 (2017): 546-67.

[6]  Laura Radel et al., “Substance Use, the Opioid Epidemic, and the Child Welfare System: Key Findings from a Mixed Methods Study,” ASPE Research Brief, March 7, 2018, accessed at, on Oct. 23, 2018.

[7]  Bridget Freisthler et al., “Drug Use, the Drug Environment, and Child Physical Abuse and Neglect,” Child Maltreatment 22, no. 3 (2017): 245-55.

[8]  Johns Hopkins Bloomberg School of Public Health, “A Neighborhood’s Quality Influences Children’s Behaviors Through Teens, Study Suggests” (November 2017), accessed at, on Oct. 23, 2018.

[9]  Beth Jarosz and Alicia VanOrman, “Accidental Poisoning Deaths—Mostly Drug Overdoses—Exceed Homicides of U.S. Young Adults,” Population Reference Bureau, Feb. 22, 2016, accessed at, on Oct. 23, 2018.

[10]  Commonwealth of Kentucky, Cabinet for Health and Family Services, “Sobriety Treatment and Recovery Team,” accessed at, on Oct. 23, 2018.