Genetic Risk Scores for Cannabis Disorder Did Not Predict When African American Youth Started Using | Research

The takeaway

In 1,017 African American youth, polygenic risk scores for cannabis use disorder did not predict age of cannabis initiation, though parental monitoring interacted with genetic risk in an unexpected pattern.

behavioral geneticists, substance use prevention researchers, health equity researchers

What the researchers found

Cannabis use disorder and nicotine dependence polygenic risk scores were not associated with initiation of these substances. Unexpectedly, higher alcohol use disorder PRS predicted later alcohol initiation (HR 0.78). For cannabis, high parental monitoring was associated with earlier initiation among those with high genetic risk, while low monitoring predicted earlier initiation among those with low genetic risk.

Why it matters

Most genetic risk studies are conducted in European-ancestry populations. This study provides critical data on whether polygenic risk scores derived from current research predict substance use in African American youth, where they largely did not.

The numbers in context

1,017 participants, 56% female. CUD PRS: not significant for cannabis initiation. Nicotine PRS: not significant. AUD PRS: HR 0.78 (later alcohol initiation). CUD PRS x monitoring interaction: high PRS + high monitoring = earlier cannabis; low PRS + low monitoring = earlier cannabis.

How the study worked

Longitudinal study of 1,017 African American participants (56% female) from an elementary school prevention trial. At age 20, substance initiation ages were reported and genetic samples collected. At age 12, parental monitoring and census-tract disadvantage were measured. Cox Proportional Hazard Models tested PRS and environment interactions.

What this study cannot tell us

PRS derived largely from European-ancestry GWAS may perform poorly in African American populations. Self-reported initiation ages subject to recall bias. Parental monitoring measured at age 12 may not capture later changes. Single geographic region.

How to read the evidence

Longitudinal design with adequate sample and genetic data, but PRS ancestry limitations and self-reported outcomes place at moderate.

When this study was published

Longitudinal data from elementary school through young adulthood.

The bigger picture

The null findings for CUD and nicotine PRS suggest current genetic risk scores may not transfer well across ancestry groups, highlighting the need for more diverse GWAS studies. The unexpected monitoring interaction challenges simple models of genetic and environmental risk.

Questions still open

Would African American-derived PRS better predict initiation?
Why does high monitoring predict earlier initiation in genetically at-risk youth?

Common questions

Can genetics predict who will start using cannabis?

In this study of African American youth, current genetic risk scores for cannabis use disorder did not predict when young people started using cannabis. This may reflect limitations in how risk scores are developed across different populations.

Does parental monitoring prevent cannabis use?

The relationship was more complex than expected. Among youth with high genetic risk, higher monitoring was actually associated with earlier cannabis initiation, suggesting monitoring alone may not be sufficient for genetically at-risk youth.

Read the original research

Polygenic Risk for Substance Use Disorders as Predictors of Substance Use Initiation Among African American Youth.

Journal of studies on alcohol and drugs, 86(4), 530-541

View on PubMed ↗Publisher’s paper ↗

Citation

Sosnowski, David W; Rabinowitz, Jill A; Feder, Kenneth A; Strickland, Justin C; Hancock, Dana B; Uhl, George R; Ialongo, Nicholas S; Maher, Brion S. (2025). Polygenic Risk for Substance Use Disorders as Predictors of Substance Use Initiation Among African American Youth.. Journal of studies on alcohol and drugs, 86(4), 530-541. https://doi.org/10.15288/jsad.23-00397