If a parent had alcohol use disorder, what are the odds you'll develop alcohol use disorder yourself?
Evidence quality 4.38/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 5/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 4/5
- D4 Uncertainty
- 4/5
- D5 Scope
- 4/5
- D6 Prose
- 4/5
- D7 Perception honesty
- 4/5
- D8 Caveat completeness
- 5/5
Lifetime probability · lifetime, subgroup
1 in 3.5
29% lifetime chance
range 1 in 4.5 to 1 in 2.0
● your factors — click this risk ▾ to reveal
≈ As likely as
Perceived
There is no rigorous public-perception survey of how adult children of alcoholics estimate their own AUD risk, but the cultural script around inherited alcoholism is strong: family memoirs, recovery literature, and pop-genetics commentary often imply something close to a coin flip. The intuitive frame compresses two distinct claims — that AUD is heritable (true: roughly half the variance is genetic) and that an individual child of an AUD parent will likely develop AUD themselves (substantially overstated). The best direct measurement (COGA) puts the elevation at roughly 2x baseline, not 5x or 10x, and the majority of adult children of alcoholics never meet criteria for the disorder. The gap between inheritance-as-mechanism and inheritance-as-destiny is where the perception calibration sits.
Rough estimate: many adult children of alcoholics believe their lifetime risk is 50% or higher
Source: editorial intuition, not polled
Actual
28.8% of first-degree relatives of an alcohol-dependent proband meet lifetime DSM-IV criteria for alcohol dependence, vs 14.4% of controls (COGA, Nurnberger et al. 2004)
First-degree relatives (parents, siblings, offspring) of alcohol-dependent probands ascertained through the Collaborative Studies on the Genetics of Alcoholism (COGA); N=8,296 relatives vs N=1,654 community controls
Show derivation
The headline figure is the directly measured lifetime DSM-IV alcohol-dependence rate in COGA's first-degree-relative pool (Nurnberger et al. 2004): 28.8% of relatives vs 14.4% of community controls assessed with the same Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) instrument — a ~2.0x relative risk. This is preferred over inferring a number by multiplying NESARC-III's 29.1% DSM-5 lifetime AUD baseline by the 2x family-history effect for three reasons. (1) Apples-to-apples instrumentation: relatives and controls in COGA were diagnosed identically. (2) DSM-IV alcohol dependence is more restrictive than DSM-5 AUD (which collapsed abuse + dependence into a single disorder with a 2-criterion floor), so the COGA figure is conservative relative to a hypothetical DSM-5 re-analysis. (3) No nationally representative study has measured DSM-5 AUD prevalence specifically in adult children of AUD parents; the COGA first-degree-relative pool is the largest direct measurement available. Important framing limitation: COGA's "first-degree relatives" pools parents, siblings, and offspring. The paper does not break results down by relationship type, so the 28.8% figure is the best available proxy for offspring-specific risk, not a direct measurement of it. Adult children of AUD parents are likely near this figure given (a) shared genetic loading equivalent to siblings and (b) shared early-life environmental exposure; the COGA estimate is anchored to a population that includes them but is not exclusively them. Heritability estimates from the Verhulst, Neale & Kendler (2015) meta-analysis of 12 twin and 5 adoption studies converge on h² = 0.49 (95% CI 0.43-0.53), supporting the magnitude of the elevation as part-genetic. The uncertainty band reflects cohort-definition heterogeneity (one parent vs both, biological vs adopted-out studies), DSM edition (DSM-IV ~28% in COGA vs likely higher under DSM-5 criteria), and the offspring-specific vs first-degree-relative-pooled distinction.
Caveats: Three framing limitations matter for reading this number correctly. (1) Populati…
Three framing limitations matter for reading this number correctly. (1) Population mismatch: COGA's "first-degree relatives" pools parents, siblings, and offspring of alcohol-dependent probands. The 28.8% figure is the best direct measurement available for offspring-specific risk because the genetic loading and early-environment exposure are comparable across these relationships, but it is not a study of adult children of AUD parents exclusively. (2) Diagnostic edition: COGA used DSM-IV alcohol dependence, which is more restrictive than DSM-5 AUD (DSM-5 collapsed the earlier "abuse" and "dependence" categories into a single disorder with a 2-criterion floor). A hypothetical DSM-5 re-analysis of COGA would likely yield higher absolute prevalences in both relatives and controls, but the 2x ratio is expected to hold. (3) Ascertainment: COGA probands were recruited through treatment programs, which selects for severe and clinically identified cases; the relative risk for offspring of a parent with mild, never-treated AUD is likely smaller. The 28.8% figure should not be read as a forecast for any individual — most adult children of alcoholic parents (about 71%) do not meet lifetime criteria for alcohol dependence. Heritability is not destiny: the Verhulst meta-analysis h² of 0.49 means roughly half the variance in liability is genetic, leaving substantial room for environmental, behavioral, and policy factors to shift individual outcomes. Adoption studies have separately established that the elevation persists in offspring of AUD biological parents even when raised by non-alcohol-using adoptive families, confirming that the genetic component is real and not purely a learned behavior, but those same studies also show that the elevation is substantially attenuated in protective environments.
Risks at similar odds
Other risks with roughly the same likelihood — useful for calibration.
Major depression
What are the odds of experiencing a major depressive episode in your lifetime?
Family caregiver probability
How likely am I to become an unpaid family caregiver — and what is the mental-health toll?
Loneliness & health
What are the odds of chronic loneliness causing serious health harm over a lifetime?
Menopause CV risk acceleration
How likely is a woman to develop cardiovascular disease after menopause — and how much does menopause accelerate that risk?
Chronic painkillers
What are the odds of being harmed by taking over-the-counter painkillers regularly?
Recently viewed on this device
Stored locally — clear anytime.
Pick challenger
The clearest direct measurement comes from the Collaborative Studies on the Genetics of Alcoholism (COGA), reported by Nurnberger and colleagues in 2004. COGA assessed 8,296 first-degree relatives of alcohol-dependent probands and 1,654 community controls with the same Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) interview. Lifetime DSM-IV alcohol dependence appeared in 28.8% of relatives and 14.4% of controls — a roughly 2-fold elevation. That is the cleanest apples-to-apples comparison in the literature: same diagnostic instrument, same threshold, same era, families recruited from the same multi-site framework. The convergence with NESARC-III’s 29.1% general-population lifetime DSM-5 AUD figure is a coincidence of definitions, not a finding — DSM-5 AUD is a substantially broader category than DSM-IV alcohol dependence, so the absolute numbers happen to land near each other despite measuring different things in different populations. The signal to anchor on is the 2x ratio within COGA, not the absolute 28.8% read against the unrelated NESARC-III baseline.
That 2x elevation is part-genetic and part-environmental, and twin and adoption studies can separate the two. The Verhulst, Neale and Kendler 2015 meta-analysis of 12 twin studies and 5 adoption studies put the heritability of AUD at h² = 0.49 (95% CI 0.43-0.53), with no evidence of heterogeneity by sex. Roughly half the variance in liability is genetic. Adoption studies — which follow biological children of alcohol-dependent parents raised by non-related adoptive families — have separately shown that the elevation persists even when the early-life drinking environment is removed, confirming that the genetic component is real rather than purely a modeled behavior. The remaining ~50% of variance is environmental, which is why protective adoptive environments and policy-level interventions (drinking-onset delay, monitored adolescent alcohol access, screening in primary care) can meaningfully shift outcomes even for high-genetic-liability individuals.
The framing this number resists is “inheritance as destiny.” About 71% of COGA first-degree relatives of alcohol-dependent probands did not meet lifetime criteria for alcohol dependence. That is the modal outcome, not the exception. The cultural script around “alcoholism runs in families” often implies something closer to 50% or higher lifetime risk in adult children of alcoholics; the direct measurement is closer to 29%. The elevation is real and substantial — a 2x relative risk for a disorder with a 14% baseline is one of the larger first-degree-relative effects in psychiatric epidemiology — but it operates within a probabilistic, not deterministic, frame. Most adult children of alcoholic parents reach midlife without developing AUD themselves.
Three caveats worth foregrounding. COGA’s first-degree-relative pool combines parents, siblings, and offspring of probands; the paper does not break results out by relationship type, so the 28.8% figure is the best available proxy for offspring-specific risk rather than a direct measurement of it. DSM-IV alcohol dependence is a higher bar than DSM-5 AUD, so a re-analysis under current diagnostic criteria would likely yield higher absolute figures in both groups while preserving the 2x ratio. And COGA recruited probands through treatment programs, which selects for clinically severe cases — the elevation in the children of an undiagnosed or mild AUD parent is likely smaller than the COGA figure suggests.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
-
[1] Nurnberger JI Jr, Wiegand R, Bucholz K, et al. — Archives of General Psychiatry, 2004 — A family study of alcohol dependence: coaggregation of multiple disorders in relatives of alcohol-dependent probands
A family study of alcohol dependence: coaggregation of multiple disorders in relatives of alcohol-dependent probands- Statistic
Lifetime risk rates of 28.8% and 14.4% for DSM-IV alcohol dependence in first-degree relatives of alcohol-dependent probands and controls, respectively; relative risk approximately 2-fold- Excerpt
“"Lifetime risk rates of 28.8% and 14.4% for DSM-IV alcohol dependence in relatives of probands and controls, respectively... The risk of alcohol dependence in relatives of probands compared with controls is increased about 2-fold." ”
- Source data from
- 2004-12-01
- Accessed
- 2026-05-28 · archived copy
- Calculation
- Native numerator (28.8 per 100) is taken directly from the Nurnberger 2004 finding for first-degree relatives of alcohol-dependent probands (parents, siblings, offspring). The 14.4% control rate is the matched same-instrument baseline, yielding a ~2.0x relative risk. Sample sizes: N=8,296 relatives of probands and N=1,654 controls assessed via the SSAGA structured diagnostic interview. DSM-IV alcohol-dependence criteria were applied; the broader DSM-5 AUD criteria would likely yield somewhat higher absolute prevalences in both groups while preserving the ~2x ratio. The 28.8% figure is used as the lifetime_us_adult for this subgroup; this is the best direct measurement available because no nationally representative study has measured DSM-5 AUD specifically in offspring of AUD parents. The first-degree-relative pool includes adult children of probands but is not exclusively offspring; this is acknowledged in `assumptions` and `caveats`.
- Independence
- The Collaborative Studies on the Genetics of Alcoholism (COGA) is a NIAAA-funded multi-site project (Indiana University, Washington University in St. Louis, SUNY Downstate, University of Connecticut, University of California San Diego, University of Iowa) that ascertained alcohol-dependent probands through inpatient and outpatient treatment programs and recruited their relatives plus a matched community-control sample. Probands and controls were diagnosed with the same SSAGA instrument, making the relative-vs-control comparison methodologically clean.
-
[2] Verhulst B, Neale MC, Kendler KS — Psychological Medicine, 2015 — The heritability of alcohol use disorders: a meta-analysis of twin and adoption studies
The heritability of alcohol use disorders: a meta-analysis of twin and adoption studies- Statistic
Best-fit estimate of the heritability of AUD: h² = 0.49 (95% CI 0.43-0.53); meta-analysis of 12 twin studies and 5 adoption studies; no evidence for heterogeneity by sex- Excerpt
“"The best-fit estimate of the heritability of AUD was 0.49 [95% confidence interval (CI) 0.43-0.53]... There was no evidence for heterogeneity by study design, sex or assessment method." ”
- Source data from
- 2015-03-01
- Accessed
- 2026-05-28 · archived copy
- Calculation
- Establishes that roughly half of the variance in AUD risk is heritable, consistent with the magnitude of the COGA family-aggregation finding being part-genetic rather than purely shared-environmental. Used as supporting evidence for the mechanism behind the 2x elevation; not used to compute the headline rate. Twin and adoption studies separately estimate the genetic component because identical twins reared apart and adoptees raised by non-biological parents disentangle shared genes from shared environment; the meta-analytic h² = 0.49 means that even substantial early-life environmental exposure to a parent's drinking does not, on its own, drive the elevated risk seen in offspring.
- Independence
- This is a meta-analysis of independent twin and adoption studies conducted across multiple countries and decades, including the Virginia Twin Registry, Swedish adoption studies, and Australian twin samples. The aggregate h² estimate is methodologically independent of the COGA family-aggregation data.
-
[3] Grant BF, Chou SP, Saha TD, et al. — JAMA Psychiatry, 2015 — Epidemiology of DSM-5 Alcohol Use Disorder: Results From the National Epidemiologic Survey on Alcohol and Related Conditions III
Epidemiology of DSM-5 Alcohol Use Disorder: Results From the National Epidemiologic Survey on Alcohol and Related Conditions IIISee all 2 Likelier entries citing this source →
- Statistic
Lifetime prevalence of DSM-5 alcohol use disorder among US adults: 29.1% overall; 12-month prevalence 13.9% (NESARC-III, N=36,309)- Excerpt
“"In 2012-2013, US prevalences of DSM-5 12-month and lifetime AUD among adults 18 years and older were 13.9% and 29.1%, respectively. Prevalence was generally highest for men (17.6% and 36.0%, respectively), White and Native American respondents, and younger and never-married adults." ”
- Source data from
- 2015-08-01
- Accessed
- 2026-05-28 · archived copy
- Calculation
- Used as the US adult general-population reference for context. The coincidence between NESARC-III lifetime DSM-5 AUD (29.1%) and the COGA first-degree-relative DSM-IV alcohol-dependence figure (28.8%) is a methodological artifact, not a finding: DSM-5 AUD has a much broader definition than DSM-IV alcohol dependence, so the absolute numbers happen to land near each other despite measuring different things in different populations. The clean apples-to-apples comparison is COGA relatives 28.8% vs COGA controls 14.4%, both DSM-IV alcohol dependence, both SSAGA instrument — that is where the 2x elevation comes from.
- Independence
- NESARC-III was conducted by the National Institute on Alcohol Abuse and Alcoholism (NIAAA) using probability sampling of the US non-institutionalized civilian population and the AUDADIS-5 structured diagnostic interview. Methodologically independent from COGA (which is a treatment-ascertained family study) and from the heritability meta-analysis.
-
[4] Grant BF — Alcohol Health and Research World, 1998 — The impact of a family history of alcoholism on the relationship between age at onset of alcohol use and DSM-IV alcohol dependence: results from the National Longitudinal Alcohol Epidemiologic Survey
The impact of a family history of alcoholism on the relationship between age at onset of alcohol use and DSM-IV alcohol dependence: results from the National Longitudinal Alcohol Epidemiologic Survey- Statistic
Family history of alcoholism is associated with higher lifetime DSM-IV alcohol-dependence prevalence; earlier age at first drink is independently associated with higher lifetime risk (NLAES, nationally representative)- Excerpt
“"People with a family history of alcoholism had a higher prevalence of lifetime alcohol dependence than did people without such a history. Respondents with an earlier age of drinking onset were more likely to become alcohol dependent compared with respondents with a later age of drinking onset." ”
- Source data from
- 1998-01-01
- Accessed
- 2026-05-28 · archived copy
- Calculation
- Provides nationally representative cross-validation that family history of alcoholism elevates lifetime alcohol-dependence prevalence in the respondent, consistent with the direction of the COGA finding. Also establishes that early age at first drink interacts with family history to compound risk — supporting the framing that the elevation is probabilistic and modifiable by behavior, not deterministic. Not used to derive the headline rate; the NLAES instrument and sampling differ from COGA/SSAGA, so the figures are not directly substitutable.
- Independence
- The National Longitudinal Alcohol Epidemiologic Survey (NLAES, 1991-1992) was conducted by NIAAA using the AUDADIS instrument, independent of COGA's SSAGA-based family study.