Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

The causal impact of older siblings’ academic achievement on younger siblings’ risk for drug use disorder : instrumental variable and propensity score analyses

Kendler, Kenneth S. ; Ohlsson, Henrik LU ; Fagan, Abigail A. ; Sundquist, Jan LU and Sundquist, Kristina LU (2026) In Psychological Medicine 56.
Abstract

Background Drug use disorder (DUD) clusters in families due partly to shared environment, including sibling influences. Low academic achievement (AA) in adolescence increases DUD risk. This study examined whether low AA in an older sibling causally increases DUD risk in younger siblings. Methods We studied all Swedish full sibling pairs (n = 309,666) born 1972–1985 and ≤ 5 years apart. Older sibling AA was assessed at age 16. Using Month-of-Birth (MoB) as an instrument, we conducted instrumental variable (IV) analyses and propensity score (PS) models to evaluate the causal impact of older sibling AA on younger sibling DUD risk, assessed by DUD registration in national medical, criminal, or pharmacy registries. Results Older sibling AA... (More)

Background Drug use disorder (DUD) clusters in families due partly to shared environment, including sibling influences. Low academic achievement (AA) in adolescence increases DUD risk. This study examined whether low AA in an older sibling causally increases DUD risk in younger siblings. Methods We studied all Swedish full sibling pairs (n = 309,666) born 1972–1985 and ≤ 5 years apart. Older sibling AA was assessed at age 16. Using Month-of-Birth (MoB) as an instrument, we conducted instrumental variable (IV) analyses and propensity score (PS) models to evaluate the causal impact of older sibling AA on younger sibling DUD risk, assessed by DUD registration in national medical, criminal, or pharmacy registries. Results Older sibling AA significantly predicted younger sibling DUD risk across models. Beta coefficients (±95% CI) were 2.04 (1.97–2.12) in raw analysis, 1.88 (0.74–3.02) in IV, and 1.26 (1.17–1.34) in PS models. Together with the strong first-stage association, the IV estimates remain positive under small departures from the ideal identifying assumptions. Effect sizes declined with increasing sibling age differences (p = 0.036 for IV; p < 0.0001 for PS) and were strongest in male–male pairs (IV: 4.01 [1.42–6.61]; PS: 1.74 [1.55–1.93]). Mediation by older sibling DUD was modest. Conclusions Findings from two causal inference approaches support a largely causal link between low AA in an older sibling and increased DUD risk in younger siblings. Stronger effects in close-aged and male–male pairs further support this conclusion. Interventions to improve AA in older siblings may yield indirect preventive benefits for younger siblings.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
academic achievement, adolescence, drug use disorder, genetic risk
in
Psychological Medicine
volume
56
article number
e77
publisher
Cambridge University Press
external identifiers
  • scopus:105034364323
  • pmid:41891148
ISSN
0033-2917
DOI
10.1017/S0033291726103729
language
English
LU publication?
yes
additional info
Publisher Copyright: © The Author(s), 2026. Published by Cambridge University Press.
id
3c898e0d-07bb-4f83-ba25-503daa6440e2
date added to LUP
2026-05-25 15:15:25
date last changed
2026-08-05 02:36:11
@article{3c898e0d-07bb-4f83-ba25-503daa6440e2,
  abstract     = {{<p>Background Drug use disorder (DUD) clusters in families due partly to shared environment, including sibling influences. Low academic achievement (AA) in adolescence increases DUD risk. This study examined whether low AA in an older sibling causally increases DUD risk in younger siblings. Methods We studied all Swedish full sibling pairs (n = 309,666) born 1972–1985 and ≤ 5 years apart. Older sibling AA was assessed at age 16. Using Month-of-Birth (MoB) as an instrument, we conducted instrumental variable (IV) analyses and propensity score (PS) models to evaluate the causal impact of older sibling AA on younger sibling DUD risk, assessed by DUD registration in national medical, criminal, or pharmacy registries. Results Older sibling AA significantly predicted younger sibling DUD risk across models. Beta coefficients (±95% CI) were 2.04 (1.97–2.12) in raw analysis, 1.88 (0.74–3.02) in IV, and 1.26 (1.17–1.34) in PS models. Together with the strong first-stage association, the IV estimates remain positive under small departures from the ideal identifying assumptions. Effect sizes declined with increasing sibling age differences (p = 0.036 for IV; p &lt; 0.0001 for PS) and were strongest in male–male pairs (IV: 4.01 [1.42–6.61]; PS: 1.74 [1.55–1.93]). Mediation by older sibling DUD was modest. Conclusions Findings from two causal inference approaches support a largely causal link between low AA in an older sibling and increased DUD risk in younger siblings. Stronger effects in close-aged and male–male pairs further support this conclusion. Interventions to improve AA in older siblings may yield indirect preventive benefits for younger siblings.</p>}},
  author       = {{Kendler, Kenneth S. and Ohlsson, Henrik and Fagan, Abigail A. and Sundquist, Jan and Sundquist, Kristina}},
  issn         = {{0033-2917}},
  keywords     = {{academic achievement; adolescence; drug use disorder; genetic risk}},
  language     = {{eng}},
  month        = {{03}},
  publisher    = {{Cambridge University Press}},
  series       = {{Psychological Medicine}},
  title        = {{The causal impact of older siblings’ academic achievement on younger siblings’ risk for drug use disorder : instrumental variable and propensity score analyses}},
  url          = {{http://dx.doi.org/10.1017/S0033291726103729}},
  doi          = {{10.1017/S0033291726103729}},
  volume       = {{56}},
  year         = {{2026}},
}