Curated analytic data and Python code for childhood adversity, adult threatening life events, symptom burden and maladjustment during substance-related rehabilitation

Published: 29 September 2026| Version 1 | DOI: 10.17632/s5mrcmztc7.1
Contributor:
Diego Rivera-Porras

Description

This deposit contains the curated de-identified analytic dataset and Python code supporting the cross-sectional analyses reported in “Childhood adversity, adult threatening life events, symptom burden and maladjustment during substance-related rehabilitation: a cross-sectional study”. The source study included 88 adults receiving substance-related rehabilitation care. MI, LTE and SCL-90-R variables are available for 88 participants; complete-domain ACE-B and ACE-F scores are available for 84. Primary models use ordinary least-squares regression with HC3 robust standard errors and adjust for LTE, age, prior psychiatric history, current employment and categorical programme phase. Direct identifiers and the private source-row linkage have been removed. Because individual-level health, substance-use and adversity variables remain sensitive, the files should be deposited using Restricted Access / Restrict access to files.

Files

Institutions

Categories

Psychology, Public Health, Clinical Psychiatry, Substance Use Epidemiology

Licence