Scholarlink

Background <p>Early life adversity (ELA) is associated with increased internalizing and externalizing behavior in high-income cohorts. However, it is unclear whether these associations generalize to low- and middle-income countries (LMICs). We examined the association between ELA and internalizing and externalizing behaviors in children across three LMIC cohorts.</p> Methods <p>Data from three cohorts, the Brazilian High-Risk Cohort (BHRC; <i>n</i> = 1111), Sustainable Programme Incorporating Nutrition and Games (SPRING; <i>n</i> = 601), and Drakenstein Child Health Study (DCHS; <i>n</i> = 708) were pooled. Children (7–10 years) were assessed using the Strengths and Difficulties Questionnaire (SDQ). ELA was measured using a cumulative index. Ordinary Least Squares linear regression models were run to examine the associations of adversity with internalizing and externalizing behaviors while adjusting for child age, sex, and cohort, and to test interactions with child age, sex, and cohort. 2420 children (53.9% boys) were included in analyses.</p> Results <p>SDQ measurement invariance was not established across cohorts. ELA associations were strongest in BHRC (SDQ total: B = 6.78, <i>p</i> &lt; .001; internalizing: B = 3.3, <i>p</i> &lt; .001; externalizing: B = 3.48, <i>p</i> &lt; .001), attenuated in SPRING (SDQ total: B = 1.54, <i>p</i> = .014; internalizing: B = 0.81, <i>p</i> &lt; .001; externalizing: B = 0.73, <i>p</i> = .007), and not observed in DCHS (SDQ total: B = 0.12, <i>p</i> = .78; internalizing: B = -0.18, <i>p</i> = .42; externalizing: B = 0.3, <i>p</i> = .29). Interaction analyses indicated cohort heterogeneity and age-related effects, with inconsistent sex moderation across cohorts and outcomes.</p> Conclusions <p>In Global South cohorts, while there is an association between ELA and psychopathology, there are also important context-dependent differences across cohorts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Association of early life adversity with internalizing and externalizing behavior across 3 low- and middle-income countries

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月24日

查看全文
  • Sheri-Michelle Koopowitz,
  • Richard Shadwell,
  • Nadia Hoffman,
  • Heather J. Zar,
  • Giovanni Salum,
  • Pedro M. Pan,
  • Gauri Divan,
  • Supriya Bhavnani,
  • Dan J. Stein

摘要

Background

Early life adversity (ELA) is associated with increased internalizing and externalizing behavior in high-income cohorts. However, it is unclear whether these associations generalize to low- and middle-income countries (LMICs). We examined the association between ELA and internalizing and externalizing behaviors in children across three LMIC cohorts.

Methods

Data from three cohorts, the Brazilian High-Risk Cohort (BHRC; n = 1111), Sustainable Programme Incorporating Nutrition and Games (SPRING; n = 601), and Drakenstein Child Health Study (DCHS; n = 708) were pooled. Children (7–10 years) were assessed using the Strengths and Difficulties Questionnaire (SDQ). ELA was measured using a cumulative index. Ordinary Least Squares linear regression models were run to examine the associations of adversity with internalizing and externalizing behaviors while adjusting for child age, sex, and cohort, and to test interactions with child age, sex, and cohort. 2420 children (53.9% boys) were included in analyses.

Results

SDQ measurement invariance was not established across cohorts. ELA associations were strongest in BHRC (SDQ total: B = 6.78, p < .001; internalizing: B = 3.3, p < .001; externalizing: B = 3.48, p < .001), attenuated in SPRING (SDQ total: B = 1.54, p = .014; internalizing: B = 0.81, p < .001; externalizing: B = 0.73, p = .007), and not observed in DCHS (SDQ total: B = 0.12, p = .78; internalizing: B = -0.18, p = .42; externalizing: B = 0.3, p = .29). Interaction analyses indicated cohort heterogeneity and age-related effects, with inconsistent sex moderation across cohorts and outcomes.

Conclusions

In Global South cohorts, while there is an association between ELA and psychopathology, there are also important context-dependent differences across cohorts.