Depression is the leading cause of disability worldwide, affecting millions of people every year. While treatments like therapy and medication can help, they don’t work for everyone, and many people struggle to access them. Increasingly, researchers are asking a different question: what if we could prevent depression before it takes hold? One promising way forward is to identify and tackle the factors that put people at higher risk in the first place.
One of the clearest candidates is childhood maltreatment. Experiences such as abuse and neglect can leave deep and lasting marks, shaping both physical and mental health well into adulthood. Years of research show that people who were maltreated as children are more likely to struggle with depression, and when they do, their symptoms often start earlier, last longer, and come back more frequently (Nanni et al., 2012; see Camille’s recent Mental Elf blog on maltreatment and multimorbidity here). They are also more likely to face other mental health difficulties alongside depression.
Crucially, childhood maltreatment is not inevitable. It can be reduced through policy, prevention programmes, and support for families. Modelling work in Australia suggests that eliminating maltreatment could prevent over a million cases of depression and other disorders (Grummitt et al., 2024). Against this backdrop, Watson and colleagues (2025) set out to provide the most comprehensive summary yet, of how strongly maltreatment in childhood is linked to depression later in life, with the aim of determining whether prevention of child maltreatment may be theoretically feasible to reduce the prevalence of depression worldwide.
With depression affecting millions globally, researchers are increasingly shifting focus from treatment to prevention; finding ways to tackle the factors that put people at risk before illness takes hold.
Methods
The authors systematically searched three databases for peer-reviewed English-language studies that examined whether experiencing maltreatment before age 18 was linked to depression in adulthood. Data extraction was carried out by a single author. Where possible, the team focused on extracting results that adjusted for other factors that might influence depression.
A random-effects meta-analysis was used to combine results on the association between child maltreatment and depression across studies. The authors also examined how different screening methods for maltreatment and depression affected findings, and ran tests for potential bias and study quality (assessed using the Newcastle-Ottawa scale). Studies were included in the meta-analysis that either reported odds ratios (ORs) or provided enough information for the authors to calculate them.
Results
Study characteristics
The authors identified 77 studies that collectively included over half a million participants (n = 516,302). Nearly half of these studies were conducted in the United States (40%), meaning American samples were overly represented. Most studies adjusted for sex, age, and education, but other confounders like income, socioeconomic status, and ethnicity were less common, and no studies adjusted for genetic factors.
Most studies (n = 66) relied on adult recollection of childhood experiences, while nine used structured clinical interviews and two drew on prospective child protection records. Depression was most commonly assessed using self-report questionnaires (n = 49), with diagnostic interviews used less often (n = 28).
Meta-analysis
The main meta-analysis pooled 87 effect sizes from 44 studies. The findings were striking: people who experienced childhood maltreatment were about two and a half times more likely to have depression as adults (OR = 2.49, 95% CI [2.25 to 2.76]). This means that maltreatment in childhood substantially increases the risk of depression later in life.
Heterogeneity (i.e., variation between studies) was very high (I² = 93.5%), which means the strength of this association varied across studies. However, no single study was driving the result, and when only the highest-quality studies were included, the association remained strong (OR = 2.36, 95% CI [2.06 to 2.71]).
When looking at differences in measurement, depression identified by questionnaires showed slightly stronger associations with maltreatment (OR = 2.60) compared to depression diagnosed via clinical interviews (OR = 2.34), though the authors did not state whether this difference was statistically significant. Similarly, maltreatment reported through questionnaires gave higher odds of depression (OR = 2.58) compared to clinical interviews (OR = 2.13), but this difference was not significant. Subgroup analyses by specific tools did not reveal any further notable differences.
According to this review, people who experienced childhood maltreatment were more than twice as likely to develop depression in adulthood, highlighting the lasting effects of early adversity.
Conclusions
The authors conclude that childhood maltreatment is strongly and consistently associated with adult depression, with affected individuals facing more than double the risk compared to those without maltreatment histories.
This pattern held even when accounting for study quality and potential publication biases, reinforcing the robustness of the findings.
Although the association remained regardless of the screening measure used, self-report questionnaires for both maltreatment and depression showed consistently stronger associations. Overall, the authors conclude that,
there is strong evidence suggesting that the experience of CM [child maltreatment] may be one of the many factors contributing to the development of depression in adulthood. As a preventable phenomenon, CM may play an important role among these contributors as a modifiable risk factor for depression.
The association between childhood maltreatment and adult depression held across different screening methods, and after accounting for study quality and publication bias.
Strengths and limitations
This review extends and updates previous meta-analyses by comparing different ways of measuring both childhood maltreatment and depression. Importantly, the results were robust to publication bias and study quality checks, which strengthens confidence in the overall pattern. The review also followed PRISMA guidelines and supplemented database searches with hand-searching of reference lists, increasing the likelihood that relevant studies were comprehensively identified. However, there are several important limitations to bear in mind.
Firstly, there are several sources of bias that could be shaping the association between childhood maltreatment and depression, including recall bias (adults asked to recall childhood experiences of maltreatment, measured at the same time as depression) and common method bias (self-report measures). Neither of these sources of bias were discussed in the paper. The authors also describe nine studies as “prospective” (pp.22) because they used clinical interviews, but without further elaboration on the timing or temporal sequence of these assessments, only the two studies using child protection records can be considered genuinely prospective. Greater transparency about the timing of maltreatment and depression assessments would have helped clarify these issues and reduce ambiguity.
Only including English language studies is likely to have missed a significant body of research that could have impacted on the findings of this review.
Another limitation lies in the statistical approach. The authors report multiple effect sizes from single studies that are therefore not independent (e.g., separate ORs for each Adverse Childhood Experiences score within the same dataset). However, they did not use a multilevel meta-analysis model, which is designed to account for the statistical dependencies between such effect sizes. Failing to do so may have led to inflated precision of the overall results. Additionally, while the authors extracted data on individual subtypes of maltreatment, these were not included in the meta-analysis; a multilevel model could have enabled this, leading to an overall richer picture.
A final limitation acknowledged by the authors concerns the conclusions drawn about the size of the association between childhood maltreatment and depression, given how studies handled confounding factors. Although many adjusted for basic demographics, very few accounted for more complex confounds related to environmental or genetic influences. A meta-analysis of studies using quasi-experimental designs (Baldwin et al., 2021) has shown that when genetic and environmental confounds are considered, the association between maltreatment and mental health is reduced by nearly half. This suggests that a substantial part of the observed association may reflect pre-existing vulnerabilities rather than a direct effect of maltreatment. If this is the case, reducing maltreatment alone may not lead to the large reductions in depression that the size of the association may initially suggest.
Taken together, these limitations raise questions about how far this meta-analysis can deliver on its stated aim of testing whether preventing childhood maltreatment would be a viable way to reduce global prevalence of depression. Importantly, this reflects limitations in the underlying evidence base rather than in the meta-analysis itself. While there is no doubt that maltreatment is harmful and preventable, the exact size of its impact on depression is likely to be smaller and more complex than the pooled odds ratios here suggest.
Childhood maltreatment is undoubtably damaging and preventable, but its influence on depression may be smaller and more complicated than this meta-analysis suggests.
Implications for practice
The findings from this meta-analysis reinforce previous evidence showing a strong link between childhood maltreatment and adult depression, highlighting maltreatment as an important potentially modifiable risk factor. Even if part of this link reflects genetic and environmental confounds, there remains a clear moral obligation to reduce the likelihood that children will experience maltreatment, and research that highlights the importance of prevention is therefore crucial.
Recognising that some of the association may be explained by the wider environments in which maltreatment occurs also does not contradict the case for prevention; instead, it suggests that prevention strategies may be most effective when they also target these wider conditions. Tackling the structural conditions that make families vulnerable to maltreatment, such as poverty, parental stress or housing instability, may have particularly wide-reaching benefits by simultaneously reducing maltreatment and other co-occurring risks for depression at the same time. Policies that reduce poverty and financial strain, or that strengthen family and community support systems, are promising examples. Not only do these policies reduce the likelihood of childhood maltreatment but also help to address the broader constellation of co-occurring risks that contribute to depression. By focusing on improving the social and economic conditions in which families live, these policies and interventions have the potential to deliver wide reaching mental health benefits in addition to fostering the kind of safe, supportive environments that every child deserves to grow up in.
More research is clearly needed to understand how much of the link between childhood maltreatment and later mental health problems is actually causal. To do this, studies must make use of designs that can help disentangle cause from background vulnerability – for example through quasi-experimental of genetically informed approaches. Future research also must take better account of the environments in which maltreatment occurs, by accounting for factors such as socioeconomic disadvantage, parental mental health difficulties, and family conflict, that can increase the likelihood of maltreatment and the risk of depression. Improving causal evidence in this way is essential for guiding prevention efforts and ensuring that policies and interventions target the most effective and modifiable pathways to better mental health.
Prevention may be most effective when it targets not only maltreatment itself, but also the wider social conditions – like poverty and family stress – that increase the risk of both maltreatment and depression.
Statement of interests
None declared.
Links
Primary paper
Watson, C. B., Sharpley, C. F., Bitsika, V., Evans, I., & Vessey, K. (2025). A Systematic Review and Meta‐Analysis of the Association Between Childhood Maltreatment and Adult Depression. Acta Psychiatrica Scandinavica, 151(5), 572-599. https://doi.org/10.1111/acps.13794
Other references
Baldwin, J. R., Wang, B., Karwatowska, L., Schoeler, T., Tsaligopoulou, A., Munafò, M. R., & Pingault, J. B. (2023). Childhood maltreatment and mental health problems: A systematic review and meta-analysis of quasi-experimental studies. American Journal of Psychiatry, 180(2), 117-126. https://doi.org/10.1176/appi.ajp.20220174
Grummitt, L., Baldwin, J. R., Lafoa’i, J., Keyes, K. M., & Barrett, E. L. (2024). Burden of mental disorders and suicide attributable to childhood maltreatment. JAMA Psychiatry, 81(8), 782-788. https://doi.org/10.1001/jamapsychiatry.2024.0804
Nanni, V., Uher, R., & Danese, A. (2012). Childhood maltreatment predicts unfavorable course of illness and treatment outcome in depression: a meta-analysis. American Journal of Psychiatry, 169(2), 141-151. https://doi.org/10.1176/appi.ajp.2011.11020335
Souama, C. (2025). Under the skin: How childhood maltreatment may trigger lifelong multimorbidity. The Mental Elf.
