We conducted a thorough systematic review and meta-analysis, adhering to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) 2020 [20] (Supplementary File 1). The study was registered in the international prospective register of systematic reviews in health and social care (PROSPERO) database (registration number: CRD42024532631). The registration ensures transparency and minimizes bias by documenting the review’s protocol before data collection and analysis.
Search strategy
Our study involved a thorough and extensive search for publications focusing on post-traumatic stress disorder (PTSD), with a particular emphasis on the pediatric population. We conducted this search across multiple prominent databases, including PubMed, Embase, Scopus, Science Direct, and the search engines, Google Scholar and Google. The search encompassed articles published from 2014 until May 15, 2024.
To refine our search and improve its precision, we employed relevant MeSH headings and carefully selected keywords that were specifically related to PTSD in pediatrics. The literature search was conducted across multiple prominent databases, including PubMed, Embase, Scopus, Science Direct, as well as the search engines Google Scholar and Google. A comprehensive set of MeSH terms and keywords were used to capture all relevant studies. This included terms related to post-traumatic stress disorder, such as “Acute Post-Traumatic Stress Disorder”, “Chronic Post-Traumatic Stress Disorder”, “Delayed Onset Post-Traumatic Stress Disorder”, “Moral Injury”, “Neuroses, Post-Traumatic”, “Neuroses, Posttraumatic”, “PTSD”, “Post Traumatic Stress Disorder”, “Post-Traumatic Stress Disorders”, “Posttraumatic Stress Disorders”, “Stress Disorder, Post Traumatic”, and “Stress Disorders, Posttraumatic”.
To capture information on prevalence and associated factors, additional terms were included, such as “Prevalence”, “Epidemiology”, “Risk Factors”, “Incidence”, and “Associated Factors”.
These PTSD-related and prevalence/associated factor terms were combined with terms for the population of interest, including “Pediatric”, “Pediatrics”, “Paediatric”, “Children”, “Adolescents”, “Child”, and “Kids”.
Finally, a comprehensive set of geographic terms was used to capture studies from across the African continent, including countries such as Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cape Verde, Cameroon, Central African Republic, Chad, Comoros, Congo Democratic Republic, Congo, Cote d’Ivoire, Djibouti, Egypt, Equatorial Guinea, Eritrea, Eswatini, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya, Lesotho, Liberia, Libya, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique, Namibia, Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, Somalia, South Africa, South Sudan, Sudan, Tanzania, Togo, Tunisia, Uganda, Zambia, and Zimbabwe.
By utilizing this approach, we aimed to optimize the retrieval of articles that were most relevant to our study objectives. Through this comprehensive and targeted search strategy, we intended to gather a robust collection of publications that would provide valuable insights into the understanding and management of PTSD in the pediatric population. In addition to electronic searches, we examined the reference lists of all included primary studies. This step was crucial to ensure that no relevant studies were inadvertently missed. In cases where we encountered articles that were inaccessible, we proactively reached out to the authors via email to request the full texts. When we encountered inaccessible articles, we contacted the authors directly to request the full texts. For two articles that met our inclusion criteria, we successfully obtained the full texts from the corresponding authors.
Eligibility criteria
Inclusion criteria for this systematic review and meta-analysis were: (1) studies of individuals under the age of 18 years, (2) studies that reported the prevalence of PTSD as defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria, and (3) studies published between 2014 and May 15, 2024. To ensure the relevance and currency of our findings, we restricted our literature search to the past 10 years. This decision was based on several factors: the types and severity of traumatic events experienced by children, such as natural disasters, armed conflicts, and abuse, have varied over the past decade, impacting PTSD prevalence. Additionally, significant advancements in the understanding, screening, diagnosis, and treatment of PTSD in children and adolescents have occurred, potentially influencing reported prevalence rates. Furthermore, societal attitudes towards mental health and the availability of mental health resources for children and adolescents in Africa have evolved, affecting the identification and reporting of PTSD cases. By focusing on the most recent 10-year period, we aim to provide the most relevant and reliable estimates of PTSD prevalence and associated factors in pediatric populations.
We excluded studies conducted in languages other than English. Additionally, we omitted case reports, correspondence, reviews, editorials, and duplicate studies.
Outcomes of the study
The primary outcome of interest for this study was the prevalence of post-traumatic stress disorders (PTSD) in pediatrics. Additionally, we investigated the factors associated with PTSD as secondary outcomes.
Screening of articles
Duplicates were first removed from all citations identified through our search strategy. Two authors (TTT and BY) independently screened the titles, abstracts, and full texts to identify eligible studies. Any discrepancies were discussed and resolved by consensus. The Rayyan programme [21] was used for screening of articles.
Quality assessments
The methodological quality and risk of bias of the included studies were evaluated using the Newcastle-Ottawa Scale, a tool specifically designed for assessing the quality of cross-sectional studies [22]. This scale provides a comprehensive assessment of study selection, comparability of groups, and the ascertainment of the outcome of interest. The Newcastle-Ottawa Scale assigns scores ranging from 0 to 10 points, with the following categories: Very Good Studies (9–10 points), Good Studies (7–8 points), Satisfactory Studies (5–6 points), and Unsatisfactory Studies (0–4 points) (Supplementary File 2).
Data extraction
Data were extracted from included articles using a piloted form. We collected information about the characteristics of studies (author, country, design, year of publication, sample size, prevalence of PTSD and predictors). The data were extracted independently by two reviewers (TTT and BY).
Statistical analysis
The meta-analysis was conducted using Comprehensive Meta-Analysis software, version 3. A random-effects model was used to estimate the pooled proportion of PTSD and predictors in pediatric population in Africa. The summary effect estimates and 95% confidence intervals (CI) for the outcomes were represented with a forest plot. Heterogeneity between studies was examined using Cochran’s Q test and quantitatively measured by the index of heterogeneity squared (I2) statistics and corresponding 95% confidence intervals (CI) [23]. Heterogeneity was considered low, moderate or high when I2 values were below 25%, between 25% and 75%, and above 75%, respectively [23]. To evaluate the stability and reliability of our results, we performed a comprehensive sensitivity analysis and the detailed findings from this analysis, including any significant changes or consistencies in the results, are thoroughly documented. The sources of heterogeneity were explored through subgroup analysis using study characteristics as covariates. Potential publication bias was assessed using funnel plots and asymmetry was evaluated with Egger’s method.
Meta-Regression Analysis Considerations: According to the Cochrane Handbook for Systematic Reviews and Meta-Analyses, meta-regression is appropriate when there is substantial heterogeneity between studies and a sufficient number of studies to reliably estimate the effects of potential moderator variables [24]. However, for the current review, we determined that meta-regression would not provide robust or reliable insights. This decision was based on the relatively small number of studies available for the variables of interest, which was below the Cochrane-recommended minimum of 10 studies per covariate [24]. Attempting meta-regression with such a limited dataset would likely result in unstable and unreliable estimates. Therefore, we instead focused our analysis on a comprehensive subgroup and sensitivity analysis, as reported in the results section. As the evidence base expands in the future, meta-regression may become a more viable approach to explore the influence of key moderating factors on PTSD prevalence in African pediatric populations.
link