{
  "abstract": "Background Depression following traumatic brain injury (TBI) is a major contributor to disability, however, published estimates vary widely and no comprehensive meta-analysis of prevalence has yet been conducted. Consequently, we synthesised prevalence rates of depression in TBI patients including all international studies and also tested for potential moderators via sub-group analyses and meta- regressions.Methods The review was pre-registered on PROSPERO (CRD42024499164) and followed recommendations from PRISMA 2020 guidelines. We searched CINHAL, PsycINFO (Ovid) and MEDLINE (Ovid) from 1960 onwards including studies that sampled at least 100 adult participants and met criteria for a diagnosis of TBI. All severities were included ranging from mild to severe. We excluded studies where TBI exposure could not be distinguished from non-TBI conditions within one group. We extracted data from 165 studies to examine the prevalence of depression in adult TBI patients of all severities. We used a random-effects model using the DerSimonian-Laird estimator to estimate prevalence with 95% confidence intervals (CI) and the meta-analysis was conducted in R using the meta and dmetar packages. Subgroup and meta-regression analyses examined the effects of TBI severity, setting, population, study design, measure, sample age, gender balance, year of publication and risk of bias. Publication bias tests and sensitivity analyses ensured robustness. A detailed quality assessment was carried out using Kmet et al (2004) quality scale. All coding information is available online.Results Our pooled sample size of 2,648,289 generated a prevalence estimate of 27% (95% CI: 25-29%) with high, significant heterogeneity ( I2 = 100%, τ2 = 0.0182, p = 0). Subgroup analysis found that self-report measures and studies conducted in the USA were associated with increased prevalence of depression. Meta-regressions found age to be a small but significant moderator with increasing age leading to decreased prevalence (-0.0027, 95% CI: -0.0039 to -0.0015, p = <0.001). Evidence for moderation by TBI severity was mixed with no clear severity – prevalence relationship. No evidence of publication bias was found. Outlier and influence analyses found prevalence to be largely unaffected when influential studies were removed (26.46%). Our quality assessment found the majority of studies (n = 93) to be of moderate quality.Conclusion Post-TBI depression is prevalent but varies globally and is influenced by assessment method. Further investigations into reasons for lack of moderation by severity should be conducted.",
  "authors": [
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "Oriella Stellakis"
    },
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "Grace Revill"
    },
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "Aoibhe DeBurca"
    },
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "Kate DeBurca"
    },
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "Alkistis Saramandi"
    },
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "King-Chi Yau"
    },
    {
      "affiliations": [
        "University College London, Department of Clinical, Educational and Health Psychology"
      ],
      "name": "Vaughan Bell"
    }
  ],
  "title": "#8196 A systematic review and meta-analysis of the prevalence of depression after traumatic brain injury (TBI)",
  "uid": "0e897455-0ea8-56ba-91ef-3af83a0898ee"
}