OSCR

Personality change after traumatic brain injury: a systematic review and meta-analysis.

Code ↔ Paper

1 match between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 1 match
  1. [1] § Results › Broad personality change ↔ Burns_et_al_analysis.ipynb, lines 447–462 · score 0.52 · personality change prevalence, meta regression adjusted, broad personality change, risk, bias, sensitivity

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Jupyter notebook · 637 lines · 18 KB · MIT · 1 match

  1. # %% [markdown]
  2. # Analysis code for
  4. # %% [markdown]
  5. # # Personality Change After Traumatic Brain Injury: A Systematic Review and Meta-analysis
  7. # %%
  8. # Load the libraries
  9. library(dplyr)
  10. library(readxl)
  11. library(meta)
  12. library(dmetar)
  13. library(ggplot2)
  14. library(metafor)
  15. library(metasens)
  16. library(metaforest)
  18. # Note: the dmetar package is not available on CRAN.
  19. # Install instructions from github here: https://dmetar.protectlab.org/
  21. # %% [markdown]
  22. # Load data file
  24. # %%
  25. # Define the data directory path
  26. data_dir <- "/home/main/Dropbox/Studies/LaurenPersonalitySysReview/Analysis/"
  27. output_dir <- "/home/main/Dropbox/Studies/LaurenPersonalitySysReview/Analysis/"
  29. # Load the data into main_df
  30. main_df <- read_excel(paste(data_dir, "Burns_TBI_PersChange_Data.xlsx", sep = ""), sheet = "Sheet1")
  32. # Copy main_df to pc_df
  33. pc_df <- main_df
  35. # %% [markdown]
  36. # Data cleaning and variable typing
  38. # %%
  39. # Change key variables to numeric
  40. pc_df$TBI_N <- as.numeric(pc_df$TBI_N)
  41. pc_df$PersChange_N <- as.numeric(pc_df$PersChange_N)
  43. # Prepare labels for the left column
  44. pc_df$Authors <- paste(pc_df$Authors, " (", pc_df$Year, ")", sep = "")
  46. # %% [markdown]
  47. # ### Run main meta-analysis
  49. # %%
  50. pc_meta <- metaprop(PersChange_N, TBI_N,
  51. studlab = Authors,
  52. sm = "PFT",
  53. method.tau = "PM",
  54. method.ci = "NAsm",
  55. data = pc_df)
  57. # %% [markdown]
  58. # Text summary of meta-analysis
  60. # %%
  61. summary(pc_meta)
  63. # %% [markdown]
  64. # Forest plot
  66. # %%
  67. # Set image size
  68. options(repr.plot.width = 9, repr.plot.height = 6, repr.plot.res = 210)
  70. # Forest plot
  71. forest(pc_meta,
  72. studlab = Authors,
  73. common = FALSE,
  74. test.effect.common = FALSE,
  75. test.overall.common = FALSE,
  76. leftlabs = c("Study", "Events", "Total N"),
  77. rightlabs = c("Prev %", "95% CIs", "Weight"),
  78. rightcols = c("effect", "ci", "w.random"),
  79. digits = 1,
  80. pscale = 100,
  81. xlim = c(10, 100),
  82. colgap.forest.left = unit(5, "mm"),
  83. colgap.forest.right = unit(0, "mm"),
  84. sortvar = TE
  85. )
  87. # %% [markdown]
  88. # ### Identify sources of heterogeneity
  90. # %%
  91. set.seed(42)
  93. # Compute PFT effect sizes (consistent with main meta-analysis)
  94. mf_df <- escalc(measure = "PFT", xi = PersChange_N, ni = TBI_N, data = pc_df)
  96. # Factorise categorical moderators
  97. mf_df[c("Continent","DesignCategory","OutcomeMeasureStatus","OutcomeType")] <-
  98. lapply(mf_df[c("Continent","DesignCategory","OutcomeMeasureStatus","OutcomeType")], factor)
  100. # MetaForest using moderators with complete data (k = 17)
  101. mf <- MetaForest(yi ~ Year + RiskOfBiasPercent + Continent + DesignCategory +
  102. OutcomeMeasureStatus + OutcomeType,
  103. data = mf_df, vi = "vi", whichweights = "random",
  104. num.trees = 10000, method = "PM")
  106. # Bootstrap preselection: identify moderators reliably above noise (100 replications)
  107. mf_sel <- preselect(mf, replications = 100, algorithm = "bootstrap")
  108. cat("Moderators surviving preselection:\n")
  109. print(preselect_vars(mf_sel, cutoff = 0.5))
  111. # Variable importance plot
  112. options(repr.plot.width = 6, repr.plot.height = 4, repr.plot.res = 210)
  113. VarImpPlot(mf)
  115. # %%
  116. # Split dataframes into broad personality change and secondary PD for subsequent analyses
  117. pc_df_perschange <- pc_df %>% filter(OutcomeType == "Personality change")
  118. pc_df_secondarypd <- pc_df %>% filter(OutcomeType == "Secondary PD")
  120. # %% [markdown]
  121. # ### Separate broad personality change vs secondary personality disorder analysis
  123. # %% [markdown]
  124. # Broad personality change meta-analysis
  126. # %%
  127. # Meta-analysis of broad personality change
  128. bpc_meta <- metaprop(PersChange_N, TBI_N,
  129. studlab = Authors,
  130. subgroup = OutcomeType,
  131. sm = "PFT",
  132. method.tau = "PM",
  133. method.ci = "NAsm",
  134. data = pc_df_perschange)
  136. # Set image size
  137. options(repr.plot.width = 9, repr.plot.height = 4, repr.plot.res = 210)
  139. # Forest plot
  140. forest(bpc_meta,
  141. subgroup.name = "Outcome",
  142. overall = FALSE,
  143. overall.hetstat = FALSE,
  144. hetstat.subgroup = TRUE,
  145. common = FALSE,
  146. common.subgroup = FALSE,
  147. test.effect.common = FALSE,
  148. test.overall.common = FALSE,
  149. test.effect.subgroup.common = FALSE,
  150. test.subgroup.common = FALSE,
  151. label.test.subgroup.random = "Subgroup difference ",
  152. leftlabs = c("Study", "Events", "Total N"),
  153. rightlabs = c("Prev %", "95% CIs", "Weight"),
  154. rightcols = c("effect", "ci", "w.random"),
  155. digits = 1,
  156. pscale = 100,
  157. xlim = c(10, 100),
  158. colgap.forest.left = unit(5, "mm"),
  159. colgap.forest.right = unit(0, "mm"),
  160. test.subgroup = FALSE
  161. )
  163. # %% [markdown]
  164. # Personality disorder diagnosis meta-analysis
  166. # %%
  167. # Meta-analysis of personality disorder diagnosis
  168. pdx_meta <- metaprop(PersChange_N, TBI_N,
  169. studlab = Authors,
  170. subgroup = OutcomeType,
  171. sm = "PFT",
  172. method.tau = "PM",
  173. method.ci = "NAsm",
  174. data = pc_df_secondarypd)
  176. # Set image size
  177. options(repr.plot.width = 9, repr.plot.height = 4, repr.plot.res = 210)
  179. # Forest plot
  180. forest(pdx_meta,
  181. subgroup.name = "Outcome",
  182. overall = FALSE,
  183. overall.hetstat = FALSE,
  184. hetstat.subgroup = TRUE,
  185. common = FALSE,
  186. common.subgroup = FALSE,
  187. test.effect.common = FALSE,
  188. test.overall.common = FALSE,
  189. test.effect.subgroup.common = FALSE,
  190. test.subgroup.common = FALSE,
  191. label.test.subgroup.random = "Subgroup difference ",
  192. leftlabs = c("Study", "Events", "Total N"),
  193. rightlabs = c("Prev %", "95% CIs", "Weight"),
  194. rightcols = c("effect", "ci", "w.random"),
  195. digits = 1,
  196. pscale = 100,
  197. xlim = c(10, 100),
  198. colgap.forest.left = unit(5, "mm"),
  199. colgap.forest.right = unit(0, "mm"),
  200. test.subgroup = FALSE
  201. )
  203. # %% [markdown]
  204. # Combined figure
  206. # %%
  207. # Meta-analysis with subgroup by outcome type
  208. pc_outcome_meta <- metaprop(PersChange_N, TBI_N,
  209. studlab = Authors,
  210. subgroup = OutcomeType,
  211. sm = "PFT",
  212. method.tau = "PM",
  213. method.ci = "NAsm",
  214. data = pc_df)
  217. # Set image size
  218. options(repr.plot.width = 9, repr.plot.height = 7, repr.plot.res = 210)
  220. # Forest plot
  221. forest(pc_outcome_meta,
  222. subgroup.name = "Outcome",
  223. overall = FALSE,
  224. overall.hetstat = FALSE,
  225. hetstat.subgroup = TRUE,
  226. common = FALSE,
  227. common.subgroup = FALSE,
  228. test.effect.common = FALSE,
  229. test.overall.common = FALSE,
  230. test.effect.subgroup.common = FALSE,
  231. test.subgroup.common = FALSE,
  232. label.test.subgroup.random = "Subgroup difference ",
  233. leftlabs = c("Study", "Events", "Total N"),
  234. rightlabs = c("Prev %", "95% CIs", "Weight"),
  235. rightcols = c("effect", "ci", "w.random"),
  236. digits = 1,
  237. pscale = 100,
  238. xlim = c(10, 100),
  239. colgap.forest.left = unit(5, "mm"),
  240. colgap.forest.right = unit(0, "mm"),
  241. test.subgroup = FALSE
  242. )
  244. # %% [markdown]
  245. # ### Subgroup analyses
  247. # %% [markdown]
  248. # Study design: broad personality change studies
  250. # %%
  251. # Meta-analysis with study design subgroup
  252. bpc_design_meta <- metaprop(PersChange_N, TBI_N,
  253. studlab = Authors,
  254. subgroup = StudyDesignBrief,
  255. sm = "PFT",
  256. method.tau = "PM",
  257. method.ci = "NAsm",
  258. data = pc_df_perschange)
  261. # Set image size
  262. options(repr.plot.width = 9, repr.plot.height = 6, repr.plot.res = 210)
  264. # Forest plot
  265. forest(bpc_design_meta,
  266. subgroup.name = "Design",
  267. common = FALSE,
  268. common.subgroup = FALSE,
  269. test.effect.common = FALSE,
  270. test.overall.common = FALSE,
  271. test.effect.subgroup.common = FALSE,
  272. test.subgroup.common = FALSE,
  273. label.test.subgroup.random = "Subgroup difference ",
  274. leftlabs = c("Study", "Events", "Total N"),
  275. rightlabs = c("Prev %", "95% CIs", "Weight"),
  276. rightcols = c("effect", "ci", "w.random"),
  277. digits = 1,
  278. pscale = 100,
  279. xlim = c(10, 100),
  280. colgap.forest.left = unit(5, "mm"),
  281. colgap.forest.right = unit(0, "mm")
  282. )
  284. # %% [markdown]
  285. # Study design: secondary PD diagnosis studies
  287. # %%
  288. # Meta-analysis with study design subgroup
  289. pdx_design_meta <- metaprop(PersChange_N, TBI_N,
  290. studlab = Authors,
  291. subgroup = StudyDesignBrief,
  292. sm = "PFT",
  293. method.tau = "PM",
  294. method.ci = "NAsm",
  295. data = pc_df_secondarypd)
  298. # Set image size
  299. options(repr.plot.width = 9, repr.plot.height = 6, repr.plot.res = 210)
  301. # Forest plot
  302. forest(pdx_design_meta,
  303. subgroup.name = "Design",
  304. common = FALSE,
  305. common.subgroup = FALSE,
  306. test.effect.common = FALSE,
  307. test.overall.common = FALSE,
  308. test.effect.subgroup.common = FALSE,
  309. test.subgroup.common = FALSE,
  310. label.test.subgroup.random = "Subgroup difference ",
  311. leftlabs = c("Study", "Events", "Total N"),
  312. rightlabs = c("Prev %", "95% CIs", "Weight"),
  313. rightcols = c("effect", "ci", "w.random"),
  314. digits = 1,
  315. pscale = 100,
  316. xlim = c(10, 100),
  317. colgap.forest.left = unit(5, "mm"),
  318. colgap.forest.right = unit(0, "mm")
  319. )
  321. # %% [markdown]
  322. # Validated or unvalidated outcome measure: broad personality change studies
  324. # %%
  325. # Meta-analysis with measure status subgroup
  326. bpc_measurestatus_meta <- metaprop(PersChange_N, TBI_N,
  327. studlab = Authors,
  328. subgroup = OutcomeMeasureStatus,
  329. sm = "PFT",
  330. method.tau = "PM",
  331. method.ci = "NAsm",
  332. data = pc_df_perschange)
  335. # Set image size
  336. options(repr.plot.width = 9, repr.plot.height = 6, repr.plot.res = 210)
  338. # Forest plot
  339. forest(bpc_measurestatus_meta,
  340. subgroup.name = "Measure status",
  341. common = FALSE,
  342. common.subgroup = FALSE,
  343. test.effect.common = FALSE,
  344. test.overall.common = FALSE,
  345. test.effect.subgroup.common = FALSE,
  346. test.subgroup.common = FALSE,
  347. label.test.subgroup.random = "Subgroup difference ",
  348. leftlabs = c("Study", "Events", "Total N"),
  349. rightlabs = c("Prev %", "95% CIs", "Weight"),
  350. rightcols = c("effect", "ci", "w.random"),
  351. digits = 1,
  352. pscale = 100,
  353. xlim = c(10, 100),
  354. colgap.forest.left = unit(5, "mm"),
  355. colgap.forest.right = unit(0, "mm")
  356. )
  358. # %% [markdown]
  359. # Validated or unvalidated outcome measure: personality disorder diagnosis studies
  361. # %%
  362. # Meta-analysis with measure status subgroup
  363. pdx_measurestatus_meta <- metaprop(PersChange_N, TBI_N,
  364. studlab = Authors,
  365. subgroup = OutcomeMeasureStatus,
  366. sm = "PFT",
  367. method.tau = "PM",
  368. method.ci = "NAsm",
  369. data = pc_df_secondarypd)
  372. # Set image size
  373. options(repr.plot.width = 9, repr.plot.height = 7, repr.plot.res = 210)
  375. # Forest plot
  376. forest(pdx_measurestatus_meta,
  377. subgroup.name = "Measure status",
  378. common = FALSE,
  379. common.subgroup = FALSE,
  380. test.effect.common = FALSE,
  381. test.overall.common = FALSE,
  382. test.effect.subgroup.common = FALSE,
  383. test.subgroup.common = FALSE,
  384. label.test.subgroup.random = "Subgroup difference ",
  385. leftlabs = c("Study", "Events", "Total N"),
  386. rightlabs = c("Prev %", "95% CIs", "Weight"),
  387. rightcols = c("effect", "ci", "w.random"),
  388. digits = 1,
  389. pscale = 100,
  390. xlim = c(10, 100),
  391. colgap.forest.left = unit(5, "mm"),
  392. colgap.forest.right = unit(0, "mm")
  393. )
  395. # %% [markdown]
  396. # ### Meta-regressions
  398. # %%
  399. extract_reg <- function(meta_obj, moderator, label) {
  400. fit <- tryCatch(
  401. suppressWarnings(metareg(meta_obj, as.formula(paste("~", moderator)))),
  402. error = function(e) NULL
  403. )
  404. if (is.null(fit)) return(data.frame(Moderator = label, k = NA, Estimate = NA, CI = NA, p = NA))
  405. s <- coef(summary(fit))[2, ]
  406. data.frame(
  407. Moderator = label,
  408. k = fit$k,
  409. Estimate = round(s$estimate, 3),
  410. CI = paste0("[", round(s$ci.lb, 3), ", ", round(s$ci.ub, 3), "]"),
  411. p = round(s$pval, 3)
  412. )
  413. }
  415. mods <- list(
  416. list("mean_age", "Mean age"),
  417. list("PercentFemales", "% Female"),
  418. list("FollowUpMonths", "Follow-up (months)"),
  419. list("Percent_mTBI", "% mild TBI"),
  420. list("Percent_modTBI", "% moderate TBI"),
  421. list("Percent_sevTBI", "% severe TBI"),
  422. list("Year", "Year"),
  423. list("RiskOfBiasPercent", "Risk of bias (%)")
  424. )
  426. # %% [markdown]
  427. # Broad personality change
  429. # %%
  430. bpc_table <- do.call(rbind, lapply(mods, function(m) extract_reg(bpc_meta, m[[1]], m[[2]])))
  431. bpc_table
  433. # %% [markdown]
  434. # Secondary personality disorder diagnosis
  436. # %%
  437. pdx_table <- do.call(rbind, lapply(mods, function(m) extract_reg(pdx_meta, m[[1]], m[[2]])))
  438. pdx_table
  440. # %% [markdown]
  441. # Entire sample
  443. # %%
  444. entire_sample_table <- do.call(rbind, lapply(mods, function(m) extract_reg(pc_meta, m[[1]], m[[2]])))
  445. entire_sample_table
  447. # %% [markdown]
  448. # ### Robustness and sensitivity analyses
  450. # %% [markdown]
  451. # #### Meta-regression adjusted estimate taking into account risk of bias association on full sample
  453. # %% [markdown]
  454. # Broad personality change estimate
  456. # %%
  457. # Broad personality change prevalence association with risk of bias
  458. bpc_rob_reg <- metareg(bpc_meta, ~ RiskOfBiasPercent)
  460. # Predict prevalence at 100% risk of bias score (ideal/no bias)
  461. predict(bpc_rob_reg, newmods = 100, transf = transf.ipft.hm,
  462. targ = list(ni = mean(bpc_meta$n)))
  464. # %% [markdown]
  465. # Secondary personality disorder diagnosis
  467. # %%
  468. # Secondary personality disorder prevalence association with risk of bias
  469. pdx_rob_reg <- metareg(pdx_meta, ~ RiskOfBiasPercent)
  471. # Predict prevalence at 100% risk of bias score (ideal/no bias)
  472. predict(pdx_rob_reg, newmods = 100, transf = transf.ipft.hm,
  473. targ = list(ni = mean(bpc_meta$n)))
  475. # %% [markdown]
  476. # #### Publication bias
  478. # %% [markdown]
  479. # Doi Plot and LFK index: broad personality change studies
  481. # %%
  482. # Filter for only broad personality studies
  483. bpc_df <- pc_df %>%
  484. filter(OutcomeType == "Personality change")
  486. # Re-run meta
  487. spd_meta <- metaprop(PersChange_N, TBI_N,
  488. studlab = Authors,
  489. sm = "PFT",
  490. method.tau = "PM",
  491. method.ci = "NAsm",
  492. data = bpc_df)
  495. # Doi plot and LFK index
  496. options(repr.plot.width = 7, repr.plot.height = 5, repr.plot.res = 210)
  497. doiplot(spd_meta)
  498. lfkindex(spd_meta)
  500. # %% [markdown]
  501. # Trim and fill to revised estimate of prevalence based on imputed missing studies
  503. # %%
  504. bpc_trimfill <- trimfill(bpc_meta)
  505. summary(bpc_trimfill)
  507. # %% [markdown]
  508. # Doi Plot and LFK index: secondary personality disorder studies
  510. # %%
  511. # Filter for only broad personality studies
  512. spd_df <- pc_df %>%
  513. filter(OutcomeType == "Secondary PD")
  515. # Re-run meta
  516. spd_meta <- metaprop(PersChange_N, TBI_N,
  517. studlab = Authors,
  518. sm = "PFT",
  519. method.tau = "PM",
  520. method.ci = "NAsm",
  521. data = spd_df)
  524. # Doi plot and LFK index
  525. options(repr.plot.width = 7, repr.plot.height = 5, repr.plot.res = 210)
  526. doiplot(spd_meta)
  527. lfkindex(spd_meta)
  529. # %% [markdown]
  530. # ### Influence diagnostics
  532. # %% [markdown]
  533. # Outliers: Broad personality change studies
  535. # %%
  536. dmetar_bpc_output <- dmetar::find.outliers(bpc_meta)
  538. # Print outliers from random effects model
  539. dmetar_bpc_output$out.study.random
  541. # Store recalculated meta minus outliers (dmetar automatically recalculates this and stores it in dmetar_fo_output$m.random
  542. bpc_meta_minus_outliers <- dmetar_bpc_output$m.random
  544. # %%
  545. # Set image size
  546. options(repr.plot.width = 9, repr.plot.height = 5, repr.plot.res = 210)
  548. # Show recalculated meta with
  549. forest(bpc_meta_minus_outliers,
  550. common = FALSE,
  551. test.effect.common = FALSE,
  552. test.overall.common = FALSE,
  553. test.subgroup = FALSE,
  554. print.Q = TRUE,
  555. leftlabs = c("Study", "Events", "Total N"),
  556. rightlabs = c("Prev %", "95% CIs", "Weight"),
  557. rightcols = c("effect", "ci", "w.random"),
  558. digits = 1,
  559. pscale = 100,
  560. xlim = c(10, 100),
  561. colgap.forest.left = unit(5, "mm"),
  562. colgap.forest.right = unit(0, "mm")
  563. )
  565. # %% [markdown]
  566. # Leave one out sensitivity diagnostics: Broad personality change studies
  568. # %%
  569. # Leave1Out sensitivity analysis
  570. metainf(bpc_meta, pooled = "random")
  572. # %% [markdown]
  573. # Outliers: Personality disorder diagnosis studies
  575. # %%
  576. dmetar_pdx_output <- dmetar::find.outliers(pdx_meta)
  578. # Print outliers from random effects model
  579. dmetar_pdx_output$out.study.random
  581. # Store recalculated meta minus outliers (dmetar automatically recalculates this and stores it in dmetar_fo_output$m.random
  582. pdx_meta_minus_outliers <- dmetar_pdx_output$m.random
  584. # %%
  585. # Set image size
  586. options(repr.plot.width = 9, repr.plot.height = 5, repr.plot.res = 210)
  588. # Show recalculated meta with
  589. forest(pdx_meta_minus_outliers,
  590. common = FALSE,
  591. test.effect.common = FALSE,
  592. test.overall.common = FALSE,
  593. test.subgroup = FALSE,
  594. print.Q = TRUE,
  595. leftlabs = c("Study", "Events", "Total N"),
  596. rightlabs = c("Prev %", "95% CIs", "Weight"),
  597. rightcols = c("effect", "ci", "w.random"),
  598. digits = 1,
  599. pscale = 100,
  600. xlim = c(10, 100),
  601. colgap.forest.left = unit(5, "mm"),
  602. colgap.forest.right = unit(0, "mm")
  603. )
  605. # %% [markdown]
  606. # Leave one out sensitivity diagnostics: Personality disorder diagnosis studies
  608. # %%
  609. # Leave1Out sensitivity analysis
  610. metainf(pdx_meta, pooled = "random")
  612. # %% [markdown]
  613. # ### Analysis platform details and software versions
  615. # %%
  616. version
  618. # %%
  619. packageVersion("dplyr")
  621. # %%
  622. packageVersion("readxl")
  624. # %%
  625. packageVersion("meta")
  627. # %%
  628. packageVersion("metafor")
  630. # %%
  631. packageVersion("dmetar")
  633. # %%
  634. packageVersion("metasens")
  636. # %%
  637. packageVersion("metaforest")

Burns_et_al_analysis.ipynb at commit a14bc00, under MIT · at the source

Overview

Authors: Lauren Burns1, Kelly Jones2, Keishema Kerr1, Nora Brennan3, Natalie Clapshaw3, Huw Green4, Hannah Farrimond5, Claire Stone3, Sam Wilkinson5, Members of Headway London3, Vaughan Bell1,6
ORCID iDs: Vaughan Bell
  1. Clinical, Educational and Health Psychology, University College London, London, UK
  2. University of Swansea, Swansea, UK
  3. Headway London, London, UK
  4. Department of Neuropsychology, Addenbrooke’s Hospital, Cambridge, UK
  5. Department of Social and Political Sciences, Philosophy, and Anthropology, University of Exeter, Exeter, UK
  6. Dept of Neuropsychiatry, South London and Maudsley NHS Foundation Trust, London, UK
Institutions: University College London (United Kingdom); Swansea University (United Kingdom); Addenbrooke's Hospital (United Kingdom); University of Exeter (United Kingdom); South London and Maudsley NHS Foundation Trust (United Kingdom)
Journal: Journal of neurology, volume 273, issue 8, article 502
Dates: received 18 February 2026; accepted 17 July 2026; published online 3 August 2026; in print 2026
Type: Review · Language: English
License: CC BY
Identifiers: DOI 10.1007/s00415-026-14028-0 · PMID 42547613 · PMCID PMC13433385 · OpenAlex W7172314308
Open access: hybrid, a free copy (OpenAlex)
Status: code verified
Categories: human (organism), traumatic brain injury (population), clinical / translational (subfield)
Keywords: Traumatic brain injury, Personality change, Organic personality disorder, Secondary personality disorder, Neuropsychiatry, Disinhibition
MeSH: Brain Injuries, Traumatic*, Personality Disorders*, Humans (* major topic)
Topic: Traumatic Brain Injury Research (Epidemiology, Medicine), according to OpenAlex
Citations: not cited yet (Europe PMC); 56 references in the paper

Abstract

Background: Personality change is a debilitating consequence of traumatic brain injury (TBI), yet its prevalence, characteristics, and treatment remain poorly understood.

Methods: We completed a pre-registered (CRD42023440990) systematic review and meta-analysis searching four databases (MEDLINE, PsycINFO, EMBASE and CINAHL) for primary studies assessing personality change after TBI. We synthesized conceptualization, prevalence, longitudinal outcome, lesion location, and treatment. Prevalence was estimated using a random effect meta-analysis using the Paule–Mandel estimator, with subgroup, meta-regression and robustness analyses.

Results: A total of 101 studies were included in this review, seventeen of which were suitable for meta-analysis. Personality change was defined inconsistently although common symptoms involved the emergence or increase of affective, behavioral, and social disturbances, including irritability, depression, emotional instability, anger outbursts, social withdrawal, anxiety, impulsivity, restlessness, aberrant motor behaviors, and aggression. The prevalence of secondary personality disorder was estimated as 29.1% (CIs 22.5% – 36.2%) and prevalence of broad personality change was 68.1% (CIs 53.4% – 81.2%). Robustness analyses showed that the estimate for broad personality change should be treated with caution as it was unstable when adjusted for risk of bias and potential publication bias. Follow-up studies, although of varying quality, consistently showed personality change remained stable over long follow-up periods. The relationship between personality change and specific lesion locations in TBI remains unclear, likely due to the poor methodological quality of studies examining this association. Perhaps most concerning, there is limited evidence and very few systematic studies addressing treatment.

Conclusion: Personality change is a common and persistent consequence of TBI. Varying definitions, and the lack of high-quality lesion mapping studies and systematic investigations into treatment highlights critical gaps in understanding and management.

Supplementary Information: The online version contains supplementary material available at 10.1007/s00415-026-14028-0.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 1 match between paragraphs and lines of code.

vaughanbell/personality-change-TBI-meta

License: MIT
State: the link answers, verified on 26 September 2026
Evidence: files inventoried
Commit: a14bc00250ad6ab433bba439ca7e29903f17bad8, 25 June 2026
Languages: Jupyter (1)
Size: 4 files, 1 script
Software Heritage: not archived
Found in: “Data availability”
Holds: README, 1 notebook
Not found: license file, CITATION.cff, environment file, tests, continuous integration, documentation
Tools: ggplot2 (1 file), metafor (1 file), tidyverse (1 file)
Availability: 1 check, the latest on 26 September 2026: the link answers
  • 26 September 2026: the link answers
2 files
At the source: github.com/vaughanbell/personality-change-TBI-meta/

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 1 script, each with its path and the digest of its content;
  • 1 match between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data availability

All data and analysis code used in this study are available at https://github.com/vaughanbell/personality-change-TBI-meta/.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 3, 28 September 2026

  • Publisher: n/a → Springer Science+Business Media

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 11 authors, 6 keywords, 3 MeSH terms, 52 references.

Cite

This paper

Burns, L., Jones, K., Kerr, K., Brennan, N., Clapshaw, N., Green, H., Farrimond, H., Stone, C., Wilkinson, S., Members of Headway London, & Bell, V. (2026). Personality change after traumatic brain injury: a systematic review and meta-analysis. Journal of neurology, 273(8), 502. https://doi.org/10.1007/s00415-026-14028-0

BibTeX

@article{burns2026personality,
author = {Burns, Lauren and Jones, Kelly and Kerr, Keishema and Brennan, Nora and Clapshaw, Natalie and Green, Huw and Farrimond, Hannah and Stone, Claire and Wilkinson, Sam and Members of Headway London and Bell, Vaughan},
title = {{Personality change after traumatic brain injury: a systematic review and meta-analysis}},
journal = {Journal of neurology},
year = {2026},
month = aug,
volume = {273},
number = {8},
pages = {502},
publisher = {Springer Science+Business Media},
issn = {0340-5354},
doi = {10.1007/s00415-026-14028-0},
url = {https://doi.org/10.1007/s00415-026-14028-0},
pmid = {42547613},
pmcid = {PMC13433385}
}

RIS

TY - JOUR
AU - Burns, Lauren
AU - Jones, Kelly
AU - Kerr, Keishema
AU - Brennan, Nora
AU - Clapshaw, Natalie
AU - Green, Huw
AU - Farrimond, Hannah
AU - Stone, Claire
AU - Wilkinson, Sam
AU - Members of Headway London
AU - Bell, Vaughan
TI - Personality change after traumatic brain injury: a systematic review and meta-analysis
T2 - Journal of neurology
J2 - J Neurol
PY - 2026
DA - 2026/08/03
VL - 273
IS - 8
SP - 502
SN - 0340-5354
PB - Springer Science+Business Media
DO - 10.1007/s00415-026-14028-0
UR - https://doi.org/10.1007/s00415-026-14028-0
LA - en
ER -

CSL-JSON

{
"id": "10.1007/s00415-026-14028-0",
"type": "article-journal",
"title": "Personality change after traumatic brain injury: a systematic review and meta-analysis",
"container-title": "Journal of neurology",
"author": [
{
"family": "Burns",
"given": "Lauren"
},
{
"family": "Jones",
"given": "Kelly"
},
{
"family": "Kerr",
"given": "Keishema"
},
{
"family": "Brennan",
"given": "Nora"
},
{
"family": "Clapshaw",
"given": "Natalie"
},
{
"family": "Green",
"given": "Huw"
},
{
"family": "Farrimond",
"given": "Hannah"
},
{
"family": "Stone",
"given": "Claire"
},
{
"family": "Wilkinson",
"given": "Sam"
},
{
"family": "Members of Headway London"
},
{
"family": "Bell",
"given": "Vaughan"
}
],
"container-title-short": "J Neurol",
"volume": "273",
"issue": "8",
"page": "502",
"DOI": "10.1007/s00415-026-14028-0",
"PMID": "42547613",
"PMCID": "PMC13433385",
"ISSN": "0340-5354",
"publisher": "Springer Science+Business Media",
"URL": "https://doi.org/10.1007/s00415-026-14028-0",
"language": "en",
"issued": {
"date-parts": [
[
2026,
8,
3
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1128/msystems.00416-26 [code]
Integrative multicohort analysis reveals consistent sex differences in gut microbiota of multiple sclerosis patients.
Journal: mSystems
In common: metafor, ggplot2, tidyverse, clinical / translational
[2] doi:10.1186/s13195-026-02036-1 [code]
Genetic drivers of progression in Alzheimer's disease are distinct from disease risk.
Journal: Alzheimer's research & therapy
In common: metafor, ggplot2, tidyverse, clinical / translational
[3] doi:10.1093/ageing/afag263 [code]
Cardiometabolic medication exposures and cognitive outcomes in Alzheimer's disease.
Journal: Age and ageing
In common: metafor, ggplot2, tidyverse
[4] doi:10.1186/s40478-026-02415-7 [code]
A standardized framework resolves ambiguity in motor neuron loss across neurodegenerative diseases.
Journal: Acta neuropathologica communications
In common: metafor, ggplot2, tidyverse
[5] doi:10.3390/ijms27136068 [code]
Loss of Neuropeptide Y Signaling Accompanies the Neural-to-Mesenchymal Transcriptional Transition in Glioblastoma: A Multi-Scale Transcriptomic Analysis.
Journal: International journal of molecular sciences
In common: metafor, ggplot2, tidyverse
[6] doi:10.1038/s41562-026-02486-5 [code]
Genome-wide association studies of infant and toddler temperament in European and multi-ancestry populations.
Journal: Nature human behaviour
In common: metafor, ggplot2, tidyverse
[7] doi:10.1038/s41467-026-74565-0 [code]
The functional neurobiology of dispositions towards negative emotions.
Journal: Nature communications
In common: metafor, ggplot2, tidyverse
[8] doi:10.1038/s41467-026-73865-9 [code]
Histamine shapes the neurocomputational dynamics of human learning.
Journal: Nature communications
In common: metafor, ggplot2, tidyverse
[9] doi:10.1093/braincomms/fcag146 [code]
Convergent structural brain alterations in chronic pain: a multi-metric individual participant data meta-analysis.
Journal: Brain communications
In common: metafor, ggplot2, tidyverse
[10] doi:10.1007/s11357-026-02195-x [code]
The aging epigenome: integrative analyses reveal intersection with Alzheimer's disease.
Journal: GeroScience
In common: metafor, ggplot2, tidyverse

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.