CoChat
Sign Up
Cited by 107Personality Disorders and PsychopathologyDeath Anxiety and Social ExclusionPersonality Traits and Psychology

The link between rejection sensitivity and borderline personality disorder: A systematic review and meta-analysis

Open access
Foxhall, Mia, Hamilton‐Giachritsis, Catherine, Button, Katherine·2019·British Journal of Clinical Psychology
DOI 10.1111/bjc.12216
Open this paper in CoChat — free accountOpen PDF

Abstract

People with Borderline Personality Disorder (BPD) may experience heightened rejection sensitivity (RS). Pooled effect sizes suggest RS is linked with BPD (r = .326), with strong effect sizes when comparing clinical and control samples (r = .655).

Integrity signals

Registry

Crossref · 10.1111/bjc.12216

Cites

85 of 85 references verified0 citation errors · 0 not checkable

Statistics

17 tests recomputed · 0 with the reported and recomputed p on opposite sides of .05

Claims the paper makes

29 of 88 cited sources checked · 20 supported · 0 flagged · accuracy not yet measured

  • Findingcitation

    Five studies indicated physical abuse significantly correlated with rejection sensitivity.

    “Five studies indicated physical abuse significantly correlated with RS (Bungert, Liebke, et al., 2015;Erozkan, 2015;Feldman & Downey, 1994;Hernandez, Trout, & Liu, 2016;Kahya, 2018) and three studies indicated physical neglect correlated significantly (Bungert, Liebke, et al., 2015;Goodman et al., 2014;Kahya, 2018).”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Bungert, Liebke, et al., 2015 source

      The Results section ('RSQ, childhood maltreatment, and borderline symptom severity'; page not provided) reports that rejection sensitivity was positively linked to every childhood-maltreatment subscale, including physical abuse and physical neglect, in healthy controls. It does not give a significance value for the physical-abuse subscale, and reports no physical-abuse correlation in either BPD group. It does report a significant physical-neglect correlation in the acute BPD group (r = .27, p < .05). (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites Erozkan, 2015 source
    • cites Feldman & Downey, 1994 source
  • Findingcitation

    One study reported no significant relationship between physical abuse and rejection sensitivity in a non-clinical population.

    “However, one study reported no significant relationship between physical abuse and RS in a non-clinical population (Goodman et al., 2014) and another reported no significant relationship between physical abuse or neglect in a mixed population (Masland, 2018).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Goodman et al., 2014 source

      The Results say that correlations between rejection sensitivity and physical abuse are presented in Table 2, and Limitations & Future Directions identifies the participants as a nonclinical college sample. The available passages do not give the physical-abuse/rejection-sensitivity correlation or its significance, so they do not establish the claimed nonsignificant relationship. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
  • Premisecitation

    Early rejection experiences proposed to underlie rejection sensitivity may overlap within invalidating environments frequently observed in the childhood of people with Borderline Personality Disorder.

    “Furthermore, the early rejection experiences proposed to underlie RS may overlap within the invalidating environments frequently observed in the childhood of people with BPD (Ball & Links, 2009;Crowell, Beauchaine, & Linehan, 2009).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites Ball & Links, 2009 source
    • cites Crowell, Beauchaine, & Linehan, 2009 source
  • Premisecitation

    Rejection sensitivity may give rise to significant psychopathology.

    “RS has been linked with reduced self-esteem (Watson & Nesdale, 2012) and may give rise to significant psychopathology (Gao, Assink, Cipriani, & Lin, 2017;Pietrzak et al., 2005;Rosenbach & Renneberg, 2011).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Gao, Assink, Cipriani, & Lin, 2017 source
    • cites Pietrzak et al., 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Rosenbach & Renneberg, 2011 source
  • Findingcitation

    Three studies indicated physical neglect correlated significantly with rejection sensitivity.

    “Five studies indicated physical abuse significantly correlated with RS (Bungert, Liebke, et al., 2015;Erozkan, 2015;Feldman & Downey, 1994;Hernandez, Trout, & Liu, 2016;Kahya, 2018) and three studies indicated physical neglect correlated significantly (Bungert, Liebke, et al., 2015;Goodman et al., 2014;Kahya, 2018).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Bungert, Liebke, et al., 2015 source
    • cites Goodman et al., 2014 source

      The Results (page unavailable) say that correlations between rejection sensitivity and physical neglect are presented in Table 2, but the passages available from our copy do not give the correlation or its significance. Searches for the physical-neglect–RS correlation and Table 2 did not retrieve those values, so this copy cannot establish whether Goodman et al. found a significant correlation. (LLM, conf=0.69)

      Experimental — accuracy not yet measured
    • cites Kahya, 2018

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. 24
  • May combine two claims. 38

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Quality was assessed using adapted versions of the Newcastle-Ottawa Quality Assessment Scale for case control and cross-sectional studies.

    “Quality was assessed using adapted versions of the Newcastle-Ottawa Quality Assessment Scale for case control (Wells et al., 2000) and cross-sectional studies (Herzog et al., 2013), assessing for quality in participant selection, comparability of cases, and measurement and analysis of outcomes.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Wells et al., 2000

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Herzog et al., 2013 source

  • May combine two claims.

    Premisecitation

    The prior meta-analysis only included correlational data where participants were more often drawn from non-clinical populations.

    “However, the paper only included correlational data where participants were more often drawn from non-clinical populations.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Gao et al., 2017 source

  • May combine two claims.

    Findingcitation

    Emotional abuse and/or neglect correlated with rejection sensitivity in a mixed sample of people with Borderline Personality Disorder and/or major depressive disorder, but not in a pure Borderline Personality Disorder sample or community sample with clinically significant levels of Borderline Personality Disorder.

    “EAN correlated with RS in a mixed sample of people with BPD and/or MDD (Chesin et al., 2015), but not in a pure BPD sample (Bungert, Liebke, et al., 2015) or community sample with clinically significant levels of BPD (Masland, 2016).”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Chesin et al., 2015 source
    • cites Bungert, Liebke, et al., 2015 source
    • cites Masland, 2016

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingcitation

    Physical neglect was measured in one study and a small correlation was found with rejection sensitivity.

    “Physical neglect was only measured in one study due to low internal reliability and a small correlation was found with RS (Bungert, Liebke, et al., 2015).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Bungert, Liebke, et al., 2015 source

      The Methods (Measurements; page unavailable) report that physical neglect was measured, with internal consistency of α = .66. The Results (RSQ, childhood maltreatment, and borderline symptom severity; page unavailable) report a small correlation with rejection sensitivity in the acute BPD group (r = .27). The paper does not establish that physical neglect was measured in only one study or that low reliability was the reason. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingcitation

    One study indicated higher rejection sensitivity in a sample with Avoidant Personality Disorder compared with Borderline Personality Disorder, but this was non-significant.

    “One study indicated higher RS in a sample with Avoidant Personality Disorder compared with BPD, but this was non-significant (Berenson, Gregory, et al., 2016).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Berenson, Gregory, et al., 2016 source

  • May combine two claims.

    Study design and conductreasoning

    Four Borderline Personality Disorder samples partially overlap; all three are reported in the narrative review and the largest sample was included in the meta-analysis.

    “Four BPD samples partially overlap (Bungert, Koppe, et al., 2015;Bungert, Liebke, et al., 2015;Thome et al., 2016;Liebke et al. 2018).As the authors contacted were not certain of the extent of this, all three are reported in the narrative review and highlighted grey in tables.The study with the largest sample was included in the meta-analysis (Bungert, Liebke, et al., 2015).”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Bungert, Koppe, et al., 2015 source

      The Discussion (page unavailable) describes an earlier paper as reporting social-interaction findings from the present study, supporting overlap with at least one other publication. The cited paper does not establish that all four named BPD samples partially overlap. The statements about the narrative review and meta-analysis describe the citing authors’ own methods, not findings attributed to this source. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Bungert, Liebke, et al., 2015 source
    • cites Thome et al., 2016 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Two studies used screening tools for the control samples.

    “However, two studies used screening tools for the control (Fertuck, Grinband, & Stanley, 2013;Jobst et al., 2016) and some studies did not provide enough information to determine (Bungert, Koppe, et al., 2015;Bungert, Liebke, et al., 2015;Rosenbach & Renneberg, 2015;Staebler et al., 2011;Winter, Koplin, & Lis, 2015).”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites Fertuck, Grinband, & Stanley, 2013 source
    • cites Jobst et al., 2016 source

  • May combine two claims.

    Findingcitation

    The interaction between rejection sensitivity and emotional abuse and/or neglect significantly predicted Borderline Personality Disorder symptom frequency in non-clinical populations, although the latter suggested an attenuated effect for those reporting greater than average abuse.

    “In contrast, two studies measuring EAN in isolation report the interaction between RS and EAN significantly predicted BPD diagnosis in clinical populations (Chesin et al., 2015), and BPD symptom frequency in non-clinical populations (Goodman et al., 2014), although the latter suggested an attenuated effect for those reporting greater than average abuse.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Goodman et al., 2014 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Findingcitation

    Rejection sensitivity did not mediate the link between childhood abuse on Borderline Personality Disorder in a clinical group, or Borderline Personality Disorder symptom severity in a non-clinical group.

    “One study concluded that RS did not mediate the link between childhood abuse on BPD in a clinical group, or BPD symptom severity in a non-clinical group (Bungert, Liebke, et al., 2015).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Bungert, Liebke, et al., 2015 source

  • May combine two claims.

    Findingcitation

    The finding of higher rejection sensitivity in Borderline Personality Disorder versus mood disorder was not replicated in one other study, although the sample of people with major depressive disorder was small.

    “This finding was not replicated in one other study (Beeney, Levy, Gatzke-Kopp, & Hallquist, 2014), although the sample of people with MDD was small.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Beeney, Levy, Gatzke-Kopp, & Hallquist, 2014 source

  • May combine two claims.

    Findingcitation

    Shared variance between RS and BPD symptoms did not reach 40% in one study and 10% in another.

    “For instance, shared variance between variables, including RS and BPD symptoms, did not reach 40% in one study (Boldero et al., 2009) and 10% in another (Tragesser, Lippman, Trull, & Barrett, 2008), and not all participants with BPD reported elevated RS (Winter et al., 2015).”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites Boldero et al., 2009 source
    • cites Tragesser, Lippman, Trull, & Barrett, 2008 source
    • cites Winter et al., 2015 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    This review is protocol-driven.

    “This review is protocol-driven and extends upon previous reviews (Gao et al., 2017;Rosenbach & Renneberg, 2011) by adding additional data-sets, identifying overlapping samples, and assessing the full model of RS as it applies to BPD.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Gao et al., 2017 source
    • cites Rosenbach & Renneberg, 2011 source

  • May combine two claims.

    Findingcitation

    Two studies indicated that sexual abuse did correlate with rejection sensitivity, although the effect was small in the higher quality study.

    “Two others indicated that sexual abuse did correlate with RS (Erozkan, 2015), although the effect was small in the higher quality study (Kahya, 2018).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Erozkan, 2015 source
    • cites Kahya, 2018

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Findingreasoning

    Outcomes from the first search violated the assumption of independence and were too heterogeneous for meaningful comparison, so meta-analysis was not conducted.

    “Accordingly, meta-analysis was not conducted as outcomes violated assumption of independence and were considered too heterogeneous for meaningful comparison.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Four data-sets identified the effect of rejection sensitivity on Borderline Personality Disorder was mediated by executive control; rejection sensitivity was related to number of Borderline Personality Disorder features in individuals low on executive control.

    “Four data-sets identified the effect of RS on BPD was mediated by executive control; RS was related to number of BPD features in individuals low on executive control (Ayduk et al., 2008;De Panfilis, Meehan, Cain, & Clarkin, 2016).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Ayduk et al., 2008 source
    • cites De Panfilis, Meehan, Cain, & Clarkin, 2016 source

  • May combine two claims.

    Findingown data

    When compared with healthy controls, people with Borderline Personality Disorder had significantly higher rejection sensitivity, with large effect sizes (d = .83-3.25), not impacted by quality.

    “When compared with healthy controls, people with BPD had significantly higher RS.Large effect sizes were detected (d = .83-3.25), and not impacted by quality (see Table 5).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    Two studies did not find a significant effect in five different community samples and a clinical sample, though the effect size remained small to moderate (r = .16-.35).

    “Two studies did not find a significant effect in five different community samples and a clinical sample, though the effect size remained small to moderate (r = .16-.35; Brown, 2014;Liebke et al, 2018).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites Brown, 2014

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Liebke et al, 2018 source

  • May combine two claims.

    Findingcitation

    In Chesin et al.'s (2015) study, 50% of the Borderline Personality Disorder sample had concurrent major depressive disorder, whilst only 32% of the major depressive disorder sample had current major depressive disorder.

    “However, 50% of the BPD sample had concurrent MDD in Chesin et al.'s (2015) study, whilst only 32% of the MDD sample had current MDD.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Chesin et al., 2015 source

  • May combine two claims.

    Findingown data

    In non-clinical samples, rejection sensitivity correlated with Borderline Personality Disorder features, with variation in effect sizes (r = .11-.63).

    “In non-clinical samples, RS correlated with BPD features, with variation in effect sizes (r = .11-.63).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Trim and fill correction was undertaken and 11 studies were added; following correction, effect size was moderate.

    “Trim and fill correction was undertaken and 11 studies were added (Figure 5).Following correction, effect size was moderate.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    Large effect sizes were found in studies of moderate quality, however quality did not differentiate moderate and small effects.

    “Large effect sizes were found in studies of moderate quality, however quality did not differentiate moderate and small effects.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingcitation

    Only one study considered individual symptoms of Borderline Personality Disorder; in this study 'dysfunctional responses to emotion' accounted for a large portion of the effect of rejection sensitivity on PAI-BOR.

    “Only one study considered individual symptoms of BPD; in this study 'dysfunctional responses to emotion' accounted for large portion of effect of RS on PAI-BOR (Peters, Smart, & Baer, 2015).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Peters, Smart, & Baer, 2015 source

  • May combine two claims.

    Findingcitation

    People with acute Borderline Personality Disorder had higher rejection sensitivity than people with remitted Borderline Personality Disorder, approaching significance, although this attenuated when controlling for symptom severity.

    “In one study, people with acute BPD had higher RS than people with remitted BPS, approaching significance, although this attenuated when controlling for symptom severity.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingcitation

    Rejection from the same sex parent predicted more variance in rejection sensitivity.

    “Parental rejection significantly correlated with RS on the two occasions it was measured (r = .27-.45) (Rosenbach & Renneberg, 2014), with one study suggesting that rejection from the same sex parent predicted more variance in RS (Ibrahim, Rohner, Smith, & Flannery, 2015).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Ibrahim, Rohner, Smith, & Flannery, 2015 source

  • May combine two claims.

    Findingown data

    In the meta-analysis, BPD versus healthy control had an effect size of .784 (95% CI .633-.936), Z = 10.15, p < .001, mean r = .655, I2 = 80.3%, and Egger's test p = .285.

    “BPD vs. healthy control 12 .784.633-.936 10.15*** .65580.3% .285”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In the meta-analysis, BPD versus clinical control had an effect size of .294 (95% CI .111-.478), Z = 3.14, p < .01, mean r = .286, I2 = 76%, and Egger's test p = .143.

    “BPD vs. clinical control 7 .294.111-.478 3.14** .28676% .143”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Separate meta-analyses based on study design confirmed a small to moderate relationship between RS and BPD in correlational studies and a large effect size in studies comparing BPD groups with a healthy control.

    “Outcomes from separate meta-analyses based on study design confirmed a small to moderate relationship between RS and BPD in correlational studies and a large effect size in studies comparing BPD groups with a healthy control, although the effect size reduced when community samples with a high number of BPD features were used.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The included studies' measures of RS and BPD assess the same constructs sufficiently to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    Meta-regression for case-control studies with healthy controls indicated that percentage of females and population type (mixed vs. community) predicted heterogeneity, with studies using community samples incurring smaller effect sizes.

    “Meta-regression for case-control studies with healthy controls indicated that percentage of females and population type (mixed vs. community) predicted heterogeneity, with studies using community samples incurring smaller effect sizes.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The included studies' measures of RS and BPD assess the same constructs sufficiently to be combined in meta-analysis.

  • May combine two claims.

    Findingown data

    In the corrected main meta-analysis, the effect size was .431, mean Z .338, 95% CI .377-.485 and .279-.397, Z values 15.74 and 11.23, mean r .406 and .326, I2 79.4%, and Egger's test p = .012.

    “Main analyses Corrected 41 52 .431.338.377-.485 .279-.397 15.74*** 11.23*** .406.32679.4% .012”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Inter-rater agreement was moderate for the first search (k = .53).

    “Inter-rater agreement was moderate for the first search (k = .53)and good for the second (k = .74).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Statistical tests were usually reported adequately and confidence intervals were included in four studies.

    “Finally, statistical tests were usually reported adequately and confidence intervals were included in four studies.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Where partially overlapping samples were indicated, the largest sample was included.

    “Where partially overlapping samples were indicated, the largest sample was included.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Trim and fill statistics were calculated to correct publication bias where relevant.

    “Trim and fill statistics (Duval & Tweedie, 2000) were calculated to correct this, where relevant.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Duval & Tweedie, 2000 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Rejection sensitivity has been linked with reduced self-esteem.

    “RS has been linked with reduced self-esteem (Watson & Nesdale, 2012) and may give rise to significant psychopathology (Gao, Assink, Cipriani, & Lin, 2017;Pietrzak et al., 2005;Rosenbach & Renneberg, 2011).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Watson & Nesdale, 2012 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    A second reviewer screened 20% of abstracts for each search.

    “A second reviewer screened 20% of abstracts for each search.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The remaining samples could not be considered representative, limiting generalisability.

    “The remaining could not be considered representative, limiting generalisability.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    A second-rater repeated data extraction and quality assessment for 20% of the papers and perfect agreement was achieved.

    “A second-rater repeated data extraction and quality assessment for 20% of the papers and perfect agreement was achieved.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    In question 1, nine studies were rated good quality and five were moderate.

    “In terms of quality assessment, nine studies were rated good quality and five were moderate (see Supplementary Material).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Funnel plots and Eggers test were computed to assess risk of publication bias.

    “Funnel plots were created and Eggers test computed to assess for risk of publication bias (Egger, Smith, Schneider, & Minder, 1997).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Egger, Smith, Schneider, & Minder, 1997 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Only one study reported sample size justification.

    “Three samples were deemed representative and only one study reported sample size justification.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The review protocol was registered on PROSPERO.

    “Details of the protocol were registered on PROSPERO and can be accessed at: https://www.crd.york.ac.uk/prospero/display_record.asp?ID=CRD42017065936.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductreasoning

    Some papers employed multiple statistics or control groups, which may violate the assumption of independence; separate analyses were run for correlational data, case-control data with healthy controls, case-control data with clinical controls, and an overall meta-analysis.

    “Some papers employed multiple statistics or control groups, which may violate assumption of independence.Accordingly, separate analyses were run for correlational data, case-control data with healthy controls, case-control data with clinical controls and an overall meta-analysis (excluding overlapped samples; prioritising correlational data).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    A second reviewer assessed 20% of full texts for each search, with perfect agreement.

    “Remaining full texts were reviewed for inclusion.A second reviewer assessed 20% for each search.Perfect agreement was achieved for both.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Inter-rater agreement was good for the second search (k = .74).

    “Inter-rater agreement was moderate for the first search (k = .53)and good for the second (k = .74).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Rejection sensitivity is proposed to develop in response to childhood rejecting environments, including those characterised by neglect and abuse.

    “RS is proposed to develop in response to childhood rejecting environments, including those characterised by neglect and abuse, and has an adaptive purpose of keeping individuals safe (Pietrzak, Downey, Ayduk, & Baldwin, 2005).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Pietrzak, Downey, Ayduk, & Baldwin, 2005

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Study design and conductown data

    Visual inspection of the funnel plot suggested asymmetry and this was confirmed with Eggers test (p < .001).

    “Visual inspection of the funnel plot (Figure 4) suggested asymmetry and this was confirmed with Eggers test (p < .001).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Visual inspection of the funnel plot indicated some asymmetry and this was confirmed with Eggers test (p = .03), suggesting publication bias.

    “Similarly, visual inspection of the funnel plot indicated some asymmetry and this was confirmed with Eggers test (p = .03),suggesting publication bias.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    Quality assessment rated two studies as excellent, twelve as good, nineteen as moderate and one as low quality.

    “With regards quality assessment, two studies were rated as excellent, twelve as good, nineteen were moderate and one was low quality (see Supplementary Material).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    Differences between the current and previous effect sizes may be related to trim-and-fill outcomes.

    “The current study's effect sizes prior to correction for publication bias were similar (r = .431),therefore differences may be related to trim-and-fill outcomes.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    RS appears to be an important factor linked with BPD and may offer a target for intervention.

    “Accordingly, RS appears to be an important factor linked with BPD and may offer a target for intervention.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Neither meta-analyses indicated publication bias against small studies (p > .05), so corrections were not performed.

    “Neither meta-analyses indicated publication bias against small studies (p > .05),so corrections were not performed (see Supplementary Material for forest and funnel plots).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Publication bias was corrected using trim and fill.

    “Filled funnel plot following trim and fill corrections”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    A small number of studies were included in meta-regression, limiting the power of statistical analysis.

    “However, a small number of studies were included in meta-regression, limiting the power of statistical analysis.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductnone given

    Few studies commented on power analyses and sample sizes were particularly small for clinical samples.

    “Overall, few studies commented on power analyses and sample sizes were particularly small for clinical samples.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    All outcomes were self-reported and measures of childhood rejecting experiences were retrospective, increasing risk of response bias and inaccurate reporting.

    “Firstly, all outcomes were self-reported and measures of childhood rejecting experiences were retrospective, increasing risk of response bias and inaccurate reporting.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductnone given

    A representative sample was rarely recruited, leading to an over-representation of females, young adults and students.

    “A representative sample was rarely recruited, leading to an over-representation of females, young adults and students.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The corrected meta-analytic results were adjusted for publication bias.

    “Corrected results refer to outcomes corrected for publication bias.”
    Quote in the paperNot a substantial claim
57 of 57

  • The paper relies on this without support.

    Research into the link between childhood rejection and rejection sensitivity is sparse and hampered by limitations in defining and measuring early childhood rejection.

    “However, research into the link between childhood rejection and RS appears sparse and hampered by limitations in defining and measuring early childhood rejection.”

  • The authors state this as an assumption.

    The full rejection sensitivity model's assumption that high rejection sensitivity results from early childhood experiences is valid enough to test.

    “Finally, the full model posits that high RS results from early childhood experiences, therefore it seems important to identify whether this assumption is valid to help understand the relationship between BPD and RS more fully.”

  • The paper relies on this without support.

    The included studies' measures of RS and BPD assess the same constructs sufficiently to be combined in meta-analysis.

    “Outcomes from separate meta-analyses based on study design confirmed a small to moderate relationship between RS and BPD in correlational studies and a large effect size in studies comparing BPD groups with a healthy control, although the effect size reduced when community samples with a high number of BPD features were used.”
3 of 3

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Emotional and physical neglect have been shown to be a unique risk factor for BPD.

    “As noted above, emotional and physical neglect have been shown to be a unique risk factor for BPD [48, 136, 194].”

    From Complex PTSD and borderline personality disorder (2021) · cites this paper as [48]

    The “Link with BPD” section reports that emotional abuse and neglect, in interaction with rejection sensitivity, predicted BPD diagnosis or symptoms in two studies. The “Abuse” section reports mixed associations between physical neglect and rejection sensitivity, not a distinct risk for BPD. The review does not establish emotional and physical neglect as unique risk factors for BPD. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • BPD occurs in adults who have experienced childhood maltreatment primarily but only if they have limitations in the kinds of executive control necessary to modulate emotional distress.

    “This is consistent with findings that BPD occurs in adults who have experienced childhood maltreatment primarily but only if they have limitations in the kinds of executive control necessary to modulate emotional distress [48].”

    From Complex PTSD and borderline personality disorder (2021) · cites this paper as [48]

    The ‘Elevated RS and past childhood rejecting experiences’ section says childhood rejection may contribute to rejection sensitivity and proposes that greater executive control may buffer development of BPD. The ‘Correlational outcomes’ section reports an association between rejection sensitivity and BPD features in people with low executive control. These are qualified, largely correlational findings; the review does not establish that BPD occurs in maltreated adults primarily or only when they have limited executive control. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • A meta-analysis concluded that a combination of childhood emotional abuse and neglect with limited executive control in adulthood was associated with the development of BPD, and that previously maltreated adults with higher levels of executive control tend to experience relational mistrust and self-blame (which are characteristic of cPTSD/DSO) rather than BPD.

    “A meta-analysis similarly concluded that a combination of childhood emotional abuse and neglect with limited executive control in adulthood was associated with the development of BPD, and that previously maltreated adults with higher levels of executive control tend to experience relational mistrust and self-blame (which are characteristic of cPTSD/DSO) rather than BPD [48].”

    From Complex PTSD and borderline personality disorder (2021) · cites this paper as [48]

    Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured

Statistics that don’t add up

Each reported p-value recomputed from its test statistic and degrees of freedom.

  • Inconsistent p-value: t(374) = 3.54, p = .005

    Minor

    The paper reports p = .005 here; recomputing from the test statistic and its degrees of freedom gives p < .001 (two-tailed). The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p = .005 · recomputed p < .001

  • Inconsistent p-value: z = 2.12, p = .004

    Minor

    The paper reports p = .004 here; recomputing from the test statistic and its degrees of freedom gives p = .034 (two-tailed). The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p = .004 · recomputed p = .034

2 of 2

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Doesn’t add upt(374) = 3.54reported p = .005 · recomputed p < .001

    “t(374) = 3.54, p = .005”

  • Doesn’t add upz = 2.12reported p = .004 · recomputed p = .034

    “z = 2.12, p = .004”

  • ConsistentF(3, 37) = 13.18reported p < .001 · recomputed p < .001

    “F(3,37) = 13.18,p < .001”

  • ConsistentF(1, 33) = 39.38reported p < .001 · recomputed p < .001

    “F(1, 33) = 39.38,p < .001”

  • ConsistentF(1, 37) = 7.94reported p = .008 · recomputed p = .008

    “F (1, 37) = 7.94, p = .008”

  • Consistentt(142) = -1.382reported p = .169 · recomputed p = .169

    “t (142) = -1.382,p = .169”

  • Consistentt(100) = 4.81reported p < .0001 · recomputed p < .001

    “t(100) = 4.81, p < .0001”

  • Consistentt(100) = 1.94reported p = .055 · recomputed p = .055

    “t(100) = 1.94, p = .055”

  • ConsistentF(2, 59) = 5.99reported p < .005 · recomputed p = .004

    “F(2,59) = 5.99, p < .005”

  • Consistentt(64) = -3.686reported p < .001 · recomputed p < .001

    “t (64) = -3.686,p < .001”

  • Consistentz = 2.36reported p = .018 · recomputed p = .018

    “z = 2.36, p = .018”

  • Consistentz = 2.16reported p = .031 · recomputed p = .031

    “z=2.16,p = .031”

  • Consistentt(82) = -3.28reported p = .002 · recomputed p = .002

    “t(82) = -3.28,p = .002”

  • Consistentt(27) = 4.96reported p = 0 · recomputed p < .001

    “t (27)=4.96,p = .000”

  • Consistentt(35) = 3.4reported p = .002 · recomputed p = .002

    “t (35) = 3.4, p = .002”

  • ConsistentF(2, 85) = 19.52reported p < .001 · recomputed p < .001

    “F(2, 85) =19.52, p < .001”

  • Consistentz = -2.9reported p = .002 · recomputed p = .004

    “(z = -2.9,p = .002”

    The paper states an exact rank test

17 of 17

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Two values are attributed to the earlier Gao meta-analysis

    Minor

    The paper first attributes an r of .413 to Gao et al. (2017), but later describes that review's effect size as .437.

  • A quality score of five is labeled good despite the stated scale

    Minor

    The methods classify scores of four or five as moderate quality, but the Berenson et al. (2018) table entry labels a score of five good.

  • Overlapping-sample count changes within one passage

    Minor

    The passage identifies four partially overlapping BPD samples, then says all three are reported in the narrative review.

  • Discussion mistakes Fisher's Z for the corrected correlation

    Minor

    The discussion calls .338 the publication-bias-corrected correlation, but Table 6 identifies .338 as the corrected mean Fisher's Z and gives the corrected mean correlation as .326.

  • Uncorrected correlation is reported as its Fisher's Z value

    Minor

    The discussion labels .431 as an uncorrected correlation. Table 6 lists .431 under mean Fisher's Z and .406 under mean r for that analysis.

  • Bungert study has an incompatible sample size in Table 5

    Minor

    Table 5 assigns the Bungert, Liebke case-control study an N of 52. Its three participant groups listed elsewhere contain 77, 15, and 75 people, so even the BPD group alone exceeds 52.

6 of 6

Cites

Flagged first, then by citations

Research works printed but not linked: 6

  1. *Berlingo M. T.(2015).Ostracism and rejection sensitivity: Are the sensitive really sensitive?(Doctoral dissertation). Washington State University Washington USA.
  2. *Brown W. C.(2014).Borderline features rejection sensitivity and romantic relationships: A daily diary study of romantic partners(Doctoral dissertation). University of Missouri Missouri USA.
  3. Higgins J. P., 2011, Cochrane handbook for systematic reviews of interventions (Vol. 4)
  4. *Masland S. R.(2016).Trustworthiness Appraisal in Borderline Personality Disorder (Doctoral dissertation). Harvard University Massachusetts USA.
  5. *Skinner S.(2014).Responses to social rejection: The role of Borderline Personality Disorder traits (Doctoral dissertation). The University of North Carolina at Greensboro North Carolina USA.
  6. StataCorp, 2017, Stata statistical software: Release 15

Cited by

Sorted by citations

Related

Authors
Foxhall, M., Hamilton‐Giachritsis, C., & Button, K.
Affiliations
  • University of Bath (GB)
  • Department of Psychology University of Bath UK
Corresponding author
Foxhall, Mia
ORCID
Venue
British Journal of Clinical Psychology
Publisher
Wiley
Published
2019-03-21
Volume
58
Issue
3
Pages
289-326
ISSN
  • 0144-6657 (ISSN-L)
  • 2044-8260
DOI
10.1111/bjc.12216
OpenAlex
W2924722563
Semantic Scholar
d414123fdb89908bb93cf3788d8f22671c9e6833
Cited by
107
References
84
Citations by year
2026: 13 · 2025: 14 · 2024: 10 · 2023: 14 · 2022: 34 · 2021: 11 · 2020: 7 · 2019: 4
FWCI
2.21
Type
journal article
Language
en
Open access
Licence
Locations
  • British Journal of Clinical Psychology (publishedVersion)
  • PubMed (publishedVersion)
  • Pure (University of Bath) (submittedVersion, other-oa)
Topics
  • Social Sciences › Psychology › Clinical Psychology › Personality Disorders and Psychopathology
  • Social Sciences › Psychology › Social Psychology › Death Anxiety and Social Exclusion
  • Social Sciences › Psychology › Clinical Psychology › Personality Traits and Psychology
Keywords
rejection sensitivity, borderline personality disorder, childhood rejection, emotional abuse, emotional neglect, executive function, mentalization, theory of mind, interpersonal problems, therapeutic formulation, therapeutic relationship
MeSH terms
Adult, Borderline Personality Disorder, Child, Child Abuse, Emotions, Female, Humans, Male, Rejection, Psychology, Executive Function, Self Report

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-11.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-11.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review