Invisible barriers: investigating implicit bias in healthcare delivery and clinical psychology training pathways
Invisible barriers: investigating implicit bias in healthcare delivery and clinical psychology training pathways
Implicit bias has been identified as a potential contributor to disparities in healthcare delivery and workforce representation. This thesis explores implicit bias across two domains: its relationship with differential healthcare outcomes (Chapter 1) and its potential influence on access to clinical psychology training in the UK (Chapter 2).
Chapter 1 presents a systematic review of 30 studies examining whether healthcare professionals hold implicit biases and whether these biases are associated with differential healthcare outcomes. The review identified consistent evidence of implicit bias across domains such as race/ethnicity, socioeconomic status, bodyweight, gender, and sexual orientation, with biases almost always directed against minoritised groups. However, the relationship between implicit bias and healthcare outcomes was complex and context-dependent. Although 21 studies reported group-based disparities in treatment or outcomes, only 12 found a direct or partial association with implicit bias. The review highlights the need for more ecologically valid, intersectional, and geographically diverse research using validated bias measures.
Chapter 2 reports an experimental study investigating whether the race of a mock interviewee influences interviewers’ candidate ratings for clinical psychology training, and whether unconscious bias training impacts this. Participants (n = 55), all eligible to sit on UK DClinPsy selection panels, were randomised to interview either a White or Asian actor, with or without prior bias training. Despite pre-existing pro-White bias on the Implicit Association Test (IAT), Asian candidates received significantly higher scores on several selection domains, with no main effect of training. While this finding was not explained by bias scores, it raises questions about overcorrection or social desirability. Training increased bias awareness but did not reduce implicit bias or significantly influence interview behaviour.
Together, these findings highlight the prevalence of implicit bias and its varied impact on practice. Addressing healthcare inequities and workforce diversity requires both individual and systemic approaches, including bias-aware training, structural reform, and more rigorous evidence.
University of Southampton
Cockburn, Sarah-Jane
996a2094-4872-4b6d-843f-f432d7b6efaa
Hodgkinson, Melanie
72964a09-0d9c-4941-91ee-a37d33a81d7b
Brignell, Catherine
ec44ecae-8687-4bbb-bc81-8c2c8f27febd
2025
Cockburn, Sarah-Jane
996a2094-4872-4b6d-843f-f432d7b6efaa
Hodgkinson, Melanie
72964a09-0d9c-4941-91ee-a37d33a81d7b
Brignell, Catherine
ec44ecae-8687-4bbb-bc81-8c2c8f27febd
Hodgkinson, Melanie
72964a09-0d9c-4941-91ee-a37d33a81d7b
Brignell, Catherine
ec44ecae-8687-4bbb-bc81-8c2c8f27febd
Cockburn, Sarah-Jane, Hodgkinson, Melanie and Brignell, Catherine
(2025)
Invisible barriers: investigating implicit bias in healthcare delivery and clinical psychology training pathways.
University of Southampton, Doctoral Thesis, 173pp.
Record type:
Thesis
(Doctoral)
Abstract
Implicit bias has been identified as a potential contributor to disparities in healthcare delivery and workforce representation. This thesis explores implicit bias across two domains: its relationship with differential healthcare outcomes (Chapter 1) and its potential influence on access to clinical psychology training in the UK (Chapter 2).
Chapter 1 presents a systematic review of 30 studies examining whether healthcare professionals hold implicit biases and whether these biases are associated with differential healthcare outcomes. The review identified consistent evidence of implicit bias across domains such as race/ethnicity, socioeconomic status, bodyweight, gender, and sexual orientation, with biases almost always directed against minoritised groups. However, the relationship between implicit bias and healthcare outcomes was complex and context-dependent. Although 21 studies reported group-based disparities in treatment or outcomes, only 12 found a direct or partial association with implicit bias. The review highlights the need for more ecologically valid, intersectional, and geographically diverse research using validated bias measures.
Chapter 2 reports an experimental study investigating whether the race of a mock interviewee influences interviewers’ candidate ratings for clinical psychology training, and whether unconscious bias training impacts this. Participants (n = 55), all eligible to sit on UK DClinPsy selection panels, were randomised to interview either a White or Asian actor, with or without prior bias training. Despite pre-existing pro-White bias on the Implicit Association Test (IAT), Asian candidates received significantly higher scores on several selection domains, with no main effect of training. While this finding was not explained by bias scores, it raises questions about overcorrection or social desirability. Training increased bias awareness but did not reduce implicit bias or significantly influence interview behaviour.
Together, these findings highlight the prevalence of implicit bias and its varied impact on practice. Addressing healthcare inequities and workforce diversity requires both individual and systemic approaches, including bias-aware training, structural reform, and more rigorous evidence.
Text
SJC_Thesis_with Amendments_PDFA3_FINAL
- Version of Record
Restricted to Repository staff only
Text
SJC_Thesis_with AmendmentsPDFA3(2)_REDACTED
- Version of Record
Text
Final-thesis-submission-Examination-Miss-Sarah-Jane-Cockburn
Restricted to Repository staff only
More information
Published date: 2025
Identifiers
Local EPrints ID: 502428
URI: http://eprints.soton.ac.uk/id/eprint/502428
PURE UUID: b8847399-4316-4c77-8869-62f156b5d143
Catalogue record
Date deposited: 26 Jun 2025 16:31
Last modified: 11 Sep 2025 03:23
Export record
Contributors
Author:
Sarah-Jane Cockburn
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics