HCAT's evidence base spans 60+ peer-reviewed studies across 20+ countries — from the tool's development and independent validations to adaptations, translations, and applications in hospitals, general practice and national systems. This page lists them all.

Featured research

The HCAT paper (2016)

Development and reliability testing of HCAT on a national sample of NHS hospital complaints. HCAT reliably codes seven problem categories at four severity levels, plus stage of care and harm — “not only the first reliable tool for coding complaints… the first tool to measure the severity of complaints”. Free to download and use.

Gillespie & Reader (2016), BMJ Quality & Safety

View article

Hot spots and blind spots (2018)

Analysis of 1,110 HCAT-coded complaints from 56 NHS trusts showed how aggregated complaints reveal “hot spots” where harm clusters onto stages of care, and “blind spots” other monitoring misses: problems at admission and discharge, systemic problems, and errors of omission. Benchmark data and analysis scripts are downloadable as supplementary files.

Gillespie & Reader (2018), The Milbank Quarterly

View article

Complaint severity predicts hospital mortality (2021)

Across 59 NHS hospital trusts, the severity of clinical problems in HCAT-coded complaints was the only patient-generated measure associated with hospital-level mortality — and it explained variance beyond staff-generated data, while staff incident-report rates showed no association at all. Patients provide valid, supplementary data on unsafe care.

Reader & Gillespie (2021), Journal of Applied Psychology

View article

Cross-country replication in Ireland (2022)

A national study of 641 complaints from 16 Irish public hospitals reproduced the UK hot-spot and blind-spot patterns almost exactly, with very high coding reliability (Gwet's AC1 = 0.93). Evidence that HCAT transfers across healthcare systems.

O'Dowd et al. (2022), International Journal for Quality in Health Care

View article

Whole-hospital adoption at Odense (2023)

Odense University Hospital analysed 6,366 complaints (13,156 problems) with disproportionality analysis borrowed from pharmacovigilance, ranking where problems cluster across departments. Hospital management then adopted HCAT for all complaints from 2022 — “likely… the world's largest sample of HCAT-scored complaints”.

Bogh et al. (2023), International Journal for Quality in Health Care

View article

The dark figures of complaints (2025)

The first study to quantify how many negative experiences never become complaints: for every problem in a filed maternity complaint, between 4.8 and 13 more went unreported, depending on the issue. Formal complaints are the tip of an iceberg — and low-harm problems are the most under-reported.

Clausen et al. (2025), Acta Obstetricia et Gynecologica Scandinavica

View article

HCAT and AI

Complaint volumes are rising faster than teams can read them. In 2023–24 the NHS in England received 241,922 complaints — up 37% since 2013–14 — and, at an hour per complaint, manually analysing a single year's complaints would take roughly 35,000 hours, or about 95 staff-years (Reader, 2025). Can AI help? Two independent 2025 studies tested large language models against trained human HCAT coders, with convergent results. Bunt et al. (2025) found GPT-4o only moderately valid on HCAT harm coding (weighted κ 0.49–0.57, against a human ICC of 0.86), systematically overestimating harm. Koh et al. (2025) tested three LLMs against HCAT(GP) coding in Singapore primary care: substantial agreement on domain (κ ≈ 0.62), but poor agreement on severity and harm (κ 0.11–0.24). Our testing with more recent models shows significant improvements for both harm and severity. If you would like to discuss an AI-assisted version of HCAT, contact us.

Validations, adaptations & translations

VariantSettingCountryYearAuthorsLink
Persian localisationTranslation and expert Delphi validationIran2025Vafaee Najar et al.Article
Learner-mistreatment reporting frameworkMedical educationCanada2025Rachul et al.Article
End-of-life care complaint analysis toolNHS end-of-life care (licensed derivative)England2024Walker & DowlingArticle
Japanese translation of the taxonomyFive-year analysis at a Tokyo tertiary teaching hospitalJapan2024Uramatsu et al.Article
HCAT (unmodified)National validation across 16 public hospitalsIreland2022O'Dowd et al.Article
HCAT (unmodified)Great Ormond Street specialist paediatric hospitalUK2022Wray, Williams & OldhamArticle
HCAT-DKDanish translation — the first non-English editionDenmark2021Morsø et al.Article
HCAT(GP)General practice; public training videoIreland2021O'Dowd, Lydon & O'ConnorArticle
HCAT (unmodified)Reliability testing on patient compensation claimsDenmark2019Bogh et al.Article
Out-of-hours GP modificationOut-of-hours general practiceIreland2018Wallace et al.Article
RSDM-PCT dental forkDental educationUSA2017Fredericks-Younger et al.Article
Compliments / praise extensionComplaints and praise at an Italian hospitalItaly2017Mattarozzi et al.Article
Dutch regulatory variantHealthcare inspectorateNetherlands2016Bouwman et al.Article

HCAT Research

FocusSettingCountryYearAuthorsLink
AIValidating large language models against HCAT harm codingUK2025Bunt et al.Article
ApplicationMaternity survey: dark figures of unreported problemsDenmark2025Clausen et al.Article
ReviewScoping review: complaints and claims in quality improvementInternational2025Hansen et al.Article
AIPrimary care; LLMs tested against HCAT(GP) codersSingapore2025Koh et al.Article
ApplicationStage-specific harm in a new hospital (coded in Korean)South Korea2025Kwon et al.Article
ApplicationTime-series analysis of complaints in new hospitalsSouth Korea2025Kwon et al.Article
ApplicationOut-of-hours primary care compensation claimsDenmark2025Køpfli et al.Article
ApplicationPhysician narratives of complaints during residencyCanada2025McDougall et al.Article
CommentaryEditorial: challenges and opportunities in learning from complaintsUK2025ReaderArticle
ApplicationShanghai hotline complaints with spatial analysisChina2025Wang et al.Article
CommentaryPerspective: from complaint analysis to quality improvementDenmark2024Birkeland, Bogh & MorsøArticle
ApplicationOnline complaints about city hospitals during COVID-19Türkiye2024Çınar et al.Article
ApplicationNourishment problems at a hospitalDenmark2024Clausen et al.Article
AILarge language models drafting complaint repliesSingapore2024Yong et al.Article
ApplicationCompliments about nursing and midwifery careAustralia2023Baker, White & RedleyArticle
ApplicationOdense University Hospital: disproportionality analysis of 6,366 complaintsDenmark2023Bogh et al.Article
ApplicationLaser treatment center, plastic surgery hospitalChina2023Ding et al.Article
AIOnline patient feedback as a safety valve (automated language analysis)England2023Gillespie & ReaderArticle
ApplicationGovernment complaint hotlineChina2023Han et al.Article
ApplicationOdense University Hospital: implementation, training and dashboardDenmark2023Morsø et al.Article
ApplicationObstetric care complaintsDenmark2023Walløe et al.Article
ApplicationCollege complaints against resident physiciansCanada2022Crosbie et al.Article
ApplicationGeneral hospital (Spanish-language)Peru2022del Río-Mendoza et al.Article
ApplicationEmergency-care compensation claims: hot spots and blind spotsDenmark2022Morsø et al.Article
ApplicationHCAT(GP) applied to real general-practice complaintsIreland2022O'Dowd et al.Article
ApplicationPost-operative complaints in spine surgeryUSA2022Ronald et al.Article
ApplicationNursing and midwifery complaints; HCAT in VHIMSAustralia2022Street, Russell & ConsidineArticle
ApplicationIntegrating complaints and staff incident reports, London teaching hospitalUK2022Van Dael et al.Article
ApplicationNational complaints-handling policy case studyEngland2022van Dael et al.Article
ApplicationPandemic-era complaints at a medical centerTaiwan2022Wang et al.Article
ApplicationTaxonomic review at Mayo Clinic hospitalsUSA2021Elias et al.Article
ApplicationPatient letters of complimentEngland2021Gillespie & ReaderArticle
ApplicationNeurosurgery: compliments and complaintsUK2021Jones et al.Article
ApplicationComplaint severity predicts hospital mortality, 59 NHS trustsUK2021Reader & GillespieArticle
ApplicationCompensation claims after emergency-care reorganisationDenmark2020Birkeland et al.Article
ApplicationENT departments at two teaching hospitalsUK2020Bujoreanu et al.Article
ReviewSystematic review of general-practice complaintsInternational2020O'Dowd et al.Article
ReviewRealist review: learning from complaints in healthcareUK2020van Dael et al.Article
ApplicationLinking complaints, incidents and adverse events (surgery)Netherlands2019de Vos et al.Article
ApplicationMetropolitan maternity serviceAustralia2019Nowotny et al.Article
ApplicationRadiology complaints, nine-year seriesNetherlands2019van den Berg et al.Article
ApplicationHot spots and blind spots, 56 NHS trusts (open data + R scripts)England2018Gillespie & ReaderArticle
ApplicationIntensive care vs general wards, academic medical centerTaiwan2018Jerng et al.Article
ApplicationImage-guided interventional radiologyUSA2018Salazar, Burk & AbujudehArticle
ApplicationOutpatient cancer care (Dana-Farber)USA2017Mack et al.Article
DevelopmentHCAT development and reliability testingUK2016Gillespie & ReaderArticle
ApplicationSerious complaints to the NSW healthcare commissionAustralia2016Harrison et al.Article
DevelopmentSystematic review: 88,069 complaints, three-domain taxonomyInternational2014Reader, Gillespie & RobertsArticle