Robert J. Hancox
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Robert J. Hancox has a total of 529 co-authors. Here are the top co-authors by number of shared publications:
- Richard Beasley
Medical Research Institute of New Zealand, Wellington, New Zealand
13 shared publications Malcolm R. Sears
Department of Medicine, McMaster University, Hamilton, Ontario, Canada
12 shared publicationsRichie Poulton
Dunedin Multidisciplinary Health and Development Research Unit, Department of Psychology, University of Otago, Dunedin, New Zealand
10 shared publications- Andrew Robert Gray
Biostatistics Centre University of Otago Dunedin New Zealand
8 shared publications Catherina L. Chang
Waikato Hospital, Hamilton, New Zealand
8 shared publications- Catherina L. Chang
Department of Respiratory Medicine, Waikato Hospital, Hamilton, New Zealand
7 shared publications
Latest Publications
- 5 September 2026
- Published by European Respiratory Society (ERS)
- Vol. 68 (suppl 70) , PA2589
Blood neutrophil-to-lymphocyte ratio and eosinophils as predictors of 10-year mortality after exacerbations of COPD
Background: Blood neutrophil-to-lymphocyte ratios (NLR) and eosinophils are markers of type 1 and type 2 airway inflammation in exacerbations of COPD, their indication of long-term prognosis is unclear. Aims: To assess associations between NLR, eosinophils and 10-year mortality after COPD hospitalisation, and phenotype persistence across consecutive exacerbations. Methods: 371 patients admitted to Waikato Hospital for COPD were categorised as high NLR (>3.5) and/or high eosinophils (>0.3×10^9/L). Kaplan-Meier curves and Cox proportional hazards were used to assess mortality. Results: Patients with high NLR (78% admissions) tended to have worse FEV1 and lower eosinophils. Patients with high eosinophils (14%), were younger and had lower NLR. Patients with high eosinophils had lower 10-year mortality (HR=0.66; p=0.017), whereas patients with a high NLR had higher mortality (HR=1.39, p=0.021). Associations were attenuated and not statistically significant after adjusting for age, FEV1% predicted and current smoking. 50% of patients with high eosinophils remained high at their next admission. Similarly, 81% patients with high NLR remained high at the next admission. erj;68/suppl_70/PA2589/F1 F1 F1 Conclusion: High NLR at COPD exacerbation are associated with a worse long-term prognosis, but high eosinophils indicate a better prognosis. These associations are only partly explained by age, severity of airway obstruction and smoking status. These inflammatory phenotypes tended to persist in subsequent exacerbations, but did not predict exacerbation risk.
- 5 September 2026
- Published by European Respiratory Society (ERS)
- Vol. 68 (suppl 70) , PA2589
Blood neutrophil-to-lymphocyte ratio and eosinophils as predictors of 10-year mortality after exacerbations of COPD
Background: Blood neutrophil-to-lymphocyte ratios (NLR) and eosinophils are markers of type 1 and type 2 airway inflammation in exacerbations of COPD, their indication of long-term prognosis is unclear. Aims: To assess associations between NLR, eosinophils and 10-year mortality after COPD hospitalisation, and phenotype persistence across consecutive exacerbations. Methods: 371 patients admitted to Waikato Hospital for COPD were categorised as high NLR (>3.5) and/or high eosinophils (>0.3×10^9/L). Kaplan-Meier curves and Cox proportional hazards were used to assess mortality. Results: Patients with high NLR (78% admissions) tended to have worse FEV1 and lower eosinophils. Patients with high eosinophils (14%), were younger and had lower NLR. Patients with high eosinophils had lower 10-year mortality (HR=0.66; p=0.017), whereas patients with a high NLR had higher mortality (HR=1.39, p=0.021). Associations were attenuated and not statistically significant after adjusting for age, FEV1% predicted and current smoking. 50% of patients with high eosinophils remained high at their next admission. Similarly, 81% patients with high NLR remained high at the next admission. erj;68/suppl_70/PA2589/F1 F1 F1 Conclusion: High NLR at COPD exacerbation are associated with a worse long-term prognosis, but high eosinophils indicate a better prognosis. These associations are only partly explained by age, severity of airway obstruction and smoking status. These inflammatory phenotypes tended to persist in subsequent exacerbations, but did not predict exacerbation risk.
- Open Access
- 2 September 2026
- Published by Elsevier in Public Health
- Vol. 259
Health and behavioural risks for adverse pregnancy and offspring outcomes across peak male fertility: Prevalence in the Australian longitudinal study on male health
- 2 September 2026
- Published by Elsevier in Public Health
- Vol. 259
Open AccessHealth and behavioural risks for adverse pregnancy and offspring outcomes across peak male fertility: Prevalence in the Australian longitudinal study on male health
- 5 March 2026
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 67 (3) , 2502359
Shaping lung health from the start: early-life trajectories and lifelong risk
Extract In their landmark study of the natural history of chronic airflow obstruction among London working men, Fletcher and Peto [1] identified two factors leading to impaired lung function in later life. The first and most important was cigarette smoking causing an accelerated age-related decline in forced expiratory volume in 1 s (FEV1). However, they also demonstrated that starting with a low lung function in early adulthood enhanced the risk. Thus, someone who already has lung function at the lower end of the normal range in their mid-20s is likely to experience disability much earlier than someone who starts with above average lung function, despite a similar rate of lung function decline. This second factor, having been largely ignored for many years, has been reaffirmed in more recent studies and is now highlighted in international guidelines [2, 3]. Attention has therefore turned to how we might maximise lung growth in childhood and adolescence. Shareable abstract Life-course trajectories of lung function are established in childhood, but show considerable potential for improvement in the first years of life https://bit.ly/4i4mqwC
- 5 March 2026
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 67 (3) , 2502359
Shaping lung health from the start: early-life trajectories and lifelong risk
Extract In their landmark study of the natural history of chronic airflow obstruction among London working men, Fletcher and Peto [1] identified two factors leading to impaired lung function in later life. The first and most important was cigarette smoking causing an accelerated age-related decline in forced expiratory volume in 1 s (FEV1). However, they also demonstrated that starting with a low lung function in early adulthood enhanced the risk. Thus, someone who already has lung function at the lower end of the normal range in their mid-20s is likely to experience disability much earlier than someone who starts with above average lung function, despite a similar rate of lung function decline. This second factor, having been largely ignored for many years, has been reaffirmed in more recent studies and is now highlighted in international guidelines [2, 3]. Attention has therefore turned to how we might maximise lung growth in childhood and adolescence. Shareable abstract Life-course trajectories of lung function are established in childhood, but show considerable potential for improvement in the first years of life https://bit.ly/4i4mqwC
- Top Cited Papers
- 21 January 2026
- Published by Elsevier in The Lancet Respiratory Medicine
- Vol. 14 (3) , 203-214
Bisoprolol to prevent adverse cardiac events (PACE) in COPD: a multicentre, double-blind, randomised, controlled, phase 3 trial
- 21 January 2026
- Published by Elsevier in The Lancet Respiratory Medicine
- Vol. 14 (3) , 203-214
Top Cited PapersBisoprolol to prevent adverse cardiac events (PACE) in COPD: a multicentre, double-blind, randomised, controlled, phase 3 trial
- 23 December 2025
- Published by Springer Nature in Clinical Child and Family Psychology Review
- Vol. 29 (2) , 276-298
Preconception Predictors of Next Generation Early Relational Health: A Living Review of Prospective Cohort Studies
Intergenerational studies suggest that the developmental foundations of children’s early relational health are seeded well before conception. Here, we present studies that report on associations between factors in the parental preconception life course (across childhood, adolescence, and young adulthood), and indicators of the child’s early relational ecology (e.g., infant-to-parent attachment, parent-to-infant bonds). We searched MEDLINE, PsycINFO and Embase databases for peer-reviewed articles, published in English, that reported on prospectively assessed factors at any stage of a parent’s life course prior to conception, and associations with offspring relational health from conception to end of age 3 years. No date restrictions were imposed. A total of 13,594 articles were screened of which 30 met inclusion criteria, reporting data from 17 separate cohorts. Next generation child relational health was assessed using both coded observations and parent reports, with most studies focused on the mother–child relationship (70%). Most preconception predictors were assessed at the individual level (63%) or within the family microsystem (53%) and were measured during adolescence (60%). Few studies assessed other microsystems or broader ecological systems, with no studies investigating workplaces, online interactions, or cultural belief systems as predictors of next generation relational health. We conclude that longitudinal cohorts that track individuals into parenthood can provide critical insights into the preconception origins of early relational health; however, the existing body of prospective studies is heterogeneous and reflects a nascent field of enquiry. We present five recommendations for future investment in public health approaches that promote early relational health.
- 23 December 2025
- Published by Springer Nature in Clinical Child and Family Psychology Review
- Vol. 29 (2) , 276-298
Preconception Predictors of Next Generation Early Relational Health: A Living Review of Prospective Cohort Studies
Intergenerational studies suggest that the developmental foundations of children’s early relational health are seeded well before conception. Here, we present studies that report on associations between factors in the parental preconception life course (across childhood, adolescence, and young adulthood), and indicators of the child’s early relational ecology (e.g., infant-to-parent attachment, parent-to-infant bonds). We searched MEDLINE, PsycINFO and Embase databases for peer-reviewed articles, published in English, that reported on prospectively assessed factors at any stage of a parent’s life course prior to conception, and associations with offspring relational health from conception to end of age 3 years. No date restrictions were imposed. A total of 13,594 articles were screened of which 30 met inclusion criteria, reporting data from 17 separate cohorts. Next generation child relational health was assessed using both coded observations and parent reports, with most studies focused on the mother–child relationship (70%). Most preconception predictors were assessed at the individual level (63%) or within the family microsystem (53%) and were measured during adolescence (60%). Few studies assessed other microsystems or broader ecological systems, with no studies investigating workplaces, online interactions, or cultural belief systems as predictors of next generation relational health. We conclude that longitudinal cohorts that track individuals into parenthood can provide critical insights into the preconception origins of early relational health; however, the existing body of prospective studies is heterogeneous and reflects a nascent field of enquiry. We present five recommendations for future investment in public health approaches that promote early relational health.
- Open Access
- 17 November 2025
- Published by Springer Nature in Communications Medicine
- Vol. 5 (1)
Quantifying the impact of early life growth adversity on later life health
Background: Early-life growth adversity is important to later-life health, but precision assessment in adulthood is challenging. We evaluated whether the difference between attained and genotype-predicted adult height (“height-GaP”) would associate with prospectively ascertained early-life growth adversity and later-life all-cause and cardiovascular mortality. Methods: Data were first analyzed from the Avon Longitudinal Study of Parents and Children (ALSPAC; n = 4582; 56/43% female/male) and UKBiobank ( n = 483,385; 54/46% female/male). Genotype-predicted height was calculated using a multi-ancestry polygenic height score. Height-GaP was calculated as the difference between measured and genotype-predicted adult height. Early-life growth conditions were ascertained prospectively via standardized procedures (ALSPAC) and mortality via death register (UKBiobank). Regression models examined: (i) adult height-GaP as the outcome with early-life growth conditions as predictors; and (ii) mortality as the outcome with adult height-GaP as predictor. All models were adjusted for age, sex, genotype-predicted height and genetic ancestry. Analyses were replicated in the Dunedin Multidisciplinary Health and Development Study (DMHDS; n = 855; 49/51% female/male) and the Multi-Ethnic Study of Atherosclerosis (MESA; n = 6352; 52/48% female/male). Results: Here we show that among ALSPAC participants (median [IQR] age: 24 [18-25] years at height-GaP assessment), lower gestational age at birth, greater pre- and post-natal deprivation indices, tobacco smoke exposure and less breastfeeding are associated with larger adult height-GaP deficit ( p < 0.01). Among UKBiobank participants (mean ± SD age: 56 ± 8 years at height-GaP assessment), height-GaP deficit is associated with death from all-causes (adjusted hazard ratio comparing highest-to-lowest height-GaP deficit quartile [aHR]: 1.25 95%CI: 1.21–1.29), atherosclerotic cardiovascular disease (aHR: 1.32 95%CI: 1.23–1.42) and coronary heart disease (aHR: 1.64 95%CI: 1.49-1.81). Early- and later-life height-GaP associations replicate in DMHDS and MESA. Conclusions: This study introduces a precision index of early-life growth adversity deployable in adulthood to investigate the developmental origins of longevity and improve health equity across the life course.
- 17 November 2025
- Published by Springer Nature in Communications Medicine
- Vol. 5 (1)
Open AccessQuantifying the impact of early life growth adversity on later life health
Background: Early-life growth adversity is important to later-life health, but precision assessment in adulthood is challenging. We evaluated whether the difference between attained and genotype-predicted adult height (“height-GaP”) would associate with prospectively ascertained early-life growth adversity and later-life all-cause and cardiovascular mortality. Methods: Data were first analyzed from the Avon Longitudinal Study of Parents and Children (ALSPAC; n = 4582; 56/43% female/male) and UKBiobank ( n = 483,385; 54/46% female/male). Genotype-predicted height was calculated using a multi-ancestry polygenic height score. Height-GaP was calculated as the difference between measured and genotype-predicted adult height. Early-life growth conditions were ascertained prospectively via standardized procedures (ALSPAC) and mortality via death register (UKBiobank). Regression models examined: (i) adult height-GaP as the outcome with early-life growth conditions as predictors; and (ii) mortality as the outcome with adult height-GaP as predictor. All models were adjusted for age, sex, genotype-predicted height and genetic ancestry. Analyses were replicated in the Dunedin Multidisciplinary Health and Development Study (DMHDS; n = 855; 49/51% female/male) and the Multi-Ethnic Study of Atherosclerosis (MESA; n = 6352; 52/48% female/male). Results: Here we show that among ALSPAC participants (median [IQR] age: 24 [18-25] years at height-GaP assessment), lower gestational age at birth, greater pre- and post-natal deprivation indices, tobacco smoke exposure and less breastfeeding are associated with larger adult height-GaP deficit ( p < 0.01). Among UKBiobank participants (mean ± SD age: 56 ± 8 years at height-GaP assessment), height-GaP deficit is associated with death from all-causes (adjusted hazard ratio comparing highest-to-lowest height-GaP deficit quartile [aHR]: 1.25 95%CI: 1.21–1.29), atherosclerotic cardiovascular disease (aHR: 1.32 95%CI: 1.23–1.42) and coronary heart disease (aHR: 1.64 95%CI: 1.49-1.81). Early- and later-life height-GaP associations replicate in DMHDS and MESA. Conclusions: This study introduces a precision index of early-life growth adversity deployable in adulthood to investigate the developmental origins of longevity and improve health equity across the life course.
- Open Access
- 29 September 2025
- Published by European Respiratory Society (ERS) in ERJ Open Research
- Vol. 12 (1)
Cost-effectiveness of as-needed budesonide-formoterol in adults with mild asthma: the Novel START trial
Background: The Global Initiative for Asthma recommends as-needed combination inhaled corticosteroid-formoterol reliever for mild asthma. Its cost-effectiveness is uncertain outside double-blind regulatory trials. The study objectives were to assess the cost-effectiveness of as-needed budesonide-formoterol versus maintenance budesonide plus as-needed salbutamol, and versus as-needed salbutamol, in averting exacerbations in adults with mild asthma during 12-month follow-up of the New Zealand participants (n=551) of the open-label Novel START multicountry clinical trial. Methods: New Zealand health-system and societal perspectives were adopted using 2025 resource costs. Healthcare utilisation comprised electronically monitored inhaler use (with multiple imputation for missing data), asthma-related pharmaceuticals, primary-care visits, emergency-department visits and hospitalisations. Regression methods were used for cost-effectiveness analysis. Scenarios included generic substitution and prescription-charge exemption. Results: As-needed budesonide-formoterol, compared with maintenance budesonide plus as-needed salbutamol, is more effective at reducing severe exacerbations and provides an annual health-system cost saving of NZD 12 (95% CI 0–28) per patient, increasing to NZD 22 (95% CI −15–58) if a societal perspective is adopted. Compared with as-needed salbutamol alone, as-needed budesonide-formoterol costs the health system NZD 38 and NZD 111, respectively, to prevent an exacerbation and a severe exacerbation. From a societal perspective it is cost-saving NZD 70 (95% CI 2–134). Findings were robust to scenario analyses. Conclusion: For patients with mild asthma, as-needed budesonide-formoterol 200/6 μg is cost-saving from health-system, patient and societal perspectives, while improving outcomes, dominating maintenance budesonide plus as-needed salbutamol. Compared with as-needed salbutamol, as-needed budesonide-formoterol is cost-effective from a health-system perspective and dominates from a societal perspective.
- 29 September 2025
- Published by European Respiratory Society (ERS) in ERJ Open Research
- Vol. 12 (1)
Open AccessCost-effectiveness of as-needed budesonide-formoterol in adults with mild asthma: the Novel START trial
Background: The Global Initiative for Asthma recommends as-needed combination inhaled corticosteroid-formoterol reliever for mild asthma. Its cost-effectiveness is uncertain outside double-blind regulatory trials. The study objectives were to assess the cost-effectiveness of as-needed budesonide-formoterol versus maintenance budesonide plus as-needed salbutamol, and versus as-needed salbutamol, in averting exacerbations in adults with mild asthma during 12-month follow-up of the New Zealand participants (n=551) of the open-label Novel START multicountry clinical trial. Methods: New Zealand health-system and societal perspectives were adopted using 2025 resource costs. Healthcare utilisation comprised electronically monitored inhaler use (with multiple imputation for missing data), asthma-related pharmaceuticals, primary-care visits, emergency-department visits and hospitalisations. Regression methods were used for cost-effectiveness analysis. Scenarios included generic substitution and prescription-charge exemption. Results: As-needed budesonide-formoterol, compared with maintenance budesonide plus as-needed salbutamol, is more effective at reducing severe exacerbations and provides an annual health-system cost saving of NZD 12 (95% CI 0–28) per patient, increasing to NZD 22 (95% CI −15–58) if a societal perspective is adopted. Compared with as-needed salbutamol alone, as-needed budesonide-formoterol costs the health system NZD 38 and NZD 111, respectively, to prevent an exacerbation and a severe exacerbation. From a societal perspective it is cost-saving NZD 70 (95% CI 2–134). Findings were robust to scenario analyses. Conclusion: For patients with mild asthma, as-needed budesonide-formoterol 200/6 μg is cost-saving from health-system, patient and societal perspectives, while improving outcomes, dominating maintenance budesonide plus as-needed salbutamol. Compared with as-needed salbutamol, as-needed budesonide-formoterol is cost-effective from a health-system perspective and dominates from a societal perspective.
- Open Access
- 21 August 2025
- Published by European Respiratory Society (ERS) in ERJ Open Research
- Vol. 12 (2)
Acute effects of low-dose bisoprolol on lung function and blood pressure in COPD patients
Background and objective: Recent observational data suggest that cardioselective β-blockers like bisoprolol are safe and beneficial for patients with COPD. However, the acute effects of bisoprolol on lung and cardiovascular function in these patients is unclear, a gap that this study aimed to address. Methods: This was a subanalysis of pre-randomisation screening visit data from the ongoing Preventing Adverse Cardiac Events (PACE) in COPD randomised controlled trial. If all other eligibility criteria were met, participants were orally administered an unblinded 1.25 mg tablet of bisoprolol. Post-bronchodilator spirometry, heart rate and blood pressure were monitored at 0, 30 (cardiovascular parameters only), 60 and 120 min. For this subanalysis, respiratory intolerance was defined as a decrease in forced expiratory volume in 1 s (FEV 1 ) (L) ≥200 mL and ≥12% from the 0-min FEV 1 (L) value; and cardiovascular intolerance was defined as systolic blood pressure (SBP) falling below 100 mmHg at 1 or 2 h. Results: Of 359 consented participants, 292 conducted the test-dose procedure. 13 (4.5%) were respiratory intolerant and six (2.1%) were cardiovascular intolerant at 1 or 2 h. No participant was intolerant for both. There was no significant difference in FEV 1 (L) or SBP at baseline At 120 min the intolerant group's mean FEV 1 had significantly decreased to 1.05 L (95% CI 0.86–1.25 L; p95% of COPD patients.
- 21 August 2025
- Published by European Respiratory Society (ERS) in ERJ Open Research
- Vol. 12 (2)
Open AccessAcute effects of low-dose bisoprolol on lung function and blood pressure in COPD patients
Background and objective: Recent observational data suggest that cardioselective β-blockers like bisoprolol are safe and beneficial for patients with COPD. However, the acute effects of bisoprolol on lung and cardiovascular function in these patients is unclear, a gap that this study aimed to address. Methods: This was a subanalysis of pre-randomisation screening visit data from the ongoing Preventing Adverse Cardiac Events (PACE) in COPD randomised controlled trial. If all other eligibility criteria were met, participants were orally administered an unblinded 1.25 mg tablet of bisoprolol. Post-bronchodilator spirometry, heart rate and blood pressure were monitored at 0, 30 (cardiovascular parameters only), 60 and 120 min. For this subanalysis, respiratory intolerance was defined as a decrease in forced expiratory volume in 1 s (FEV 1 ) (L) ≥200 mL and ≥12% from the 0-min FEV 1 (L) value; and cardiovascular intolerance was defined as systolic blood pressure (SBP) falling below 100 mmHg at 1 or 2 h. Results: Of 359 consented participants, 292 conducted the test-dose procedure. 13 (4.5%) were respiratory intolerant and six (2.1%) were cardiovascular intolerant at 1 or 2 h. No participant was intolerant for both. There was no significant difference in FEV 1 (L) or SBP at baseline At 120 min the intolerant group's mean FEV 1 had significantly decreased to 1.05 L (95% CI 0.86–1.25 L; p95% of COPD patients.
- 14 July 2025
- Published by Wiley in Respirology
- Vol. 30 (10) , 1003-1004
Cheap, Convenient, and Quality Healthcare—Pick Any Two
- 14 July 2025
- Published by Wiley in Respirology
- Vol. 30 (10) , 1003-1004
Cheap, Convenient, and Quality Healthcare—Pick Any Two
- Open Access
- 21 June 2025
- Published by Wiley in Internal Medicine Journal
- Vol. 55 (8) , 1339-1349
Variation in clinical presentation, complications and outcomes for Māori and Pacific peoples among hospitalised adults with COVID‐19 in 2022, Aotearoa New Zealand
Background: Pacific region‐specific data on the clinical course of COVID‐19 are limited. We aimed to describe clinical features and outcomes from Aotearoa New Zealand patients, focusing on Māori and Pacific peoples.Methods: We conducted a retrospective cohort study among adults (≥16 years) hospitalised due to COVID‐19 at 11 hospitals from January to May 2022. We included all Māori and Pacific patients and every second non‐Māori, non‐Pacific (NMNP) patient using data from chart review and national datasets.Results: Of 2319 patients, 582 (25%) were Māori, 914 (39%) Pacific peoples and 862 NMNP (median age 52, 57 and 63 years respectively). Vaccination coverage (≥2 doses) was 73.4% (n = 437) for Māori, 76.7% (n = 701) for Pacific peoples (n = 701) and 84.8% (n = 731) for NMNP. Among 832 (35.9%) with complications, Māori had a greater risk than NMNP of acute kidney injury (risk ratio (RR) 1.87, P < 0.001), cardiac arrhythmia (RR = 1.60, P = 0.023), shock (RR = 2.64, P = 0.005), myocardial infarction (RR 2.21, P = 0.042), cardiac arrest (RR 2.68, P = 0.046) and acute respiratory distress syndrome (RR = 2.81, P = 0.008). Pacific patients experienced a greater risk than NMNP of acute kidney injury (RR = 2.18, P < 0.001) and pneumonia (RR = 1.32, P = 0.047) and a lower risk of thromboembolism (RR = 0.35, P = 0.004) and myocarditis/pericarditis (RR = 0.23, P = 0.003). During admission, 23 (3.3%) Māori, 36 (3.9%) Pacific and 28 (3.2%) NMNP patients died, with no difference in age‐standardised mortality.Conclusions: The clinical course of patients hospitalised by COVID‐19 varied between ethnic groups, likely reflecting differential access to social determinants of health. Healthcare services that respond to this variability are needed to achieve the highest attainable health for all.
- 21 June 2025
- Published by Wiley in Internal Medicine Journal
- Vol. 55 (8) , 1339-1349
Open AccessVariation in clinical presentation, complications and outcomes for Māori and Pacific peoples among hospitalised adults with COVID‐19 in 2022, Aotearoa New Zealand
Background: Pacific region‐specific data on the clinical course of COVID‐19 are limited. We aimed to describe clinical features and outcomes from Aotearoa New Zealand patients, focusing on Māori and Pacific peoples.Methods: We conducted a retrospective cohort study among adults (≥16 years) hospitalised due to COVID‐19 at 11 hospitals from January to May 2022. We included all Māori and Pacific patients and every second non‐Māori, non‐Pacific (NMNP) patient using data from chart review and national datasets.Results: Of 2319 patients, 582 (25%) were Māori, 914 (39%) Pacific peoples and 862 NMNP (median age 52, 57 and 63 years respectively). Vaccination coverage (≥2 doses) was 73.4% (n = 437) for Māori, 76.7% (n = 701) for Pacific peoples (n = 701) and 84.8% (n = 731) for NMNP. Among 832 (35.9%) with complications, Māori had a greater risk than NMNP of acute kidney injury (risk ratio (RR) 1.87, P < 0.001), cardiac arrhythmia (RR = 1.60, P = 0.023), shock (RR = 2.64, P = 0.005), myocardial infarction (RR 2.21, P = 0.042), cardiac arrest (RR 2.68, P = 0.046) and acute respiratory distress syndrome (RR = 2.81, P = 0.008). Pacific patients experienced a greater risk than NMNP of acute kidney injury (RR = 2.18, P < 0.001) and pneumonia (RR = 1.32, P = 0.047) and a lower risk of thromboembolism (RR = 0.35, P = 0.004) and myocarditis/pericarditis (RR = 0.23, P = 0.003). During admission, 23 (3.3%) Māori, 36 (3.9%) Pacific and 28 (3.2%) NMNP patients died, with no difference in age‐standardised mortality.Conclusions: The clinical course of patients hospitalised by COVID‐19 varied between ethnic groups, likely reflecting differential access to social determinants of health. Healthcare services that respond to this variability are needed to achieve the highest attainable health for all.