Use this URL to cite or link to this record in EThOS:
Title: Mechanisms underlying between-hospital variation in mortality after emergency laparotomies in England and Wales : a structure-process analysis
Author: Oliver, C. M.
ISNI:       0000 0004 7231 528X
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Emergency laparotomies are highly invasive abdominal operations that are performed commonly across the globe for potentially life-threatening conditions. Up to 18% of patients die within the first month of surgery and the sequelae may represent significant burdens to patients, healthcare systems and wider societies long beyond the operative period. Recent observations of marked between-hospital variation in mortality after emergency laparotomy offer opportunities to improve the quality of care and survival of these patients across the globe. However, the causes of between-hospital variation are poorly understood and methods for identifying high-risk patients poorly evidenced. The aims of this thesis are to explore the complex interactions between organisational structures, processes of care and patient-level risk in order to determine the contributions of modifiable factors to patient outcomes after emergency laparotomy. My research comprises three parts: Firstly, univariate analysis of data submitted to the National Emergency Laparotomy Audit (NELA) first organisational and patient audits in order to identify and characterise variation in structural provisions for and delivery of care to emergency laparotomy patients. Secondly, a systematic review to identify the best validated risk assessment tools for emergency laparotomy, informing the selection of patient risk factors included in NELA's first patient audit and my subsequent analyses. Finally, statistical modelling to identify casemix adjusted between-hospital variation in postoperative mortality; and multivariable and mixed effects modelling to identify and compare the effects of processes of care and organisational structures associated with postoperative mortality, controlling for patient-level factors. Data submitted by participants at 190 hospitals to the first NELA organisational audit demonstrated variation in the provision of recommended structures for the care of emergency general surgery (EGS) patients. Provisions were more comprehensive at large and tertiary surgical referral centres. A systematic review identified 20 studies assessing 25 risk assessment tools in adult emergency laparotomy cohorts. APACHE II and P-POSSUM were the most widely studied prognostic models, but poor data reporting precluded comparisons of performance. POSSUM data items were included in the first NELA patient audit. Following exclusions, the first NELA patient cohort comprised 20,183 patients. Overall inpatient 30-day mortality was 11.3%, marked between-hospital casemix variation was demonstrated and delivery of processes of care varied considerably between patient subgroups and between hospitals, but was poorly characterised by measured hospital characteristics. Following the derivation and internal validation of a novel casemix adjustment model, substantial between-hospital variation in casemix adjusted mortality was demonstrated in the first NELA patient audit cohort. Multiple logistic regression modelling identified only three processes as independent predictors of postoperative mortality in the NELA patient cohort: postoperative critical care admission (odds ratio (OR) 1.6 (95%CI 1.4-1.8, p<0.005)), preoperative risk documentation (OR 1.1 (1.1-1.3, p<0.05)) and postoperative review of older patients (>70 years) by a medicine for the care of the older person (MCOP) physician (OR 0.3 (0.2-0.4 p<0.005)). Collectively, patient risk factors modelled up to 27% of the overall variation in mortality. Finally, mixed effects analysis demonstrated significant between-hospital variation in inpatient 30-day mortality that persisted after controlling for patient-level risk factors and perioperative processes of care. Hospital size and specialty (tertiary GI surgical referral centre) status modelled a small but significant proportion of this variation. Mortality rates were significantly higher at the smallest hospitals and significantly lower at specialty centres. Further work is required to evaluate whether comprehensive risk evaluation to inform the targeted delivery of augmented care bundles to high-risk patients can improve quality of care and postoperative survival and reduce the costs associated with emergency laparotomies. In the context of ageing populations and policy discussions regarding the reconfiguration of EGS services, the effect of MCOP input and associations with hospital size and specialist status merit urgent investigation.
Supervisor: Moonesinghe, S. R. ; Grocott, M. P. W. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available