, 16:O7 | Cite as

Understanding patterns of adverse events after surgery and their impact on recovery

  • Rachel L Nash
  • Barnaby C Reeves
  • Gianni D Angelini
  • Chris A Rogers
Oral presentation


Objective Measure Vital Sign Severe Complication Surgery Patient Preliminary Investigation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Clinical trial reports include data on adverse events experienced by study participants. Typically the frequency of each event is reported, but patterns of events occurring in one participant are rarely described. We explore associations between complications after surgery, and their impact on time-to-discharge as a first step to deriving an objective measure of recovery from surgery.


The occurrences of different complications have been explored in a cohort of cardiac surgery patients, all of whom participated in an RCT. Multiple correspondence analysis (MCA) and latent class analysis (LCA) were used to identify associations and underlying “classes” of individuals based on complication profiles.


Data on 1453 patients from 6 clinical trials were collated. Sixteen complications were investigated; 44% of patients were complication-free, 31% experienced one complication, 14% two, and 16% three or more. Preliminary investigations showed that patients who experienced more severe complications (e.g. stroke) often had other, less severe, complications as well. Using LCA, three classes were identified; the class labelled ‘poor recovery’ had a high probability of serious complications (> 30%). As expected, post-operative stay was longest in patients assigned to the ‘poor’ recovery group (median 16 days, 95% CI [13,20] versus 9 [8,10] and 6 [6,6] in the ‘moderate’ and ‘good’ recovery groups, respectively).


Many patients experience multiple complications. We have identified three classes with good face validity. The next step will be to investigate associations with routine data on vital signs (heart rate, temperature etc.), to identify which, if any can discriminate between recovery “classes”.

‘This work was carried out under an NIHR Research Methods Fellowship. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the HTA, NIHR, NHS or the Department of Health.’

Copyright information

© Nash et al. 2015

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Authors and Affiliations

  • Rachel L Nash
    • 1
  • Barnaby C Reeves
    • 1
  • Gianni D Angelini
    • 2
  • Chris A Rogers
    • 1
  1. 1.Clinical Trials and Evaluation Unit, School of Clinical SciencesUniversity of BristolBristolUK
  2. 2.Bristol Heart Institute, School of Clinical SciencesUniversity of BristolBristolUK

Personalised recommendations