You are here

Article Text

Article menu

Download PDFPDF

Oral Presentations
78 Sex differences in intensive care mortality during infancy and childhood: a study of English linked administrative data
Free
  1. Ofran Almossawi1,
  2. Simon Nadel2,
  3. Katie Harron3,
  4. Richard Feltbower4,
  5. Bianca De Stavola3
  1. 1Great Ormond Street Hospital for Children NHS Foundation Trust and University College London, UK
  2. 2Imperial College Healthcare NHS Trust, UK
  3. 3UCL Great Ormond Street Institute of Child Health, UK
  4. 4University of Leeds (PICANet), UK

Abstract

Introduction Infant and child mortality are important indicators of population health. Despite their sizeable reductions over the last two decades, distinct inequalities remain, as child mortality rates vary not only by age and income status, but also by sex, with higher mortality rates generally found in males.

In the population of children admitted to Paediatric Intensive Care Units (PICU), our previous research on infants showed that mortality in PICU is higher for females relative to males, despite more male infants being admitted to PICU.

Aim To study the reversal in PICU mortality rates for males and females

Methods Using data from Paediatric Intensive Care Audit Network (PICANet) and Hospital Admission Episodes, we created a linked cohort of 64,495 children born in England between 2002 and 2019, who had at least one PICU admission between January 2010 and December 2019. The linked cohort contains maternal, birth, and children’s hospital admission records (prior to PICU admission), and mortality records from the Office of National Statistics.

Restricting the focus on the PICU population creates associations between predictors of admission (including sex) that can lead to bias. We used causal graphs (figure 1) to identify the direct and indirect paths linking sex and mortality in PICU. To overcome the high dimensionality of the identified potential confounders, we used machine learning methods to specify the propensity score function used to estimate the causal effect of being female on PICU mortality. We used doubly robust methods and targeted learning to overcome bias resulting from potential model mis-specification.

Results The sex specific death rate in PICU was higher for females, risk difference of 0.35% (95%CI: 0.10%, 0.61%), figure 2. The Mortality gap continued to diverge for up to 5 years after discharge from PICU, logrank P=0.001, (figure 3).

Conclusion Female sex is a risk for higher PICU mortality.

Abstract 78 Figure 1
Abstract 78 Figure 1
Abstract 78 Figure 2
Abstract 78 Figure 2
Abstract 78 Figure 3
Abstract 78 Figure 3

Acknowledgements for funding or support This research was funded by an NIHR fellowship

https://doi.org/10.1136/bmjpo-2023-GOSH.5

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.