Characteristics of deaths occurring in hospitalised children: changing trends

Despite a gradual shift in the focus of medical care among terminally ill patients to a palliative model, studies suggest that many children with life-limiting chronic illnesses continue to die in hospital after prolonged periods of inpatient admission and mechanical ventilation.

To (1) examine the characteristics and location of death among hospitalised children, (2) investigate yearwise trends in these characteristics and (3) test the hypothesis that professional ethical guidance from the UK Royal College of Paediatrics and Child Health (1997) would lead to significant changes in the characteristics of death among hospitalised children.

Routine administrative data from one large tertiary-level UK children's hospital was examined over a 7-year period (1997-2004) for children aged 0-18 years. Demographic details, location of deaths, source of admission (within hospital vs external), length of stay and final diagnoses (International Classification of Diseases-10 codes) were studied. Statistical significance was tested by the Kruskal-Wallis analysis of ranks and median test (non-parametric variables), chi(2) test (proportions) and Cochran-Armitage test (linear trends).

Of the 1127 deaths occurring in hospital over the 7-year period, the majority (57.7%) were among infants. The main diagnoses at death included congenital malformations (22.2%), perinatal diseases (18.1%), cardiovascular disorders (14.9%) and neoplasms (12.4%). Most deaths occurred in an intensive care unit (ICU) environment (85.7%), with a significant increase over the years (80.1% in 1997 to 90.6% in 2004). There was a clear increase in the proportion of admissions from in-hospital among the ICU cohort (14.8% in 1998 to 24.8% in 2004). Infants with congenital malformations and perinatal conditions were more likely to die in an ICU (OR 2.42, 95% CI 1.65 to 3.55), and older children with malignancy outside the ICU (OR 6.5, 95% CI 4.4 to 9.6). Children stayed for a median of 13 days (interquartile range 4.0-23.25 days) on a hospital ward before being admitted to an ICU where they died.

A greater proportion of hospitalised children are dying in an ICU environment. Our experience indicates that professional ethical guidance by itself may be inadequate in reversing the trends observed in this study.