Open Access Research article

Variation of health-related quality of life assessed by caregivers and patients affected by severe childhood infections

Wantanee Kulpeng12*, Vorasith Sornsrivichai2, Virasakdi Chongsuvivatwong2, Waranya Rattanavipapong1, Pattara Leelahavarong1, John Cairns3, Yoel Lubell4 and Yot Teerawattananon1

Author Affiliations

1 Health Intervention and Technology Assessment Program (HITAP), Ministry of Public Health, Nonthaburi, Thailand

2 Epidemiology Unit, Faculty of Medicine, Prince of Songkla University, Songkla, Thailand

3 London School of Hygiene and Tropical Medicine, London, United Kingdom

4 Mahidol Oxford Tropical Medicine Research Unit, Faculty of Tropical Medicine, Mahidol University, Bangkok, Thailand

For all author emails, please log on.

BMC Pediatrics 2013, 13:122  doi:10.1186/1471-2431-13-122

Published: 13 August 2013



The agreement between self-reported and proxy measures of health status in ill children is not well established. This study aimed to quantify the variation in health-related quality of life (HRQOL) derived from young patients and their carers using different instruments.


A hospital-based cross-sectional survey was conducted between August 2010 and March 2011. Children with meningitis, bacteremia, pneumonia, acute otitis media, hearing loss, chronic lung disease, epilepsy, mild mental retardation, severe mental retardation, and mental retardation combined with epilepsy, aged between five to 14 years in seven tertiary hospitals were selected for participation in this study. The Health Utilities Index Mark 2 (HUI2), and Mark 3 (HUI3), and the EuroQoL Descriptive System (EQ-5D) and Visual Analogue Scale (EQ-VAS) were applied to both paediatric patients (self-assessment) and caregivers (proxy-assessment).


The EQ-5D scores were lowest for acute conditions such as meningitis, bacteremia, and pneumonia, whereas the HUI3 scores were lowest for most chronic conditions such as hearing loss and severe mental retardation. Comparing patient and proxy scores (n = 74), the EQ-5D exhibited high correlation (r = 0.77) while in the HUI2 and HUI3 patient and caregiver scores were moderately correlated (r = 0.58 and 0.67 respectively). The mean difference between self and proxy-assessment using the HUI2, HUI3, EQ-5D and EQ-VAS scores were 0.03, 0.05, -0.03 and -0.02, respectively. In hearing-impaired and chronic lung patients the self-rated HRQOL differed significantly from their caregivers.


The use of caregivers as proxies for measuring HRQOL in young patients affected by pneumococcal infection and its sequelae should be employed with caution. Given the high correlation between instruments, each of the HRQOL instruments appears acceptable apart from the EQ-VAS which exhibited low correlation with the others.

Infection; Chronic conditions; Child; Health-related quality of life; Utility; Proxy