Use of concentration indices of Hospital Episode Statistics data to monitor inequality in English children and young people (1999–2000 to 2009–10)

Abstract

Listen

Translate article

BackgroundNew methods are needed to monitor inequality in child health. The National Health Service (NHS) reforms introduced a statutory duty to reduce inequality at a time when UNICEF predicts a substantial increase in child poverty. The government believes that the previous focus on infant mortality was too narrow and that inequality should be measured across a range of indicators that matter most for children and young people. We propose that the concentration index of hospital activity could be a valuable aid to monitor one aspect of inequality with routinely collected data. As illustration, inequality trends in inpatient activity among children and young people for 10 years are analysed. MethodsHospital Episode Statistics data from 1999–2000 and 2009–10 were accessed via the NHS Information Centre. Activity was measured by finished consultant episodes and socioeconomic status by index of multiple deprivation decile, derived from the patient's postcode. The proportion of inpatient activity accounted for by each deprivation decile was calculated, stratifying by sex and age (<1, 1–4, 5–9, 10–14, and 15–19 years). Age-specific population estimates and deprivation scores were then accessed from the Office of National Statistics for all 6781 medium super output areas (MSOAs) in England. Linking and ranking of these data allowed estimation of the proportion of age-specific populations within each deprivation decile. We then calculated concentration indices and standard errors in Microsoft Excel using formulae published by the World Bank (appendix). FindingsComparing the concentration indices, inequality in inpatient care increased for infants (boys in 1999: −0·046, 95% CI −0·042 to −0·050; boys in 2009: −0·071, −0·067 to −0·075; girls in 1999: −0·040, −0·036 to −0·044; girls in 2009: −0·067, −0·063 to −0·071). For children aged 1–4 years, inequality was unchanged for boys (1999: −0·067, −0·065 to −0·069; 2009: −0·066, −0·064 to −0·068) but increased for girls (1999: −0·064, −0·063 to −0·066; 2009: −0·070, −0·068 to −0·072). Inequality decreased in all other groups, most substantially in girls aged 15–19 years (1999: −0·245, −0·243 to −0·247; 2009: −0·183, −0·181 to −0·185). InterpretationThe negative concentration indices suggest that children and young people from more deprived areas account for a greater proportion of inpatient care than do those living in more affluent areas. These data show converging inequality trends in different ages of childhood and adolescence during the 10 year period. Inequality in the need for hospital treatment was lowest in the early years, but increased during the 10 years. Inequality in late adolescence was much greater than at younger ages, especially for girls, but significantly reduced. Despite the policy focus on outcomes that matter most for children and young people, financial constraints have reduced the available data. Concentration indices use routinely collected, robust data and can be calculated at national, regional, or local level. They can conveniently be adapted to assess inequality in specific activity—eg, asthma, injuries, or alcohol-related admissions. Although they are affected by access factors as well as objective patient need, they might be a valuable aid for public health professionals and commissioners who wish to reduce inequality, contain costs of acute care, or interpret regional variation in hospital activity, particularly when used in conjunction with the recently published NHS Atlas of Variation in Healthcare for Children and Young People. FundingNone.