PT - JOURNAL ARTICLE AU - Y Thakker AU - T A Sheldon AU - R Long AU - R MacFaul TI - Paediatric inpatient utilisation in a district general hospital. AID - 10.1136/adc.70.6.488 DP - 1994 Jun 01 TA - Archives of Disease in Childhood PG - 488--492 VI - 70 IP - 6 4099 - http://adc.bmj.com/content/70/6/488.short 4100 - http://adc.bmj.com/content/70/6/488.full SO - Arch Dis Child1994 Jun 01; 70 AB - Paediatric inpatient utilisation in a district general hospital was studied for 20 general practices covering a population of 26,433 children. The factors influencing the rate and route of admission (general practitioner (GP) or accident and emergency department) were analysed for 894 emergency non-traumatic admissions over a 12 month period. The overall rate of acute, nontraumatic admission was 33.8/1000; 35% of these admissions were via the accident and emergency department. Asthma was the most common reason for admission (16.1%); 56.9% of the admissions resulted from respiratory tract illness and 44% were for an infective illness. There was a significant variation in the route and rate of admission across practices. Admission rates ranged from 10 to 70/1000 children under 15 and the proportion via the accident and emergency department from 19% to 85%. The proportion of admissions via the accident and emergency department for each practice was highly negatively correlated with the number of GPs in the practice, the number of children under 15 registered, and positively correlated with the unemployment rate attributed to the list. Using multiple logistic regression analysis, the risk of being admitted via the accident and emergency department relative to GP admission was shown to be higher for older children (odds ratio for each year of age 1.05) and less for children registered with large practices with more GPs (odds ratio for each extra GP 0.36) or practices with more children under 15 (odds ratio per extra child 0.9991). Access to hospital as measured by isochrone bars and social characteristics of the ward of residence of each child admitted were not associated with the route of admission. The admission rate for each practice was positively, but not statistically significantly, associated with the unemployment rate attributed to the list, the unemployment rate of the ward where the practice was located, and the percentage of admissions via accident and emergency, and negatively associated with the percentage of the list under 15 years.