Introduction Recent studies have shown that readmission rates in paediatrics are increasing. Hospital readmissions form a part of our clinical care quality indicators, outcome frameworks and clinical dashboards. This study tries to evaluate the characteristics of paediatric readmissions in a standard district general hospital in United Kingdom. We focused on the first 72 hours from discharge as we felt that this time frame was most likely to include readmissions that were avoidable
Methods Children who were readmitted, between 1 May and 30th of June 2011, within 72 hours after discharge, were identified using the admissions database. Their medical records were reviewed to determine the characteristics and reasons for the readmission.
Results Of the 626 paediatric medical acute hospital admissions that occurred during this two-month period, 27 children were readmitted. 33 (5.3% of the initial admissions) episodes of representation occurred in these 27 children. 20 (74%) children represented with the same diagnosis, 3 (11.1%) presented with new, unrelated illness, 2 (7.4%) children were planned readmissions following the acute admission and 2 (7.4%) represented due to parental anxiety. In 11 (33%) of episodes, patients were reassured and discharged. In 22 (66%) episodes observation was required. 17 of 33 (51%) episodes of representations were during out of hours. 25 patients were readmitted once, 1 patient was readmitted twice and another was readmitted 4 times. 25 (93%) of the patients presented with acute illness – gastroenteritis, respiratory tract infections and head injury were the commonest. Only two patients had underlying chronic health issues. 70 % of these patients were given open access to the ward as a safety net at the time of initial discharge.
Conclusion Six (18%) of our readmissions were unavoidable. Providing open access as a safety net was assumed to provide reassurance and reduce representations, and yet 70 % of our patients who represented had this facility. We recommend performing a larger study to look at whether such open access makes a difference to the risk of representation