Investigating increased admissions to neonatal intensive care in England between 1995 and 2006: data linkage study using Hospital Episode Statistics

Table 8 Changes in the number of births in HES data over time

HES data set ^a	1995 ^b				2006 (<26 weeks) ^c				Percentage change ^d	2006 (<27 weeks) ^e
	Live	Still	Not	Total	Live	Still	Not	Total		Live	Still	Not	Total
	births	births	known		births	births	known			births	births	known
Reported	621	213	33	867	887	121	180	1188	37 %	1856	201	278	2535
“True”	396	16	10	422	699	127	187	1013	140 %	1158	213	291	1662
“Confirmed”	282	13	5	300	412	81	75	568	89 %	684	134	114	932
“Misclassified”	339	200	28	567	475	40	105	620	9 %	1172	67	364	1603

Changes in the number of births in Hospital Episode Statistics (HES) data between 1995 and 2006: reported, “true”, “confirmed” and “misclassified” data
^aFor each year, data sets were created based upon : a) gestational age as reported in the original HES data; b) only the “true” data identified by the data linkage exercise (i.e. contained in both HES and EPICure); c) HES data “confirmed” by the “true” data; and, d) “misclassified” data, which are those reported by HES but that were not identified as “true” during data linkage
^bIn 1995, data were available from March 1^st – December 31^st for babies of <26 completed weeks gestational age
^cComparison data sets from 2006 were created to include babies born between 1^st March and 31^st December at less than 26 weeks gestational age
^dThe total percentage increase in all births is presented
^eThe complete data sets from 2006 include births of <27 completed weeks gestational age from the entire year

ISSN: 1471-2288