Identification of risk factors for hospital admission using multiple-failure survival models: a toolkit for researchers

Table 2 Extracts from the HCS-HES database corresponding to participants with different patterns of follow-up

Extract 2: Intervals at risk and indicators of failure events in the linked HCS-HES survival file
ID	Gender	Clinic	Start	End	Admission	Death	Failure	Risk_set
1	Male	17-Sep-00	17-Sep-00	31-Mar-10	0	0	.	1
2	Female	14-Jan-00	14-Jan-00	23-Jun-07	0	1	1	1
3	Female	23-Feb-01	23-Feb-01	06-Mar-06	1	0	1	1
3	Female	23-Feb-01	09-Mar-06	31-Mar-10	0	0	.	2
4	Male	20-May-01	20-May-01	08-Apr-03	1	0	1	1
4	Male	20-May-01	27-Apr-03	24-May-03	1	0	1	2
4	Male	20-May-01	05-Jun-03	31-Mar-10	0	0	0	3
5	Female	18-Sep-00	18-Sep-00	28-Jan-04	1	0	1	1
5	Female	18-Sep-00	15-Feb-04	13-Jan-05	1	0	1	2
5	Female	18-Sep-00	16-Jan-05	01-Mar-05	0	1	1	3

'ID' represents the individual the data correspond to
‘Clinic’ specifies the date of the HCS baseline clinic. This was the date that each individual first entered the study and, therefore, the first time they were regarded as being at risk of admission/death
'Start' specifies the beginning of the time interval when the individual was at risk of admission/death. This can be the date the individual attended the clinic or the date of discharge from a previous admission
'End' specifies the end of the time interval when the individual was at risk of admission/death. This can be the end of follow-up or the date of an admission/death
'Admission' and 'Death' indicate the occurrence of these events at the end of the risk interval and 'Failure' indicates the occurrence of admission/death
In Table 1 and 2, ID numbers and dates of admission, discharge and death are fictitious and are purely for illustrative purposes

ISSN: 1471-2288