The effect of using NHS number as the unique identifier for patients who self-harm: a multi-centre descriptive study

Table 1 Characteristics of self-harm patients associated with non-capture of NHS number: all three centres

		Oxford (N = 1000)			Manchester (N = 1000)			Leeds (N = 1000)
		N	%*	Statistic	N	%*	Statistic	N	%*	Statistic
Total attendances without NHS number		269	27		272	27		449	45
Gender	Female	159	26	X² = 1.6	143	25	X² = 2.0	241	42	X² = 4.1
	Male	110	29	df = 1	129	30	df = 1	208	49	df = 1
				P = 0.21			P = 0.16			P = 0.04
Age	Up to 24	109	32	X² = 10.6	92	28	X² = 2.3	144	46	X² = 4.7
	25 – 54	148	26	df = 2	168	28	df = 2	284	46	df = 2
	55 Plus	12	15	P = 0.005	11	19	P = 0.32	20	32	P = 0.10
Method of self-harm	Self-poisoning	206	27	X² = 0.8	219	27	X² = 0.0	307	40	X² = 41.5
	Self-injury	48	30	df = 2	45	27	df = 2	113	67	df = 2
	Both	15	25	P = 0.67	7	27	P = 1.0	29	45	P < 0.001
Time of presentation	9 am – 5 pm	80	27	X² = 0.01	92	32	X² = 4.1	129	48	X² = 1.5
	Out of hours	167	27	df = 1	177	25	df = 1	318	44	df = 1
				P = 0.91			P = 0.04			P = 0.22
Admitted to general	Yes	226	26	X² = 1.8	125	27	X² = 0.4	202	29	X² = 223.6
hospital	No	43	32	df = 1	96	25	df = 1	247	80	df = 1
				P = 0.18			P = 0.55			P < 0.001
No fixed abode	NFA	11	32	X² = 0.6	19	86	X² = 40.2	12	60	X² = 1.9
	Address	252	26	df = 1	250	26	df = 1	429	44	df = 1
				P = 0.43			P < 0.001			P = 0.17

Completeness of independent variables: data at all centres at least 97% complete for "gender", "age" and "method of harm"; "time of presentation" at least 85%; "admitted to general hospital" at least 92%; "no fixed abode" at least 86%.
* percentage of category by variable with no NHS number