Table 1.

Completeness and quality of data from hospital episode statistics (values are numbers unless stated otherwise)

	Values for England financial year 2002-3 (range for strategic health authorities)
Total finished and unfinished consultant episodes available for analysis after removal of duplicates*	13 427 090
Total finished consultant episodes available for analysis after removal of duplicates†	12 692 451
Independent count of hospital activity (finished consultant episodes) submitted by NHS trusts	12 632 619
Surplus records submitted to hospital episode statistics	59 832
Continuous inpatient spells created from finished and unfinished episodes	12 016 995
% of finished consultant episodes ending in death in hospital with primary diagnosis missing or invalid	2.6 (0.3-6.0)
% of all finished consultant episodes used for hospital case fatality indicator with primary diagnosis missing or invalid	2.1 (0.2-6.1)
% of all episodes used for hospital case fatality indicator with missing or invalid codes in at least one hospital episode statistics field used for the indicator (similar figures for the stroke admissions and primary care indicators)	2.9 (0.6-8.4)
% of all episodes used for stroke case fatality and timely discharge indicators with missing or invalid codes in at least one hospital episode statistics field used for the indicator (similar figures for the readmissions indicator)	3.6 (0.7-9.3)
% of all episodes used for patterns of mental health care indicator with missing or invalid codes in at least one hospital episode statistics field used for the indicator	3.0 (0.6-8.5)

^*Includes regular day and night episodes.

^†Excludes regular day and night episodes.