Table.
Standardised case fatality rates for patients admitted for stroke, by hospital of admission during 1978-87 and 1988-97*
| Hospital
|
No of admissions
|
Case fatalities (95% CI)
|
|||
|---|---|---|---|---|---|
| In-hospital deaths <30 days
|
Deaths anywhere
|
||||
| <30 days
|
<90 days
|
<365 days
|
|||
| 1978-87 | |||||
| I† | 2 933 | 29.7 (27.6 to 31.8) | 32.5 (30.3 to 34.7) | 40.9 (38.4 to 43.5) | 50.3 (47.5 to 53.1) |
| II | 936 | 33.7 (30.0 to 37.5) | 37.5 (33.5 to 41.5) | 47.3 (42.8 to 51.8) | 57.0 (52.0 to 61.9) |
| III | 1 138 | 37.1 (33.2 to 41.0) | 41.2 (37.0 to 45.3) | 50.4 (45.8 to 55.0) | 58.6 (53.6 to 63.5) |
| IV | 1 422 | 41.7 (38.1 to 45.2) | 47.3 (43.5 to 51.1) | 54.6 (50.5 to 58.7) | 63.1 (58.7 to 67.5) |
| V | 566 | 44.8 (37.6 to 52.0) | 47.2 (39.8 to 54.5) | 58.3 (50.0 to 66.6) | 64.7 (56.1 to 73.3) |
| VI | 1 249 | 35.2 (31.9 to 38.5) | 36.5 (33.1 to 39.8) | 43.6 (39.9 to 47.3) | 53.3 (49.2 to 57.3) |
| VII | 1 187 | 43.4 (39.3 to 47.5) | 47.1 (42.9 to 51.3) | 54.5 (49.9 to 59.1) | 61.7 (56.9 to 66.5) |
| VIII | 2 104 | 34.5 (32.0 to 37.0) | 36.8 (34.1 to 39.4) | 44.6 (41.7 to 47.5) | 55.9 (52.7 to 59.1) |
| IX | 853 | 43.4 (38.2 to 48.7) | 45.2 (39.9 to 50.5) | 52.1 (46.4 to 57.7) | 60.4 (54.3 to 66.5) |
| X | 467 | 41.9 (35.7 to 48.1) | 44.8 (38.4 to 51.2) | 52.4 (45.4 to 59.4) | 59.3 (51.9 to 66.7) |
| Subtotal II-IX | 9 922 | 38.6 (37.4 to 39.9) | 41.8 (40.5 to 43.1) | 49.7 (48.3 to 51.1) | 58.8 (57.3 to 60.4) |
| Total‡ | 12 855 | 36.6 (35.5 to 37.7) | 39.7 (38.5 to 40.8) | 47.7 (46.4 to 48.9) | 56.9 (55.5 to 58.2) |
| Unadjusted case fatality rate | 35.4 (34.5 to 36.2) | 38.4 (37.5 to 39.2) | 45.9 (45.1 to 46.8) | 54.6 (53.7 to 55.5) | |
| χ2, df=162§ | 278.1 | 305.6 | 307.8 | 281.2 | |
| 1988-97 | |||||
| I† | 4 289 | 25.7 (24.2 to 27.3) | 28.6 (27.0 to 30.3) | 35.5 (33.7 to 37.3) | 44.8 (42.8 to 46.8) |
| II | 1 233 | 35.4 (32.1 to 38.7) | 36.4 (33.0 to 39.7) | 43.0 (39.3 to 46.7) | 51.9 (47.8 to 55.9) |
| III | 1 718 | 23.1 (20.7 to 25.4) | 36.3 (33.4 to 39.2) | 42.9 (39.7 to 46.1) | 52.4 (48.9 to 55.9) |
| IV | 2 624 | 29.9 (27.8 to 32.0) | 33.1 (30.8 to 35.3) | 41.5 (39.0 to 44.0) | 50.8 (48.1 to 53.6) |
| V and VI | 2 522 | 30.4 (28.3 to 32.6) | 33.0 (30.8 to 35.3) | 39.6 (37.2 to 42.1) | 45.9 (43.3 to 48.6) |
| VII and VIII | 3 657 | 32.2 (30.4 to 34.1) | 34.3 (32.4 to 36.2) | 42.0 (39.9 to 44.1) | 50.0 (47.7 to 52.2) |
| IX and X | 2 475 | 30.1 (27.9 to 32.2) | 34.0 (31.7 to 36.3) | 43.0 (40.4 to 45.5) | 50.8 (48.0 to 53.6) |
| XI | 1 357 | 28.7 (25.9 to 31.4) | 30.9 (28.0 to 33.9) | 37.6 (34.3 to 40.8) | 46.3 (42.7 to 49.9) |
| XII | 1 350 | 32.8 (29.7 to 35.9) | 34.7 (31.5 to 37.9) | 43.6 (40.0 to 47.2) | 52.8 (48.8 to 56.7) |
| Subtotal II-XII | 16 936 | 30.5 (29.6 to 31.3) | 34.0 (33.1 to 34.9) | 41.6 (40.6 to 42.6) | 49.9 (48.8 to 51.0) |
| Total‡ | 21 225 | 29.5 (28.8 to 30.2) | 32.9 (32.1 to 33.7) | 40.3 (39.5 to 41.2) | 48.9 (47.9 to 49.8) |
| Unadjusted case fatality rate | 29.7 (29.1 to 30.3) | 33.1 (32.5 to 33.8) | 40.7 (40.0 to 41.3) | 49.3 (48.6 to 49.9) | |
| χ2, df=144§ | 282.9 | 250.9 | 266.7 | 263.8 | |
Standardised for age group and sex by the direct method, with all patients admitted for stroke in the former Oxford region in 1978-1997 as the standard for making comparisons over time and between hospitals.
Differences between the teaching hospital (hospital I) and other hospitals combined at 30 days were 9.3% (95% confidence interval 7.3% to 11.3%, P<0.001) in 1978-87 and 5.4% (3.9% to 6.9%, P<0.001) in 1988-97. Differences between the teaching hospital and the other hospitals combined at 30-365 days (that is, excluding deaths in the first 30 days) were −0.8% (−2.4 to 0.8, P=0.31) in 1978-87 and −0.3% (−1.5 to 0.9, P=0.63) in 1988-97.
Between 1978-87 and 1988-97, the total case fatality rates at 30 days decreased by 6.8% (5.7% to 7.9%, P<0.001) and for deaths from 30-365 days after the acute event decreased by 1.2% (0.4% to 2.0%, P<0.01).
Obtained through logistic regression analysis. Provides an overall test of the variation in case fatality rates between all individual hospitals: comparisons between hospitals in all eight sets of case fatality rates were highly significant (P<0.001). Corresponding χ2 statistics that compared only the non-teaching hospitals were 187.1, 199.2, 206.9, and 180.4 for the four successive periods after admission for 1978-87 (df144) and 209.3, 166.8, 178.2, and 202.4 for 1988-97 (df126); all were significant (P<0.01 in seven cases and P<0.05 in one case). Chi-square statistics for variation in case fatality rates comparing all hospitals from 30 days to 365 days (that is, excluding deaths in the first 30 days) were non-significant in 1978-87 (183.2, df162, P=0.24) and 1988-97 (161.8, df144, P=0.29).