Table 2.
Characteristics of cohort over study period (June 2010-December 2018) and before and after implementation of the Transitions Program, including both index and non-index hospital admissions. Values are percentages unless stated otherwise (numbers in parentheses denote the range across all 21 study hospitals)
| Characteristics | Total | Pre-implementation | Post-implementation | P value | Standardized mean difference |
|---|---|---|---|---|---|
| No of hospital admissions | 1 584 902 | 1 161 452 | 423 450 | ||
| No of patients | 753 587 | 594 053 | 266 478 | ||
| Inpatient* | 82.8 (69.7-90.6) | 84.4 (73.1-90.7) | 78.5 (57.7-90.3) | <0.001 | −0.151 |
| Observation | 17.2 (9.4-30.3) | 15.6 (9.3-26.9) | 21.5 (9.7-42.3) | <0.001 | 0.151 |
| Inpatient stay <24 hours† | 5.2 (3.3-6.7) | 5.1 (3.8-6.5) | 5.6 (1.8-9.1) | <0.001 | 0.041 |
| Transport-in‡ | 4.5 (1.4-8.7) | 4.5 (1.7-8.8) | 4.5 (0.4-8.4) | 0.56 | −0.001 |
| Mean age | 65.3 (62.2-69.8) | 65.1 (61.9-69.6) | 65.8 (62.8-70.4) | <0.001 | 0.038 |
| Men | 47.5 (43.4-53.8) | 47.0 (42.4-53.5) | 48.9 (45.3-54.9) | <0.001 | 0.037 |
| Kaiser Foundation health plan member | 93.5 (75.3-97.9) | 93.9 (80.0-98.0) | 92.5 (61.7-97.6) | <0.001 | −0.052 |
| Met strict membership definition§ | 80.0 (63.4-84.9) | 80.6 (67.6-85.5) | 78.5 (51.3-83.8) | <0.001 | −0.053 |
| Met regulatory definition§ | 61.9 (47.2-69.7) | 63.9 (50.2-72.2) | 56.5 (38.7-66.6) | <0.001 | −0.152 |
| Admission through emergency department | 70.4 (56.7-82.0) | 68.9 (56.0-80.3) | 74.4 (58.4-86.6) | <0.001 | 0.121 |
| Median Charlson comorbidity score¶ | 2.0 (2.0-3.0) | 2.0 (2.0-2.0) | 3.0 (2.0-3.0) | <0.001 | 0.208 |
| Charlson comorbidity score ≥4 | 35.2 (29.2-40.7) | 33.2 (28.2-39.8) | 40.9 (33.0-46.2) | <0.001 | 0.161 |
| Mean COPS2** | 45.6 (39.1-52.4) | 43.5 (38.4-51.5) | 51.2 (39.7-55.8) | <0.001 | 0.159 |
| COPS2 ≥65 | 26.9 (21.5-32.0) | 25.3 (21.0-31.6) | 31.1 (22.5-35.4) | <0.001 | 0.129 |
| Mean admission LAPS2†† | 58.6 (48.0-67.6) | 57.6 (47.4-65.8) | 61.3 (50.2-72.8) | <0.001 | 0.092 |
| Mean discharge LAPS2 | 46.7 (42.5-50.8) | 46.3 (42.5-50.8) | 47.6 (42.3-52.9) | <0.001 | 0.039 |
| LAPS2 ≥110 | 12.0 (7.8-16.0) | 11.6 (7.5-15.2) | 12.9 (8.3-18.4) | <0.001 | 0.039 |
| Full code at discharge | 84.4 (77.3-90.5) | 84.5 (77.7-90.5) | 83.9 (75.9-90.5) | <0.001 | −0.016 |
| Mean length of stay (days) | 4.8 (3.9-5.4) | 4.9 (3.9-5.4) | 4.7 (3.9-5.6) | <0.001 | −0.034 |
| Discharge disposition‡‡: | 0.082 | ||||
| Regular discharge | 72.7 (61.0-86.2) | 73.3 (63.9-85.9) | 71.0 (52.1-86.9) | <0.001 | |
| Home health | 16.1 (6.9-23.3) | 15.2 (6.9-22.6) | 18.5 (7.0-34.5) | <0.001 | |
| Regular skilled nursing facility | 9.9 (5.9-14.3) | 10.0 (6.0-15.2) | 9.5 (5.6-12.4) | <0.001 | |
| Custodial skilled nursing facility | 1.3 (0.7-2.5) | 1.5 (0.8-2.7) | 0.9 (0.4-1.8) | <0.001 | |
| Hospice referral | 2.6 (1.7-4.4) | 2.6 (1.7-4.6) | 2.7 (1.5-4.0) | <0.001 | 0.007 |
| Outcomes§§: | |||||
| Inpatient mortality | 2.8 (2.1-3.3) | 2.8 (2.1-3.3) | 2.8 (1.8-3.3) | 0.17 | −0.003 |
| 30 day mortality | 6.0 (4.0-7.3) | 6.1 (4.1-7.6) | 5.9 (3.9-6.8) | <.01 | −0.006 |
| Any readmission | 14.5 (12.7-17.2) | 14.3 (12.3-17.3) | 15.1 (13.3-17.0) | <0.001 | 0.021 |
| Any non-elective readmission | 12.4 (10.4-15.4) | 12.2 (10.2-15.5) | 13.1 (10.8-15.4) | <0.001 | 0.029 |
| Non-elective inpatient readmission | 10.5 (8.2-12.6) | 10.4 (8.1-12.8) | 10.8 (8.6-12.9) | <0.001 | 0.012 |
| Non-elective observation readmission | 2.4 (1.4-3.7) | 2.2 (1.2-3.4) | 3.0 (1.9-5.6) | <0.001 | 0.049 |
| 30 day post-discharge mortality | 4.0 (2.6-5.2) | 4.1 (2.7-5.4) | 3.9 (2.3-4.9) | <0.001 | −0.007 |
| Composite outcome | 15.2 (12.9-18.8) | 15.0 (12.9-19.1) | 15.8 (13.3-18.0) | <0.001 | 0.023 |
COPS2=comorbidity point score, version 2; LAPS2=laboratory based acute physiology score, version 2.
Rates employ hospital admission episodes (which can include linked stays for patients who were transported) as the denominator.
Hospital episodes where patients transitioned from observation to inpatient status are classified as inpatients.
Refers to patients whose linked hospital admission episode began at a hospital not owned by Kaiser Foundation Hospitals. These hospital admissions, involving 60 326 patients, had increased inpatient (4.9%) and 30 day (9.2%) mortality, compared with 2.8% and 6.0% in the rest of the Kaiser Permanente Northern California cohort.
The Healthcare Effectiveness Data and Information Set (HEDIS) membership definition restricts the denominator to patients with continuous health plan membership in the 12 months preceding and the 30 days after hospital discharge, with a maximum gap in coverage of 45 days in the preceding 12 months. The public reporting definition only includes patients meeting HEDIS membership criteria and excludes hospital admissions for observation and inpatient admissions with length of stay <24 hours.
Scores were computed using the methodology of Deyo et al.43
COPS2 is assigned based on all diagnoses incurred by a patient in the 12 months preceding the index hospital admission. The univariate relationship of COPS2 with 30 day mortality is: 0-39, 1.7%; 40-64, 5.2%; ≥65, 9.0%.
LAPS2 is assigned based on a patient’s worst vital signs, pulse oximetry, neurological status, and 16 laboratory test results in the preceding 24 hours (discharge LAPS2) or 72 hours (admission LAPS2). The univariate relationship of an admission LAPS2 with 30 day mortality is: 0-59, 1.0%; 60-109, 5.0%; ≥110, 13.7%; for discharge LAPS2, the relationship is 0-59, 2.2%; 60-109, 8.1%; ≥110, 20.5%.
Disposition among patients discharged alive from hospital. Hospice referral is independent of discharge disposition.
Only patients who survived to discharge are included in the post-discharge outcomes. Non-elective readmissions are those which began in the emergency department, or were experienced by patients with an ambulatory care sensitive condition or with an admission LAPS2 score ≥60, as described in Escobar et al.21 The composite outcome corresponds to non-elective readmission to hospital or death within 30 days after discharge. These figures do not include transports-in (see ‡).