Background
Methods
Setting and study participants
Study design, participant recruitment, and experimental allocation
Study interventions: development and implementation of the study algorithm and protocol
Intervention implementation
Data sources and study variables
Data analysis
Study protocol modifications
Results
Study participants
Total study population (n, health units) | Intervention group (n, health units) | Control group (n of health units) | |
---|---|---|---|
Population served by the LPHAs | |||
Large (Population > 400,000) | 7 | 4 | 3 |
Medium (Population 150,000 to 399,999) | 5 | 3 | 2 |
Small (Population < 150,000 | 3 | 2 | 1 |
Emergency department syndromic system | |||
ACES system | 13 | 9 | 4 |
LPHA specific system | 2 | 0 | 2 |
Total | 15 | 9 | 6 |
Dataset development via inductive thematic analysis
Characterization of aberrant events received by public health unit study participants
Total (n, alerts) | Intervention group (n, alerts) | Control group (n, alerts) | |
---|---|---|---|
Alerts received by local public health agency study participants | 1969 | 942 | 1027 |
Year 1 | 801 | 409 | 392 |
Year 2 | 1168 | 533 | 635 |
Alerts checked for data misclassification, data stream issues, or error-related issues | 305 (15%) | 224 (24%) | 81 (8%) |
Year 1 | 193 (24%) | 156 (38%) | 37 (9%) |
Year 2 | 112 (10%) | 68 (13%) | 44 (7%) |
Alerts checked for epidemiological clustering | 1538 (78%) | 664 (70%) | 874 (85%) |
Year 1 | 635 (79%) | 253 (62%) | 382 (97%) |
Year 2 | 903 (77%) | 411 (77%) | 492 (77%) |
Alerts warranting public health action | 714 (36%) | 165 (18%) | 549 (53%) |
Year 1 | 319 (40%) | 11 (3%) | 308 (79%) |
Year 2 | 395 (34%) | 154 (29%) | 241 (38%) |
Missing | 380 (19%) | 168 (18%) | 212 (21%) |
Steps taken to investigate syndromic surveillance aberrant events
LPHA responses to syndromic surveillance aberrant events
Total (n, alerts) | Intervention group (n, alerts) | Control group (n, alerts) | |
---|---|---|---|
Number of alerts | 1,969 | 942 | 1,027 |
Year 1 | 801 | 409 | 392 |
Year 2 | 1,168 | 533 | 625 |
Any action | 946 (48%) | 382 (41%) | 564 (55%) |
Year 1 | 397 (50%) | 74 (18%) | 323 (82%) |
Year 2 | 549 (47%) | 308 (58%) | 241 (38%) |
Watchful waiting | 647 (33%) | 263 (28%) | 384 (37%) |
Year 1 | 275 (34%) | 18 (4%) | 257 (66%) |
Year 2 | 372 (32%) | 245 (46%) | 127 (20%) |
Internal notification | 297 (15%) | 146 (15%) | 151 (15%) |
Year 1 | 109 (14%) | 55 (13%) | 54 (14%) |
Year2 | 188 (16%) | 91 (17%) | 97 (15%) |
External notification | 171 (9%) | 84 (9%) | 87 (8%) |
Year 1 | 66 (8%) | 5 (1%) | 61 (16%) |
Year 2 | 105 (9%) | 79 (15%) | 26 (4%) |
Predictors of outcomes related to public health responses to syndromic alerts
Outcomes | ||||||
---|---|---|---|---|---|---|
(Odds Ratio, 95% Confidence Interval) | ||||||
Any response | Any response excluding watchful waiting | Watchful waiting | Internal notification | External notification | ||
Intervention | Year 1 | 0.024 (<0.01, 0.51)a
| 0.43 (0.06, 2.92) | 0.04 (0.001, 1.58) | 2.94 (0.12, 72.2) | 0.04 (<0.01, 0.37)a
|
Year 2 | 0.20 (0.02, 2.38) | 0.69 (0.16, 3.03) | 0.19 (<0.01, 55.8) | 0.98 (0.08, 12.4) | 1.27 (0.19, 8.34) | |
Control | Reference | Reference | Reference | Reference | Reference | |
Medium LPHA | Year 1 | 0.83 (0.15, 45.9) | 0.91 (0.06, 1.39) | 4.25 (0.04, 423) | 0.35 (0.01, 21.2) | 3.66 (0.18, 74.0) |
Year 2 | 0.29 (0.012, 7.12) | 0.45 (0.06, 3.2) | 11.7 (0.009, 15,000) | 0.78 (0.04, 17.5) | 1.18 (0.10, 13.9) | |
Large LPHA | Reference | Reference | Reference | Reference | Reference | |
Small LPHA | Year 1 | 0.352 (0.13, 9.30) | 1.02 (0.12, 8.44) | 1.28 (0.02, 69.2) | 0.61 (0.02, 15.9) | 3.83 (0.29, 50.5) |
Year 2 | 1.65 (0.11, 24.0) | 0.56 (0.11, 2.86) | 32.1 (0.05, 19,908) | 0.19 (0.01, 3.33) | 1.62 (0.21, 12.6) | |
Large LPHA | Reference | Reference | Reference | Reference | Reference | |
Sustained alert | Year 1 | 1.81 (1.10, 2.99)a
| 2.06 (1.23, 3.45)a
| 0.62 (0.24, 1.56) | 2.21 (1.27, 3.84)a
| 7.83 (3.2, 19.2)a
|
Year 2 | 0.24 (0.16, 0.35)a
| 0.92 (0.61, 1.39) | 0.15 (0.10, 0.23)a
| 0.75 (0.47, 1.19) | 0.59 (0.37, 0.94)a
| |
New alert | Reference | Reference | Reference | Reference | Reference |
Outcomes | |||
---|---|---|---|
(Odds Ratio, 95% Confidence Interval) | |||
Any response including watchful waiting | Any response excluding watchful waiting | ||
Intervention | Year 1 | 0.020 (<0.01, 0.18)a
| 0.41 (0.06, 2.78) |
Year 2 | 0.18 (0.016, 2.12) | 0.72 (0.18, 3.00) | |
Control | Reference | Reference | |
Medium LPHA | Year 1 | 0.38 (0.02, 7.33) | 0.79 (0.06, 10.2) |
Year 2 | 0.21 (0.01, 5.41) | 0.52 (0.08, 3.44) | |
Large LPHA | Reference | Reference | |
Small LPHA | Year 1 | 1.12 (0.01, 12.7) | 1.30 (0.14, 11.9) |
Year 2 | 2.49 (0.16, 39.1) | 0.47 (0.09, 2.43) | |
Large LPHA | Reference | Reference | |
Sustained alert | Year 1 | 1.87 (1.13,3.08)a
| 2.07 (1.24, 3.46)a
|
Year 2 | 0.24 (0.16, 0.35)a
| 0.92 (0.61, 1.39) | |
New alert | Reference | Reference | |
Views syndromic surveillance as a primary data source | Year 1 | 47.5 (4.2, 536)a
| 2.10 (0.26, 16.6) |
Year 2 | 3.86 (0.28, 52.9) | 0.60 (0.12, 2.88) | |
Secondary | Reference | Reference |