Table 2 Effects of training plus collaborative improvement on data quality: results of segmented regression modelling
|  | Completeness Estimate (95% confidence interval) %-points | Accuracy Discordance between data sources†Estimate (95% confidence interval) %-points | ||
|---|---|---|---|---|
All fields | All clinically-relevant fields | Malaria cases (OPD vs. report) | Test positivity rate (lab vs. report) | |
Baseline | 15.0 (7.7, 22.2)*** | 30.5 (27.8, 33.1)*** | 6.7 (0.8, 12.6)* | 20.1 (13.1, 27.2)*** |
Baseline slope (per month) | 0.7 (− 0.4, 1.7) | 0.9 (0.6, 1.3)*** | 0.1 (− 0.8, 0.9) | − 0.6 (− 1.6, 0.4) |
Immediate change after TCI | 70.4 (60.8, 80.0)*** | 68.2 (64.6, 71.7)*** | − 0.9 (− 8.7, 6.9)†| − 15.8 (− 25.1, − 6.5)**†|
Change in slope after TCI (per month) | − 1.6 (− 2.8, − 0.3)** | − 1.0 (− 1.5, − 0.6)*** | − 0.3 (− 1.4, 0.7)†| 0.7 (− 0.6, 1.9)†|
TCI single effect (absolute difference at follow-up mid-point) | 60.1 (46.9, 73.2)*** | 61.6 (56.6, 66.7)* | – | – |
TCI single effect (relative difference at follow-up mid-point) | – | – | − 47.4 (− 177.0, 82.2) | − 57.4 (− 105.5, − 9.3)** |