Table 3. Changes in level and trend of malaria, non-malaria and overall visits at three health centres combined, results from a segmented linear regression model, unadjusted and adjusted results.
| Unadjusted | Adjusted ^ | |||||
|---|---|---|---|---|---|---|
| Malaria visits $ | Non-malaria visits | Overall visits | Malaria visits $ | Non-malaria visits | Overall visits | |
| Constant (β 0) | 207.7** | 106.0*** | 296.3*** | 204.0** | 106.3*** | 294.4*** |
| (70.9) | (23.3) | (46.5) | (70.8) | (23.7) | (45.5) | |
| Secular trend (β 1) | 19.2* | 5.6 | 32.7*** | 15.7 | 6.0 | 29.8*** |
| (8.9) | (3.2) | (6.3) | (8.9) | (3.4) | (6.5) | |
| Change in level after intervention starts (β 2) | -245.0*** | -39.8 | -427.9*** | -218.7** | -42.6 | -408.3*** |
| (65.9) | (27.1) | (54.0) | (62.9) | (28.3) | (54.3) | |
| Change in slope after intervention starts (β 3) | -27.1* | -1.8 | -32.3*** | -24.7* | -2.1 | -30.0*** |
| (11.0) | (3.5) | (7.0) | (11.1) | (3.6) | (7.0) | |
| Rainy season (β 4) | 51.8* | -6.1 | 42.8 | |||
| (23.7) | (14.6) | (28.1) | ||||
Standard errors in parenthesis
*p<0.05
** p<0.01
*** p<0.001
$model includes an adjustment for autocorrelation
^adjusted for bimodal rainy seasons.