Table 4.
Crude and adjusted multilevel models of the association between the provision of the YFS programme and clinic visit score
| Clinic visit score | ||||
|---|---|---|---|---|
|
|
||||
| Mean (SD) | Regression coefficient | 95% CI | p | |
| Univariate multilevel model | ||||
| YFS provided | ||||
| No | 0.06 (1.47) | Reference | ||
| Yes | −0.13 (1.82) | −0.12 | −0.86, 0.62 | 0.748 |
| Multivariable multilevel model | ||||
| YFS provided | ||||
| No | 0.06 (1.47) | Reference | ||
| Yes | −0.13 (1.82) | −0.18 | −0.95, 0.60 | 0.656 |
| Simulated client gender | ||||
| Female | 0.81 (1.06) | Reference | ||
| Male | −0.93 (1.68) | −1.52 | −2.65, −0.38 | 0.009 |
| Healthcare worker age | ||||
| 20–29 | 1.38 (0.67) | |||
| 30–39 | 0.02 (1.37) | −0.59a | −1.15, −0.02 | 0.041 |
| 40+ | −0.72 (2.01) | |||
For ordered categorical exposures, where tests for trend performed better than categorical tests in univariate, multilevel models, the variable was fitted as a continuous variable for use in multivariable, multilevel models.