Table 3.
Mixed effects regression models estimating associations between intervention and participation on change in HbA1c
| Change in HbA1c (mmol/mol) | |||||||||
| Model 1 † | Model 2 † | Model 3 † | |||||||
|
| |||||||||
| Effect | B(95%CI) | SE | P | B(95% CI) | SE | P | B(95% CI) | SE | P |
|
| |||||||||
| Adapted-Family | −5 (−10, −1) | 2.3 | 0.029 | 2 (−5, 9) | 3.5 | 0.633 | −2 (−8, 5) | 3.4 | 0.655 |
| Exposure | −1 (−2, 0) | 0.4 | 0.013 | −1 (−1, 0) | 0.3 | 0.046 | |||
| Age | −0 (0, 0) | 0.1 | 0.254 | ||||||
| Male | 1 (−4, 6) | 2.5 | 0.645 | ||||||
| Married | 0 (−5, 5) | 2.6 | 0.995 | ||||||
| Employed | 0 (−5, 5) | 2.6 | 0.953 | ||||||
| Education | 1 (−2, 3) | 1.6 | 0.376 | ||||||
| Medication | −3 (−7, 2) | 2.4 | 0.279 | ||||||
|
| |||||||||
| Change in HbA1c (%) | |||||||||
| Model 1 † | Model 2 † | Model 3 † | |||||||
|
| |||||||||
| Effect | B(95%CI) | SE | P | B(95% CI) | SE | P | B(95% CI) | SE | P |
|
| |||||||||
| Adapted-Family | −0.5 (−0.9, −0.1) | 0.21 | 0.029 | 0.2 (−0.5, 0.8) | 0.32 | 0.633 | −0.1 (−0.8, 0.5) | 0.31 | 0.655 |
| Exposure | −0.1 (−0.2, 0.0) | 0.04 | 0.013 | −0.1 (−0.1, 0.0) | 0.03 | 0.046 | |||
| Age | −0.0 (−0.0, 0.0) | 0.01 | 0.254 | ||||||
| Male | 0.1 (−0.3, 0.6) | 0.23 | 0.645 | ||||||
| Married | 0.0 (−0.5, 0.5) | 0.24 | 0.995 | ||||||
| Employed | 0.0 (−0.5, 0.5) | 0.24 | 0.953 | ||||||
| Education | 0.1 (−0.2, 0.3) | 0.15 | 0.376 | ||||||
| Medication | −0.2 (−0.7, 0.2) | 0.22 | 0.279 | ||||||
Statistically significant P values are bolded (p<0.05). DSME, diabetes self-management education; HbA1c, glycated haemoglobin; SE=standard error.
†
Adjusted for baseline HbA1c.