Table 5.
Associations between breakfast consumption, meal frequency and glycaemic control.
| Adjusted means | B | 95% CI | p | ||
|---|---|---|---|---|---|
| Breakfast | No BF | ||||
| HbA1c < 59 mmol/mol (<7.5%) | 0.510 | 0.262–0.994 | 0.048 | ||
| HbA1c | 64.1 ± 0.5 | 66 ± 1.7 | 1.447 | −2.090–4.983 | 0.423 |
| Mean of the BG measurements | 8.2 ± 0.1 | 8.8 ± 0.2 | 0.609 | 0.105–1.112 | 0.018 |
| CV of the BG measurements | 0.41 ± 0.01 | 0.43 ± 0.02 | 0.019 | −0.015–0.053 | 0.280 |
| ≥8 meals | ≤5 meals | ||||
| HbA1c < 59 mmol/mol (<7.5%) | 0.741 | 0.472–1.164 | 0.194 | ||
| HbA1c | 62 ± 1.0 | 66 ± 0.9 | 3.944 | 1.254–6.633 | 0.004 |
| Mean of the BG measurements | 8.0 ± 0.1 | 8.6 ± 0.1 | 0.588 | 0.209–0.967 | 0.002 |
| CV of the BG measurements | 0.42 ± 0.01 | 0.40 ± 0.01 | −0.022 | −0.049–0.005 | 0.113 |
Models are adjusted for sex, diabetes duration, smoking, energy intake, physical activity, and mode of insulin administration. Logistic regression analysis for the dichotomised variable (HbA1c < 59 mmol/mol), and generalised linear regression analysis for the continuous variables. BF, breakfast; BG, blood glucose; CV, coefficient of variation. In the logistic regression analyses, breakfast = 0, no breakfast = 1; ≥ 8 meals = 0, ≤ 5 meals = 1.