Table 3.
Clinical and laboratory values at the end of the AAD diet and DASH-type diet, respectively, and changes (unit and percent) from AAD to DASH-type diet in all participants
| Variables | AAD | DASH-type | Unit changea | % Change | P-valueb |
|---|---|---|---|---|---|
| Body weight (kg) | 80 ± 16 | 79 ± 15 | −0.4 ± 1 | −0.5 ± 2 | 0.0002/0.010/0.009 |
| Diastolic blood pressure (mmHg) | 75 ± 7 | 74 ± 7 | −1.2 ± 4 | −1.4 ± 5 | 0.0003/0.004/0.003 |
| Systolic blood pressure (mmHg) | 117 ± 10 | 115 ± 9 | −2.3 ± 6 | −1.7 ± 5 | <0.0001/<0.0001/0.006 |
| Total cholesterol (mg/dl) | 186 ± 32 | 169 ± 31 | −16 ± 19 | −8 ± 9 | <0.0001 |
| LDL cholesterol (mg/dl) | 115 ± 28 | 103 ± 26 | −12 ± 15 | −10 ± 12 | <0.0001 |
| HDL cholesterol (mg/dl) | 56 ± 15 | 50 ± 13 | −5 ± 6 | −8 ± 10 | <0.0001 |
| Triglycerides (mg/dl) | 64 (49–85) | 70 (51–95) | 4 ± 22 | 8 ± 26 | 0.026/0.090/0.088 |
| Lipoprotein(a) (mg/dl) | 44 (22–80) | 58 (28–94) | 11 ± 11 | 24 ± 25 | <0.0001 |
| Apolipoprotein A-1 (mg/dl) | 136 ± 31 | 127 ± 29 | −9 ± 11 | −6 ± 8 | <0.0001 |
| Apolipoprotein B-100 (mg/dl) | 90 ± 20 | 85 ± 20 | −5 ± 10 | −6 ± 10 | <0.0001 |
| Glucose (mg/dl) | 93 ± 11 | 92 ± 10 | −0.9 ± 5 | −0.7 ± 5 | 0.027/0.045/0.045 |
| Insulin (μU/ml) | 12 ± 7 | 12 ± 7 | −0.2 ± 4 | 5 ± 44 | 0.559/0.731/0.750 |
Data at the end of each diet period (AAD and DASH-type) are shown as mean ± SD with the exception of triglycerides and Lp(a) levels which are shown as median (25th–75th percentiles). Data for changes (unit and percent) are shown as mean ± SD for all variables. Unit and percent changes were calculated using the data collected at the end (week five) of each diet period.
AAD, average American diet; DASH, Dietary Approaches to Stop Hypertension.
Unit changes are in kg, mmHg, and μU/ml for body weight, blood pressure, and insulin, respectively. For all cholesterol and apolipoprotein values, unit changes are shown in mg/dl.
P-values are for significance for absolute and percent changes. Three P-values from paired t test, mixed effect model with 1 random effect (subject), and mixed effect model with 2 random effects (subject and siblings). For mixed effect models, P-value is for treatment and period interaction. When period effect is not statistically significant, P-value for the treatment term is reported. If the three P-values were identical, a single value was reported. Since Lp(a) levels were not assessed at baseline, two repeated measures were included in the mixed effect model for longitudinal data analysis; for all other variables, three measurements were included.