Table 3.
Primary and secondary outcomes involving energy expenditure, physical activity, and metabolic hormones*
| Variables | No | Mean (SD) pre-randomization start of trial | Change: average(midpoint of test phase, end of test phase)−start of trial | ||
|---|---|---|---|---|---|
| Mean (95% CI) | P value between groups | Trend estimate (95% CI)† | |||
| Primary outcome: total energy expenditure (kcal/d) | |||||
| Intention-to-treat: | |||||
| High | 54 | 2640 (62) | −19 (−104 to 66) | 0.002 | 52 (23 to 82) |
| Moderate | 52 | 2504 (65) | 71 (−12 to 155) | ||
| Low | 56 | 2713 (64) | 190 (109 to 270) | ||
| Per protocol: | |||||
| High | 38 | 2711 (77) | −102 (−201 to −2) | <0.001 | 69 (36 to 103) |
| Moderate | 39 | 2577 (72) | 29 (−64 to 123) | ||
| Low | 43 | 2758 (70) | 176 (87 to 265) | ||
| Resting energy expenditure (kcal/d) | |||||
| Intention-to-treat: | |||||
| High | 54 | 1603 (24) | 34 (10 to 57) | 0.47 | 5 (−3 to 13) |
| Moderate | 51 | 1576 (25) | 46 (23 to 69) | ||
| Low | 56 | 1615 (24) | 54 (32 to 76) | ||
| Per protocol: | |||||
| High | 38 | 1601 (28) | 20 (−8 to 48) | 0.18 | 8 (−1 to 18) |
| Moderate | 38 | 1597 (27) | 28 (2 to 54) | ||
| Low | 43 | 1608 (26) | 53 (28 to 78) | ||
| Total physical activity (counts/d, 000s)‡ | |||||
| Intention-to-treat: | |||||
| High | 54 | 476.6 (23.3) | −26.3 (−52.0 to −0.6) | 0.13 | 4.8 (−3.9 to 13.6) |
| Moderate | 52 | 463.8 (24.9) | −42.4 (−67.7 to −17.1) | ||
| Low | 55 | 495.8 (23.9) | −6.9 (−31.0 to 17.1) | ||
| Per protocol: | |||||
| High | 38 | 493.2 (28.8) | −29.1 (−59.0 to 0.7) | 0.17 | 4.4 (−5.6 to 14.3) |
| Moderate | 39 | 481.3 (27.0) | −48.3 (−76.3 to −20.3) | ||
| Low | 42 | 521.3 (26.3) | −11.6 (−38.2 to 14.9) | ||
| Moderate to vigorous intensity physical activity (min/d)§ | |||||
| Intention-to-treat: | |||||
| High | 54 | 31.6 (2.6) | −3.6 (−6.3 to −0.9) | 0.09 | 0.7 (−0.2 to 1.6) |
| Moderate | 52 | 31.3 (2.7) | −4.8 (−7.5 to −2.1) | ||
| Low | 55 | 30.0 (2.6) | −0.9 (−3.4 to 1.6) | ||
| Per protocol: | |||||
| High | 38 | 33.4 (3.0) | −4.3 (−7.4 to −1.1) | 0.06 | 0.9 (−0.1 to 2.0) |
| Moderate | 39 | 33.0 (2.8) | −5.2 (−8.1 to −2.2) | ||
| Low | 42 | 32.2 (2.8) | −0.5 (−3.3 to 2.3) | ||
| Sedentary time (min/d)¶ | |||||
| Intention-to-treat: | |||||
| High | 54 | 592.1 (14.2) | 8.6 (−7.7 to 25.0) | 0.12 | −2.7 (−8.3 to 2.9) |
| Moderate | 52 | 604.7 (14.8) | 20.9 (4.8 to 37.0) | ||
| Low | 55 | 597.0 (14.6) | −2.3 (−17.6 to 13.0) | ||
| Per protocol: | |||||
| High | 38 | 593.6 (17.2) | 2.1 (−17.7 to 22.0) | 0.31 | 0.9 (−5.8 to 7.5) |
| Moderate | 39 | 611.0 (16.1) | 21.4 (2.8 to 40.0) | ||
| Low | 42 | 589.4 (15.7) | 5.6 (−12.0 to 23.1) | ||
| Skeletal muscle work efficiency at 10 W (%)** | |||||
| Intention-to-treat: | |||||
| High | 53 | 12.2 (0.3) | −0.1 (−0.8 to 0.5) | 0.66 | 0.1 (−0.1 to 0.3) |
| Moderate | 51 | 11.7 (0.4) | −0.0 (−0.6 to 0.6) | ||
| Low | 55 | 12.2 (0.3) | 0.3 (−0.3 to 0.9) | ||
| Per protocol: | |||||
| High | 38 | 12.1 (0.4) | −0.1 (−0.9 to 0.6) | 0.46 | 0.1 (−0.1 to 0.4) |
| Moderate | 38 | 11.9 (0.4) | −0.0 (−0.7 to 0.6) | ||
| Low | 42 | 12.2 (0.4) | 0.5 (−0.2 to 1.1) | ||
| Ghrelin (pg/mL and % change)†† | |||||
| Intention-to-treat: | |||||
| High | 54 | 693.2 (51.4) | -4.9 (−8.4 to −1.2) | 0.02 | −1.9 (−3.2 to −0.6) |
| Moderate | 51 | 640.1 (49.9) | −8.7 (−12.0 to −5.3) | ||
| Low | 56 | 598.2 (45.4) | −11.8 (−14.8 to −8.6) | ||
| Per protocol: | |||||
| High | 38 | 689.5 (63.0) | −5.9 (−10.1 to −1.5) | 0.02 | −2.1 (−3.7 to −0.6) |
| Moderate | 38 | 620.6 (52.8) | −8.0 (−11.8 to −4.0) | ||
| Low | 43 | 603.0 (49.0) | −13.5 (−16.9 to −10.0) | ||
| Leptin (ng/mL and % change)†† | |||||
| Intention-to-treat: | |||||
| High | 54 | 10.9 (1.6) | 34.2 (21.8 to 47.7) | 0.07 | −3.3 (−6.8 to 0.1) |
| Moderate | 51 | 9.8 (1.5) | 34.8 (22.6 to 48.2) | ||
| Low | 56 | 9.6 (1.4) | 17.9 (7.7 to 29.1) | ||
| Per protocol: | |||||
| High | 38 | 11.8 (2.2) | 47.6 (33.9 to 62.8) | 0.009 | −4.9 (−8.4 to −1.5) |
| Moderate | 38 | 8.6 (1.5) | 42.0 (29.4 to 55.8) | ||
| Low | 43 | 9.0 (1.5) | 21.9 (11.7 to 33.0) | ||
Means and changes were constructed and compared from repeated measures analysis of variance, unadjusted except for structural design variables (study site, cohort, enrolment wave).
Linear trend across diet groups for every 10% decrease in the contribution of carbohydrate to total energy intake. Estimates are equivalent to comparing the high with low carbohydrate diet, divided by 4 (with these two diet groups differing in carbohydrate by 40%).
Quantified based on triaxial counts, representing a composite vector magnitude of three orthogonal planes (vertical, anteroposterior, mediolateral).43 Average accelerometer wear time was mean 14.9 (SD 1.2) hours per day.
Quantified using vertical axis count thresholds of Troiano et al.44
Defined as <100 counts per minute for vertical axis counts.43
Efficiency is expressed as percentage ratio of power generated (with conversion of Watts to kcal/min using a factor of 0.01433) to energy expenditure above resting (kcal/min).45 46 Data not collected at midpoint of test phase. Change: end of test phase−start of trial. There were no significant group effects at 10 W (presented here), 25 W, and 50 W.
Hormone levels were log transformed for analysis. For reporting, the adjusted mean and standard error were retransformed to the original units (exp(mean log)±exp(mean log)×(exp(SE log)–1)), and changes were expressed in percentage units (100%×(exp(change in log)–1)).