Skip to main content

Table 3 Change in BMI Z-scores with exercise with interaction term analyses (subjects ≥ 10 years old)

From: Self-reported exercise and longitudinal outcomes in cystic fibrosis: a retrospective cohort study

Models: n = 366 subjects

Estimated BMI Z-score at time = 0 without exercise*

Estimated BMI Z-score at time = 0 with exercise*(pvalue)

Baseline decline in BMI Z-score per year (pvalue)

Change in BMI Z-score over baseline per year with exercise (pvalue)

Change in BMI Z-score over baseline per year for interaction (pvalue)

Any Exercise

Unadjusted Model (No Covariates)

-0.36

-0.30 (0.65)

-0.05 (0.08)

0.03 (0.29)

-

Adjusted Model

-0.30

-0.30 (0.96)

-0.05 (0.08)

0.03 (0.30)

-

Adjusted Model + Interaction for Sex

-0.29

-0.30 (0.93)

-0.05 (0.08)

0.02 (0.48)

0.02 (0.36)

Adjusted Model + Interaction for Age

-0.32

-0.32 (0.91)

-0.05 (0.08)

0.001 (0.98)

0.08 (0.001)

Adjusted Model + Interaction for Baseline BMI Z-Score

-0.24

-0.25 (0.92)

-0.05 (0.08)

0.03 (0.92)

0.04 (0.07)

Competitive Exercise

Unadjusted Model (No Covariates)

-0.41

-0.18 (0.021)

-0.03 (0.025)

0.02 (0.27)

-

Adjusted Model

-0.30

-0.31 (0.72)

-0.03 (0.024)

0.03 (0.25)

-

Adjusted Model + Interaction for Sex

-0.30

-0.31 (0.72)

-0.03 (0.024)

0.02 (0.38)

0.01 (0.80)

Adjusted Model + Interaction for Age

-0.30

-0.31 (0.69)

-0.03 (0.024)

0.005 (0.85)

0.06 (0.06)

Adjusted Model + Interaction for Baseline BMI Z-Score

-0.24

-0.26 (0.67)

-0.03 (0.023)

-0.01 (0.68)

0.06 (0.044)

  1. *Estimated intercepts were calculated using the regression coefficients assuming the subject was male and had the mean values for baseline BMI Z-score (-0.30 ± 0.95), baseline FEV1 percent predicted (80.3 ± 22.0), age of diagnosis (3.5 ± 7.2 years), and age at time of exercise ascertainment (19.4 ± 9.3 years)(n = 366). The estimated BMI Z-score values are not statistically different for the adjusted models owing to the adjustment for baseline BMI Z-scores.