Table 3.

Significant results of mixed effects logistic regression for nutrition and medication

Covariates	Phase 1 p value	Phase 2 p value	Phase 1 beta value	Phase 2 beta value
Aa
Ratio of lipids to total calories (g/cal)	0.1913	0.0002	− 57.74	119.8
Ratio of proteins to total calories (g/cal)	0.0092	0.0053	110.1	− 70.12
Proportion of calories enteral	0.8786	< 0.0001	− 0.2702	− 6.039
Diuretics past week	0.007	0.1053	− 1.457	0.5927
PMA (week)	0.032	0.1674	− 0.3831	0.1403
Bb
Ratio of proteins to total calories (g/cal)	0.9455	0.0259	− 84.96	− 98.05
Proportion of calories enteral	0.7124	0.0017	15.29	− 6.428
Antibiotics past week	0.8903	0.0014	9.91	1.903
PMA (week)	0.3467	0.0011	− 7.108	− 0.4875

^aResults of mixed-effects logistic regression analysis between nutrition/medication and microbiome phases during the EARLY period (< 34 weeks PMA). Phases 1 and 2 are considered as binary outcome variables (yes/no) and are analyzed separately. Beta values are the estimated regression coefficients and p values are computed from the likelihood ratio tests. For clarity, only significant associations are reported in this table. Full results are reported in Additional file 9: Table S8A

^bResults of mixed-effects logistic regression analysis between nutrition/medication and microbiome phases during the LATE period (≥ 34 weeks PMA). Phases 1 and 2 are considered as binary outcome variables (yes/no) and are analyzed separately. Beta values are the estimated regression coefficients and p values are computed from the likelihood ratio tests. For clarity, only significant associations are reported in this table. Full results are reported in Additional file 9: Table S8B