Table 3. Participant characteristics at baseline: Participants vs. drop-outs¹.

	Completers (n = 317)	Drop-outs (n = 143)	p-value4
Age at random assignment, years2	21.4 [2.9]	21.9 [3.0]	0.06
Highest educational level3, %			0.16
Elementary school	2.3	2.9
Middle school	56.8	47.1
High school	20.3	20.3
Occupational school or higher	20.6	29.7
Living arrangement3, %			0.03
With parents-in-law	71.3	59.0
With husband, only	10.5	16.5
With parents	18.2	24.5
Work as farmers3, %	77.8	69.8	0.07
Anthropometry2
Weight, kg	45.9 [4.8]	45.9 [5.3]	0.91
Height, cm	152.8 [5.1]	152.4 [5.8]	0.55
AMA, cm2	23.4 [4.8]	21.8 [5.5]	0.004
AFA, cm2	16.4 [4.4]	17.3 [4.9]	0.05
BMI, kg/m2	19.7 [1.8]	19.7 [1.9]	0.73
BMI < 18.53, %	28.4	24.5	0.39

¹ Sample sizes for each variable vary slightly because of item-specific missing data.

² Values are means [SD]

³ Values are % of the total

⁴ Comparisons between the completer and the drop-out groups: t-tests were used for comparing means of normally distributed continuous variables, the Mann-Whitney U test was used to compare means with a non-normal distribution for continuous variables, and a chi-square test was used to compare percentages of the population. Pair-wise differences statistically significant under the Benjamini-Hochberg procedure to maintain an overall false discovery rate ≤5% in bold.

AMA, arm muscle area; AFA, arm fat area; BMI, body mass index.