Structure formats of randomised controlled trial abstracts: a cross-sectional analysis of their current usage and association with methodology reporting

Table 4 Reporting of each quality item and supplementary item by structure format, presented with unadjusted and adjusted odds ratios (ORs)

Items/Supplementary items	N (%)			Crude OR (95% CI)	Adjusted OR (95% CI); P value^a
Items/Supplementary items	Overall (n = 341)	IMRaD (n = 176)	HS (n = 165)	Crude OR (95% CI)	Adjusted OR (95% CI); P value^a
1. Design	120 (35.2)	54 (30.7)	66 (40.0)	1.51 (0.96, 2.36)	1.51 (0.77, 2.93); 0.228
2. Participant	327 (95.9)	166 (94.3)	161 (97.6)	2.43 (0.75, 7.89)	2.43 (0.80, 7.33); 0.116
3. Setting	182 (53.4)	65 (36.9)	117 (70.9)	4.16 (2.64, 6.56)	4.16 (1.74, 9.97); 0.001
4. Interventions	268 (78.6)	138 (78.4)	130 (78.8)	1.02 (0.61, 1.72)	1.02 (0.64, 1.62); 0.924
5. Outcome	177 (51.9)	72 (40.9)	105 (63.6)	2.53 (1.63, 3.91)	2.53 (1.30, 4.92); 0.006
5a. Time point	138 (40.5)	49 (27.8)	89 (53.9)	3.04 (1.94, 4.76)	3.04 (1.49, 6.19); 0.002
5b. No. of outcomes ^b	123 (36.1)^c	93 (52.8)^c	30 (18.2)^c	1.87 (0.88, 3.97)^f	1.83 (0.69, 4.87); 0.224
	177 (51.9)^d	72 (40.9)^d	105 (63.6)^d
	41 (12.0)^e	11 (6.3)^e	30 (18.2)^e
6. Random assignment	336 (98.5)	172 (97.7)	164 (99.4)	3.81 (0.42, 34.48)	3.81 (0.47, 30.76); 0.209
6a. Unit of randomisation	246 (72.1)	141 (80.1)	105 (63.6)	0.43 (0.27, 0.71)	0.43 (0.25, 0.77); 0.004
7. Sequence generation	16 (4.7)	11 (6.3)	5 (3.0)	0.47 (0.16, 1.38)	0.47 (0.10, 2.25); 0.343
8. Allocation concealment	7 (2.1)	5 (2.8)	2 (1.2)	0.42 (0.08, 2.19)	0.42 (0.06, 2.76); 0.366
9. Blinding (Masking)	72 (21.1)	38 (21.6)	34 (20.6)	0.94 (0.56, 1.59)	0.94 (0.42, 2.14); 0.887
9a. Generic blinding	57 (16.7)	28 (15.9)	29 (17.6)	1.13 (0.64, 1.99)	1.13 (0.51, 2.47); 0.765

IMRaD Introduction, methods, results, and discussion format, HS Highly structured format
^aDerived from GEE analyses adjusting for potential clustering effect among abstracts published in the same journal, with individual quality item as the dependent variable (adequately reported vs. not adequately reported) and journal as the grouping factor;
^bA continuous variable, not dichotomous;
^cN (%) for category No. = 0; ^dN (%) for category 1 ≤ No. ≤ 2; ^eN (%) for category No. > 2;
^fDerived from binary logistic regression (reference group: IMRaD format; dependent variable coding: [0] 1 ≤ No. ≤ 2, [1] No. > 2)

ISSN: 1471-2288