“Publish or perish”—presentations at annual national orthopaedic meetings and their correlation with subsequent publication

Table 2 Analysis of subspecialties as potential predictive factors for publication

Specialty	No. published/no. presented (%)	% published abstracts ^a	% unpublished abstracts ^b	Odds ratio (OR)	Confidence interval (CI)	p value
Foot/ankle	18/56 (32.1)	14.4	11.9	1.205	0.716–2.029	0.485
Hand/wrist	4/6 (66.7)	3.2	0.6	5.088	0.944–27.429	0.056^c
Hip/knee	54/209 (25.8)	43.2	48.7	0.886	0.704–1.116	0.293
Infectious disease	2/19 (10. 5)	1.6	5.3	0.299	0.070–1.277	0.080
Metabolic	3/15 (20.0)	2.4	3.8	0.636	0.174–2.261	0.472
Oncology	9/20 (45.0)	7.2	3. 5	2.081	0.884–4.901	0.088
Paediatrics	7/24 (29.2)	5.6	5.3	1.048	0.445–2.464	0.915
Shoulder/elbow	15/39 (38. 5)	12.0	7. 5	1. 590	0.863–2.929	0.137
Spine	13/51 (25.5)	10.4	11.9	0.870	0.480–1.578	0.646
Sports/arthroscopy	9/38 (23.7)	7.2	9.1	0.790	0.385–1.620	0. 516
Trauma	21/111 (18.9)	16.8	28.3	0. 530^d	0.310–0.904^d	0.02^d
Others	2/8 (25.0)	1.6	1.9	0.848	0.173–4.145	1.000^c

^aOut of all 125 published abstracts, ^bout of 318 unpublished abstracts, ^cFisher’s exact test used, ^dafter multivariate binary logistic regression analysis.

ISSN: 1749-799X