Table 3.

Logistic Regression Identifying the Effects of Features on High Participation Scholarly Activity Programs^*

Program Features	Univariate Analysis		Multivariable Analysis
	OR (95% C.I.)	p-value	OR (95% C.I.)	p-value
Participation in scholarly activity is required: Yes	5.8 (1.2–27.3)	0.027	5.5 (1.03–30.0)	0.047
No	1.0 (Ref)
Program director believes having all residents present their scholarly project is extremely or very important: Yes	4.7 (1.6–13.4)	0.004	4.7 (1.5–15.1)	0.01
No	1.0 (Ref)
>25% of faculty have mentored residents in the last 3 years: Yes	4.8 (1.7–13.7)	0.004	3.6 (1.2–11.4)	0.027
No	1.0 (Ref)
Regional Location: Northeast	1.0 (Ref)
Midwest	1.8 (0.5–6.6)	0.37
South	3.2 (0.9–11.3)	0.07
West	0.5 (0.05–4.7)	0.51
Protected time for faculty to support resident scholarly activity is available: Yes	8.5 (0.9–80.2)	0.062
No	1.0 (Ref)
>75%residents satisfied with scholarly activity opportunities provided in more recent ACGME resident survey: Yes No	2.9 (0.9–9.7) 1.0 (Ref)	0.082

^*“High Participation” represents the top quartile of programs, or >85% participation by residents in conducting original research studies, systematic reviews or meta-analyses, or case reviews or series with references