A social network intervention to improve connectivity and burnout among psychiatry residents in an academic institution: a quasi-experimental study

Table 2 The intercept and coefficients of linear mixed-effect regressions for different outcomes and their statistical significance

	PGY-2 residents (one-mode^a) Networks Normalized Indegrees			Faculty members (two-mode^b) Networks Normalized Indegrees			Maslach Burnout Inventory			PHEEM Score
	Personal Support	Clinical Advice	Educational Advice	Personal Support	Clinical Advice	Educational Advice	Emotional Exhaustion	Depersonalization	Personal Accomplishment	PHEEM Score
Baseline: Non-active members (intercept)	0.450	0.231	0.206	0.032	0.051	0.039	10.90	2.20	38.90	90.14
Baseline: Difference of active^c vs. non-active members	-0.003	0.054	0.052	0.065	0.086*	0.098**	-0.04	3.37*	-4.90*	-1.71
Non-active members: Difference of follow-up vs. baseline	-0.067	-0.034	0.017	-0.009	0.004	-<0.000	11.69**	0.78	-4.56*	-8.44*
Active members: Difference of follow-up vs. baseline	-0.122**	-0.085*	-0.077	0.068**	-0.033	-0.012	5.85	< 0.01	5.14*	12.42**
Difference in differences: Differential changes in active vs. non-active over time	-0.054	-0.051	-0.095	0.078**	-0.037	-0.011	-5.83	-0.78	9.71**	20.86**

Definitions: Normalized indegree = Total number of nominations an individual receives in a network; reported as a fraction of total possible nominations.
* = Statistically significant with P < 0.05
** = Statistically significant with P < 0.01
^a Social networks that both nominators and nominees are PGY-2 residents
^b Social networks that nominators are PGY-2 residents, but nominees are faculty members (who did not participate in the survey themselves)
^c PGY-2 residents and faculty members who acted as moderators of intervention (interest groups)

ISSN: 1472-6920