Table 12. Other studies.

First author, year	Study design, intervention length, number of sites, baseline sample size	Control condition	Education material	Training	Reminders	Audit and feedback	Mentoring or support	Champion/s	Team meetings	Policy/procedure changes	Organizational restructure	Intervention description	Staff direct behavior outcomes	Staff indirect outcomes	Resident outcomes
Molloy, 2000	Paired, cluster randomized RCT, 1 wk, 78 wks, 6 sites, CG: 656 res; IG: 636 res		+	+				+				Training for nurses on advance directives, as health care facilitators, approaches to educating staff, residents and families, and assessing capacity to complete directives. Nurses trained staff and emergency workers. Videos describing program provided. Refresher sessions for new staff and to maintain awareness of already trained staff.	Compared to CG, IG had higher proportion of Advanced Care Directives completed.		Compared to CG, over time IG homes reported fewer hospitalizations per resident and they had a lower mean number of hospital days. Indirect outcomes: differences between groups on satisfaction with health care in competent or incompetent residents or in the proportion of deaths.
Jones, 2004	NRCT, 26 wks, 12 homes, 1899 res total (IG and CG n not specified)	Usual care	+	+		+	+	+		+		Training for all staff, particularly nurses. 3 member internal pain teams (IPTs) formed who developed pain vital sign assessment and documentation. Feedback reports provided. Resident educational video provided on admission. Physicians offered video and pamphlet, CME credits malpractice insurance premium discount. Pain expert available, documentation developed by staff for their facility.	No differences between groups on non-MDS pain assessments and pain reassessments.		No reduction in percentage of residents reporting pain or reporting moderate/ severe pain in IG homes. Improvement in percentage of residents reporting constant pain in intervention homes.
Irvine, 2012	Cluster randomized RCT, 2 wks, 14 wks follow-up, 6 sites, CG: 45 staff; IG: 58 staff	Delayed treatment		+								Internet based training including videos on fundamental de-escalation skills with residents exhibiting aggressive behavior, about hits, hits with fists or arms, hair grabs, wrist grabs.		No differences in self-efficacy or empathy.	Compared to CG, the IG decreased over time in number of assaults reported.
Teresi, 2013	Cluster randomized (region-based) RCT, 4 wks, 52 wks follow-up, 47 sites, CG: 500 staff, 685 res; IG: 525 staff, 720 res	Training in filling out behavior recognition and documentation sheets		+								Staff trained in identification and intervention (and reporting) with respect to resident to resident elder mistreatment.	Compared to CG, IG had higher levels of recognition and documentation of resident to resident mistreatment over time.
Beeckman, 2013	Cluster randomized RCT, 16 wks, 11 sites, CG: 239 res, 53 staff; IG: 225 res, 65 staff	Pressure ulcer prevention protocol hard copy and lecture for all staff	+	+	+	+		+	+			Electronic decision support system, PrevPlan. Training about pressure ulcer prevention. Monitoring and feedback on adequacy of pressure ulcer prevention, knowledge, attitudes. Reminders. Key nurse introduced, inventory and feedback on quality and availability of current material for pressure ulcers, support the acquisition of new pressure ulcer preventive materials.	Compared to CG, IG more likely to provide fully adequate prevention when in a chair, and in the proportion of residents with some prevention, but no difference between groups over time for fully adequate prevention in bed.		Compared to CG, IG had significantly lower pressure ulcer prevalence (categories I–IV). No difference when only considering categories II-IV.

NRCT, non-randomized controlled trial; RCT, randomized controlled trial; CG, control group; IG, intervention group; IG1, intervention group 1; IG2. Intervention group 2; wks, weeks; res, residents.