intervention components, significant concerns regarding the fidelity of implementation of the intervention, small sample sizes and risk of bias. As a result, while many interventions were studied, relatively few clear conclusions can be drawn from the evidence thus far. The remainder of this report is divided into three sections. The first is an update to the original report, consisting of summaries of the evidence synthesis on interventions discussed in the original report, organized by intervention as they were in the original report. The second section consists of summaries of the evidence for ‘new’ interventions that were not included in the original report, presented alphabetically. The last section is an appendix that details our methodology and data extraction. Findings: Interventions Studied in Original Report The original project reported on 13 different interventions. These are updated below in the same order as they were in the original, from the strongest to the weakest body of evidence. 2018 EVIDENCE UPDATE Managing Agitation and Aggression in Long-Term Care Residents with Dementia 6 Music Therapy We identified eight additional systematic reviews studying Music Therapy (MT) of moderate (16– 18,22,23,27) or high methodological (11,12) quality. The updated evidence is not as consistent as it was in the original report. Three reviews demonstrated statistically significant decreases in agitation and/or aggression in LTC residents with dementia (18,23,27). However, a Cochrane Review and an extensive AHRQ Review found statistically insignificant benefits and concluded that the evidence indicates MT is not more effective than usual care (12,17). And one review concluded there was insufficient evidence to draw any conclusions (22). This confusion appears to originate from the methodological limitations in the original primary studies, mainly poor experimental design and heterogeneity of the design and implementation of the interventions. Several reviews highlighted the low quality of the evidence, citing potential risk of bias, poor safeguards for intervention fidelity, small sample sizes and lack of follow-up (12,17,27). Further, MT as a single intervention category incorporates a range of different approaches and implementations. Differences in variables, such as the design of the MT intervention, and the frequency and the duration of intervention exposure, may be expected to produce different estimates of effect size. In particular, Tsoi et al. point to evidence indicating that “receptive” MT may be more effective than “active” MT (27). Receptive MT consists of individuals listening to personalized music selections as an activity in itself or while engaging in unrelated activities, e.g., listening to music while eating. Individuals or groups engaged in active MT do not just listen but are also engaged in related music activities, e.g., playing instruments or singing in a group. The Cochrane review by Van der Steen (12) included two receptive MT studies that scored higher than two active MT studies, one for agitation and one for anxiety. This new finding is consistent with the finding of a high-quality review by Whear et al. (37) reported in our original report ; but Konno et al. (22) do not agree. An additional potential confounding variable to note is the implementation of MT with or without a protocol. Livingston et al. (23) found that MT delivered with specific protocols was effective for reducing emergent agitation and decreasing symptomatic agitation (but not “additional” agitation or agitation among residents with severe dementia). Neither of the recent reviews identified in this update that concluded that MT was not effective (12,17) carried out sub-group analyses for these different types of MT interventions. The differing conclusions of the review literature make it difficult to assess the strength of the body of evidence, which is most accurately labelled as inconsistent and “Very Weak.” The conclusion of the most recent, high-quality Cochrane Review is that MT is not more effective than usual care represents a significant caution (12). Despite these negative findings, the reader should note that there remains a substantive and varied review literature that endorses MT to reduce aggression/agitation among LTC residents with moderately severe dementia (16,18,23,27,37,38). No reviews have found MT to reduce aggression/agitation for residents with severe dementia. As described above, the most likely explanation for the inconsistency is that the diversity of MT interventions and implementations, combined with the relatively low volume and substandard quality of research trials, has produced conflicting findings in the review literature. Given the relatively low cost and few barriers to implementation, decision makers may still want to consider trying versions of MT for 2018 EVIDENCE UPDATE Managing Agitation and Aggression in Long-Term Care Residents with Dementia 7 which the evidence is strongest, i.e., receptive MT based on an established protocol for residents with moderately severe dementia. Staff Training Evidence within the intervention category of “Staff Training” shares some of the same complexities and difficulties as it does for MT, namely wide variation in approaches and implementations. The original report included reviews that considered various staff training interventions grouped together. Those reviews