RESUMO
Background: Bedside "sitters" are often used for patients at high risk for falls, but they are expensive and their effectiveness is unclear. Purpose: To review evidence about the effect of sitters and alternatives to sitters on patient falls in acute care hospitals. Data Sources: PubMed searches to 8 October 2019, other databases from inception to December 2018, citation searches on key articles, and a Google search (22 October 2019). Study Selection: English-language studies of any design that assessed the effect of adding sitters to usual care or compared alternatives to sitters (for example, video monitors or "close observation units") for adult patients on general wards of acute care hospitals and reported falls as a primary outcome. Data Extraction: Dual-reviewer extraction of study data and risk of bias; single reviewer with group discussion for GRADE (Grading of Recommendations Assessment, Development and Evaluation) certainty of evidence. Data Synthesis: Of 20 studies meeting inclusion criteria, 2 added sitters to usual care and 18 compared alternatives to sitters. There were no randomized trials, 11 time-series studies, 1 retrospective quasi-experimental study, and 8 pre-post studies. All studies had at least 1 methodological limitation. Two studies provided very-low-certainty evidence that adding sitters reduced falls. Eight studies provided moderate-certainty evidence that interventions that included video monitoring reduced sitter use and either did not affect or reduced the number of falls. Very-low-certainty evidence suggested that interventions that included nurse assessment tools (3 studies) or a close observation unit (2 studies) were effective alternatives to sitters. Limitation: No studies had low risk of bias, publication bias is likely, and studies may have been missed. Conclusion: Despite a compelling rationale, evidence is scant that adding sitters to usual care reduces falls. Primary Funding Source: Veterans Affairs Quality Enhancement Research Initiative. (PROSPERO: CRD42019127424).