Interventions for preventing falls in older people in care facilities and hospitals

Ian D Cameron; Suzanne M Dyer; Claire E Panagoda; Geoffrey R Murray; Keith D Hill; Robert G Cumming; Ngaire Kerse

doi:10.1002/14651858.CD005465.pub4

. 2018 Sep 7;2018(9):CD005465. doi: 10.1002/14651858.CD005465.pub4

Interventions for preventing falls in older people in care facilities and hospitals

Ian D Cameron ^1,^✉, Suzanne M Dyer ², Claire E Panagoda ³, Geoffrey R Murray ⁴, Keith D Hill ⁵, Robert G Cumming ⁶, Ngaire Kerse ⁷

Editor: Cochrane Bone, Joint and Muscle Trauma Group

PMCID: PMC6148705 PMID: 30191554

Abstract

Background

Falls in care facilities and hospitals are common events that cause considerable morbidity and mortality for older people. This is an update of a review first published in 2010 and updated in 2012.

Objectives

To assess the effects of interventions designed to reduce the incidence of falls in older people in care facilities and hospitals.

Search methods

We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (August 2017); Cochrane Central Register of Controlled Trials (2017, Issue 8); and MEDLINE, Embase, CINAHL and trial registers to August 2017.

Selection criteria

Randomised controlled trials of interventions for preventing falls in older people in residential or nursing care facilities, or hospitals.

Data collection and analysis

One review author screened abstracts; two review authors screened full‐text articles for inclusion. Two review authors independently performed study selection, 'Risk of bias' assessment and data extraction. We calculated rate ratios (RaR) with 95% confidence intervals (CIs) for rate of falls and risk ratios (RRs) and 95% CIs for outcomes such as risk of falling (number of people falling). We pooled results where appropriate. We used GRADE to assess the quality of evidence.

Main results

Thirty‐five new trials (77,869 participants) were included in this update. Overall, we included 95 trials (138,164 participants), 71 (40,374 participants; mean age 84 years; 75% women) in care facilities and 24 (97,790 participants; mean age 78 years; 52% women) in hospitals. The majority of trials were at high risk of bias in one or more domains, mostly relating to lack of blinding. With few exceptions, the quality of evidence for individual interventions in either setting was generally rated as low or very low. Risk of fracture and adverse events were generally poorly reported and, where reported, the evidence was very low‐quality, which means that we are uncertain of the estimates. Only the falls outcomes for the main comparisons are reported here.

Care facilities

Seventeen trials compared exercise with control (typically usual care alone). We are uncertain of the effect of exercise on rate of falls (RaR 0.93, 95% CI 0.72 to 1.20; 2002 participants, 10 studies; I² = 76%; very low‐quality evidence). Exercise may make little or no difference to the risk of falling (RR 1.02, 95% CI 0.88 to 1.18; 2090 participants, 10 studies; I² = 23%; low‐quality evidence).

There is low‐quality evidence that general medication review (tested in 12 trials) may make little or no difference to the rate of falls (RaR 0.93, 95% CI 0.64 to 1.35; 2409 participants, 6 studies; I² = 93%) or the risk of falling (RR 0.93, 95% CI 0.80 to 1.09; 5139 participants, 6 studies; I² = 48%).

There is moderate‐quality evidence that vitamin D supplementation (4512 participants, 4 studies) probably reduces the rate of falls (RaR 0.72, 95% CI 0.55 to 0.95; I² = 62%), but probably makes little or no difference to the risk of falling (RR 0.92, 95% CI 0.76 to 1.12; I² = 42%). The population included in these studies had low vitamin D levels.

Multifactorial interventions were tested in 13 trials. We are uncertain of the effect of multifactorial interventions on the rate of falls (RaR 0.88, 95% CI 0.66 to 1.18; 3439 participants, 10 studies; I² = 84%; very low‐quality evidence). They may make little or no difference to the risk of falling (RR 0.92, 95% CI 0.81 to 1.05; 3153 participants, 9 studies; I² = 42%; low‐quality evidence).

Hospitals

Three trials tested the effect of additional physiotherapy (supervised exercises) in rehabilitation wards (subacute setting). The very low‐quality evidence means we are uncertain of the effect of additional physiotherapy on the rate of falls (RaR 0.59, 95% CI 0.26 to 1.34; 215 participants, 2 studies; I² = 0%), or whether it reduces the risk of falling (RR 0.36, 95% CI 0.14 to 0.93; 83 participants, 2 studies; I² = 0%).

We are uncertain of the effects of bed and chair sensor alarms in hospitals, tested in two trials (28,649 participants) on rate of falls (RaR 0.60, 95% CI 0.27 to 1.34; I² = 0%; very low‐quality evidence) or risk of falling (RR 0.93, 95% CI 0.38 to 2.24; I² = 0%; very low‐quality evidence).

Multifactorial interventions in hospitals may reduce rate of falls in hospitals (RaR 0.80, 95% CI 0.64 to 1.01; 44,664 participants, 5 studies; I² = 52%). A subgroup analysis by setting suggests the reduction may be more likely in a subacute setting (RaR 0.67, 95% CI 0.54 to 0.83; 3747 participants, 2 studies; I² = 0%; low‐quality evidence). We are uncertain of the effect of multifactorial interventions on the risk of falling (RR 0.82, 95% CI 0.62 to 1.09; 39,889 participants; 3 studies; I² = 0%; very low‐quality evidence).

Authors' conclusions

In care facilities: we are uncertain of the effect of exercise on rate of falls and it may make little or no difference to the risk of falling. General medication review may make little or no difference to the rate of falls or risk of falling. Vitamin D supplementation probably reduces the rate of falls but not risk of falling. We are uncertain of the effect of multifactorial interventions on the rate of falls; they may make little or no difference to the risk of falling.

In hospitals: we are uncertain of the effect of additional physiotherapy on the rate of falls or whether it reduces the risk of falling. We are uncertain of the effect of providing bed sensor alarms on the rate of falls or risk of falling. Multifactorial interventions may reduce rate of falls, although subgroup analysis suggests this may apply mostly to a subacute setting; we are uncertain of the effect of these interventions on risk of falling.

Plain language summary

Interventions for preventing falls in older people in care facilities and hospitals

Review question  How effective are interventions designed to reduce falls in older people in care facilities and hospitals?

Background  Falls by older people in care facilities, such as nursing homes, and hospitals are common events that may cause loss of independence, injuries, and sometimes death as a result of injury. Effective interventions to prevent falls are therefore important. Many types of interventions are in use. These include exercise, medication interventions that include vitamin D supplementation and reviews of the drugs that people are taking, environment or assistive technologies including bed or chair alarms or the use of special (low/low) beds, social environment interventions that target staff members and changes in the organisational system, and knowledge interventions. A special type of intervention is the multifactorial intervention, where the selection of single interventions such as exercise and vitamin D supplementation is based on an assessment of a person's risk factors for falling. Falls are reported in two ways in our review. One outcome is rate of falls, which is the number of falls. The other outcome is risk of falling, which is the number of people who had one or more falls.

Search date

We searched the healthcare literature for reports of randomised controlled trials relevant to this review up to August 2017.

Study characteristics  This review included 95 randomised controlled trials involving 138,164 participants. Seventy‐one trials (40,374 participants) were in care facilities, and 24 (97,790 participants) in hospitals. On average, participants were 84 years old in care facilities and 78 years old in hospitals. In care facilities, 75% were women and in hospitals, 52% were women.

Quality of the evidence  The majority of trials were at high risk of bias, mostly relating to lack of blinding. With few exceptions, the quality of evidence for individual interventions in either setting was generally rated as low or very low. Risk of fracture and adverse events were generally poorly reported and, where reported, the evidence was very low quality, which means that we are uncertain of the estimates.

Key results

There was evidence, often from single studies, for a wide range of interventions used for preventing falls in both settings. However, in the following we summarise only the falls outcomes for four key interventions in care facilities and three key interventions in hospitals.

Care facilities  We are uncertain of the effect of exercise on the rate of falls (very low‐quality evidence) and it may make little or no difference to the risk of falling (low‐quality evidence).  General medication review may make little or no difference to the rate of falls (low‐quality evidence) or the risk of falling (low‐quality evidence).  Prescription of vitamin D probably reduces the rate of falls (moderate‐quality evidence) but probably makes little or no difference to the risk of falling (moderate‐quality evidence). The population included in these studies appeared to have low vitamin D levels.  We are uncertain of the effect of multifactorial interventions on the rate of falls (very low‐quality evidence). They may make little or no difference to the risk of falling (low‐quality evidence).

Hospitals  We are uncertain whether physiotherapy aimed specifically at reducing falls in addition to usual rehabilitation in the ward has an effect on the rate of falls or reduces the risk of falling (very low‐quality evidence).  We are uncertain of the effect of bed alarms on the rate of falls or risk of falling (very low‐quality evidence).  Multifactorial interventions may reduce the rate of falls, although this is more likely in a rehabilitation or geriatric ward setting (low‐quality evidence). We are uncertain of the effect of these interventions on risk of falling.

Summary of findings

Background

Description of the condition

Studies of falls in nursing facilities show considerable variation in falls incidence rates but a “middle of the road” figure provided in a review of incidence rates is 1.7 falls per person‐year, compared with 0.65 falls per person‐year for older people living in the community (Rubenstein 2006). In a study conducted in 40 Canadian residential care facilities, 62% of participants fell over a one‐year period, with a falls rate of 2.51 falls per person per year (Kennedy 2015). It should be noted, however, that routine recording of falls incidents in standard reporting systems is likely to under‐estimate the incidence of falls (Hill 2010; Sutton 1994). In a prospective one‐year study in 528 nursing homes in Bavaria, Germany, about 75% of falls occurred in the residents' rooms or in bathrooms; 41% occurred during transfers and 36% when walking (Becker 2012). The fall rate was higher in men (2.8 falls per person year) than women (1.49 falls per person year), and falls were less common in people requiring the lowest and highest levels of care. Lord 2003 also found that fall rates were lower in frailer people who were unable to rise from a chair or stand unaided. In this group, increased age, male sex, higher care classifications, incontinence, psychoactive medication use, previous falls and slow reaction times were associated with increased falls. Systematic reviews have shown that in nursing homes, falls history, walking aid use, moderate disability, cognitive impairment, wandering, Parkinson's disease, dizziness, use of sedatives, antipsychotics, antidepressants and total number of medications used are associated with an increased risk of falling (Deandrea 2013; Muir 2012). In residents with dementia, age, use of psychotropic drugs, fair or poor general health, gait impairment and trunk restraint use are associated with an increased number of falls (Kropelin 2013).

In hospital settings, a falls incidence of 5.71 falls per 1000 bed days has been found in 16 US general medical surgical and speciality units (Shorr 2012), 6.45 falls per 1000 bed days in 24 Australian medical and surgical wards (Barker 2016), 10.9 falls per 1000 bed days in eight Australian rehabilitation/geriatric units (Hill 2015) and 17.1 falls per 1000 bed days in psychogeriatric wards (Nyberg 1997). In elderly care wards in an UK district general hospital in 2004, the reported rate was as high as 18.0 falls per 1000 bed days (Healey 2004). A similar rate has been reported in some high‐risk wards in Australia (Barker 2016).

Systematic reviews have shown that risk factors for falls in hospital inpatients are falls history, age, cognitive impairment, sedative and antidepressant use, gait instability, agitated confusion and urinary incontinence (Deandrea 2013; Oliver 2004). For older patients in rehabilitation hospital settings, risk factors include carpet flooring, vertigo, being an amputee, confusion, cognitive impairment, stroke, sleep disturbance, anticonvulsants, tranquillisers, antihypertensive medications, previous falls and need for transfer assistance (Vieira 2011).

There is considerable mortality and morbidity associated with falls in care facilities and hospitals. A study in 24 Australian medical and surgical wards reported a fall injury rate of 2.36 per 1000 bed days (Barker 2016). A study in both these settings reported an incidence of 533 per 1000 person years for all injuries, 20 per 1000 person years for hip fracture, and 270 per 1000 person years for head injuries, for which 13% (14/107) required medical attention (Nurmi 2002). Overall, men were 1.5 times more likely to be injured than women. Older people who sustain a hip fracture while in hospital have been shown to have poor outcomes compared with people sustaining similar fractures in the community (Murray 2007). Falls have been reported to be the most common cause of death from an external cause in residents of care facilities (Ibrahim 2015).

Description of the intervention

The majority of falls are caused by complex combinations of factors operating at the time of each fall event. Interventions may target risk factors in participants or target staff and clinicians with the aim of improving clinical practice or the organisation of care. In some studies, single interventions have been evaluated while in others, interventions with more than one component have been evaluated. Delivery of multiple‐component interventions may be based on individual assessment of risk (a multifactorial intervention) or the same components are provided to all participants (a multiple intervention). A taxonomy has been developed to describe and classify types of intervention (Lamb 2007; Lamb 2011). Key intervention categories include exercise, medication (drug target) interventions which include interventions targeting vitamin D and medication reviews, environment or assistive technologies including bed/chair alarms or the use of low/low beds, social environment interventions which target staff members and changes in the organisational system, knowledge interventions and multifactorial interventions.

The majority of randomised controlled trials considered within this review provide a comparison with ‘usual care’ in the care facilities and hospitals involved. Typically, 'usual care' will include standard practices for managing commonly known, potentially modifiable, risk factors for falls and, moreover, the components of usual care will vary both over time and between settings.

Why it is important to do this review

A systematic review is required to summarise evidence of the impact of purposeful interventions designed to prevent falls, in addition to the unknown impact of routine (and probably variable) care in care facilities and hospitals. Despite routine activities attempting to reduce falls, falls are common in these settings and they result in considerable mortality and morbidity. Results will inform healthcare professionals, researchers, policy makers, informal care givers and consumers. This review is an update of a Cochrane Review first published in 2010 (Cameron 2010), and previously updated in 2012 (Cameron 2012).

Objectives

To assess the effects of interventions designed to reduce the incidence of falls in older people in care facilities and hospitals.

Methods

Criteria for considering studies for this review

Types of studies

We considered for inclusion all randomised trials, including quasi‐randomised trials (for example, alternation), cluster‐randomised trials and trials in which treatment allocation was inadequately concealed.

Types of participants

We included trials of interventions to prevent falls in older people, of either sex, in care facilities or hospitals. We considered trials for inclusion if the majority of participants were over 65 years or the mean age was over 65 years, and the majority were living in care facilities or were patients in hospital. We excluded trials conducted in places of residence that do not provide residential health‐related care or rehabilitative services, for example retirement villages or sheltered housing. Trials with participants resident in the community and in care facilities were included either in this review or in the Cochrane Review of interventions for preventing falls in older people living in the community (Gillespie 2012), depending on the proportion of participants in each setting. Inclusion in either review was determined by discussion between the authors of both reviews. Trials recording falls in both settings may be included in both reviews.

We subdivided care facilities based on level of care provided. We defined high‐level care facilities as "establishments that are primarily engaged in providing inpatient nursing and rehabilitative services for long‐term care patients. The care is generally provided for an extended period of time to individuals requiring nursing care. These establishments have a permanent core staff of registered or licensed practical nurses that, along with other staff, provide nursing care in combination with personal care" (OECD 2011). We defined intermediate‐care facilities as "institutions which provide health‐related care and services to individuals who do not require the degree of care which hospitals or skilled nursing facilities provide, but because of their physical or mental condition require care and services above the level of room and board" (NLM 2012). Some facilities provided both these levels of care. For cluster‐randomised trials, the classification of the level of care was based on the description of the facility. For individually‐randomised trials where the level of care provided by the facility was clearly described, this description informed the classification. Where the inclusion/exclusion criteria of a trial selected patients who required high or intermediate level of care from a mixed‐care facility, the classification was based upon the care needs of the individual participants.

For trials in hospitals, participants included staff or in‐patients. We excluded interventions that took place in emergency departments, outpatient departments or where hospital services were provided in community settings. We subdivided hospitals into those providing acute, and those providing subacute care. We defined subacute care as "medical and skilled nursing services provided to patients who are not in an acute phase of an illness but who require a level of care higher than that provided in a long‐term care setting" (NLM 2012).

Studies recruiting participants post‐stroke were excluded as interventions to prevent falls in this population are reviewed in a separate Cochrane Review Interventions for preventing falls in people after stroke (Verheyden 2013).

Types of interventions

Any intervention designed to reduce falls in older people compared with any other intervention, usual care or placebo. We grouped interventions using the fall‐prevention classification system (taxonomy) developed by the Prevention of Falls Network Europe (ProFaNE) (Lamb 2011). Interventions have been grouped by combination (single, multiple, or multifactorial), and then by the type of intervention (descriptors). Full details are available in the ProFaNE taxonomy manual (Lamb 2007). The possible intervention descriptors are: exercises, medication (drug target, i.e. withdrawal, dose reduction or increase, substitution, provision), surgery, management of urinary incontinence, fluid or nutrition therapy, psychological interventions, environment/assistive technology, social environment, interventions to increase knowledge, other interventions.

Types of outcome measures

We included only trials that reported raw data or statistics relating to rate or number of falls, or number of participants sustaining at least one fall during follow‐up (fallers). Trials that reported only those participants who had more than one fall were included. Trials that reported only specific types of fall (e.g. injurious falls) were not included. Trials that focused on intermediate outcomes such as improved balance or strength, and did not report falls or falling as an outcome, were excluded.

Primary outcomes

Rate of falls (falls per unit of person time that falls were monitored)
Number of fallers (risk of falling)

Secondary outcomes

Number of participants sustaining fall‐related fractures
Complications of the interventions
Economic outcomes

Search methods for identification of studies

Electronic searches

We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (to 3 August 2017), the Cochrane Central Register of Controlled Trials (CENTRAL) (2017, Issue 8), MEDLINE (including Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE Daily, Ovid MEDLINE and Versions) (1946 to 3 August 2017), Embase (1980 to 2017 Week 31), and CINAHL (1982 to 3 August 2017). We also searched ongoing trial registers via the World Health Organization's ICTRP Search Portal (3 August 2017) and ClinicalTrials.gov (3 August 2017). We did not apply any language restrictions.

For this update, the search results were limited from 2012 onwards. The search update process was run in two stages: the first search was run in February 2016 and a second top‐up search was run in August 2017. Details of the search strategies used for previous versions of the review are given in Cameron 2012.

In MEDLINE (OvidSP), subject‐specific search terms were combined with the sensitivity‐ and precision‐maximising version of the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE (Lefebvre 2011). We modified this strategy for use in CENTRAL, Embase, and CINAHL (seeAppendix 1 for all strategies).

Searching other resources

We also checked reference lists of articles and further trials were identified by contact with researchers in the field. For the first version of this review, we identified trials in care facilities and hospitals included in Gillespie 2003.

Data collection and analysis

Data collection and analysis were carried out according to methods stated in the published protocol (Cameron 2005), and subsequently amended to concur with updated methods in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a) as described in Differences between protocol and review. Data collection and analysis were carried out according to methods stated in the published protocol (Cameron 2005), which were based on the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a).

Selection of studies

From the title, abstract, or descriptors, one review author screened all abstracts to identify potentially relevant trials for full review. Two review authors screened potentially relevant abstracts. From the full text, two review authors independently assessed potentially eligible trials for inclusion and resolved disagreement by discussion, or by adjudication with a third review author. Full‐text review was undertaken using Covidence. Disagreement was resolved by discussion and consensus or third party adjudication when necessary. We contacted trial authors for additional information if necessary to assess eligibility.

Data extraction and management

Pairs of review authors independently extracted data using a pre‐tested data extraction form for studies included to 2012. For this update,again pairs of review authors independently extracted data from the identified studies using Covidence. Multiple reports from the same study were linked as a single study in Covidence and evidence from all reports were reviewed in undertaking data extraction. Where data were unclear authors were contacted whenever possible for clarification. Disagreement was resolved by discussion and consensus or third party adjudication when necessary.

Assessment of risk of bias in included studies

Pairs of review authors independently assessed risk of bias for each included study based on recommendations in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b). Assessors were not blinded to author and source institution. Review authors did not assess their own trials. Disagreement was resolved by consensus, or by third party adjudication.

We assessed risk of bias for the following domains: sequence generation (selection bias); allocation concealment (selection bias); blinding of participants and personnel (performance bias); blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), and selective reporting (reporting bias). Since all the outcomes collected in our review are susceptible to the same risk of bias, we have not assessed outcomes for risk of detection bias or completeness of outcome data separately. Additionally, we assessed bias in the recall of falls due to less reliable methods of ascertainment (Hannan 2010), and bias resulting from major imbalances in key baseline characteristics (e.g. age, gender, previous falls, medical status, dependency, cognitive function). Assessors rated the risk of bias as low, high or unclear for each domain.

We established additional criteria within currently existing domains for assessing the additional risks of bias associated with cluster randomisation (Section 16.3.2; Higgins 2011b). Thus 'recruitment bias' was considered as a component of selection bias under allocation concealment; 'baseline imbalance' resulting from small numbers of clusters was considered in bias resulting from major imbalances in key characteristics; risk of bias resulting from 'loss of clusters' was considered under incomplete outcome data; and 'incorrect analysis' that failed to take into account the effect of clustering and that could not be satisfactorily remedied was considered under selective outcome reporting. We did not assess the risk of bias relating to the 'comparability with individually‐randomised trials' as a separate item as it is impossible to establish suitable criteria for an individual trial out of context. The potential for differences in effects between cluster‐ and individually‐randomised trials was considered in our assessment of the quality of the evidence and in our Discussion.

Our criteria for 'Risk of bias' assessments are shown in Appendix 2.

Measures of treatment effect

We have reported the treatment effect for rate of falls as a rate ratio (RaR) and 95% confidence interval (CI). For number of fallers and number of participants sustaining fall‐related fractures we have reported a risk ratio (RR) and 95% CI. We used results reported at discharge from hospital for trials that continued to monitor falls after discharge.

Rate of falls

The rate of falls is the total number of falls per unit of person time that falls were monitored (e.g. falls per person year). The rate ratio compares the rate of falls in any two groups during each trial.

We used a rate ratio (for example, incidence rate ratio or hazard ratio for all falls) and 95% CI if these were reported in the paper. If both adjusted and unadjusted rate ratios were reported, we used the unadjusted estimate, unless the adjustment was for clustering. If a rate ratio was not reported but appropriate raw data were available, we used Excel to calculate a rate ratio and 95% CI. We used the reported rate of falls (falls per person year) in each group and the total number of falls for participants contributing data, or we calculated the rate of falls in each group from the total number of falls and the actual total length of time falls were monitored (person years) for participants contributing data. In cases where data were only available for people who had completed the study, or where the trial authors had stated there were no losses to follow‐up, we assumed that these participants had been followed up for the maximum possible period. Where there were no falls in one arm of a study, and a low total number of falls and/or participants (e.g. Beck 2016; Cadore 2014), the rate of falls cannot be determined. Such data were therefore not pooled, however the omission of these data from the pooled analysis is considered unlikely to change any estimate of effect.

Risk of falling

For number of fallers, a dichotomous outcome, we used a risk ratio as the treatment effect. The risk ratio compares the number of people who fell once or more (fallers) in the intervention and control arms of each trial.

We used a reported estimate of risk (hazard ratio for first fall, risk ratio (relative risk), or odds ratio) and 95% CI if available. If both adjusted and unadjusted estimates were reported we used the unadjusted estimate, unless the adjustment was for clustering. If an odds ratio was reported, or there was no effect estimate and 95% CI, and appropriate data were available, we calculated a risk ratio and 95% CI using the csi command in Stata or in Review Manager. For the calculations, we used the number of participants contributing data in each group if this was known; if not reported, we used the number randomised to each group.

Secondary outcomes

For the number of participants sustaining one or more fall‐related fractures, we used a risk ratio as described in 'Risk of falling' above.

Unit of analysis issues

For trials that were cluster randomised, for example by care facility or ward, we performed adjustments for clustering (Higgins 2011c), if this was not done in the published report. We used intra‐cluster correlation coefficients reported by Dyer 2004 (falls per person year 0.100, number of residents falling 0.071, and residents sustaining a fracture 0.026).

For trials with multiple intervention groups, we either combined the groups or included only one pair‐wise comparison (intervention versus control) in any analysis in order to avoid the same group of participants being included twice.

For trials that excluded the intervention period from the falls outcomes, we did not pool the outcomes data with other studies.

Dealing with missing data

Only the available data were used in the analyses; we did not impute missing data.

Assessment of heterogeneity

We assessed heterogeneity within a pooled group of trials using a combination of visual inspection of the graph along with consideration of the Chi² test (with statistical significance set at P < 0.10), and the I² statistic (Higgins 2003). We based our interpretation of the I² results on that suggested by Higgins 2011a: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity; and 75% to 100% may represent very substantial ('considerable') heterogeneity.

Assessment of reporting biases

To explore the possibility of publication and other reporting biases, we constructed funnel plots for analyses that contained more than 10 studies.

Data synthesis

We classified interventions into those taking place in care facilities and those taking place in hospitals, and pooled these separately because participant characteristics and the environment warrants different types of interventions in the different settings, possibly implemented by people with different skill mixes.

Within these categories, we grouped the results of trials with comparable interventions and participant characteristics, and compiled forest plots using the generic inverse variance method in Review Manager. This method enabled pooling of the adjusted and unadjusted treatment effect estimates (rate ratios or risk ratios) that were reported in the paper, or we calculated from data presented in the paper (seeMeasures of treatment effect). Where the total number of patients, rather than admissions, could not be determined, we did not pool these data with other studies. Where the reported trial outcomes did not include falls during the intervention period, we did not pool these data with those of other trials.

Where appropriate, we pooled results of comparable studies using both fixed‐effect and random‐effects models. We chose the model to report by careful consideration of the extent of heterogeneity and whether it can be explained by factors such as the number and size of included studies, or the level of care provided. We used 95% CIs throughout. We considered, on a case by case basis, not pooling data where there was considerable heterogeneity (I² statistic value of greater than 75%) that could not be explained by the diversity of methodological or clinical features among trials. Where it was inappropriate to pool data, we still presented trial data in the analyses or tables for illustrative purposes and reported these in the text.

Subgroup analysis and investigation of heterogeneity

We minimised heterogeneity as much as possible by grouping trials as described previously (using ProFaNE categories of interventions). We categorised broad interventions further by grouping subtypes of interventions according to ProFaNE (e.g. for exercise interventions). We explored heterogeneity by carrying out subgroup analyses based on level of care and level of cognition at enrolment in care facilities and hospitals where possible. We subdivided the care facilities into high, intermediate or mixed levels of care. The levels of care of the facilities reflect the levels of dependence of the participants. In hospitals, the level of care was subdivided by acute versus subacute or mixed levels of care. We also carried out subgroup analyses by stratification of intervention types according to ProFaNE (e.g. for exercise types, medication target interventions), and type of fracture. Subgroup analyses based upon the individual components of the multifactorial interventions was precluded by the study design and reporting. Data were inadequate for conducting a subgroup analysis by level of frailty of the participants in trials of exercise in care facilities.

We grouped trials by level of cognition into those that included only participants with cognitive impairment versus those with no cognitive impairment, or a mixed sample at enrolment.

We used the random‐effects model to pool data in all subgroup analyses testing for subgroup differences due to the high risk of false‐positive results when comparing subgroups in a fixed‐effect model (Higgins 2011d). We used the test for subgroup differences available in Review Manager to determine whether there was evidence for a difference in treatment effect between subgroups.

Sensitivity analysis

Where there was substantial statistical heterogeneity we carried out a post‐hoc sensitivity analysis to explore the effect of removing trials from the analysis if visual inspection of the graph showed poorly overlapping confidence intervals. Where there was considered to be significant statistical heterogeneity for rate of falls but not risk of falling, sensitivity analyses were carried out to determine the likely effects of using random‐effects versus fixed‐effect meta‐analyses for the risk of falling (e.g. for exercise versus usual care in care facilities and multifactorial interventions in care facilities). We conducted post‐hoc sensitivity analyses for exercise in care facilities, excluding trials with 20 participants or less in each arm of the trial to explore the possibility of small‐trial effects, due to the observed asymmetry in the Funnel plots. We conducted a sensitivity analysis for exercise compared to usual care in care facilities including Cadore 2014, which had zero falls in the intervention arm, using one fall in the intervention arm to examine the likely effect of omitting this trial from the analysis. We also conducted a sensitivity analysis excluding one trial with a known non‐normal distribution of falls in the intervention arm from the analysis of general medication review in care facilities for the rate of falls outcomes.

Sensitivity analyses according to study quality were not possible as most studies were at potential risk of bias.

Economic issues

We have noted the results from any economic evaluations (cost‐effectiveness analysis, cost‐utility analysis) incorporated in included studies. We also extracted from each trial reporting a cost analysis, cost description or analytic model, the type of resource use reported (e.g. delivering the intervention, hospital admissions, medication use) and the cost of the items for each group.

Assessing the quality of the evidence and 'Summary of findings' tables

For each comparison, we used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to assess the quality of the body of evidence (Schünemann 2011) for each outcome listed in Types of outcome measures. For all comparisons where there were two or more trials, GRADE assessment was performed independently by two review authors and disagreement was resolved by discussion, or by adjudication with a third review author. We adopted a different approach for single trial comparisons, where we started with the assumption that the quality of evidence was likely to be very low. This reflected assumptions of downgrading at a minimum for serious risk of bias (typically performance and detection bias), for serious indirectness (trial being conducted was a single trial or setting), and for serious imprecision (failure to meet the 200 to 300 events optimal size criteria) (Guyatt 2011). Where these assumptions did not hold, we performed GRADE assessment as above. The quality rating 'high' is reserved for a body of evidence based on randomised controlled trials. We ‘downgraded’ the quality rating to 'moderate', 'low' or 'very low' depending on the presence and extent of five factors: study limitations, inconsistency of effect, imprecision, indirectness or publication bias. We used the GRADE approach to assess quality of evidence related to the primary and secondary outcomes listed in the Types of outcome measures. We prepared a 'Summary of findings' table for each of the main categories of interventions, for listed outcomes.

We selected the following comparisons for presentation in 'Summary of findings' tables as these are the most common falls prevention activities considered and applied in clinical settings. In care facilities: exercise, vitamin D supplementation, medication review and multifactorial interventions; in hospitals: exercise, bed alarms and multifactorial interventions.

Results

Description of studies

Results of the search

For this update we screened a total of 3989 records from the following databases: Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (0 records); CENTRAL (127), MEDLINE (1104), Embase (1211), CINAHL (314) the WHO ICTRP (450) and Clinicaltrials.gov (783). We also found 29 potentially eligible studies from other sources. After removal of 503 duplicates, 3515 citations were screened for inclusion.

Screening of the search update identified a total of 413 records for potential inclusion, for which full‐text reports were obtained. Thirty‐five new trials were included in this update, 27 new ongoing trials identified and seven new studies await classification. In addition, a new subgroup analysis (Stenvall 2012) from the Stenvall 2007 trial and a cost‐effectiveness analysis (Haines 2013) of Haines 2011 have been added. A flow diagram summarising the study selection process is shown in Figure 1.

Overall, there are now 95 included trials, 105 excluded studies, eight studies awaiting classification and 31 ongoing trials.

Due to the review size, not all links to references have been inserted in the text but can be viewed in Table 8.

1. Description of included studies: reference links.

Study description	Links to references
Additional studies included in this update	Care facilities N = 28: Beck 2016; Buckinx 2014; Cadore 2014; Colon‐Emeric 2013; da Silva Borges 2014; Houghton 2014; Frankenthal 2014; Fu 2015; Garcia Gollarte 2014; Huang 2016; Imaoka 2016; Irez 2011; Juola 2015; Kennedy 2015; Kovacs 2012; Kovacs 2013; Peyro Saint Paul 2013; Potter 2016; Salvà 2016; Saravanakumar 2014; Sitja Rabert 2015; Streim 2012; Tuunainen 2013; Van de Ven 2014; Van het Reve 2014; Walker 2015; Whitney 2017; Yokoi 2015 Hospitals N = 7: Aizen 2015; Barker 2016; Hill 2015; Michalek 2014; Shorr 2012; Treacy 2015; Wolf 2013
Design	Cluster randomised N = 42: Aizen 2015; Barker 2016; Beck 2016; Becker 2003; Chenoweth 2009; Choi 2005; Colon‐Emeric 2013; Cox 2008; Crotty 2004b; Cumming 2008; Dyer 2004; Dykes 2010; Garcia Gollarte 2014; Haines 2010; Healey 2004; Hill 2015; Houghton 2014; Jensen 2002; Juola 2015; Kennedy 2015; Kerse 2004; Kerse 2008; Koh 2009; Lapane 2011; Law 2006; McMurdo 2000; Meyer 2009; Michalek 2014; Neyens 2009; Patterson 2010; Ray 1997; Rosendahl 2008; Salvà 2016; Sambrook 2012; Shorr 2012; Van de Ven 2014; Van Gaal 2011a; Van Gaal 2011b; Walker 2015; Ward 2010; Whitney 2017; Yokoi 2015
Setting (country)	Australia (N = 17): Barker 2016; Chenoweth 2009; Crotty 2004a; Crotty 2004b; Cumming 2008; Flicker 2005; Grieger 2009; Haines 2004; Haines 2010; Haines 2011; Hill 2015; Mador 2004; Potter 2016; Sambrook 2012; Saravanakumar 2014; Treacy 2015; Ward 2010  Belgium (N = 1): Buckinx 2014 Brazil (N = 1): da Silva Borges 2014 Denmark (N = 1): Beck 2016 Canada (N = 3): Kennedy 2015; Klages 2011; Mayo 1994 China (N = 1): Fu 2015 Finland (N = 3): Juola 2015; Sihvonen 2004; Tuunainen 2013 France (N = 3): Chapuy 2002; Peyro Saint Paul 2013; Toulotte 2003 Germany (N = 4): Becker 2003; Meyer 2009; Michalek 2014; Wolf 2013 Hungary (N = 2): Kovacs 2012; Kovacs 2013 Israel (N = 2): Aizen 2015; Frankenthal 2014 Korea (N = 1): Choi 2005 Japan (N = 5): Imaoka 2016; Sakamoto 2006; Sakamoto 2012; Shimada 2004; Yokoi 2015 The Netherlands (N = 5): Faber 2006; Neyens 2009; Van de Ven 2014; Van Gaal 2011a; Van Gaal 2011b New Zealand (N = 2): Kerse 2004; Kerse 2008 Singapore (N = 2): Ang 2011; Koh 2009 Spain (N = 5): Cadore 2014; Garcia Gollarte 2014; Salvà 2016; Serra‐Rexach 2011; Sitja Rabert 2015 Sweden (N = 3): Jensen 2002; Rosendahl 2008; Stenvall 2007 Switzerland (N = 2): Bischoff 2003; Van het Reve 2014 Taiwan (N = 1): Huang 2016 Turkey (N = 1): Irez 2011 United Kingdom (N = 14): Burleigh 2007; Cox 2008; Houghton 2014; Donald 2000; Dyer 2004; Healey 2004; Jarvis 2007; Law 2006; McMurdo 2000; Patterson 2010; Shaw 2003; Walker 2015; Whitney 2017; Zermansky 2006 USA (N = 16): Broe 2007; Buettner 2002; Clifton 2009; Colon‐Emeric 2013; Dykes 2010; Lapane 2011; Mulrow 1994; Nowalk 2001; Ray 1997; Rubenstein 1990; Schnelle 2003; Schoenfelder 2000; Shorr 2012; Streim 2012; Tideiksaar 1993; Wald 2011
Setting	Care facilities N = 71 High level nursing care N = 17: Beck 2016; Becker 2003; Bischoff 2003; Broe 2007; Chenoweth 2009; Clifton 2009; Crotty 2004a; Fu 2015; Imaoka 2016; Meyer 2009; Mulrow 1994; Neyens 2009; Ray 1997; Schnelle 2003; Schoenfelder 2000; Van de Ven 2014; Van Gaal 2011a; Intermediate level care N = 17: Buckinx 2014; Chapuy 2002; Choi 2005; da Silva Borges 2014; Dyer 2004; Irez 2011; Jensen 2002; Kerse 2008; Kovacs 2012; McMurdo 2000; Sakamoto 2006; Sakamoto 2012; Sambrook 2012; Serra‐Rexach 2011; Sihvonen 2004; Van het Reve 2014; Yokoi 2015 Mixed levels of care N = 37: Buettner 2002; Cadore 2014; Colon‐Emeric 2013; Cox 2008; Crotty 2004b; Houghton 2014; Faber 2006; Flicker 2005; Frankenthal 2014; Garcia Gollarte 2014; Grieger 2009; Huang 2016; Juola 2015; Kennedy 2015; Kerse 2004; Klages 2011; Kovacs 2013; Lapane 2011; Law 2006; Nowalk 2001; Patterson 2010; Peyro Saint Paul 2013; Potter 2016; Rosendahl 2008; Rubenstein 1990; Salvà 2016; Saravanakumar 2014; Shaw 2003; Shimada 2004; Sitja Rabert 2015; Streim 2012; Toulotte 2003; Tuunainen 2013; Walker 2015; Ward 2010; Whitney 2017; Zermansky 2006 Hospitals N = 24 Acute care N = 10: Ang 2011; Barker 2016; Dykes 2010; Koh 2009; Mador 2004; Shorr 2012; Stenvall 2007; Tideiksaar 1993; Van Gaal 2011b; Wald 2011 Subacute care N = 12: Aizen 2015; Burleigh 2007; Donald 2000; Haines 2004; Haines 2010; Healey 2004; Hill 2015; Jarvis 2007; Mayo 1994; Michalek 2014; Treacy 2015; Wolf 2013 Acute and subacute care N = 2: Cumming 2008; Haines 2011
Care facilities	Exercises N = 23: Buckinx 2014; Buettner 2002; Cadore 2014; Choi 2005; da Silva Borges 2014; Faber 2006; Fu 2015; Irez 2011; Kerse 2008; Kovacs 2012; Kovacs 2013; Mulrow 1994; Nowalk 2001; Rosendahl 2008; Sakamoto 2006; Saravanakumar 2014; Schoenfelder 2000; Serra‐Rexach 2011; Shimada 2004; Sihvonen 2004; Sitja Rabert 2015; Toulotte 2003; Tuunainen 2013; Yokoi 2015

Trial	Intervention	Control	Comment
Buckinx 2014	Whole body vibration exercise programme	Usual care: no change to lifestyle
Buettner 2002	Supervised group exercises	Usual care
Cadore 2014	Multicomponent exercises. Twice‐weekly, 40‐minute duration	Usual care: “mobility” exercises (30 minutes per day at least 4 days per week), small active and passive movements applied as stretches in a rhythmic fashion
Choi 2005	Tai Chi	Usual care: routine activities, without participation in any regular exercise classes
da Silva Borges 2014	Ballroom dancing	Usual care: agreed not to engage in any regular physical activity
Faber 2006	A functional balance, strength and mobility programme 3D (balance)	Usual care: no change usual pattern of activity	Both comparisons of interventions vs usual care considered under exercise vs usual care. Comparisons of interventions arms considered under comparisons of different exercise categories.
Fu 2015	Wii balance training (1 hour 3 x week)	Different exercise: Balance training (Otago) (1 hour, 3 x week)
Imaoka 2016	Reduced exercise ‐ individualised exercise only.	Different exercise: groups plus individualised exercises (described by study authors as usual care)
Irez 2011	Combination exercises: Pilates	Usual care: no Pilates, instructed not to change current activity levels
Kerse 2008	Activity programme	Usual care
Kovacs 2012	Multimodal exercise – Otago Exercise programme	Different exercise: Osteoporosis exercise programme, includes balance and strengthening exercises
Kovacs 2013	Multimodal exercise – Otago Exercise programme	Usual care: social activities such as board games, listening to music
Mulrow 1994	Tailored exercises	Usual care: friendly visit, usually involved reading to participant, avoided physical activity
Nowalk 2001	Supervised exercise Tai Chi Plus control (basic enhanced programme)	Usual care: basic enhanced programme including falls‐prevention programme with 3 education sessions and a walking programme	Results for interventions vs usual care as reported by study authors presented in Analysis 1.2 as data not suitable for calculation of RaR or RR.
Rosendahl 2008	Functional exercise programme	Usual care: Seated activities, including watching films, reading, singing
Sakamoto 2006	Single leg practice 1 min / leg, 3 x daily	Usual care: no details
Saravanakumar 2014	Tai Chi Flexibility (yoga)	Different exercise: "staying active": includes games, group activities, a gym with bike and activities such as walking and gardening	All comparisons presented under comparisons of different exercise categories.
Schoenfelder 2000	Ankle‐strengthening exercise	Usual care: little information
Serra‐Rexach 2011	Training sessions + usual care physiotherapy	Different exercise: usual care physiotherapy (40 to 45 minutes/day 5 x weekly)‐ stretches, aerobic exercise such as walking (though low intensity)
Shimada 2004	Gait exercises + usual exercises	Different exercise: physiotherapy for pain, stretches, low‐ and high‐intensity resistance training, gait training, stairs, lower limb function
Sihvonen 2004	Balance training (visual feedback)	Usual care: little information
Sitja Rabert 2015	Whole body vibration + exercise static and dynamic balance and strength exercise)	Different exercise: same exercise programme done on land
Toulotte 2003	Supervised exercises	Usual care: continued daily routine
Tuunainen 2013	Group strength training: Progressive resistance, supervised group training, 1 hour, 2x weekly Balance and strength training	Different exercise: self‐administered training (1 hour, 2 x weekly): Stretching, crouching and rising administered by nurses written instructions from physiotherapist	All comparisons presented under comparisons of different exercise categories.
Yokoi 2015	Group supervised seated stick exercises 25 minutes, 2 x weekly (included daily house‐keeping and hobbies for both exercise and control group)	Usual care: activities of daily living and 10‐minute group stretching exercises continued. No other exercises were conducted.

Study	Medication review	Control	Comment
Crotty 2004a	Additional pharmacist	Usual care
Crotty 2004b	Additional pharmacist	Usual care
Frankenthal 2014	Medication review	No interventional recommendations made by pharmacist to chief physician
Garcia Gollarte 2014	Physician education on drug use in older people, plus medication review in 10%	No intervention or information about an educational intervention	Falls data excludes the intervention period; not suitable for pooling
Houghton 2014	Multiprofessional medication review	Usual care (support from the NHS)
Juola 2015	Nursing education to reduce medication use	Usual care
Lapane 2011	Clinical informatics tool for medication review: providing reports to pharmacists and nursing staff to assist identifying residents at risk for delirium and falls. Reports generated within 24 hours of admission, used during monthly medication review and at time of Minimum Data Set reporting or when falls or delirium triggered resident assessment protocols.	Usual care (includes monthly medication review by pharmacist)
Patterson 2010	Pharmacist review of psychoactive medications	Usual care
Peyro Saint Paul 2013	Ceasing medication to avoid hyponatraemia	Usual care	Unusual study, not pooled with others
Potter 2016	Deprescribing	Medication review without deprescribing
Streim 2012	Deprescribing antidepressants	Continue taking antidepressants	Data not suitable for pooling.
Zermansky 2006	Medication review by pharmacist	Usual care

Risk of Bias	Low	High	Unclear
Sequence generation (selection bias)	69% (66/95)	2% (2/95)	28% (27/95)
Allocation (selection bias)	45% (43/95)	15% (14/95)	40% (38/95)
Blinding of participants and personnel (performance bias)	7% (7/95)	91% (86/95)	2% (2/95)
Blinding of outcome assessors (detection bias)	11% (10/95)	68% (65/95)	21% (20/95)
Incomplete outcome data (attrition bias)	63% (60/95)	27% (26/95)	9% (9/95)
Selective reporting (reporting bias)	53% (50/95)	8% (8/95)	39% (37/95)
Method of ascertaining falls	47% (45/95)	28% (27/95)	24% (23/95)
Baseline imbalance	54% (51/95)	27% (26/95)	19% (18/95)
Other bias	92% (87/95)	2% (2/95)	6% (6/95)

Exercise compared with usual care for falls prevention in care facilities
Population and setting: older (≥ 65 years) residents of care facilities Intervention: exercise Comparison: usual care
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk Usual care	Corresponding risk Exercise
Rate of falls Length of follow‐up: 3 to 12 months	Low‐risk population¹		RaR 0.93 (0.72 to 1.20)	2002 (10 studies)	+ooo VERY LOW⁷	These results were heterogeneous: subgroup analysis by type of exercise did not explain the heterogeneity. Four additional trials (N = 130) with data not suitable for pooling reported a reduction in the rate of falls.
	1000 per 1000 py	930 (720 to 1200) per 1000 py
	High‐risk population²
	3500 per 1000 py	3255 (2520 to 4200) per 1000 py
Risk of falling Length of follow‐up: 3 to 12 months	Low‐risk population³		RR 1.02 (0.88 to 1.18)	2090 (10 studies)	++oo LOW⁸	1 additional trial (2 comparisons, N = 110) reported no significant difference in the risk of falling.
	250 per 1000	255 (220 to 295) per 1000
	Moderate‐risk population⁴
	500 per 1000	510 (440 to 590) per 1000
	High‐risk population⁵
	700 per 1000	714 (616 to 826) per 1000
Risk of fracture Length of follow‐up: 6 months	Average risk population⁶		RR 0.88 (0.25 to 3.14)	183 1 study	+ooo VERY LOW⁹	This outcome poorly reported.
	42 per 1000	37 (11 to 132) per 1000
Adverse events Length of follow‐up: 4 to 12 months	See comment	See comment	Not estimable.	1032 (4 studies)	+ooo VERY LOW ¹⁰	1 serious adverse event reported (death due to a ruptured abdominal aortic aneurysm one week after the follow‐up tests, association could not definitely be ruled out) in 1 trial (183 participants). Three trials reported no differences in adverse events: 1 trial (639 participants) reporting aches and pains, P = 0.75 1 trial (194 participants) reported no statistical difference in severe soreness (10 exercise versus 11 control), severe bruises (2 versus 1), severe fatigue (4 versus 1) 1 trial reported no adverse events
Illustrative risks for the control group were derived from all or subgroups of trials in care facilities reporting the outcome. The exact basis for the assumed risk* for each outcome is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; py: person years; RaR: Rate Ratio; RR: Risk Ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

General medication review compared with usual care for falls prevention in care facilities
Population and setting: older (≥ 65 years) residents of care facilities Intervention: general medication review (NB: the primary aim of all medication review is to reduce psychoactive medications) Comparison: usual care
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Usual care	General medication review
Rate of falls Length of follow‐up: 6 to 12 months	Low‐risk population¹		RaR 0.93 (0.64 to 1.35)	2409 (6 studies)	++oo LOW⁷	The approaches taken in the six pooled studies were: medication review meeting involving clinical pharmacist, pharmacy technician, care home staff and GP(s) medication review with recommendations to chief physician based on STOPP/START criteria nurse education on harmful medications in older people monthly reviews of psychoactive medications medication review and deprescribing GP record + consultation with patient and care 2 additional RCTs found no strong evidence for an effect on the rate of falls (1 RCT of education of physicians on drug use in older people (716 participants, falls only reported following the intervention period); 1 trial of antidepressant deprescribing (36 participants randomised)).
	1000 per 1000 py	930 (640 to 1350)per 1000 py
	High‐risk population²
	3500 per 1000 py	3255 (2240 to 4725)per 1000 py
Risk of falling Length of follow‐up: 6 to 12 months	Low‐risk population³		RR 0.93 (0.80 to 1.09)	5139 (6 studies)	++oo LOW⁸	The approaches taken in the six studies were: pharmacist transition coordinator for patients discharged from hospital to nursing care facilities for the first time a pharmacist‐led outreach programme (audit + feedback + education of staff regarding medications and falls risk) nurse education on harmful medications in older people GRAM software for decision support for prescribing practices GP and a geriatrician / pharmacologist independently identifying deprescribing targets using a list of potentially inappropriate medicines vs medication review without deprescribing review of GP record + consultation with patient and carer 1 additional RCT of education of physicians on drug use in older people (716 participants) found no strong evidence for an effect on the risk of falling following the intervention period.
	250 per 1000	233 (200 to 273) per 1000
	Moderate‐risk population⁴
	500 per 1000	465 (400 to 545) per 1000
	High‐risk population⁵
	700 per 1000	651 (560 to 763) per 1000
Risk of fracture Length of follow‐up: 12 months	Average risk population⁶		RR 1.60 (0.28 to 9.16)	93 (1 trial)	+ooo VERY LOW⁹	Intervention was GP and a geriatrician/pharmacologist independently identifying deprescribing targets using a list of potentially inappropriate medicines vs medication review without deprescribing.
	42 per 1000	67 (12 to 614) per 1000
Adverse events Length of follow‐up: 12 months	Average risk population¹⁰		RR 1.07 (0.23 to 5.01)	93 (1 trial)	+ooo VERY LOW⁹	Serious vascular events in both trial arms and significant withdrawal reactions in 2 intervention participants (Potter 2016).
	60 per 1000	64 (14 to 301) per 1000
Illustrative risks for the control group were derived from all or subgroups of trials in care facilities reporting the outcome. The exact basis for the assumed risk for each outcome is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; py: person years; RaR: Rate Ratio; RR:** Risk Ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Vitamin D supplementation compared with no vitamin D supplementation for falls prevention in care facilities
Population and setting: older (≥ 65 years) residents of care facilities¹ Intervention: vitamin D supplementation (vitamin D or vitamin D + calcium) Comparison: usual care (or calcium supplementation)
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Vitamin D
Rate of falls Length of follow‐up: 3 to 24 months	Low‐risk population²		RaR 0.72 (0.55 to 0.95)	4512 (4 studies)	+++o MODERATE⁸	Studies included two studies of vitamin D3 + calcium versus calcium, and 2 studies of vitamin D2 versus usual care or placebo.
	1000 per 1000 py	720 (550 to 950)per 1000 py
	High‐risk population³
	3500 per 1000 py	2520 (1925 to 3325)per 1000 py
Risk of falling Length of follow‐up: 3 to 24 months	Low‐risk population⁴		RR 0.92 (0.76 to 1.12)	4512 (4 studies)	+++o MODERATE⁹	Studies included two studies of vitamin D3 + calcium versus calcium, and 2 studies of vitamin D2 versus usual care or placebo.
	250 per 1000	230 (190 to 280)per 1000
	Moderate‐risk population⁵
	500 per 1000	460 (380 to 515)per 1000
	High‐risk population⁶
	700 per 1000	644 (532 to 784)per 1000
Risk of fracture Length of follow‐up: 3 to 24 months	Average risk population⁷		RR 1.09 (0.58 to 2.03)	4464 (3 studies)	+ooo VERY LOW¹⁰	These studies represent only a subset of studies evaluating the effect of vitamin D on fractures. Included studies were two studies of vitamin D3 + calcium versus calcium, and 1 study of vitamin D2 versus usual care.
	42 per 1000	46 (24 to 85) per 1000
Adverse events Length of follow‐up: 3 to 24 months	ND¹²	ND¹²	RR 4.84 (0.24 to 98.90)	747 (2 studies)	+ooo VERY LOW¹¹	No serious events reported. Studies tested supplementation with 800 IU oral cholecalciferol (vitamin D3) and 1000 IU oral ergocalciferol (vitamin D2) daily. Data derived from just 2 cases of increased constipation in the intervention arm in 1 study (N = 122). No adverse events recorded in the other study (N = 625)
Illustrative risks for the control group were derived from all or subgroups of trials in care facilities reporting the outcome. The exact basis for the assumed risk* for each outcome is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; ND: not done; py: person years; RaR: Rate Ratio; RR: Risk Ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Multifactorial interventions compared with usual care for falls prevention in care facilities
Population and setting: older (≥ 65 years) residents of care facilities Intervention: multifactorial interventions (two or more categories of intervention given based on individual risk profile) Comparison: usual care (without intervention)¹
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Control	Vitamin D
Rate of falls Length of follow‐up: 6 to 12 months	Low‐risk population²		RaR 0.88 (0.66 to 1.18)	3439 (10 studies)	+ooo VERY LOW⁸	One additional study (31 participants) of exercise plus nutritional support reported zero falls in the intervention arm and two in the control arm.
	1000 per 1000 py	720 (550 to 950)per 1000 py
	High‐risk population³
	3500 per 1000 py	2520 (1925 to 3325)per 1000 py
Risk of falling Length of follow‐up: 6 to 12 months	Low‐risk population⁴		RR 0.92 (0.81 to 1.05)	3153 (9 studies)	++oo LOW⁹	One additional study (482 participants) reported a reduction in the proportion of recurrent fallers (difference 19%, 95% CI 2% to 36%: P = 0.03).
	250 per 1000	230 (190 to 280)per 1000
	Moderate‐risk population⁵
	500 per 1000	460 (380 to 515)per 1000
	High‐risk population⁶
	700 per 1000	644 (532 to 784)per 1000
Risk of fracture Length of follow‐up: 6 to 12 months	Average risk population⁷		RR 0.79 (0.30 to 2.07)	2160 (5 studies)	+ooo VERY LOW¹⁰
	42 per 1000	34 (13 to 87) per 1000
Adverse events Length of follow‐up: 11 weeks to 12 months	See comment	See comment	Not estimable.	312 (3 studies)	+ooo VERY LOW¹¹	One trial reported a case of a fall in the intervention arm; two studies reported no adverse events.
Illustrative risks for the control group were derived from all or subgroups of trials in care facilities reporting the outcome. The exact basis for the assumed risk* for each outcome is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; py: person years; RaR: Rate Ratio; RR: Risk Ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Additional exercise plus physiotherapy compared with usual physiotherapy for falls prevention in hospitals
Population and setting: older (≥ 65 years) patients in hospital settings Intervention: additional exercise plus physiotherapy Comparison: usual physiotherapy
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Usual physiotherapy	Additional Exercise
Rate of falls Length of follow‐up: inpatient stay (mean 29 days) or 2 weeks	Low‐risk population¹		RaR 0.59 (0.26 to 1.34)	215 (2 studies)	+ooo VERY LOW⁷	One study compared additional exercises versus conventional physiotherapy alone, and 1 study tested additional group standing balance circuit classes
	1300 per 1000 py	767 (338 to 1742) per 1000 py
	Moderate‐risk population²
	3500 per 1000 py	2065 (910 to 4690) per 1000 py
	High‐risk population³
	6000 per 1000 py	3540 (1560 to 8040) per 1000 py
Risk of falling Length of follow‐up: inpatient stay (mean 29 days) or 8 weeks	Low‐risk population⁴		RR 0.36 (0.14 to 0.93)	83 (2 studies)	+ooo VERY LOW⁸	One study compared additional exercises versus conventional physiotherapy alone, and 1 study tested additional daily physiotherapy sessions
	30 per 1000	11 (4 to 28) per 1000
	Moderate‐risk population⁵
	150 per 1000	54 (21 to 140) per 1000
	High‐risk population⁶
	340 per 1000	122 (48 to 316) per 1000
Risk of fracture	See comment	See comment	See comment			No data available
Adverse events Length of follow‐up: 2 weeks	0 events	0 events	Not estimable	161 (1 study)	+ooo VERY LOW⁹	One study reported no adverse events, two studies did not report this outcome
Illustrative risks for the control group were derived from all or subgroups of trials in hospitals reporting the outcome. The exact basis for the assumed risk* for each outcome is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; py: person years; RaR: Rate Ratio; RR: Risk Ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Bed alarms compared with usual care for falls prevention in hospitals
Population and setting: older (≥ 65 years) patients in hospital settings Intervention: bed alarms Comparison: usual care
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect  (95% CI)	No of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk	Corresponding risk
	Usual care	Bed alarms
Rate of falls Length of follow‐up: inpatient stay (mean 19 days; not known)	Low‐risk population¹		RaR 0.60 (0.27 to 1.34)	28,649 (2 studies)	++oo VERY LOW⁷	One cluster‐randomised study tested education and support on using bed/chair alarms; and one study tested sensor alarms fitted to patients' upper leg at rest time. A third study (n = 70) reported no difference in the number of falls (data not suitable for pooling).
	1300 per 1000 py	780 (351 to 1742)per 1000 py
	Moderate‐risk population²
	3500 per 1000 py	2100 (945 to 4690) per 1000 py
	High‐risk population³
	6000 per 1000 py	3600 (1620 to 8040) per 1000 py
Risk of falling Length of follow‐up: inpatient stay (mean 19 days; not known)	Low‐risk population⁴		RR 0.93 (0.38 to 2.24)	28,649 (2 studies)	+ooo VERY LOW⁸	One cluster‐randomised study tested education and support on using bed/chair alarms; and one study tested sensor alarms fitted to patients' upper leg at rest time
	30 per 1000	28 (11 to 67) per 1000
	Moderate‐risk population⁵
	150 per 1000	140 (57 to 336) per 1000
	High‐risk population⁶
	340 per 1000	316 (129 to 762) per 1000
Risk of fracture	See comment	See comment	See comment			No data available.
Adverse events Length of follow‐up: inpatient stay (mean 19 days; not known)	0 events	0 events	Not estimable.	27,742 (2 studies)	+ooo VERY LOW⁹	2 trials reported that there were no adverse events
Illustrative risks for the control group were derived from all or subgroups of trials in hospitals reporting the outcome. The exact basis for the assumed risk* for each outcome is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).  CI: Confidence interval; py: person years; RaR: Rate Ratio; RR: Risk Ratio;
GRADE Working Group grades of evidence  High quality: Further research is very unlikely to change our confidence in the estimate of effect.  Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.  Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.  Very low quality: We are very uncertain about the estimate.

Rate of falls and number of fallers: trials with incomplete data
Study	Intervention	Comparator	Participants (N)	Study findings (NR = not reported)
Buettner 2002	Exercise: Supervised group exercises, combination exercises.	Usual care	27	Rate of falls: Falls were reduced but the treatment effect estimate and confidence interval were not reported in the published study or research monograph. Risk of falling: NR
Cadore 2014	Exercise: Multicomponent exercise programme including gait/balance and strength/resistance training	Usual care including mobility exercises	24	Rate of falls: Over 12 weeks there were no falls in the multicomponent arm in comparison to a rate of falls of 0.8 falls per patient per month in the mobility exercises arm of the study (P < 0.001). Participants were aged > 85 years. Risk of falling: NR
da Silva Borges 2014	Exercise: Ballroom dancing (3D exercises; EG)	No regular physical activity (CG)	59	Rate of falls: The authors reported " fewer falls in the EG post‐test compared to the CG post‐test (p<0.0001)." Risk of falling: NR
Nowalk 2001	Exercise: 1. "Fit NB Free" Individually tailored combination exercises. 2. "Living and Learning/Tai Chi"	Usual routine activities	110	Rate of falls: NR Risk of falling: No significant difference in risk of falling (time to first fall) between either intervention group and the usual care group (P = 0.29).
Toulotte 2003	Exercise: Supervised exercises, combination exercises.	Usual care	20	Rate of falls: The authors reported that falls were reduced but a falls rate could not be determined from the published data. Risk of falling: NR

Study setting/type	Study ID	Gait/balance/ functional training	Strength/ resistance training	Flexibility	3D (Tai Chi, dance etc)	General physical activity	Endurance	Other
CARE FACILITIES
Single	Buckinx 2014	****	****					**** (WBV)
	Buettner 2002	****	****	****		****		****
	Cadore 2014	****	****	****
	Choi 2005				****
	da Silva Borges 2014			****	****
	Faber 2006 (FW)	****
	Faber 2006 (IB)	****	****	****	****
	Fu 2015							****
	Imaoka 2016							****
	Irez 2011	****	****	****
	Kerse 2008	****
	Kovacs 2012	****	****	****		****
	Kovacs 2013	****	****	****		****
	Mulrow 1994	****	****	****
	Nowalk 2001 (FNBF)		****	****
	Nowalk 2001 (LL/TC)				****
	Rosendahl 2008	****	****
	Sakamoto 2006	****
	Saravanakumar 2014 (Tai Chi)				****
	Saravanakumar 2014 (Yoga)			****
	Schoenfelder 2000		****			****
	Serra‐Rexach 2011		****	****			****
	Shimada 2004	****
	Sihvonen 2004	****
	Sitja Rabert 2015	****	****					**** (WBV)
	Toulotte 2003	****	****	****
	Tuunainen 2013 (MF)		****
	Tuunainen 2013(MFB)	****	****
	Yokoi 2015	****		****

Multiple	Huang 2016	****	****
	Imaoka 2016						****
	Schnelle 2003		****			****

Multifactorial	Beck 2016	****	****
	Becker 2003	****	****
	Dyer 2004	****	****	****		****
	Jensen 2002	****	****

Study setting/type	Study ID	Furnishing/adaptations	Personal mobility aids	Communication/signalling aids	Body worn care/protection aids	Other environmental
CARE FACILITIES
Single	Clifton 2009			****

Multifactorial	Becker 2003	****	****		****
	Dyer 2004	****
	Jensen 2002	****	****	****	****
	Kerse 2004	****	****		****
	McMurdo 2000	****
	Neyens 2009	****	****
	Ray 1997	****	****
	Rubenstein 1990	****
	Salvà 2016^a
	Shaw 2003	****	****		****
	Walker 2015	****	****		****
	Whitney 2017	****	****	****

HOSPITALS
Single	Donald 2000 (FL)	****
	Mayo 1994			****
	Haines 2010	****
	Shorr 2012			****
	Tideiksaar 1993			****
	Wolf 2013			****

Multifactorial	Aizen 2015	****	****		****
	Barker 2016	****		****
	Cumming 2008	****	****	****
	Haines 2004			****	****
	Healey 2004	****		****	****	****
	Stenvall 2007					**** Home visit by OT and/or PT

Setting/ Combination	Study ID	Vitamin D	Calcium	Other bone health medication	Antidepressants	Antipsychotics/ neuroleptics	Medication review	Other
CARE FACILITIES
Single	Bischoff 2003	****
	Broe 2007	****
	Chapuy 2002	****	****
	Crotty 2004a						**** Pharm
	Crotty 2004b						**** Pharm
	Houghton 2014						**** MultiP
	Flicker 2005	****
	Frankenthal 2014						**** Pharm
	Garcia Gollarte 2014						**** Educ
	Grieger 2009^a	****	****
	Imaoka 2016^a	****	****
	Juola 2015^b							****
	Kennedy 2015^c	****	****	****
	Law 2006	****
	Patterson 2010^d					****
	Peyro Saint Paul 2013^e						****
	Potter 2016
	Sambrook 2012	**** (UV)
	Streim 2012						**** Depresc
	Zermansky 2006						**** Pharm

Multiple	Imaoka 2016^a
	Sambrook 2012	**** (UV)	****

Multifactorial	Dyer 2004						****
	Jensen 2002						****
	McMurdo 2000						****
	Neyens 2009						****
	Ray 1997					****
	Rubenstein 1990						****
	Salvà 2016						****
	Shaw 2003						****
	Walker 2015
	Whitney 2017						****

HOSPITALS
	Burleigh 2007	****
	Michalek 2014						****

Study ID	Intervention group: falls per person year	Control group: falls per person year	Intervention group: number of fallers	Intervention group: number in analysis	Intervention group: proportion of fallers	Control group: number of fallers	Control group: number in analysis	Control group: proportion of fallers
CARE FACILITIES
Beck 2016	0	0.43	‐‐‐	9	‐‐‐	‐‐‐	22
Becker 2003	1.40	2.56	188	509	0.37	247	472	0.52
Becker 2003 (Cognitively impaired)	1.10	2.71	50	150	0.33	98	169	0.58
Becker 2003 (Not cognitively impaired)	1.42	2.04	93	215	0.43	91	191	0.48
Bischoff 2003	‐‐‐	‐‐‐	14	62	0.23	18	60	0.30
Broe 2007 (800 IU)	0.28	1.00	5	23	0.22	11	25	0.44
Buckinx 2014	1.16	1.21	15	31	0.48	17	31	0.55
Buettner 2002	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐
Cadore 2014	0	9.6	‐‐‐	11	‐‐‐	‐‐‐	13	‐‐‐
Chapuy 2002	‐‐‐	‐‐‐	251	393	0.64	118	190	0.62
Chenoweth 2009	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐
Choi 2005	‐‐‐	‐‐‐	9	29	0.31	15	30	0.50
Clifton 2009	2.45	3.79	‐‐‐	43	‐‐‐	‐‐‐	43	‐‐‐
Colon‐Emeric 2013	2.06^a	2.64^a	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐
Cox 2008^b	‐‐‐	‐‐‐	‐‐‐	3315	‐‐‐	‐‐‐	2322	‐‐‐
Crotty 2004a	‐‐‐	‐‐‐	19	44	0.43	16	44	0.36
Crotty 2004b	‐‐‐	‐‐‐	97	381	0.26	73	334	0.22
da Silva Borges 2014	‐‐‐	‐‐‐	‐‐‐	30	‐‐‐	‐‐‐	29	‐‐‐
Houghton 2014	3.32	3.0	‐‐‐	381	‐‐‐	‐‐‐	445	‐‐‐
Dyer 2004	2.17	4.02	56	102	0.55	51	94	0.54
Faber 2006 (FW)	3.3	2.5	40	64	0.63	48	90	0.53
Faber 2006 (IB)	2.4	2.5	45	78	0.58	48	90	0.53
Faber 2006 (FW + IB)	2.8	2.5	85	142	0.60	48	90	0.53
Flicker 2005	1.26	1.90	170	313	0.54	185	312	0.59
Frankenthal 2014	0.80	1.30	‐‐‐	160	‐‐‐	‐‐‐	146	‐‐‐
Fu 2015	0.54	1.52	‐‐‐	30	‐‐‐	‐‐‐	30	‐‐‐
Garcia Gollarte 2014	1.28	1.72	82	344	0.24	104	372	0.28
Grieger 2009	0.60	1.60	11	48	0.23	12	43	0.28
Huang 2016 (CB)	0.00	1.67	0	25	0.00	7	24	0.29
Huang 2016 (CB + EX)	0.00	1.67	0	24	0.00	7	24	0.29
Imaoka 2016 (RED EX)	‐‐‐	‐‐‐	7	22	0.32	9	17	0.53
Imaoka 2016 (Vit D)	‐‐‐	‐‐‐	6	17	0.35	9	17	0.53
Imaoka 2016 (multiple)	‐‐‐	‐‐‐	4	19	0.21	9	17	0.53
Irez 2011	1.60	5.63	‐‐‐	30	‐‐‐	‐‐‐	30	‐‐‐
Jensen 2002	2.45	3.03	82	188	0.44	109	196	0.56
Jensen 2002 (MMSE < 19)	3.50	3.34	37	69	0.54	62	102	0.61
Jensen 2002 (MMSE ≥ 19)	1.77	2.90	42	112	0.38	43	79	0.54
Juola 2015	2.25	3.25	42	93	0.45	60	96	0.63

Date	Event	Description
7 September 2018	Amended	NIHR acknowledgement added
10 August 2018	New search has been performed	For this update, the following changes were made. Search updated to August 2017. 35 new trials added. The classification of social environment interventions has been reconsidered. Stenvall 2007 has been reclassified as a social environment intervention (previously multifactorial). Koh 2009 and Van Gaal 2011b are still classified within the social environment ProFaNE category but considered as organisational service model change rather than staff training. Trials on medication review in care facilities reclassified according to medication target, rather than according to the type of health professional performing the review. Additional subgroup analysis by level of care conducted for multifactorial interventions in hospitals. Background section revised and citations updated. Risk of bias conducted for additional items for previously included trials according to current Cochrane guidelines. Overall quality of evidence for main comparisons assessed according to GRADE. A new cost‐effectiveness analysis of Haines 2011 (Haines 2013) has been added. Exercise interventions are reported according to the ProFaNE exercise category and the comparator arm of the trial.
10 August 2018	New citation required and conclusions have changed	New evidence, the reclassification of some intervention categories and the implementation of new methods, including assessment of the quality of the evidence using GRADE, has resulted in some changed conclusions. Changes made to authorship, including addition of new authors.
27 February 2013	Feedback has been incorporated	Changes relate to two pieces of feedback, received 19 December 2013 and 12 February 2013. Two Summary [of feedback] and Reply entries were added to the Feedback section. There were no changes to the review in relation to the second piece of feedback. Changes in relation to the first piece included: 1. Appendix 6 was revised and Appendices 7 and 8 were deleted.  2. A new Appendix 7, containing raw data, was added.  3. Sections of the review (principally, the 'Description of studies') were revised to reflect these changes.
9 November 2012	New search has been performed	For this update, published in Issue 12, 2012, the following changes were made: 1. Search updated to March 2012  2. Twenty additional trials (35,270 participants) included in this update  3. One previously included trial recruiting people post stroke (Barreca 2004) excluded, as no longer within the scope of this version of the review  4. Kerse 2008 reclassified as an exercise intervention (formerly multifactorial)  5. Additional trials testing multifactorial interventions with results for subgroups with and without cognitive impairment  6. Evidence relating to additional interventions, these include: patient education in hospital (Ang 2011; Haines 2011), dementia care mapping (Chenoweth 2009), motion sensors (Clifton 2009), decision‐support software (Dykes 2010; Lapane 2011), multivitamin supplementation (Grieger 2009), low‐low beds (Haines 2010), multisensory stimulation (Klages 2011), guideline implementation (Koh 2009; Van Gaal 2011a; Van Gaal 2011b), a fall risk assessment tool (Meyer 2009), increased sunlight exposure (Sambrook 2012), lavender oil stimulation (Sakamoto 2012), an acute care service for elderly people (Wald 2011)  7. One newly included trial included a cost‐effectiveness analysis (Chenoweth 2009)  8. Background section revised and citations updated  9. 'Risk of bias' item relating to 'Allocation concealment' split into two: 'Sequence generation' and 'Allocation concealment' and applied to all included studies  10. Subgroup analyses revised
9 November 2012	New citation required and conclusions have changed	1. In response to the external referee's comments, the title of this review has been changed to reflect the fact that facilities which do not include nursing care are also included in this review.  2. Change in conclusion for multifactorial interventions in care facilities from no evidence of effect to a suggestion of possible benefits. Evidence from one trial for the effectiveness of an educational session targeting identified risk factors in acute hospital setting.
30 November 2009	Amended	Correction of two minor errors
23 September 2009	Amended	The published review 'Interventions for preventing falls in elderly people' (Gillespie 2003) is not being updated. Due to its size and complexity it was split into two reviews: 'Interventions for preventing falls in older people living in the community' and 'Interventions for preventing falls in older people in nursing care facilities and hospitals'
1 April 2009	Amended	Converted to new review format

Bias	Judgement of risk of bias: LOW, HIGH, or UNCLEAR
Random sequence generation Relating to selection bias (biased allocation to interventions) due to inadequate generation of a randomised sequence	According to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions.
Allocation concealment Relating to selection bias (biased allocation to interventions) due to inadequate concealment of allocations prior to assignment	According to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions. In cluster randomised trials, if patients were recruited following allocation of the cluster, this was considered as high risk. The timing of recruitment of individuals to clusters was considered within this domain.
Blinding of participants and personnel  Relating to performance bias due to knowledge of the allocated  interventions by participants and personnel carrying out the interventions	According to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions.
Blinding of outcome assessment Relating to detection bias due to knowledge of the allocated interventions by outcome assessors	According to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions.
Incomplete outcome data Relating to attrition bias due to amount, nature or handling of  incomplete outcome data	According to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions. For cluster‐randomised trials, potential bias due to loss of clusters was considered within this domain.
Selective outcome reporting Relating to bias due to the selective reporting or non reporting of findings	According to recommendations in the Cochrane Handbook for Systematic Reviews of Interventions. Where no protocol was identified, but all expected falls outcomes were reported and appropriate adjustments for clustering were performed, a ‘low risk’ rather than unclear judgement was made.
Method of ascertaining falls  Relating to bias in the recall of falls due to unreliable methods of ascertainment	All studies were assessed as follows. Judgement of ’Low risk’ if the study used a clear definition of falls plus some form of concurrent collection of data about falling, e.g. staff recorded falls daily on a hospital register.  Judgement of ’High risk’ if ascertainment relied on participant recall at longer intervals than one month during the study or at its conclusion, or if there were important differences in the methods of ascertainment of falls between study arms, or falls were poorly defined. Judgement of ’Unclear’ if there was retrospective recall over a short period only, or a definition of falls was not described, or details of ascertainment were not described, i.e. insufficient information was provided to allow a judgement of ’Low risk’ or ’High risk’.
Bias resulting from major baseline imbalances Relating to bias resulting from major imbalances in key baseline characteristics	Judgement of 'Low risk' if good comparability of groups, or confounding adjusted for in analysis. Judgement of ‘High risk’ if imbalance in characteristics likely to impact on falls rate (particularly age, previous falls/falls risk, medical status, dependency, cognitive function) and confounding not adjusted for in analysis. Judgement of ‘Unclear’ if not discussed.

Setting/ Combination	Study ID	Exercises	Medication (drug target)	Management of urinary incontinence	Fluid or nutritional therapy	Environment/ assistive technology	Social environment	Knowledge	Other
CARE FACILITIES
Single	Bischoff 2003		****
	Broe 2007		****
	Buckinx 2014	****
	Buettner 2002	****
	Cadore 2014	****
	Chapuy 2002		****
	Chenoweth 2009						****
	Choi 2005	****
	Clifton 2009					****
	Colon‐Emeric 2013						****
	Cox 2008						****
	Crotty 2004a		****
	Crotty 2004b		****
	da Silva Borges 2014	****
	Houghton 2014		****
	Faber 2006	****
	Flicker 2005		****
	Frankenthal 2014		****
	Fu 2015	****
	Garcia Gollarte 2014		****
	Grieger 2009		****
	Huang 2016 (CB)								**** Psychological
	Imaoka 2016 (RED EX)	****
	Imaoka 2016 (Vit D)		****
	Irez 2011	****
	Juola 2015		****
	Kennedy 2015		****
	Kerse 2008	****
	Klages 2011								****  Multisensory stimulation
	Kovacs 2012	****
	Kovacs 2013	****
	Lapane 2011		****				****
	Law 2006		****
	Meyer 2009						****
	Mulrow 1994	****
	Nowalk 2001	****
	Patterson 2010		****
	Peyro Saint Paul 2013		****
	Potter 2016		****
	Rosendahl 2008	****
	Sakamoto 2006	****
	Sakamoto 2012								**** Lavender patches
	Sambrook 2012 (UV)								**** Sunlight
	Saravanakumar 2014	****
	Schoenfelder 2000	****
	Serra‐Rexach 2011	****
	Shimada 2004	****
	Sihvonen 2004	****
	Sitja Rabert 2015	****
	Streim 2012		****
	Toulotte 2003	****
	Tuunainen 2013	****					****
	Van de Ven 2014
	Van Gaal 2011a						****
	Van het Reve 2014								**** Psychological
	Ward 2010						****
	Yokoi 2015	****
	Zermansky 2006		****

Multiple	Huang 2016	****							****
	Imaoka 2016	****	****
	Schnelle 2003	****		****	****
	Sambrook 2012 (UV+)		****						**** Sunlight

Multifactorial	Beck 2016	****			****
	Becker 2003	****				****	****	****
	Dyer 2004	****	****			****	****		**** Podiatry referral
	Jensen 2002	****	****			****	****
	Kerse 2004		****	****		****	****
	McMurdo 2000	****	****			****		****
	Neyens 2009	****	****			****	****
	Ray 1997		****			****	****	****
	Rubenstein 1990		****			****
	Salvà 2016^a	****	****	****		****
	Shaw 2003	****	****			****
	Walker 2015	****	****	****	****	****		****
	Whitney 2017	****	****			****	****

HOSPITALS
Single	Ang 2011							****
	Burleigh 2007		****
	Donald 2000 (2 x 2 factorial)	****				****
	Dykes 2010						****
	Haines 2010					****
	Haines 2011							****
	Jarvis 2007	****
	Koh 2009						****
	Mador 2004						****
	Mayo 1994					****
	Michalek 2014		****
	Shorr 2012					****
	Stenvall 2007						****
	Tideiksaar 1993					****
	Treacy 2015	****
	Van Gaal 2011b						****
	Wald 2011						****
	Wolf 2013					****

Multifactorial	Aizen 2015		****	****		****	****		Psychological
	Barker 2016			****		****	****
	Cumming 2008	****	****			****	****	****
	Haines 2004	****				****		****
	Healey 2004		****			****			****  Opthalmology referral
	Hill 2015						****	****

Study ID	Source for rate ratio  (falls)	Source for risk ratio  (fallers)	Source of risk ratio  (number with fractures)
Aizen 2015	ND	ND	NA
Ang 2011	NA	4	NA
Barker 2016	1b	7c	ND
Beck 2016	ND	NA	NA
Becker 2003	1b	5b	7c
Becker 2003  (Cognitively impaired/not impaired subgroup analysis)	1	5	NA
Bischoff 2003	1a	5a	7
Broe 2007 (800 IU)	1a	4a	NA
Buckinx 2014	3	7	NA
Buettner 2002	ND	NA	NA
Burleigh 2007	ND	5	7
Cadore 2014	ND	NA	NA
Chapuy 2002	NA	7	7
Chenoweth 2009	NA	ND	NA
Choi 2005	NA	7c	NA
Clifton 2009	3	NA	NA
Colon‐Emeric 2013	ND	NA	NA
Cox 2008	1ab	NA	ND
Crotty 2004a	NA	5	NA
Crotty 2004b	NA	5ab	NA
Cumming 2008	1ab	7c	7c
da Silva Borges 2014	ND	NA	NA
Donald 2000	3	5	NA
Dyer 2004	3c	6b	NA
Dykes 2010	3c	7c	NA
Faber 2006	3	4 (FW vs control and IB vs control) 4a (FW + IB vs control)	NA
Flicker 2005	1	4	7
Frankenthal 2014	3	NA	NA
Fu 2015	1a	NA	NA
Garcia Gollarte 2014	ND	ND	NA
Grieger 2009	3	7	NA
Haines 2004	3	5	7
Haines 2010	3c	NA	NA
Haines 2011	2a	6a	NA
Healey 2004	3c	NA	NA
Hill 2015	1ab	ND^a	ND^a
Houghton 2014	1b	NA	NA
Huang 2016	ND	ND	NA
Imaoka 2016	ND	ND	NA
Irez 2011	3	NA	NA
Jarvis 2007	ND	7	NA
Jensen 2002	1b	4b	6a
Jensen 2002  (MMSE < 19/ ≥ 19 subgroup analysis)	1b	7c	NA
Juola 2015	1ac	7c	NA
Juola 2015 (MMSE >15, 10‐15, <10 subgroups)	3ac	NA	NA
Kennedy 2015	3c	7c	ND
Kerse 2004	1ab	7c	NA
Kerse 2008	2b	7c	NA
Klages 2011	ND	NA	NA
Koh 2009	3c	NA	NA
Kovacs 2012	NA	5	NA
Kovacs 2013	1	5	NA
Lapane 2011	NA	4b	NA
Law 2006	3c	7c	5ab
Mador 2004	NA	7	NA
Mayo 1994	3	4	NA
McMurdo 2000	3c	7c	7c
Meyer 2009	3c	7c	7c
Michalek 2014	3c	7c	NA
Mulrow 1994	3	7	NA
Neyens 2009	1b	NA	NA
Nowalk 2001	NA	ND	NA
Patterson 2010	3c	NA	NA
Peyro Saint Paul 2013	3	7	NA
Potter 2016	3	7	7
Ray 1997	NA	ND	NA
Rosendahl 2008	1c	7c	7c
Rubenstein 1990	3	7	7
Sakamoto 2006	3	7	7
Sakamoto 2012	1	4	NA
Salvà 2016	1ab	7c	7c
Salvà 2016 (subgroup excluding dementia)	1ab	6ab	NA
Sambrook 2012	1c	7c	7c
Saravanakumar 2014	3	NA	NA
Schnelle 2003	3	7	7
Schoenfelder 2000	3	NA	NA
Serra‐Rexach 2011	ND	NA	NA
Shaw 2003	ND	5	5
Shimada 2004	3	7	NA
Shorr 2012	3ab	7c	NA
Sihvonen 2004	1a	7	NA
Sitja Rabert 2015	ND	7	7
Stenvall 2007	1	4	ND
Stenvall 2007 (dementia subgroup in Stenvall 2012)	1	7	7
Streim 2012	ND	NA	NA
Tideiksaar 1993	ND	NA	NA
Toulotte 2003	ND	NA	NA
Treacy 2015	1	NA	NA
Tuunainen 2013	3	7	NA
Van de Ven 2014	3c	NA	NA
Van Gaal 2011a	1c	NA	NA
Van Gaal 2011b	1c	NA	NA
Van het Reve 2014	3	7	NA
Wald 2011	3	NA	NA
Walker 2015	3c	NA	NA
Ward 2010	ND	NA	7c
Whitney 2017	1b	5a	7c
Wolf 2013	3	7	NA
Yokoi 2015	NA	7c	NA
Zermansky 2006	3	7	NA
^aData reported as admissions not patients Abbreviations  FW: 'Functional Walking' group  IB: 'In Balance' group MMSE: Mini Mental State Examination  800 IU: 800 International Units vitamin D group Codes for source of rate ratio:  1: incidence rate ratio reported by trial authors  2: hazard ratio/relative hazard (multiple events) reported by trial authors  3: incidence rate ratio calculated by review authors a: adjusted for confounders by trial authors  b: adjusted for clustering by trial authors  c: adjusted for clustering by review authors Codes for source of risk ratio:  4: hazard ratio/relative hazard (first fall only) reported by trial authors  5: relative risk reported by trial authors  6: odds ratio reported by trial authors  7: relative risk calculated by review authors a: adjusted for confounders by trial authors  b: adjusted for clustering by trial authors  c: adjusted for clustering by review authors NA: not applicable. Falls (for rate ratio) or fallers (for risk ratio) or number of people sustaining a fracture (for risk ratio) not reported as an outcome in the trial  ND: outcomes relating to falls or fallers or fractures were reported, but there were no useable data; results from the paper reported in the text of the review

Study ID	Intervention group: number of people with fractures	Intervention group: number in analysis	Intervention group: proportion of fracture fallers	Control group: number of people with fractures	Control group: number in analysis	Control group: proportion of fracture fallers
CARE FACILITIES
Beck 2016	‐‐‐	9	‐‐‐	‐‐‐	22	‐‐‐
Becker 2003: hip	17	509	0.033	15	472	0.032
Bischoff 2003: hip	2	62	0.032	1	60	0.017
Broe 2007 (800 IU)	‐‐‐	23	‐‐‐	‐‐‐	25
Buckinx 2014	‐‐‐	31	‐‐‐	‐‐‐	31
Buettner 2002	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐
Cadore 2014	‐‐‐	11	‐‐‐	‐‐‐	13
Chapuy 2002: NV	70	393	0.178	34	190	0.179
Chapuy 2002: hip	27	393	0.069	21	190	0.111
Chenoweth 2009	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐
Choi 2005	‐‐‐	29	‐‐‐	‐‐‐	30
Clifton 2009	‐‐‐	43	‐‐‐	‐‐‐	43
Colon‐Emeric 2013	‐‐‐	‐‐‐	‐‐‐	‐‐‐	‐‐‐
Cox 2008^a	‐‐‐	3315	‐‐‐	‐‐‐	2322
Crotty 2004a	‐‐‐	44	‐‐‐	‐‐‐	44
Crotty 2004b	‐‐‐	381	‐‐‐	‐‐‐	334
da Silva Borges 2014	‐‐‐	30	‐‐‐	‐‐‐	29
Houghton 2014	‐‐‐	381	‐‐‐	‐‐‐	445
Dyer 2004	4	102	0.039	3	94	0.032
Faber 2006 (FW)	‐‐‐	64	‐‐‐		90
Faber 2006 (IB)	‐‐‐	78	‐‐‐		90
Faber 2006 (FW + IB)	‐‐‐	142	‐‐‐		90
Flicker 2005	25	313	0.080	35	312	0.112
Frankenthal 2014	‐‐‐	160	‐‐‐		146
Fu 2015	‐‐‐	30	‐‐‐		30
Garcia Gollarte 2014	‐‐‐	344	‐‐‐		372
Grieger 2009	‐‐‐	48	‐‐‐		43
Huang 2016	‐‐‐	51	‐‐‐		24
Imaoka 2016 (RED EX)	‐‐‐	22	‐‐‐		17
Imaoka 2016 (Vit D)	‐‐‐	17	‐‐‐		17
Imaoka 2016 (multiple)	‐‐‐	19	‐‐‐		17
Irez 2011	‐‐‐	30	‐‐‐		30
Jensen 2002	3	188	0.016	12	196	0.061
Juola 2015	‐‐‐	93	‐‐‐		96
Kennedy 2015	‐‐‐	1290	‐‐‐		2727
Kerse 2004	‐‐‐	309	‐‐‐		238
Kerse 2008	‐‐‐	310	‐‐‐		329
Klages 2011	‐‐‐	‐‐‐	‐‐‐		‐‐‐
Koh 2009	‐‐‐	612	‐‐‐		510
Kovacs 2012	‐‐‐	21	‐‐‐		20
Kovacs 2013	‐‐‐	32	‐‐‐		30
Lapane 2011	‐‐‐	1769	‐‐‐		1552
Law 2006: NV	64	1762	0.036	51	1955	0.026
Law 2006: hip	24	1762	0.014	20	1955	0.010
McMurdo 2000	1	52	‐‐‐	3	38
Meyer 2009	39	574	0.068	38	551	0.069
Mulrow 1994	‐‐‐	97			97
Neyens 2009	‐‐‐	249			269
Nowalk 2001 (LL/TC)	‐‐‐	‐‐‐			‐‐‐
Nowalk 2001 (FNBF)	‐‐‐	‐‐‐			‐‐‐
Patterson 2010	‐‐‐	173			161
Peyro Saint Paul 2013	‐‐‐	4			5
Potter 2016	3	45	0.067	2	48	0.042
Ray 1997	‐‐‐	‐‐‐			‐‐‐
Rosendahl 2008	4	87	0.046	6	96	0.063
Rubenstein 1990	7	79	0.089	5	81	0.062
Sakamoto 2006: hip	1	315	0.003	1	212	0.005
Sakamoto 2012	‐‐‐	73			72
Salvà 2016	10	193	0.052	1	137	0.007
Salvà 2016 (excluding dementia)	‐‐‐	‐‐‐			‐‐‐
Sambrook 2012 (UV)	17	190	0.089	17	205	0.083
Sambrook 2012 (UV+)	13	207	0.063	17	205	0.083
Saravanakumar 2014	‐‐‐
Schnelle 2003	4	92	0.043	1	98	0.010
Schoenfelder 2000	‐‐‐	9			7
Serra‐Rexach 2011	‐‐‐	‐‐‐			‐‐‐
Shaw 2003	6	130	0.046	12	144	0.083
Shimada 2004	‐‐‐	15			11
Sihvonen 2004	‐‐‐	20			7
Sitja Rabert 2015	1	81	0.012	0	78	0
Streim 2012	‐‐‐	‐‐‐			‐‐‐
Toulotte 2003	‐‐‐	‐‐‐			‐‐‐
Tuunainen 2013 (MF)	‐‐‐	16			18
Tuunainen 2013 (MFB)	‐‐‐	14			18
Van de Ven 2014	‐‐‐	137			156
Van Gaal 2011a	‐‐‐	196			196
Van het Reve 2014	‐‐‐	54			60
Walker 2015	‐‐‐	22			20
Ward 2010	109	2802	0.039	106	2589	0.041
Whitney 2017	3	103	0.029	0	88	0
Yokoi 2015	‐‐‐	51			54
Zermansky 2006	‐‐‐	331			330
HOSPITALS
Aizen 2015	‐‐‐	200			308
Ang 2011	‐‐‐	910			912
Barker 2016	11	17698	0.0006	13	17566	0.0007
Burleigh 2007	1	100	0.010	3	103	0.029
Cumming 2008	2	2047	0.001	3	1952	0.002
Donald 2000 (FL)	‐‐‐	28			26
Donald 2000 (EX)	‐‐‐	30			24
Dykes 2010	‐‐‐	2755			2509
Haines 2004	2	310	0.006	2	316	0.006
Haines 2010	‐‐‐	6113			4986
Haines 2011 (ED)	‐‐‐	424			381
Haines 2011 (ED+)	‐‐‐	401			381
Healey 2004	‐‐‐	749			905
Hill 2015	4^b	1623	‐‐‐	6^b	1983	‐‐‐
Jarvis 2007	‐‐‐	14			15
Mador 2004	‐‐‐	36			35
Mayo 1994	‐‐‐	65			69
Michalek 2014	‐‐‐	58			56
Shorr 2012	‐‐‐	11115			17436
Stenvall 2007	0	102	0	4	97	0.041
Stenvall 2007 dementia subgroup (Stenvall 2012)	0	28	0	3	36	0.083
Tideiksaar 1993	‐‐‐	‐‐‐			‐‐‐
Treacy 2015	‐‐‐	‐‐‐			‐‐‐
Van Gaal 2011b	‐‐‐	1081			1120
Wald 2011	‐‐‐	122			95
Wolf 2013	‐‐‐	48			50
^aRaw data not available, data reported by authors as rate ratios. ^badmissions Abbreviations ED: educational materials only group  ED+: educational materials plus physiotherapist follow‐up  EX: supplementary exercises group  FL: carpet flooring group  FNBF: 'Fit NB Free' group  FW: 'Functional Walking' group (a functional balance, strength & mobility programme)  IB: 'In Balance' group  LL/TC: 'Living and learning/Tai Chi' group MF: muscle force  MFB: muscle force & balance  NV: non‐vertebral  RED EX: reduced exercise  UV: increased sunlight exposure group.  UV+: increased sunlight exposure + calcium supplementation group  Vit D: Vitamin D3 & calcium in multivitamin supplement  800 IU: 800 International Units vitamin D group

Study ID (source if not primary reference), sample, efficacy analyses, type of evaluation	Intervention(s) and comparator (N in analysis)	Perspective(s), type of currency, price year, time horizon	Cost items measured	Mean (SD) intervention cost per person	Healthcare service costs	Incremental cost per fall prevented/per QALY gained
•Buettner 2002 •Residents of 3 dementia care units (Oxford, Boston, and Palo Alto, USA) ≥ 2 falls in 1 month, mean age 83 (range 60 to 98) years •No effectiveness data available for analysis •Cost analysis	•Daily "graded" walking, "exercise for function" programme 3 x week, sensory air mat 2 x week (evenings) for 2 months vs usual care, number allocated to each group not reported (total N = 27)	•Not stated •US dollar •Not stated •2 months	•Therapist time (intervention only) •Cost of falls and injuries ("based on research data on falls")		•Treatment group USD 30,031, control group USD 79,535
•Chenoweth 2009 (Norman 2008) •Residents from 15 dementia care sites across Sydney, Australia, category 1 to 3 on Australian Resident Classification Scale (high level of care), mean age 84 (SD 7) years •No effectiveness data available for analysis •Cost‐effectiveness analysis	•Dementia care mapping (DCM) (N = 109, 5 sites) vs person centred care (PCC) (N = 98, 5 sites) vs usual care (N = 82, 5 sites) for 4 months	•Health service •Australian dollar •2008 •8 months	•Trainer time, post‐training support, staff replacement (DCM, PCC) •Pharmaceutical use	•Not reported (annual total cost per residential care setting DCM AUD 10,034, PCC AUD 2250)	•Annual pharmaceutical cost per resident AUD 545.55	•Not reported •Incremental cost per behaviour (CMAI point) averted DCM vs usual care AUD 46.89, PCC vs usual care AUD 6.43
•Clifton 2009 •Skilled nursing care‐facility residents, Eastern Washington State, USA, mean age 82 (SD 7) years •Analysis 7.1 •Analytic model	•Wear FallSaver monitor for 60 days (N = 33) vs no device for 60 days (N = 39), cross‐over trial	•Not stated •US dollar •2004 •1 year	•Annual intervention implementation for 100 residents (direct costs only) •Mean hospitalisation cost for injurious fall (from the literature)	•USD 2 per resident per day (annual cost for 100 resident facility USD 73,000)	•Assuming 35 injurious falls per 100 residents per year, annual cost savings for 100 resident facility if 12% fewer injurious falls USD 429, USD 232,953 if 50% fewer injurious falls
Houghton 2014 (Sach 2015) •Residents from care homes with average age > 65, registered with GP in local area and registered with Care Quality Commission for at least 6 months. •Analysis 5.1 •Cost analysis, detailed micro‐costing	• Multiprofessional medication review (N = 826)	•NHS and care homes •Pound sterling •2012 •1 year	•Intervention costs: personnel and resources, Staff costs for time spent on reviews, travel time and costs •Medication costs •Healthcare resource use •Hospitalisations	•GBP 104.80 (SD 50.91) per resident
•Haines 2013 (analysis of Haines 2011) •Acute and Rehabilitation hospital inpatients age ≥ 60, Brisbane and Perth, Australia, mean age 75 (SD 11) years •Analysis 20.1, Analysis 20.2 •Cost‐effectiveness analysis	•Multimedia patient education programme with physiotherapist follow up (total N = 1,206)	•Health service provider •Australian dollar •2008 •Period of hospitalisation	•Acute care costs •Rehabilitation costs •Direct falls related costs: radiological investigations, medical costs, nursing costs, medication costs, on‐call payment costs, suture procedure costs, orthoses costs, and other tests costs.		Intervention/control group costs post consent per participant (mean (SD) AUD) Subgroup cognitively intact: Intervention cost (complete programme) •Acute care AUD 10,774 (18,344) •Rehabilitation AUD 11,197 (18,906) •Direct falls related costs AUD 1 (7) Control cost •Acute care AUD 8,481  (12,856) •Rehabilitation AUD 10,964  (19,972) •Direct falls related costs intact AUD 8 (47) Subgroup cognitively impaired: Intervention cost (complete programme) •Acute care AUD 11,128 (28,570) •Rehabilitation AUD 21,740 (37,130) •Direct falls related costs AUD 187 (1,602) Control group •Acute care AUD 5,140  (8,142) •Rehabilitation AUD 26,050  (36,776) •Direct falls related costs AUD 15 (85)	For subgroup who were cognitively intact: • AUD 294 per fall prevented •AUD 526 per faller prevented
•Meyer 2009 •Nursing home residents in Hamburg, Germany, mean age 86 (SD 6) years •Analysis 8.1, Analysis 8.2, Analysis 8.3 •Cost description	•Administer standardised risk assessment tool (Downton Index) monthly (N = 574, 29 nursing homes) vs usual care (N = 551, 29 nursing homes)	•Nursing care facility •Euro •2006 •1 year	•Nurse time for training and assessing using the Downton Index	•Not reported (total during the study EUR 10,500 (USD 16,170, GBP 8160)
•Mulrow 1994 •Residents from 9 nursing homes in San Antonio, Texas, USA, dependent in ≥ 2 activities of daily living, mean age 80 (SD 8) years •Analysis 3.1, Analysis 3.2, Analysis 2.1, Analysis 2.2 •Cost analysis	•One‐on‐one physical therapy sessions (N = 97) vs friendly visits (N = 97) 3 x week for 4 months	•Not stated •US dollar •Not stated •4 months	•Intervention delivery (wages, travel expenses, equipment, overheads) •Nursing home, hospitalisation, physician and other health professional visits, emergency department visits, procedures, and medication charges	•USD 1220 (95% CI 412 to 1832) for physical therapy programme, USD 189 (95% CI 80 to 298) control group	•Healthcare charges (81% nursing home, 15% hospitalisation) USD 11,398 (95% CI 10,929 to 11,849) per participant (NS)
•Schnelle 2003 •Residents of 4 nursing homes, incontinence of urine, US, mean age 88 (SD 8) years •Analysis 11.1, Analysis 11.2, Analysis 11.3 •Cost analysis	•Low‐intensity functionally orientated exercise and incontinence care 5 days a week every 2 hours between 8:00 am and 4:00 pm for 8 months (N = 92) vs usual care (N = 98)	•Not stated •US dollar •1997/98 •8 months	•Diagnostic tests, treatment related to each acute condition (dermatological, genitourinary, gastrointestinal, respiratory and cardiovascular systems; falls; pain; psychiatric and nutritional disturbances)		•USD 24.42 per resident per week to evaluate and treat the selected conditions intervention group, USD 38.36 control group (NS)
•Van de Ven 2014 •Dementia special care units, diagnosed with dementia, ≥1 neuropsychiatric symptom, mean age 84.7 years •Analysis 8.1 •Cost analysis	•Dementia care mapping (DCM), 4 months intervention delivered twice during the study (N = 154) vs usual care (N=164)	•Healthcare perspective •US dollar •Not stated •18 months	•Intervention costs: DCM basic and advanced training, mapping exercise, inter‐rater reliability test, observation, preparing the DCM reports, feedback sessions •Hospital costs: outpatient, inpatient, emergency department & ambulance •Psychotropic drugs •Nursing home healthcare professional costs	•Intervention cost per resident per day USD 0.63 (SD 0.23)	•Healthcare consumption and drug use per resident per day at 18 months (mean(SD): intervention group USD 4.25 (0.59) vs usual care USD 4.4 (0.57)
•Wald 2011 •Medical inpatients at University of Colorado Hospital, USA, aged ≥ 70 years, mean age 81 (SD 7) years •Analysis 19.1 •Cost analysis	•Hospitalist run acute care service for elderly people (N = 122) vs usual hospital inpatient care (N = 95)	•Not stated •US dollar •2007 •6 months	•"Hospital charges"		•Mean “hospital charges” USD 24,617 (SD 15,828) intervention vs USD 21,488 (SD 13,407) usual care, P = 0.12
•Zermansky 2006 •Residents of 65 nursing care facilities in Leeds, UK taking ≥ 1 medicines, mean age 85 (interquartile range 80 to 91) years •Analysis 5.1, Analysis 5.2 •Cost analysis	•Clinical medication review by pharmacist (N = 331) vs usual general practitioner care (N = 330)	•Not stated •Pound sterling •2003 •6 months	• Pharmaceutical use		•Mean medication cost per patient per 28 days medication review group GBP 42.24 (SD 38.33) vs GBP 42.95 (SD 41.01) control group, mean difference GBP ‐0.70 (95% CI ‐7.28 to 5.71)

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Rate of falls	10	2002	Rate ratio (Random, 95% CI)	0.93 [0.72, 1.20]
2 Rate of falls and number of fallers: trials with incomplete data			Other data	No numeric data
3 Number of fallers	10	2090	Risk Ratio (Random, 95% CI)	1.02 [0.88, 1.18]
4 Number of people sustaining a fracture	1		Risk Ratio (Fixed, 95% CI)	Subtotals only
4.1 Hip fractures	1	183	Risk Ratio (Fixed, 95% CI)	0.16 [0.01, 2.81]
4.2 All fractures	1	183	Risk Ratio (Fixed, 95% CI)	0.88 [0.25, 3.14]
5 Rate of falls, excluding studies with ≤20 participants in each arm	8	1959	Rate ratio (Random, 95% CI)	0.91 [0.72, 1.15]
6 Number of fallers, excluding studies with ≤20 participants in each arm	9		Risk Ratio (Random, 95% CI)	1.04 [0.89, 1.21]
7 Adverse events: aches and pains	1	582	Risk Ratio (M‐H, Fixed, 95% CI)	1.23 [0.61, 2.48]
7.1 Severe soreness	1	194	Risk Ratio (M‐H, Fixed, 95% CI)	0.91 [0.40, 2.04]
7.2 Severe bruises	1	194	Risk Ratio (M‐H, Fixed, 95% CI)	2.0 [0.18, 21.69]
7.3 Severe fatigue	1	194	Risk Ratio (M‐H, Fixed, 95% CI)	4.0 [0.46, 35.14]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Rate of falls	5		Rate Ratio (Fixed, 95% CI)	Subtotals only
1.1 Additional gait, balance, functional training	2	56	Rate Ratio (Fixed, 95% CI)	0.62 [0.40, 0.96]
1.2 Strength/resistance vs self‐training	1	34	Rate Ratio (Fixed, 95% CI)	0.74 [0.50, 1.10]
1.3 Balance and strength vs self‐training	1	32	Rate Ratio (Fixed, 95% CI)	0.48 [0.30, 0.77]
1.4 Flexibility (Yoga) vs 'Staying active' program	1	20	Rate Ratio (Fixed, 95% CI)	0.47 [0.24, 0.91]
1.5 3D (Tai Chi) vs 'Staying active' program	1	20	Rate Ratio (Fixed, 95% CI)	0.52 [0.28, 0.98]
1.6 Flexibility (Yoga) vs 3D (Tai Chi)	1	18	Rate Ratio (Fixed, 95% CI)	1.11 [0.51, 2.37]
1.7 3D exercises ("In balance") vs Functional balance, strength & mobility	1	142	Rate Ratio (Fixed, 95% CI)	0.73 [0.60, 0.89]
1.8 Wii balance board vs Otago balance program	1	60	Rate Ratio (Fixed, 95% CI)	0.35 [0.19, 0.63]
2 Rate of falls and number of fallers: trials with incomplete data			Other data	No numeric data
3 Number of fallers	5		Risk Ratio (Fixed, 95% CI)	Subtotals only
3.1 Additional gait, balance, and functional training	2	56	Risk Ratio (Fixed, 95% CI)	0.79 [0.43, 1.45]
3.2 Strength/resistance vs self‐training	1	34	Risk Ratio (Fixed, 95% CI)	0.56 [0.30, 1.03]
3.3 Balance and strength vs self‐training	1	32	Risk Ratio (Fixed, 95% CI)	0.55 [0.29, 1.05]
3.4 Additional whole body vibration	1	159	Risk Ratio (Fixed, 95% CI)	1.28 [0.71, 2.31]
3.5 3D exercises ("In balance") vs Functional balance, strength & mobility	1	142	Risk Ratio (Fixed, 95% CI)	0.92 [0.70, 1.21]
3.6 Comparison of combination exercise programmes	1	41	Risk Ratio (Fixed, 95% CI)	0.54 [0.29, 1.01]
4 Number of people sustaining a fracture	1		Risk Ratio (Fixed, 95% CI)	Totals not selected
4.1 Total fractures	1		Risk Ratio (Fixed, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Rate of falls	10	3439	Rate ratio (Random, 95% CI)	0.88 [0.66, 1.18]
2 Number of fallers	9	3153	Risk Ratio (Random, 95% CI)	0.92 [0.81, 1.05]
3 Number of people sustaining a fracture	5	2160	Risk Ratio (Random, 95% CI)	0.79 [0.30, 2.07]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Rate of falls	2	215	Rate Ratio (Fixed, 95% CI)	0.59 [0.26, 1.34]
2 Number of fallers	2	83	Risk Ratio (Fixed, 95% CI)	0.36 [0.14, 0.93]

PERMALINK

Interventions for preventing falls in older people in care facilities and hospitals

Ian D Cameron

Suzanne M Dyer

Claire E Panagoda

Geoffrey R Murray

Keith D Hill

Robert G Cumming

Ngaire Kerse

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Rate of falls

Risk of falling

Secondary outcomes

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Economic issues

Assessing the quality of the evidence and 'Summary of findings' tables

Results

Description of studies

Results of the search

1.

1. Description of included studies: reference links.

Included studies

Design

Settings

Participants

Interventions

2. Description of interventions in studies of exercise interventions in care facilities.

3. Description of interventions in the medication review trials.

Outcomes

Excluded studies

Studies awaiting classification

Ongoing studies

Risk of bias in included studies

2.

3.

4. Summary of 'Risk of bias' assessment of included studies.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Cluster‐randomised trials

Effects of interventions

Summary of findings for the main comparison. Summary of findings: Exercise compared with usual care in care facilities.

Outcome or subgroup title	No. of studies	Statistical method	Effect size
1 Number of fallers	1	Risk Ratio (Fixed, 95% CI)	Totals not selected
1.1 Vitamin D + calcium vs calcium	1	Risk Ratio (Fixed, 95% CI)	0.0 [0.0, 0.0]
2 Number of people sustaining a fracture	1	Risk Ratio (Fixed, 95% CI)	Totals not selected
2.1 Vitamin D + calcium vs calcium	1	Risk Ratio (Fixed, 95% CI)	0.0 [0.0, 0.0]
3 Adverse events	1	Risk Ratio (M‐H, Fixed, 95% CI)	Totals not selected
3.1 Gastrointestinal complaints (nausea, vomiting, diarrhoea)	1	Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Rate of falls	5	44664	Rate ratio (Random, 95% CI)	0.80 [0.64, 1.01]
2 Number of fallers	3	39889	Risk Ratio (Random, 95% CI)	0.82 [0.62, 1.09]
3 Number of people sustaining a fracture	2		Risk Ratio (Fixed, 95% CI)	0.76 [0.14, 4.10]

Methods	Stepped‐wedge, cluster‐randomised controlled trial.
Participants	Setting: hospital, subacute, 5 geriatric rehabilitation wards, Israel. N = 508 participants; 5 clusters Sample: 52% women Age (years): mean 83.2 Baseline characteristics Individualised fall prevention programme N: 200 Age ‐ mean (SD) : 84.6 (5.6) Female ‐ N (%): 92 (46.0) Medical status defined? ‐ Y/N: Y Falls risk defined (with valid tool at baseline)? ‐Y/N: Y Dependency defined? Y/N: N Cognitive status defined? Y/N: N Usual care N: 308 Age ‐ mean (SD) : 84.1 (7.7) Female ‐ N (%): 173 (56.1) Medical status defined? ‐ Y/N: Y Falls risk defined (with valid tool at baseline)? ‐Y/N: N Dependency defined? Y/N: N Cognitive status defined? Y/N: N Inclusion criteria: Over 65 years; admitted to rehabilitation ward Exclusion criteria: Restricted to bed; refused to participate Pretreatment differences: Phase 1: Longer stay in the control group patients (P < 0.001); higher percentage of females in the control group (P = 0.03)
Interventions	Individualised fall prevention programme. Falls risk assessment and management: including medical interventions, environmental modifications, equipment modifications, cognitive and behavioural treatment, family guidance. Mobility restrictions and optimising location on weekly assessment. Environmental modifications unclear. Usual care. Any activities undertaken by the participants recommended or administered by their treating team
Outcomes	Rate of falls Rate ratio Adverse events
Duration of the study	Period of inpatient admission
Notes	Outcomes of phase one used only. Outcomes data for phase one and two only reported separately, attempts to contact authors unsuccessful. Excluded from pooling as group allocation of clusters unclear.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Insufficient information for judgement.
Allocation concealment (selection bias)	High risk	Allocation not concealed as consent only required for those receiving the intervention.
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Blinding not possible
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Senior nursing staff in control wards were aware of the study because the researchers were collecting study data. Researchers were not blinded.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Attrition balanced across groups and missing outcomes not great enough to have a clinically relevant impact on observed effect size.
Selective reporting (reporting bias)	Unclear risk	Protocol not available
Method of ascertaining falls	Low risk	Quote: "from notes in medical records themselves, and by asking a senior nurse each day about any falls on the ward in the past 24 h." Quote: "Information on falls was collected by the researchers from incident reports filed in patients’ medical records,"
Baseline imbalance	High risk	Longer length of stay in control group at baseline suggests greater dependency in this group at baseline and not adjusted for in analysis
Other bias	Unclear risk	Quote: "some falls prevention activities were already occurring in control (and intervention) wards before the start of our study. These activities continued during the study period, making it more difficult to show any effect of our interventions."   Impact of other falls intervention approaches unclear. Stepped‐wedge trial but only data from phase 1 used as falls data not reported for both phases in combination.

Methods	RCT (individually randomised)
Participants	Setting: acute care hospital, Singapore  N = 1822 participants  Sample: newly admitted patients from 8 medical wards (50% women)  Age (years): mean (SD) intervention group 70.3 (14.2), control group 69.7 (14.7)  Inclusion criteria: aged ≥ 21; Hendrich II Fall Risk Model score ≥ 5 Exclusion criteria: admitted before start of study; fallen prior to falls risk assessment
Interventions	Education + usual care: participants received one educational session (no more than 30 minutes) based on identified risk factors. Designed to increase awareness of risk of falling during hospitalisation and teach risk‐reduction strategies. Relatives of confused participants received the educational session Control: usual care and including usual fall‐prevention interventions
Outcomes	Number of people falling
Duration of the study	8 months
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Allocation of the participants to control or intervention groups was determined using block randomisation with the aid of a computer program and stratified by ward to ensure an even mix in the ward."
Allocation concealment (selection bias)	Low risk	Quote: "Sealed, opaque, serially numbered envelopes were produced from the randomizations sequence separately for each stratum."
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Judgement comment: not blinded
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	Quote: "The research investigator scanned the electronic hospital occurrence report (eHOR) daily during weekday for entries of fall incidences reported by the nurses from the wards and ascertained if the entries were on participants involved in the study." Nursing staff recording falls described as blind to group allocation. Not clear if the research investigator was blind to group allocation
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Judgement comment: all data analysed according to ITT.
Selective reporting (reporting bias)	High risk	Judgement comment: methods mention incidence of falls but only data on risk of falling reported.
Method of ascertaining falls	High risk	Judgement comment: falls not clearly defined.
Baseline imbalance	Low risk	No important differences at baseline.
Other bias	Unclear risk	Unclear impact of standard falls prevention activities.

Methods	Cluster‐randomised controlled trial
Participants	Setting: 24 acute medical and surgical wards from 6 hospitals, Australia N = 31,411 unique participants, including 3853 admitted to both intervention and control wards at different times; 24 clusters. Sample: 48.5% women Age (years): median 67 (interquartile range 51‐79) Baseline characteristics: 6‐PACK programme N: 22,670 admissions; 17,698 participants Age Median (IQR): 68 (51‐80) Female N (%): 11,476 (50.6) Medical status defined? ‐ Y/N : Y (3+ comorbidities 21.2%) Falls risk defined?‐ Y/N: N Dependency defined? ‐ Y/N: No Cognitive status defined? ‐ Y/N: Y Usual care N: 23,575 admissions; 17,566 participants Age Median (IQR): 67 (51‐79) Female N (%): 11,424 (48.5) Medical status defined? ‐ Y/N : Y (3+ comorbidities 25.3%) Falls risk defined?‐ Y/N: N Dependency defined? ‐ Y/N: N Cognitive status defined? ‐ Y/N: Y Inclusion criteria: Wards: where fall‐related injuries have been identified as a problem, acute medical or surgical wards; average patient length of stay <10 days; wards to have one or less low‐low beds to each six standard beds on medical wards and one or less low‐low beds to each, 29 standard beds on surgical wards; a fall risk assessment and/or prevention strategy checklist is not already included in the daily patient care plan documentation. Wards that have a fall risk assessment and/or prevention strategy checklist included on admission documentation but do not have a policy that this must be updated daily will not be excluded from participating in the study. Exclusion criteria: No patient level exclusion criteria. Pretreatment differences: Nil
Interventions	6‐PACK programme comprising a 9 item falls risk assessment tool and delivery of one or more of six interventions to high risk patients: 1) Placement of a ‘falls alert' sign above the patient’s bed. 2) Supervision of patients while in the bathroom. 3) Use of a low‐low bed. 4) Ensuring that the patient’s walking aid is within reach at all times. 5) Establishment of a toileting regimen. 6) Use of a bed/chair alarm when the patient is positioned in the bed/chair. Staff education integral to implementation.Nurses were asked to update the fall risk tool for each of their patients each shift and to apply a falls alert sign and one or more of the remaining 6‐PACK interventions to patients classified as being at high risk Usual care. Any standard hospital practice provided by wards as part of existing hospital policy relating to fall prevention, which may have included some components of the 6‐PACK programme and other interventions such as non‐slip socks, constant patient observers, and falls alert wrist bands.
Outcomes	Rate of falls Number of fallers (number of unique fallers provided by author correspondence) Number of injurious falls Fracture falls (number of unique patients with fractures provided by author correspondence) Multiple falls Adverse events
Duration of the study	12 months intervention period plus 3 month pre‐randomisation baseline period
Notes	ACTRN12611000332921 "The use of all 6‐PACK programme components (fall risk tool and six interventions) was threefold higher on intervention wards than on control wards (incidence rate ratio 3.05, 95% confidence interval 2.14 to 4.34; P<0.001)."
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "used the RALLOC command in Stata to develop the randomisation schedule, using a random sequence in blocks of two generated by the study statistician." Judgement comment: random sequence allocation done.
Allocation concealment (selection bias)	High risk	Quote: "Concealment of allocation was ensured, as the schedule was accessible only by the study statistician, who was not involved in ward recruitment or data collection." Judgement comment: although allocation sequence initially concealed, subjects were enrolled after cluster randomisation, and sequence would have been known at this time.
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Quote: "It will not be feasible to blind ward nurses or patients to the intervention." Judgement comment: not done.
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Quote: "Blinding of the assessors collecting the fall and falls prevention practice data was also not possible. Assessors blinded to group allocation did the secondary coding of characteristics of falls and injuries, and the primary assessor completed the coding. A statistician blinded to group allocation (RW) did the data analysis." Judgement comment: not done.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Judgement comment: no loss to follow‐up
Selective reporting (reporting bias)	Low risk	Judgement comment: all falls outcomes reported as per trial registry record.
Method of ascertaining falls	Low risk	Quote: "by daily auditing of patient medical records and verbal report of the nurse unit managers. These data will then be triangulated with hospital incident reporting and administrative patient episode datasets. Concurrent to this will be hospital‐wide education and reminders of the fall definition and incident reporting best practice, facilitated by use of an existing training package. 23 Patient" Judgement comment: multiple methods of concurrent recording of falls data used.
Baseline imbalance	Low risk	Quote: "Characteristics of admitted patients and length of stay were similar for intervention and control groups and across baseline and randomised controlled trial periods" Judgement comment: no imbalance across groups.
Other bias	Unclear risk	Unclear impact of any ongoing falls prevention activities.

Methods	RCT (cluster randomised), nursing home subgroup data.
Participants	Setting: 3 residential care homes, high‐level care, Denmark N = 31 participants; 3 clusters. Sample: 65% women Age (years): mean 88 Baseline Characteristics Multidisciplinary nutritional support N: 9 nursing home Age Range: 88.1 (9.6) Sex (% female): 6 (67) Medical status defined?: N Falls risk defined?: N Dependency defined? (ADL problem, No. social services for home help, nursing): Y Control N: 22 nursing home Age Range: 87.8 (7.0) Sex (% female): 14 (64) Medical status defined?: N Falls risk defined?: N Dependency defined? (ADL problem, No. social services for home help, nursing): Y Inclusion criteria: 65+ years, at nursing home or receiving home care (assistance with meals) with 2 points according to Eating Validation Scheme (EVS) completed by nursing staff caregivers (would benefit from intervention) able to completed planned tests Exclusion criteria: not able or willing to give informed consent Pretreatment differences: living in a nursing home: intervention 16%, control 55% (P < 0.001); 30‐seconds chair‐stand modified, mean (SD) 4.9 (3.3) intervention, 2.5 (2.7) control (P = 0.004); cognitive problem 56% intervention versus 78% control (P = 0.03)
Interventions	Multidisciplinary nutritional support. Nutrition co‐ordinator involvement, multidisciplinary project group meetings, plan of action in the municipality care register system, Exercise, nutritional support, support for dysphagia and eating problems as indicated by EVS screening. 30 to 45 minutes moderate‐intensity exercise sessions including strength and balance training twice a week, oral training supplements after exercise, weekly assessment of weight, individual dietetics treatment plan and regular reviews by dietician, multidisciplinary meeting weekly to evaluate and adjust individual treatment plans, OT involvement if indicated. Health professional involvement: Nutrition co‐ordinator, physiotherapist twice weekly, dietician performs initial interview, then regular consultations and phone or group follow‐up, occupational therapist to consults with patients who suffer from eating dependency or chewing and swallowing problems and initiate interventions if indicated. Control. Nutrition co‐ordinator involvement plus standard interventions from physiotherapist, registered dietician and occupational therapist requested through the municipality’s normal assessment and referral system will be maintained.
Outcomes	Number of falls Adverse events
Duration of the study	11 weeks
Notes	A trial of nutritional support using a structured and multidisciplinary approach, focusing on nutritional risk factors, in undernourished older adults in both home care and nursing home settings, with results reported separately.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Judgement comment: sequence generation by drawing a lot for an opaque envelope.
Allocation concealment (selection bias)	High risk	Judgement comment: randomisation by researcher not involved in the study (2016 p200). Author correspondence quote: "participants were invited by means of the staff who did not know about the result of the group allocation". and "we did not include new admissions". However: "Due to the limited knowledge about the benefit of nutritional support among home‐care clients, the aim was to randomly assign 2 of the 3 home‐care clusters to the intervention group", this is likely to enable the randomisation sequence to be predicted, concealment not possible for the final cluster.
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Quote: "The researchers for this study included the research assistants (AGC, BSH, SD‐S, and TKSM) and the primary investigator (AB), who were not blinded for the intervention." Judgement comment: blinding not done.
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Quote: "The researchers for this study included the research assistants (AGC, BSH, SD‐S, and TKSM) and the primary investigator (AB), who were not blinded for the intervention. Before starting the analysis the primary investigator (AB) was reblinded for participants’ group assignment." Judgement comment: not done, falls data collected by unblinded research nurse. Although primary investigator "reblinded" before analysis no details were reported on the method for this and it is considered likely to include a risk of residual unblinding.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Judgement comment: author correspondence clarified data and indicated one withdrawal, no other loss to follow‐up.
Selective reporting (reporting bias)	Low risk	Judgement comment: trial protocol available and falls outcomes consistently reported.
Method of ascertaining falls	Low risk	Quote: "The information was gathered by means of data from the RAI‐NH version 2.0 and RAI‐HC version 2.0 assessments and the municipality care register system. For each participant, the same trained nurse collected" Judgement comment: concurrent falls data collection with clear definition.
Baseline imbalance	High risk	Judgement comment: baseline imbalance in nursing home subgroup for cognition, no adjustment performed
Other bias	Low risk	None detected.

Methods	RCT (cluster randomised by facility).
Participants	Setting: 6 long‐term care facilities (high‐level nursing care), Germany  N = 981 participants; 6 clusters.  Sample: 79% women  Age (years): mean (SD) intervention group 83.5 (7.5), control group 84.3 (6.9)  Inclusion criteria: resident of facility. Inclusion criteria for exercise programme: able to stand while holding a chair, able to lift one foot  Exclusion criteria: none stated
Interventions	Fall prevention programme for staff and residents. Residents chose to participate in any combination of interventions for any length of time. Those choosing to participate in fall registration only also received environmental modification and modification of nursing care Staff training on risk factors and preventive measures (60 minutes), audit and monthly feedback re falls and injuries Check list of 76 environmental hazards (lighting, chair and bed height, floor surfaces, etc). Feedback to staff and administrators Resident education: all received written information, offered personal consultation by study nurse or exercise instructor Group exercise programme (progressive balance and resistance training) 75 minutes, 2 x per week Hip protectors Control: usual care, no specific program activities.
Outcomes	Rate of falls Number of people falling Number sustaining a fracture (hip fractures) Adverse events
Duration of the study	12 months
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Cluster randomisation of 6 facilities using sealed envelopes selected by an independent person. Insufficient information to permit judgement
Allocation concealment (selection bias)	Unclear risk	Allocation in sealed envelopes, but individuals admitted after group allocation by a person who may have been unblinded and may have had knowledge of participant characteristics
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Not blinded.
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Staff at facilities who recorded falls were likely to be aware of their facility's allocation status
Incomplete outcome data (attrition bias)   All outcomes	Low risk	All residents included in analysis.
Selective reporting (reporting bias)	Low risk	All expected falls outcomes completely and thoroughly reported. Adjustment for clustering conducted
Method of ascertaining falls	Low risk	Fall definition provided and concurrent recording of falls
Baseline imbalance	High risk	Greater proportion of intervention group were taking 4 or more medications
Other bias	Low risk	None identified.

Methods	RCT. Individually randomised.
Participants	Setting: 2 residential care facilities, intermediate‐level care, Belgium N = 62 participants Sample: 76% women Age (years): mean 83.2 (SD 7.9) Baseline characteristics Whole body vibration N: 31 Age ‐ mean (SD): 82.2 (9.02) Female N (%): 20 (64.5) Medical status defined? ‐ Y/N: Y Falls risk defined (Y/N)?: Y Dependency defined (Y/N)?: Y Cognitive status defined? Y/N: Y Control N: 31 Age ‐ mean (SD): 84.2 (6.83) Female N (%): 27 (87.1) Medical status defined? ‐ Y/N: Y Falls risk defined (Y/N)?: Y Dependency defined (Y/N)?: Y Cognitive status defined? Y/N: Y Inclusion criteria: residents from two nursing homes; able to remain standing; able to move with or without technical assistance Exclusion criteria: weight greater than 150 kg; electronic implants; knee or hip prostheses; epilepsy; bleeding disorders; inflammatory abdominal disorders; high risk of thromboembolism; malignancy; unconsolidated fracture; refusal of doctor or family Pretreatment differences: gender (more women in control group) P = 0.04; lower body mass in control group P < 0.01; lower MMSE in control group, P = 0.04
Interventions	Whole body vibration. Exercise programme on a sinusoidal vibration platform (Vibrosphere), standing without shoes with knees flexed, cushion placed under vibrosphere. 3 x weekly, 5 series of 15 seconds of vibrations at 30 Hz, 2mm amplitude, alternate with 30 seconds rest, total vibration time 1 minute 15 seconds, minimum 1 day between sessions. Supervised by one of 4 people, 2 physiotherapists and 2 authors. Usual care. No change to lifestyle during study, no involvement in any new type of physical activity
Outcomes	Number of falls Number of people falling
Duration of the study	6 months intervention, follow‐up to 12 months.
Notes	Compliance: 91.9% of exercise sessions performed.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "We performed the randomisation by blocks of four with a computer‐generated randomisation procedure." Judgement comment: computer‐generated randomisation.
Allocation concealment (selection bias)	Unclear risk	Quote: "An identification number and a randomisation number were created for each participant." Judgement comment: method of concealment of allocation sequence from those enrolling participants was unclear.
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Judgement comment: not done. Blinding not possible.
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Judgement comment: not done. Nurses recorded falls, they were not blinded. Blinded assessment unlikely to include falls outcomes.
Incomplete outcome data (attrition bias)   All outcomes	High risk	Judgement comment: numbers and reasons balanced between groups.
Selective reporting (reporting bias)	Low risk	Judgement comment: number of falls not defined as outcome in trial registry. Trials registry indicates 3 months outcomes. Reporting of falls data appears complete although not predefined.
Method of ascertaining falls	Low risk	Quote: "Nurses completed the fall record with the date, time, and circumstances of the falls." Judgement comment: likely that falls were recorded at time of event.
Baseline imbalance	High risk	Baseline differences in weight, gender, MMSE may impact on falls rates.
Other bias	Low risk	Judgement comment: none identified

Methods	RCT (cluster randomised by unit).
Participants	Setting: 15 residential dementia care units (high‐level nursing care), Sydney, Australia  N = 289 residents; 15 clusters  Sample: people with dementia (78% women) Age (years): mean (SD) person‐centred care group 83 (7.6), dementia‐care mapping group 84 (6.4), usual care group 83 (7.6)  Inclusion criteria (facilities): task‐focused (not person‐centred) care systems. Inclusion criteria (residents): dementia and low cognitive function; aged >60; high dependency needs; persistent need‐driven dementia compromised behaviours Exclusion criteria (residents): serious co‐morbidities complicating or masking dementia; palliative care; unremitting pain; distressing physical symptoms; respite placement
Interventions	Person‐centred care: one researcher trained 2 care staff per site in allocated method of care (see 'Notes'), worked with trained staff to implement care plans, provided two site visits to give ongoing support for staff, then regular telephone contact for 4 months Dementia care mapping: two researchers trained 2 care staff per site in allocated method of care (see 'Notes'), carried out "mapping" with trained staff, developed care plans with trained staff, trained staff helped colleagues implement plans, regular telephone contact from researchers for 4 months Usual care: non person‐centred care that is task‐focused and concerned mostly with physical care needs
Outcomes	Number of people falling
Duration of the study	8 months
Notes	Person‐centred care emphasised social interactions at affective level based on life histories; aimed to preserve personal identity and foster meaningful relationships. Dementia‐care mapping: "mapping" consisted of observation of each participant for 6 hours per day for 2 days to identify factors related to well‐being
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Allocation was done by the study statistician (MTK), who was unaware of the identity of sites, using an SAS20 program."
Allocation concealment (selection bias)	Low risk	Eligible residents were selected by facility managers or directors before randomisation of sites
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Judgement comment: not blinded
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "Treatment allocation was masked to assessors." Three separate research assistants collected outcome data from each cluster of five facilities. Staff of facilities instructed not to inform assessors of interventions
Incomplete outcome data (attrition bias)   All outcomes	High risk	Judgement comment: >20% loss from person‐centred care and usual care arms.
Selective reporting (reporting bias)	Unclear risk	Judgement comment: falls stated as outcome ‐ Incidents, quote: "and subsequent admissions to hospital were discerned from official records of incidents including residents’ falls, fractures, lacerations, bruises, medication errors, and behavioural incidents" (p320, column 1, para 2). However, falls not stated as outcome in initial trial registry record (added retrospectively)
Method of ascertaining falls	High risk	Judgement comment: falls poorly defined and multiple sites enrolled.
Baseline imbalance	Low risk	Judgement comment: differences at baseline adjusted for in analysis.
Other bias	Low risk	Judgement comment: none identified.

Methods	RCT (cluster randomised).
Participants	Setting: 2 residential care facilities (intermediate‐level care), Korea  N = 68 participants; 2 clusters.  Sample: 75% women  Age (years): mean 77.9 (range 61 to 91)  Inclusion criteria: ambulatory; age > 60; at least one fall risk factor (impaired gait, impaired balance; a fall in the last year; postural hypotension; four or more medications affecting balance)  Exclusion criteria: severe dementia; physical illness that may prevent completion of 12‐week course of exercise; involvement in any other exercise
Interventions	Supervised Tai Chi: 35‐minute group sessions with certified Tai Chi leader, 3 x per week for 12 weeks Usual routine activities
Outcomes	Number of people falling
Duration of the study	3 months
Notes	Cluster randomised, described as quasi‐experimental design with a non‐equivalent control group.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "... two facilities with similar characteristics were selected and randomly assigned to either the experimental or control group by coin tossing."
Allocation concealment (selection bias)	High risk	After first toss the allocation of the second facility would be known. No description of whether individual participant recruitment was undertaken after group allocation by a person who was unblinded and may have had knowledge of participant characteristics
Blinding of participants and personnel (performance bias)   All outcomes	High risk	Judgement comment: blinding not feasible
Blinding of outcome assessment (detection bias)   All outcomes	High risk	Staff at facilities who recorded falls were likely to be aware of their facility's allocation status
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Judgement comment: loss similar between groups.
Selective reporting (reporting bias)	Unclear risk	Judgement comment: no protocol identified.
Method of ascertaining falls	High risk	Judgement comment: falls defined but only recorded weekly.
Baseline imbalance	High risk	Judgement comment: significant difference between groups in muscle strength and balance measures ‐ addressed for balance and strength scores by using difference scores ‐ but no adjustment apparent for falls data.
Other bias	Low risk	Judgement comment: assignment predicable as cluster randomised and only 2 facilities, however this accounted for under allocation concealment. No other sources of bias identified.