Systematic Review and Meta-Analysis of Intervention Studies with General Practitioner Involvement Focused on Falls Prevention for Community-Dwelling Older People

Abstract

Objectives: Falls are a significant health problem for the ageing population. This review aimed to identify effective falls prevention interventions with involvement of general practitioners (GPs). Methods: Systematic review of randomised controlled trials conducted from 1999–2019, with meta-analysis. Searches located 2736 articles. A quality assessment was conducted of all included studies. Results: 21 randomised controlled trials met the inclusion criteria and 19 studies could be included in a meta-analysis. Overall, studies were not effective for reducing multiple falls (Relative Risk (RR) 1.16, 95% CI .97–1.39 and p = .10) or reducing one or more falls (RR .91, 95% CI: .82–1.01 and p = .08), but were effective for reducing injurious falls (RR .76, 95% CI: .66–.87 and p = .001). Discussion: Studies involving the GP in an active role and aligned with the primary care context were effective. The fidelity of interventions was limited by independent GP decisions and older participants being required to initiate the intervention.

Keywords

accidental falls primary health care healthy-ageing falls risk

Background

General practitioners (GPs) also known as family physicians have a critical role in falls prevention amongst community-dwelling older people. GPs are typically the first point of contact for people in the community to access health services. GPs can provide preventative medicine through screening for conditions, patient advice and referral to other health professionals and specialists (Croker et al., 2013; Greenfield et al., 2016). Prevention is a key component of managing falls risk (Mackenzie et al., 2013). Falls and their associated impacts continue to occur in high numbers among community-dwelling older adults, with approximately 30% of people over the age of 65 experiencing at least one fall in the previous 12 months (Australian Commission on Safety Quality in Health Care, 2009). Despite extensive research literature published over the last 20 years and strong evidence to support a range of falls prevention interventions, the prevalence of falls and fall-related injuries has remained constant (Carter et al., 1997; Gillespie et al., 2012; Smith et al., 2016). Making falls prevention services more available in primary health care settings is essential, as falls among older people in the community will continue to increase and present significant challenges for older people as the population ages (World Health Organization, 2008).

People over the age of 65 are the fastest growing population group in Australia (Australian Institute of Health and Welfare, 2017), with a projected growth of 14% by 2061 (Australian Bureau of Statistics, 2013). For an ageing population, changing health status and health needs include increased chronic disease and co-morbidities, decreased mobility and physical strength, increased frailty, declining health outcomes and increased risk of falls and fall related injuries (Stalenhoef et al., 1999; Ostaszkiewicz & O’Connell, 2005; Topinková, 2008). The risk factors associated with ageing that contribute to falls have been established. There is a two-way relationship where older age predisposes to falling and where falling predisposes to threats to an individual’s health status as they age (Boffin et al., 2014; Cesari et al., 2002; Stalenhoef et al., 1997). The consequences of falling can vary considerably and affect both physical and psychosocial aspects of the lives of older people. Increased fear of falling and reduced physical and social activity can result in reduced quality of life, isolation and lack of activity, while physical consequences range from mild to severe including fracture, hospitalisation and death (Buck et al., 2014; Peel, 2011). These unique health needs and risks in relation to falling for this population group make GPs key health personnel as the often first and most frequently seen health provider for older people living in the community (Marcinowicz et al., 2014; Monagle, 2002).

There are, however, barriers to accessing falls prevention interventions in primary care. These include practice difficulties for GPs such as time constraints, limited knowledge or education regarding falls prevention and the need to juggle competing co-morbidities of older patients, as well as personal attitudes of GPs (Child et al., 2012; Chou et al., 2006; Jones et al., 2011; Kielich et al., 2017; Mackenzie & McIntyre, 2019). In addition, GPs are reliant on their patients reporting any falls to them, unless they seek this information during a consultation. However, there are also barriers to patients discussing falls with their family doctor or accepting recommendations regarding falls prevention. These include grappling with feelings of embarrassment and fear of being judged, stereotyped or stigmatised; a fear of losing one’s independence or a fear of losing a sense of personal identity or minimising the impact of a fall on their daily life (Bunn et al., 2008; Dollard et al., 2014; Gaboreau et al., 2016; Lee et al., 2015; Whitehead et al., 2006). These barriers can make it difficult for GPs to identify older people at risk of falls and to negotiate the provision of appropriate falls prevention interventions and services (Gaboreau et al., 2016). The high prevalence of falls amongst older people as a significant health problem has driven research in falls prevention. Falls prevention for community-dwelling older people has been synthesised in previous systematic, and umbrella reviews with strong outcomes indicated for several interventions including home hazard reduction, exercise and multifactorial interventions (Gillespie et al., 2012; Stubbs et al., 2015). Gillespie et al. (2012) synthesised and reported the results for randomised controlled trials (RCTs) from various contexts such as hospitals and community aged care, including studies based in general practice. However, it remains unclear how these effective interventions are routinely carried out in the general practice context.

This study addressed the question ‘How can GPs contribute to effective falls prevention in the general practice context?’ To respond to this, it is important that effective interventions that GPs can be involved in to promote falls prevention are identified, so that current evidence about effective falls prevention interventions can be translated into primary health and falls can be reduced amongst community-dwelling older people. Therefore, the aims of this systematic review were (i) to identify falls prevention interventions where GPs had a clinical contribution and (ii) to establish which of these interventions were effective for use by GPs in practice.

Methods

Study Design and Protocol Registration

This study was conducted as a systematic review with meta-analysis. The study protocol was registered with PROSPERO (registration number CRD42017077574). The reporting of this systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and checklist for reporting systematic reviews (Moher et al., 2009).

Data Sources and Search Strategies

A computer-based search was conducted across MEDLINE, EMBASE, CINAHL, Cochrane Randomised Controlled Trial Register, AMED and supplemented by a search of Web of Science, SCOPUS, OT Seeker, and a visual inspection of the reference list of included studies in the 2012 Cochrane review (Gillespie et al., 2012) on the effectiveness of fall prevention interventions in community-dwelling older people was also conducted. Concept terms for the search were developed with the assistance of an academic liaison librarian at the University of Sydney. Keywords and MeSH terms were grouped into three concepts: (1) aged, ageing and elderly; (2) GP, general pract*, primary health practitioner, primary care practice and family practice and (3) fall*, fall risk, accidental fall and fall prevention. All terms were used across all selected databases.

Eligibility and Exclusion Criteria

Studies for inclusion were limited to randomised controlled trials published in English from 2000–2019. Additionally, eligibility was assessed against the following criteria: (1) Study participants were aged greater than 60 years, or the mean age for participants was greater than 70 years; (2) Participants were community dwelling; (3) The study investigated a falls prevention intervention in a general practice context; (4) The study required some active clinical involvement of the GP as part of the intervention for the experimental group after randomisation of participants into intervention and control groups and (5) Outcomes measured included any of the following: (i) the proportion of participants experiencing any falls, (ii) the proportion of participants experiencing multiple falls, (iii) the proportion of participants experiencing injurious falls and (iv) falls risk, where fall risk is given a score.

Studies were excluded if they only involved GPs prior to participant randomisation such as in case finding (e.g. identifying potential fallers through their computerised records), provision of medical clearance or gathering baseline characteristics of potential participants for both the control and intervention groups. Studies were also excluded where the researchers used general practice records, settings or assessment tools that were considered useable by a GP, but that did not utilise the GP in the practice context to participate in the fall intervention.

Screening and Study Selection

Articles were subjected to an initial title screening by one researcher (AB). Subsequently, an abstract screen was conducted independently by two researchers (AB and LM). Disagreements on the relevance of studies were managed through discussion and consensus between the two researchers. Where abstracts gave insufficient detail to form a decision for ongoing inclusion, full texts were inspected by both authors for consensus to be reached. A full text screen was conducted on eligible articles to generate a final list of studies for full text screening against eligibility criteria. Full text inspection was conducted by AB and ambiguous studies were referred to the second researcher (LM) for inspection and discussion in order to reach consensus regarding inclusion.

Data Collection

Data extraction was conducted by AB using a Microsoft Excel table. Data of interest included location of study, participant sample description, comparison and intervention description, number of participants at baseline and at follow-up for both intervention and control groups, the falls outcome measured, the activity required of the GP, time of outcome measurement, and any further information of note. Study characteristics were used to form preliminary groupings by outcome type and data extraction included information pertaining to participant characteristics related to eligibility for recruitment.

Risk of Bias/Quality Assessment

The quality assessment document used to appraise methodological quality of included studies was modified from the Joanna Briggs Institute Critical Appraisal Checklist for Randomised Controlled Trials (JBI CAP-RTC) (Tufanaru et al., 2017). The JBI CAP-RCT was selected for use as a standardised tool with a detailed manual explaining how to address each criterion (Tufanaru et al., 2017). Quality appraisal was conducted by (AB and LM) with discussion where consensus was required.

Analysis

Falls outcome data was analysed in three categories: (i) proportion of participants who experienced at least one fall, (ii) proportion of participants who experienced multiple falls (two or more) and (iii) proportion of participants who experienced injurious falls. Analysis was conducted using Comprehensive Meta-Analysis software with meta-analyses (forest plots) generated for the three categories of outcomes (Borenstein et al., 2010; Hoffmann et al., 2010; Gough et al., 2012). Confidence intervals and heterogeneity scores (Q-value, df(Q) and I²) were presented. Due to the varied nature of the included studies regarding intervention type and participant population (at GP and patient levels), a random effects model was used for the meta-analyses to account for the variety in study variables. All studies were assumed to have used intention-to-treat (ITT) analysis unless otherwise specified. Where studies reported censoring participant data in the analysis, this review compared outcome data supplied against the participant data at randomisation in keeping with gold standard recommendations for RCT analysis (Nich & Carroll, 2002) and to address any threats to the validity of the selected RCT results (El-Masri, 2015; Elkins & Moseley, 2015).

Ethical Considerations

Ethics approval was not required for this systematic review.

Results

Study Selection

The systematic search produced a combined total of 3891 results across the databases searched and Cochrane reference list, of which 1160 were found to be duplicates. From the 2736 results remaining, 2585 were excluded at the title screening stage including systematic reviews, clinical practice guidelines, professional body position statements, study types other than RCTs, and unrelated topics including but not limited to paediatric care, asthma, mental health and sociopolitical articles, resulting in 151 articles eligible for abstract screening. The abstract screening, including full text inspection where the abstract contained insufficient information generated a final list of 45 studies for full text screening against the inclusion criteria. Where study protocols had been identified in the database searches, additional searches of the university library catalogue were conducted to find the associated final publication of the study, if available. Finally, 21 studies were selected for inclusion in the systematic review. Reasons for exclusion of studies at the full text inspection stage are provided in the flow diagram in Figure 1.

Figure 1.

PRISMA flow chart.

Study Characteristics for 21 Studies Included in the Review

Table 1 provides relevant study characteristic data for the 21 included studies. The geographical context ranged across Europe, the USA, Australia and New Zealand, with three studies specifically related to the Australian primary health care context. For outcomes measured on follow-up, 15 studies applied outcome measures at 12 months, two studies provided data at six months only and four studies measured outcomes at greater than 12 months ranging up to five years. Two studies reported unequal allocation of participants in favour of the control group as a resource and cost saving measure. The type of GP activity varied with 15 of the included studies requiring participant GPs to follow up on recommendations provided by the study interventionists and six studies required initiation of the fall prevention intervention by the GP. Interventions investigated included electronic alert systems informing GPs of patient falls risk, medication risk screening by GP or in collaboration with a pharmacist, reduction in falls risk medications, screening by GPs, fracture risk screening, home visits by trained fall and fracture nurses, medical and occupational therapy assessment, specific falls and chronic disease clinics, green prescriptions (with and without pedometers) and assessment in secondary level health care after attending emergency departments due to falls.

Table 1.

Characteristics of Included Studies in the Review.

First author, date and study location	RCT type and sample	Control	Experimental intervention	Intervention n at baseline (n at follow-up)	Control n at baseline (n at follow-up)	Falls-related outcomes measured	GP involvement	Time of outcome measurement	Other comments
Barr et al. (2005)	2-arm RCT	Usual care	Active screening visit at GP practice, lifestyle questionnaire and quantitative ultrasound heel scan with subsequent recommendations	890 (599)	1625 (1276)	Fall rate and proportion of people reporting falls	GPs to action recommendations, e.g. vitamin D and calcium prescription	1–3 years	Unequal allocation to favour control group
UK	Community-dwelling women >70 years of age	Usual care		890 (599)	1625 (1276)	Rate of fractures		1–3 years	Falls not recorded prospectively
Coleman et al. (1999)	2-arm RCT	Usual care	Multidisciplinary primary care – chronic care clinics held every 3–4 months	73 (49)	96 (78)	Frequency of falls	GPs conducted the clinics	24 months	GP practices randomised into control or intervention; patients not randomised. Falls not recorded prospectively
USA	Community-dwelling frail older adults >65	Usual care		73 (49)	96 (78)	Frequency of falls	GPs conducted the clinics	24 months
Dapp et al. (2011)	3-arm RCT	Usual care, no HRA-O and GPs received the training	HRA-O – self-administered with tailored recommendations provided to GPs and participants. GPs received fortnightly 2 hour training sessions for reinforcing recommendations of the HRA-O	878 (587)	1702 (1376)	Multiple self-reported falls in previous 12 months	GPs received training and implemented recommendations resulting from HRA-O	12 month follow-up	Unequal allocation in favour of control. 3rd group of untrained GPs and patients recruited to act as comparison self-reported falls part of standardised assessment
Germany	Community-dwelling men and women >60	Usual care, no HRA-O and GPs received the training		878 (587)	1702 (1376)	Multiple self-reported falls in previous 12 months		12 month follow-up
De Vries et al. (2010)	2 arm RCT	Usual care	Multidisciplinary outpatient clinic with fall risk assessment and intervention. Recommendations provided to GP for actioning and follow-up	106 (81)	111 (84)	Proportion of participants who fell at least once, who fell at least twice and time to first and second fall	GPs received follow-up phone call regarding recommendations and referrals to follow-up	12 month follow-up	Participants unavailable for follow-up were censored at time of drop out
Netherlands	Community-dwelling men and women >65 who had consulted emergency department after a fall	Usual care		106 (81)	111 (84)			12 month follow-up	Falls calendar
Elley et al. (2008)	2-arm RCT	Usual care + two social visits from an accredited provider	Falls and fracture nurse working with family physician, visiting participants at home and assessment of risk of falls with referral back to GP for action and other health professionals e.g. optometrist	155 (135)	157 (145)	Proportion with at least one fall, proportion of fallers with at least two falls and proportion of fallers with injurious falls	Referrals received from nurse for action	12 month follow-up	Prospective falls calendar
New Zealand	Community-dwelling men and women >75	Usual care + two social visits from an accredited provider		155 (135)	157 (145)		Referrals received from nurse for action	12 month follow-up	Prospective falls calendar
Hendriks et al. (2008)^	2-arm RCT	Usual care	Medical and occupational therapy assessment to evaluate and address risk factors, recommendations and summary of results and sent to patient GP for action and referral	166 (124)	167 (134)	Proportion of participants with at least one fall, >1 fall and injurious falls	Received results and recommendations of assessments for action and referral	3 month and 12 month follow-up	Prospective falls calendar
Netherlands	Community-dwelling adults >65 attending emergency department after a fall	Usual care		166 (124)	167 (134)			3 month and 12 month follow-up	Prospective falls calendar
Kerse et al. (2005)	Post hoc sub-analysis of a cluster RCT	Usual care	Green prescriptions (exercise advice)	130	140	Falls as adverse events	Supply brief activity counselling and green prescriptions	12 month follow-up	Falls reported at follow-up
New Zealand	Community-dwelling, sedentary adults >65	Usual care	Green prescriptions (exercise advice)	130	140	Falls as adverse events	Supply brief activity counselling and green prescriptions	12 month follow-up	Falls reported at follow-up
Kolt et al. (2012)^	2-arm RCT	Standard green prescription	Pedometer-based green prescription with GP and physical activity counsellor advice	165 (140)	165 (130)	Falls as adverse events	GP advice regarding engagement in physical activity	3 month and 12 month follow-up	Used validated standardised tools for fall outcomes.
New Zealand	Community-dwelling men and women >65 and low activity	Standard green prescription		165 (140)	165 (130)	Falls as adverse events	GP advice regarding engagement in physical activity	3 month and 12 month follow-up	Used validated standardised tools for fall outcomes.
Lightbody et al. (2002)*	2-arm RCT	Usual care	Falls nurse intervention	171 (155)	177 (159)	Number of fallers and number of falls	Received referral from falls nurse for medication reviews	6 month follow-up	Falls diary
UK	Adults >65 attending emergency department post fall	Usual care	Falls nurse intervention	171 (155)	177 (159)	Number of fallers and number of falls	Received referral from falls nurse for medication reviews	6 month follow-up	Falls diary
Newbury et al. (2001)	2-arm RCT	Quality of life survey and usual care	Quality of life survey and 75+ health assessment	50 (45)	50 (44)	Self-reported falls	Received letter reporting identified problems and no reminders provided	12 month follow-up	Self-reported falls at follow-up
Australia	Community-dwelling adults >75	Quality of life survey and usual care	Quality of life survey and 75+ health assessment	50 (45)	50 (44)	Self-reported falls		12 month follow-up	Self-reported falls at follow-up
Palvanen et al. (2014)^	Pragmatic RCT	Usual care + general injury prevention brochure	Chaos falls clinic, injury prevention brochure and multifactorial intervention (exercise, medication review and home hazard review)	661 (589)	653 (556)	Falls, fallers and fall-related injuries	GPs part of multidisciplinary team that provided care at the clinic. Referral to usual GP	6 month and 12 month follow-up	Quarterly follow-up
Finland	High risk community-dwelling adults >70	Usual care + general injury prevention brochure		661 (589)	653 (556)	Falls, fallers and fall-related injuries		6 month and 12 month follow-up	Quarterly follow-up
Pit et al. (2007)^	Cluster RCT	Usual care and medication risk assessment	GP education, medication risk assessment and completion of medication review checklist	452 (350)	397 (309)	Proportion of participants who fell and proportion of people who experienced an injury as a result of a fall	GPs received training, completion of medication review checklist and medication risk assessment	4 month and 12 month follow-up	Self-reported falls at follow-up
Australia	Community-dwelling men and women >65	Usual care and medication risk assessment		452 (350)	397 (309)			4 month and 12 month follow-up	Self-reported falls at follow-up
Shepstone et al. (2018)	Two-arm RCT	Usual care	Screening using fracture risk assessment tool. Treatment recommended for women at high risk of hip fracture	6233 (5334)	6250 (5327)	Proportions of participants with hip fracture	GPs and patients informed of fracture risk and asked to initiate treatment	3, 6, 12, 24, 36, 48 and 60 months	Hip fracture records
UK	Community-dwelling women aged 70–85 years	Usual care		6233 (5334)	6250 (5327)	Proportions of participants with hip fracture		3, 6, 12, 24, 36, 48 and 60 months	Hip fracture records
Siegrist et al. (2016)	Cluster RCT	GPs received training and then usual care	GPs received training for falls and falls risk assessment and participants completed 16-week supervised exercise program	222 (184)	156 (116)	Proportion of participants who fell and proportion of people who experienced an injury as a result of a fall	GPs received falls and falls risk assessment training	12 month follow-up	Daily fall calendar
Germany	community-dwelling men and women >65	GPs received training and then usual care		222 (184)	156 (116)		GPs received falls and falls risk assessment training	12 month follow-up	Daily fall calendar
Spice et al. (2009)∼	3-arm RCT	Usual care	Primary care intervention and assessment by trained community nurse to identify fall risk factors with referral to primary and secondary services	136 (114)	162 (131)	Proportion of participants who fell in follow-up period	Received referrals from falls nurse to follow-up/action	12 month follow-up	Falls diary
UK	Community-dwelling adults >65 who had at least two falls in the previous 12 months	Usual care		136 (114)	162 (131)	Proportion of participants who fell in follow-up period	Received referrals from falls nurse to follow-up/action	12 month follow-up	Falls diary
Stam et al. (2018)	Cluster RCT	Usual care. Use of existing guidelines on dizziness	Medication adjustment in case of <3 fall-risk-increasing drugs by pharmacist and GP	83 (72)	85 (78)	Weekly fall frequency	Implementing recommendations	3, 6 and 12 months	Falls calendar
Netherlands	Community-dwelling >65 years who consulted their GP for dizziness	Usual care. Use of existing guidelines on dizziness		83 (72)	85 (78)	Weekly fall frequency	Implementing recommendations	3, 6 and 12 months	Falls calendar
Tamblyn et al. (2012)	Cluster RCT	GPs received commercial drug alerts	GPs received information about patient specific risk of injury via computerised system: Modifiable risk factors and psychotropic drug doses	2887 (2887)	2741 (2741)	Risk of falls related injury	GPs to take action based on alerts provided	22 month follow-up	Fall tracking unclear
Canada	Community-dwelling men and women >65	GPs received commercial drug alerts		2887 (2887)	2741 (2741)	Risk of falls related injury	GPs to take action based on alerts provided	22 month follow-up	Fall tracking unclear
Van der Meer et al. (2018)	2-arm RCT	Usual care	Medication review by pharmacist and GP on anticholinergic and sedative load	80 (65)	84 (80)	Proportion of fallers	GPs to undertake review with pharmacist in agreement with patient and implement change	3 months	Self-reported falls and hospital records
Netherlands	Community-dwelling ≥65 years using ≥5 medicines for ≥3 months, drug burden index ≥1	Usual care		80 (65)	84 (80)	Proportion of fallers		3 months	Self-reported falls and hospital records
Van Haastregt et al. (2000)	2-arm RCT	Usual care	5 home visits by community nurse over 12 months using screening protocol for fall risk factors and referrals to other health professionals including GP	159 (120)	157 (115)	Proportion of participants with at least one fall, >1 fall and injurious falls	GPs received referrals from community nurse	12 month follow-up	Weekly falls diary
Netherlands	Community-dwelling adults >70 with falls history or moderate mobility impairment	Usual care		159 (120)	157 (115)		GPs received referrals from community nurse	12 month follow-up	Weekly falls diary
Weber et al. (2008)^	Prospective RCT	Usual care	Use of existing EMR system to screen for falls risk and medication review by clinical pharmacist or geriatrician	413 (337)	207 (158)	Proportion of fallers	GPs were sent the alert and recommendations by the EMR system and GP to follow up/action	6, 9, 12 and 15 months	3-month reporting and database
USA	Community-dwelling men and women >70	Usual care		413 (337)	207 (158)	Proportion of fallers		6, 9, 12 and 15 months	3-month reporting and database
Whitehead et al. (2003)	2-arm RCT	Usual care	Falls risk assessment with results of assessment and recommendations sent to patient GP for action	70 (58)	70 (65)	At least one fall at follow-up	GPs received recommendations and assessment results to action	6 month follow-up	Falls diary
Australia	Adults >65 attending emergency department as a result of a fall, living in community or low care residential care	Usual care		70 (58)	70 (65)	At least one fall at follow-up		6 month follow-up	Falls diary

Note. ^Where outcomes were measured at several points, the 12 month outcome data was selected. *Falls data was gathered via questionnaire and prospective fall diary; for the purposes of this study only the diary data was used in line with the Prevention of Falls Network Europe recommendations (REF). ∼Secondary intervention took place in a hospital day clinic and this data was not included. GP = general practitioner; HRA-O = Health risk appraisal intervention. RCTs = randomised controlled trials; EMR = electronic medical record.

Risk of Bias/Quality Assessment of Included Studies

Overall, the methodological quality of the included studies was acceptable (Table 2), although a lack of clarity in reporting for some studies resulted in unclear quality appraisal scores (Turner et al., 2012). As these studies took place in real practice settings and relied on adherence to protocols by participating GPs and participants to maintain treatment fidelity (Dyas et al., 2014) as well as greater difficulty controlling external variables, strictly controlled experimental conditions were not possible to achieve. The appraisal allowed for the pragmatic nature of the included studies; that is the evaluation of the effectiveness of the intervention was tested in real-life clinical contexts (Fransen et al., 2007).

Table 2.

Quality Appraisal of Included Studies in the Review Based on Joanna Briggs Institute Critical Appraisal Tool for RCTs (Tufanaru et al., 2017).

First author and year	Barr et al. (2005)	Coleman et al. (1999)	Dapp et al. (2011)	De Vries et al. (2010)	Elley et al. (2008)	Hendriks et al. (2008)	Kerse et al. (2005)	Kolt et al. (2012)	Lightbody et al. (2002)	Newbury et al. (2001)	Palvanen et al. (2014)	Pit et al. (2007)	Shepstone et al. (2018)	Siegrist et al. (2016)	Spice et al. (2009)	Stam et al. (2018)	Tamblyn et al. (2012)	Van der Meer et al. (2018)	Van Haastregt et al. (2000)	Weber et al. (2008)	Whitehead et al. (2003)
Was true randomisation used for assignment to treatment groups?	+	?	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+
Was allocation to treatment groups concealed?	+	?	+	+	+	+	−	+	?	?	+	−	+	−	−	+	−	?	?	?	+
Were treatment groups similar at baseline?	+	−	+	+	+	+	+	+	+	+	+	+	+	+	+	+	−	?	+	+	−
Were participants blind to treatment assignment?	−	−	?	−	−	−	−	?	?	?	?	+	?	?	?	?	?	−	?	?	?
Were those delivering treatment blind to treatment assignment?	−	−	−	−	−	−	−	−	−	?	−	−	−	−	−	−	+	−	+	−	?
Were outcomes assessors blind to treatment assignment?	N/A	+	+	?	+	+	N/A	+	N/A	+	?	+	+	?	−	?	+	?	−	?	+
Were treatment groups treated identically other than the intervention of interest?	?	?	−	?	+	+	+	+	+	?	+	+	?	?	+	?	+	+	+	+	+
Was follow-up complete or were differences between groups adequately described and analysed?	−	?	+	?	?	+	?	?	?	?	?	?	+	?	+	+	+	+	+	?	?
Were participants analysed in the groups to which they were randomised?	+	?	+	?	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+
Were outcomes measured in the same way for treatment groups?	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	?	+	+	+
Were outcomes measured in a reliable way?	?	−	+	+	+	+	?	+	+	?	?	+	+	+	+	+	+	?	?	+	+
Was appropriate statistical analysis used?	+	+	+	−	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+
Was the trial design appropriate?	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+	+
% Score of +	54	31	77	46	77	85	54	77	62	54	62	77	77	54	69	69	77	46	69	62	69

Note. + = yes, ? = unclear and − = no.

Most studies adequately randomised and concealed allocation to control/intervention groups at the time of randomisation but were not able to conceal allocation to participants and those personnel delivering the intervention. Instances of successful outcome assessment blinded to treatment included the use of self-report postal surveys (Barr et al., 2005), chart abstraction (Coleman et al., 1999) and data entry blinded to allocation (Dapp et al., 2011; Elley et al., 2008) or use of an outside agency to complete the outcome assessment (Hendriks et al., 2008). ITT analysis was not clearly stated in all studies. Eleven (61%) studies reported that they used ITT, six (33%) did not indicate ITT, but was assumed from the follow-up data and analysis provided. De Vries et al. (2010) did not conduct intention-to-treat analysis, explicitly stating that participants were censored at the time of drop out. Participant samples varied across the studies from very small (50 participants in each of the intervention and control groups in one study to large cohorts of 6233 for the intervention group and 6250 for the control group. All studies except for Tamblyn et al. (2012) reported participant dropouts of between 10% to 30% of participants (see Table 1 for participant data).

Reliability of outcome measures was adequate for 10 (44%) studies where the use of prospective or continuous recording through the use of a fall calendar or diary was adopted in line with the Prevention of Falls Network Europe (ProFANE) recommendations (Lamb et al., 2005) or where fractures were ascertained through hospitalisation records; six studies used self-reporting via questionnaire or telephone contact at three-month intervals during the follow-up period (e.g. Barr et al., 2005; Coleman et al., 1999; Weber et al., 2008) and five studies remained unclear in their reporting regarding frequency or method of reporting/documenting falls; this was assumed to be self-reported at follow-up (Kerse et al., 2005; Newbury et al., 2001; Palvanen et al., 2014; Pit et al., 2007; Van Haastregt et al., 2000).

Studies that Could Not Be Evaluated in the Meta-Analysis

Tamblyn et al. (2012) did not report on the proportions of fallers. In this study, a drug alert intervention resulted in an overall reduction in the number of drugs and doses of psychotropic drugs for the intervention group in their study. These changes resulted in a reduction in the risk of injurious falls for the intervention group by 1.7 injuries per 1000 patients (95% CI .2/1000–3.2/1000; p = .02) when compared with the control group. Additionally, the effects of the intervention were reported as significantly greater for patients at higher risk of injury at baseline (Tamblyn et al., 2012). Patients in the intervention group who had a 10% probability of injury in the next year at baseline had a 30% reduction in their injury risk at follow-up, compared with patients in the intervention group who had a 5% probability of injury in the next year at baseline and had a 15% reduction in their injury risk at follow-up.

Stam et al. (2018) reported on fall frequency per week but did not differentiate between the intervention group and the control group, quoting an overall mean difference of 1.47 (95% CI: .55–3.89) for 83 participants. This study tested the effectiveness of a multifactorial risk factor–guided intervention for older people reporting dizziness to their GP. However, this was not effective, due in part to the intervention being too difficult for the participants to adhere to.

Proportion of Participants Experiencing at Least One Fall

A total of 18 studies including data from 186,932 participants were included in the meta-analyses that reported on the proportion of participants who had experienced at least one fall at follow-up. A fixed effects model analysis revealed a high level of heterogeneity between studies. Therefore, a random effects model was selected for analysis for this outcome (Figure 2).

Figure 2.

Studies reporting one or more falls at follow-up.

Overall, the effectiveness of GP based interventions for reducing the proportion of participants experiencing any falls was low with an insignificant 9% reduction in favour of the intervention groups (Relative Risk (RR) .91, 95% CI: .82–1.01 and p = .08). Five studies in this group were effective in reducing falls in the intervention group (Coleman et al., 1999, p = .041; Dapp et al., 2011, p = .048; Palvanen et al., 2014, p = .002; Pit et al., 2007, p = .002 and Siegrist et al., 2016, p = .017). These studies tested the following interventions: multidisciplinary primary care chronic care clinics conducted by the GP, self-administered health risk appraisal reported to the GP, multidisciplinary falls clinic including the GP, medication risk assessment by GP and falls risk assessment by GP. All other studies did not significantly contribute to reducing falls and despite the clinical involvement by GPs, interventions often used a third-party to conduct assessments who then reported to the GP for action. One study appeared to increase the risk of falls (Whitehead et al., 2003), possibly due to difficulties with the uptake of the intervention in real practice settings.

Proportion of Participants Who Experienced More than One Fall

Four studies provided data on the proportion of participants who had experienced multiple falls during the follow-up period (>1 fall) and when pooled scored low in heterogeneity (Figure 3). Both random and fixed effects models generated the same data. The analysis showed that overall the interventions in these studies were ineffective at reducing the proportion of participants who fell more than once in the intervention group.

Figure 3.

Studies reporting multiple falls at follow-up.

Proportion of Participants Who Experienced Injurious Falls

This analysis used studies that reported on the proportion of participants who experienced any injurious falls (see Figure 4). The data reported did not distinguish the severity of injurious falls, except for Shepstone et al. (2018) who reported on hip fractures as an outcome. This latter study did not report falls. There were six studies which reported on the percentage of participants who experienced injurious falls. The combined effect size of interventions reducing injurious falls in the intervention group was a significant 24% reduction in falls (RR .76, 95% CI: .66–.87 and p = .001). Of these studies, two were individually effective in reducing falls (Pit et al., 2007; Shepstone et al., 2018).

Figure 4.

Studies reporting injurious falls at follow-up.

Discussion

This review demonstrated that interventions focussing on one or more falls or multiple falls as an outcome were not effective overall, whereas those focussing on injurious falls as an outcome were effective overall. Significantly reducing injurious falls is an important outcome, particularly as a positive way to incentivise GPs to engage in fall prevention. Studies in the injurious falls meta-analysis involved one study (Shepstone et al., 2018) with a very large sample size (n = 12,483), two studies measuring fractures as outcomes (Barr et al., 2005; Shepstone et al., 2018) and one study with a five year follow-up period (Shepstone et al., 2018), which would ensure any injurious fall outcomes would be captured. The studies that contributed most to this finding included a GP education component as well as the active involvement of the GP through fall risk screening (Shepstone et al., 2018; Siegrist et al., 2016) and treatment specific to fracture risk (Shepstone et al., 2018) or reviewing medications (Pit et al., 2007; Siegrist et al., 2016). The other studies had small samples thus few injurious falls resulting in wider confidence intervals.

The non-effective result for interventions reducing falls is less clear. Some of the studies included some aspect of GP falls risk screening and assessment (Coleman et al., 1999; Dapp et al., 2011; Palvanen et al., 2014) or GP medication review (Pit et al., 2007; Siegrist et al., 2016), and most were not effective. They included a variety of interventions conducted by others such as clinics, fall and fracture nurse–led interventions (screening and assessment) and exercise, some of which according to Cochrane reviews (Sherrington et al., 2019) are evidence-based and others are not. For example the exercise intervention studies by Kolt et al. (2012) and Kerse et al. (2005), were primarily cardiovascular outdoor activity rather than compared with the balance training now known to be core to protection from falling (Sherrington et al., 2017). Interventions targeting identification of older people at risk of falls and assessment of their risk factors demonstrated efficacy over other types of intervention potentially because the interventions integrate well into the daily activities and workload of GPs. The GP role focuses on preventative care and management of chronic conditions (Freund et al., 2015) and this includes health assessment and referral to other health professionals as a fundamental part of their duties (Britt et al., 2016). GPs referring to community-based fall prevention programs and services that are evidence based and understanding the role of allied health professionals in fall prevention may be a key to better overall success.

GP education and training in falls prevention risk assessment and management has been supported in larger implementation studies. The early Stay On Your Feet program in NSW (Kempton et al., 1996) demonstrated efficacy, with approximately 80% of GPs who had participated in the program retained and used the knowledge gained during the program in subsequent years. There is a need to address where GPs have limited knowledge or education regarding falls prevention. This may involve developing effective education programs for GP and GPs in training. Preferred formats for education include in person and online delivery with the key motivator being relevance to everyday practice (Yee et al., 2014). One example of falls prevention online resources for GPs can be found at https://fallspreventiononlineworkshops.com.au.

Despite the employment of registered nurses in general practice contexts increasing in the last decade to work with GPs (Australian Medicare Local Alliance, 2012), the efficacy of the interventions using nurses to carry out the screening was not confirmed in this review. The reasons behind this outcome may not necessarily be related to the intervention itself but could be the result of factors around the necessarily pragmatic nature of RCTs in general practice (Fransen et al., 2007), where GPs may share the clinical load with their practice nurses, so specific contributions may be unclear. For instance, in a real-world context, researchers have less control over the delivery of the intervention by the GPs, and in situations where the participants are required to initiate further action, for example following up with their GP, the researchers cannot control that either. This means that the treatment fidelity of the intervention under investigation can be difficult to determine in RCTs based in general practice. For instance, for the nurse-led intervention studies, the GPs involved were distanced from the intervention, which may impact their sense of responsibility to and for study in light of heavy daily workloads and may not follow up as intended by the researchers (Dyas et al., 2014). There was little reported in the included studies about tracking intervention fidelity, and research indicates that detailed reporting of contextual factors is necessary to adequately assess the results of these types of studies (Wells et al., 2012).

Limitations of This Systematic Review

There were several limitations to this systematic review. Firstly, the risk of bias across studies needs to be considered as all of the included studies were published in English and predominantly took place in English speaking countries (the UK, Australia, the USA and New Zealand) or Western European contexts. Thus, it can be argued that the populations of the studies represented in this systematic review were subject to some geographical, and publication bias (Gough et al., 2012), which may skew the results of the meta-analysis. Secondly, the nature of research in general practice contexts, where interventions can be variable between GPs, and the focus on falls prevention may have been challenged by any patient selection bias by GPs across the studies. Most studies reported some loss of participants at follow-up, and this should be considered when interpreting the results as there is a risk of inflating the effectiveness of results where participants have dropped out. It is possible that participants who discontinue with the study do so because their mobility and health have been affected by any falls and the subsequent impacts of those falls. In this scenario, those that remain may be more physically well and less frail than those that drop out. Finally, there was high heterogeneity between the studies selected for inclusion, making interpretation of individual study results difficult. Broad age ranges from 60 years and up, and differing levels of baseline risk of falls may have impacted the appearance of efficacy of the interventions overall. Given that risk of falls increases with age (World Health Organization, 2008), selection criteria both at individual study level and in the eligibility criteria of this systematic review may mask the potential treatment effect of varied interventions.

Conclusions and Recommendations

This systematic review has found that GPs can contribute to effective falls prevention interventions by undertaking education, conducting falls risk screening and assessment and following up appropriately (e.g. referral to other health professionals). These interventions were more effective than other types of falls prevention interventions including fall nurse–led interventions, exercise interventions and multidisciplinary assessment and referral back to the GP after admission to the emergency department following a fall. These moderately effective studies align with the current body of literature regarding education programs for GPs and specific contextual factors for general practice. There were several limitations to this review and across individual studies included which may impact the results and interpretation. Falls prevention interventions are well established. Future research in falls prevention in the GP context should continue to look at overcoming the barriers for GPs in implementing falls prevention successfully, supporting GPs to reduce falls among older people in Australia by aligning the interventions with GPs and their organisational strengths.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Lynette Mackenzie

References

Australian Bureau of Statistics . (2013). Population projections 2012 to 2101, 3222.0. Commonwealth Government of Australia. https://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/3222.02012%20(base)%20to%202101?OpenDocument

Australian Commission on Safety Quality in Health Care . (2009). Guidebook for preventing falls and harm from falls in older people: Australian community care. ACSQHC. https://www.safetyandquality.gov.au/sites/default/files/migrated/Guidelines-COMM.pdf

Australian Institute of Health and Welfare . (2017). Ageing in Australia. https://www.aihw.gov.au/reports/older-people/older-australia-at-a-glance/contents/healthy-ageing

Australian Medicare Local Alliance . (2012). 2012 General Practice Nurse National Survey Report. http://healthypractices.apna.asn.au/wp-content/uploads/2015/03/General-Practice-Nurse-National-Workforce-Survey-2012.pdf

Barr

R. J.

Stewart

Torgerson

D. J.

Seymour

D. G.

Reid

D. M.

(2005). Screening elderly women for risk of future fractures-participation rates and impact on incidence of falls and fractures. Calcified Tissue International, 76, 243-248. doi:10.1007/s00223-004-0101-5

Boffin

Moreels

Vanthomme

Van Casteren

(2014). Falls among older general practice patients: A 2-year nationwide surveillance study. Family Practice, 31, 281-289. doi:10.1093/fampra/cmu002

Borenstein

Hedges

L. V.

Higgins

J. P. T.

Rothstein

H. R.

(2010). A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods, 1, 97-111. doi:10.1002/jrsm.12

Britt

Miller

Henderson

Bayram

Harrison

Valenti

Pan

Charles

Pollack

Wong

Gordon

(2016). General practice activity in Australia 2015–16. https://ses.library.usyd.edu.au/handle/2123/15514

Buck

Tian

Thompson

(2014). The cost of falls: Exploring the cost of the whole system pathway for older people in a rural community in England. Journal of Integrated Care, 22, 165-173. doi:10.1108/JICA-04-2014-0012

10.

Bunn

Dickinson

Barnett-Page

McInnes

Horton

(2008). A systematic review of older people’s perceptions of facilitators and barriers to participation in falls-prevention interventions. Ageing and Society, 28, 449-472. doi:10.1017/S0144686X07006861

11.

Carter

S. E.

Campbell

E. M.

Sanson-Fisher

R. W.

Redman

Gillespie

W. J.

(1997). Environmental hazards in the homes of older people. Age and Ageing, 26, 195-202. doi:10.1093/ageing/26.3.195

12.

Cesari

Landi

Torre

Onder

Lattanzio

Bernabei

(2002). Prevalence and risk factors for falls in an older community-dwelling population. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 57, M722-M726. doi:10.1093/gerona/57.11.m722

13.

Child

Goodwin

Garside

Jones-Hughes

Boddy

Stein

(2012). Factors influencing the implementation of fall-prevention programmes: A systematic review and synthesis of qualitative studies. Implementation Science, 7, 91. doi:10.1186/1748-5908-7-91

14.

Chou

W. C.

Tinetti

M. E.

King

M. B.

Irwin

Fortinsky

R. H.

(2006). Perceptions of physicians on the barriers and facilitators to integrating fall risk evaluation and management into practice. Journal of General Internal Medicine, 21, 117-122. doi:10.1111/j.1525-1497.2005.00298.x

15.

Coleman

E. A.

Grothaus

L. C.

Sandhu

Wagner

E. H.

(1999). Chronic care clinics: A randomized controlled trial of a new model of primary care for frail older adults. Journal of the American Geriatrics Society, 47, 775-783. doi:10.1111/j.1532-5415.1999.tb03832.x

16.

Croker

J. E.

Swancutt

D. R.

Roberts

M. J.

Abel

G. A.

Roland

Campbell

J. L.

(2013). Factors affecting patients’ trust and confidence in GPs: Evidence from the English national GP patient survey. BMJ Open, 3, Article e002762. doi:10.1136/bmjopen-2013-002762

17.

Dapp

Anders

J. A. M.

von Renteln-Kruse

Minder

C. E.

Meier-Baumgartner

H. P.

Swift

C. G.

Gillmann

Egger

Beck

J. C.

Stuck

A. E.

(2011). A randomized trial of effects of health risk appraisal combined with group sessions or home visits on preventive behaviors in older adults. The Journals of Gerontology: Series A, 66A, 591-598. doi: 10.1093/gerona/glr021

18.

De Vries

Peeters

Elders

Muller

Knol

Danner

Bouter

Lips

(2010). Multifactorial intervention to reduce falls in older people at high risk of recurrent falls: A randomized controlled trial. Archives of Internal Medicine, 170, 1110-1117. doi:10.1001/archinternmed.2010.169

19.

Dollard

Braunack-Mayer

Horton

Vanlint

(2014). Why older women do or do not seek help from the GP after a fall: A qualitative study. Family Practice, 31, 222-228. doi:10.1093/fampra/cmt083

20.

Dyas

Togher

Siriwardena

(2014). Intervention fidelity in primary care complex intervention trials: Qualitative study using telephone interviews of patients and practitioners. Quality in Primary Care, 22, 25-34.

21.

El-Masri

. (2015). Intention-to-treat analysis in RCTs. The Canadian Nurse, 111, 9.

22.

Elkins

M. R.

Moseley

A. M.

(2015). Intention-to-treat analysis. Journal of Physiotherapy, 61, 165-167. doi:10.1016/j.jphys.2015.05.013

23.

Elley

C. R.

Robertson

M. C.

Garrett

Kerse

N. M.

McKinlay

Lawton

Moriarty

Moyes

S. A.

Campbell

A. J.

(2008). Effectiveness of a falls-and-fracture nurse coordinator to reduce falls: A randomized, controlled trial of at-risk older adults. Journal of the American Geriatrics Society, 56, 1383-1389. doi:10.1111/j.1532-5415.2008.01802.x

24.

Fransen

G. A.

van Marrewijk

C. J.

Mujakovic

Muris

J. W.

Laheij

R. J.

Numans

M. E.

de Wit

N. J.

Samsom

Jansen

J. B.

Knottnerus

J. A.

(2007). Pragmatic trials in primary care. Methodological challenges and solutions demonstrated by the DIAMOND-study. BMC Medical Research Methodology, 7, 16. doi:10.1186/1471-2288-7-16

25.

Freund

Everett

Griffiths

Hudon

Naccarella

Laurant

(2015). Skill mix, roles and remuneration in the primary care workforce: Who are the healthcare professionals in the primary care teams across the world? International Journal of Nursing Studies, 52, 727-743. doi:10.1016/j.ijnurstu.2014.11.014

26.

Gaboreau

Imbert

Jacquet

J.-P.

Royer De Vericourt

Couturier

Gavazzi

(2016). Barriers to and promoters of screening for falls in elderly community-dwelling patients by general practitioners: A large cross-sectional survey in two areas of France. Archives of Gerontology and Geriatrics, 65, 85-91. doi:10.1016/j.archger.2016.03.002

27.

Gillespie

Robertson

Gillespie

Sherrington

Gates

Clemson

Lamb

(2012). Interventions for preventing falls in older people living in the community. The Cochrane Database of Systematic Reviews, 9, Article CD007146.

28.

Gough

Oliver

Thomas

(2012). An introduction to systematic reviews. SAGE.

29.

Greenfield

Foley

Majeed

(2016). Rethinking primary care’s gatekeeper role. BMJ: British Medical Journal, 354, Article i4803. doi:10.1136/bmj.i4803

30.

Hendriks

M. R. C.

Bleijlevens

M. H. C.

van Haastregt

J. C. M.

Crebolder

H. F. J. M.

Diederiks

J. P. M.

Evers

S. M. A. A.

Mulder

W. J.

Kempen

G. I. J. M.

van Rossum

Ruijgrok

J. M.

Stalenhoef

P. A..

van Eijk

J. T. M.

(2008). Lack of effectiveness of a multidisciplinary fall-prevention program in elderly people at risk: A randomized, controlled trial. Journal of the American Geriatrics Society, 56, 1390-1397. doi:10.1111/j.1532-5415.2008.01803.x

31.

Hoffmann

Bennett

Del Mar

(2010). Evidence based practice across the health professions. Churchill Livingstone/Elsevier.

32.

Jones

T. S.

Ghosh

T. S.

Horn

Smith

Vogt

R. L.

(2011). Primary care physicians perceptions and practices regarding fall prevention in adult’s 65 years and over. Accident Analysis & Prevention, 43, 1605-1609. doi:10.1016/j.aap.2011.03.013

33.

Kempton

van Beurden

Garner

(1996). Strategies to prevent falls: The stay on your feet program. Health Promotion Journal of Australia, 6, 37-43. doi:10.1093/heapro/15.1.27

34.

Kerse

Elley

C. R.

Robinson

Arroll

(2005). Is physical activity counseling effective for older people? A cluster randomized, controlled trial in primary care. Journal of the American Geriatrics Society, 53, 1951-1956. doi:10.1111/j.1532-5415.2005.00466.x

35.

Kielich

Mackenzie

Lovarini

Clemson

(2017). Urban Australian general practitioners’ perceptions of falls risk screening, falls risk assessment, and referral practices for falls prevention: An exploratory cross-sectional survey study. Australian Health Review, 41(1), 111-119. doi:10.1071/AH15152

36.

Kolt

G. S.

Schofield

G. M.

Kerse

Garrett

Ashton

Patel

(2012). Healthy steps trial: Pedometer-based advice and physical activity for low-active older adults. The Annals of Family Medicine, 10, 206-212. doi:10.1370/afm.1345

37.

Lamb

S. E.

JÃ¸rstad-Stein

E. C.

Hauer

Becker

(2005). Development of a common outcome data set for fall injury prevention trials: The Prevention of falls network Europe consensus. Journal of the American Geriatrics Society, 53, 1618-1622. doi:10.1111/j.1532-5415.2005.53455.x

38.

Lee

D.-C. A.

Day

Hill

Clemson

McDermott

Haines

T. P.

(2015). What factors influence older adults to discuss falls with their health-care providers? Health Expectations, 18, 1593-1609. doi:10.1111/hex.12149

39.

Lightbody

Watkins

Leathley

Sharma

Lye

(2002). Evaluation of a nurse-led falls prevention programme versus usual care: A randomized controlled trial. Age and Ageing, 31, 203-210. doi:10.1093/ageing/31.3.203

40.

Mackenzie

Clemson

Roberts

(2013). Occupational therapists partnering with general practitioners to prevent falls: Seizing opportunities in primary health care. Australian Occupational Therapy Journal, 60, 66-70. doi:10.1111/1440-1630.12030

41.

Mackenzie

McIntyre

(2019). How do general practitioners (GPs) engage in falls prevention with older people? A pilot survey of GPs in NHS England suggests a gap in routine practice to address falls prevention. Frontiers in Public Health, 7, 32. doi:10.3389/fpubh.2019.00032

42.

Marcinowicz

Pawlikowska

Oleszczyk

(2014). What do older people value when they visit their general practitioner? A qualitative study. European Journal of Ageing, 11, 361-367. doi:10.1007/s10433-014-0313-0

43.

Moher

Liberati

Tetzlaff

Altman

Group

(2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine, 151, 264. doi:10.7326/0003-4819-151-4-200908180-00135

44.

Monagle

(2002). Reducing falls in community dwelling elderly. The role of GP care planning. Australian Family Physician, 31, 1111-1115.

45.

Newbury

J. W.

Marley

J. E.

Beilby

J. J.

(2001). A randomised controlled trial of the outcome of health assessment of people aged 75 years and over. Medical Journal of Australia, 175, 104-107.

46.

Nich

Carroll

K. M.

(2002). ‘Intention-to-treat’ meets ‘missing data’: Implications of alternate strategies for analyzing clinical trials data. Drug and Alcohol Dependence, 68, 121-130. doi:10.1016/s0376-8716(02)00111-4

47.

O’Connell

Ostaszkiewicz

(2005). Sink or swim – ageing in Australia. Australian Health Review : A Publication of the Australian Hospital Association, 29, 146-150. doi:10.1071/AH050146

48.

Palvanen

Kannus

Piirtola

Niemi

Parkkari

Järvinen

(2014). Effectiveness of the Chaos falls clinic in preventing falls and injuries of home-dwelling older adults: A randomised controlled trial. Injury, 45, 265-271. doi:10.1016/j.injury.2013.03.010

49.

Peel

N. M.

(2011). Epidemiology of falls in older age. Canadian Journal on Aging/La Revue canadienne du vieillissement, 30, 7-19. doi:10.1017/S071498081000070X

50.

Pit

S. W.

Byles

J. E.

Henry

D. A.

Holt

Hansen

Bowman

D. A.

(2007). A quality use of medicines program for general practitioners and older people: A cluster randomised controlled trial. Medical Journal of Australia, 187, 23-30.

51.

Shepstone

Lenaghan

Cooper

Clarke

Fong-Soe-Khioe

Fordham

Gittoes

Harvey

Heawood

Holland

Howe

Kanis

Marshall

O'Neill

Peters

Redmond

Torgerson

Turner

McCloskey

(2018). Screening in the community to reduce fractures in older women (SCOOP): A randomised controlled trial. Lancet, 391, 741-747. doi:10.1016/S0140-6736(17)32640-5

52.

Sherrington

Fairhall

Wallbank

Tiedemann

Michaleff

Howard

Clemson

Hopewell

Lamb

(2019). Exercise for preventing falls in older people living in the community. The Cochrane Database of Systematic Reviews, 1, Article CD012424. doi:10.1002/14651858.CD012424.pub2

53.

Sherrington

Michaleff

Z. A.

Fairhall

Paul

S. S.

Tiedemann

Whitney

Cumming

R. G.

Herbert

R. D.

J. C. T.

Lord

S. R.

(2017). Exercise to prevent falls in older adults: An updated systematic review and meta-analysis. British Journal of Sports Medicine, 51, 1750-1758. doi:10.1136/bjsports-2016-096547

54.

Siegrist

Freiberger

Geilhof

Salb

Hentschke

Landendoerfer

Linde

Halle

Blank

W. A.

(2016). Fall prevention in a primary care setting. Deutsches Aerzteblatt Online, 113, 365-372. doi:10.3238/arztebl.2016.0365

55.

Smith

M. I.

de Lusignan

Mullett

Correa

Tickner

Jones

(2016). Predicting falls and when to intervene in older people: A multilevel logistical regression model and cost analysis. PLoS ONE, 11(7), Article e0159365. doi:10.1371/journal.pone.0159365

56.

Spice

C. L.

Morotti

George

Dent

T. H. S.

Rose

Harris

Gordon

C. J.

(2009). The Winchester falls project: A randomised controlled trial of secondary prevention of falls in older people. Age and Ageing, 38, 33-40. doi:10.1093/ageing/afn192

57.

Stalenhoef

Crebolder

Knottnerus

VanderHorst

(1997). Injuries. Incidence, risk factors and consequences of falls among elderly subjects living in the community. A criteria-based analysis. The European Journal of Public Health, 7, 328-334. doi:10.1093/eurpub/7.3.328

58.

Stalenhoef

P. A.

Diederiks

J. P. M.

De Witte

L. P.

Schiricke

K. H.

Crebolder

H. F. J. M.

(1999). Impact of gait problems and falls on functioning in independent living persons of 55 years and over: A community survey. Patient Education and Counseling, 36, 23-31. doi:10.1016/S0738-3991%2898%2900071-8

59.

Stam

van der Wouden

Hugtenburg

Twisk

van der Horst

Maarsingh

(2018). Effectiveness of a multifactorial intervention for dizziness in older people in primary care: A cluster randomised controlled trial. PLoS ONE, 13, (10), Article e0204876. doi:10.1371/journal.pone.0204876

60.

Stubbs

Brefka

Denkinger

M. D.

(2015). What works to prevent falls in community-dwelling older adults? Umbrella review of meta-analyses of randomized controlled trials. Physical Therapy, 95, 1095-1110. doi:10.2522/ptj.20140461

61.

Tamblyn

Eguale

Buckeridge

D. L.

Huang

Hanley

Reidel

Shi

Winslade

(2012). The effectiveness of a new generation of computerized drug alerts in reducing the risk of injury from drug side effects: A cluster randomized trial. Journal of the American Medical Informatics Association, 19, 635-643. doi:10.1136/amiajnl-2011-000609

62.

Topinková

(2008). Aging, disability and frailty. Annals of Nutrition & Metabolism, 52, 6-11. doi:10.1159/000115340

63.

Tufanaru

Munn

Aromataris

Campbell

Hopp

(2017). Systematic reviews of effectiveness. In Aromataris

Munn

(Eds), Joanna Briggs Institute reviewer’s manual.

64.

Turner

Shamseer

Altman

Weeks

Peters

Kober

Dias

Schulz

Plint

Moher

(2012). Consolidated standards of reporting trials (CONSORT) and the completeness of reporting of randomised controlled trials (RCTs) published in medical journals. The Cochrane Database of Systematic Reviews, 11, Article MR000030.

65.

van der Meer

H. G.

Wouters

Pont

L. G.

Taxis

(2018). Reducing the anticholinergic and sedative load in older patients on polypharmacy by pharmacist-led medication review: A randomised controlled trial. BMJ Open, 8, Article e019042. doi:10.1136/bmjopen-2017-019042

66.

van Haastregt

J. C. M.

Diederiks

Rossum

(2000). Effects of a programme of multifactorial home visits on falls and mobility impairments in elderly people at risk: Randomised controlled trial. BMJ: British Medical Journal, 321, 994-998. doi:10.1136/bmj.321.7267.994

67.

Weber

White

McIlvried

(2008). An electronic medical record (EMR)-based intervention to reduce polypharmacy and falls in an ambulatory rural elderly population. Journal of General Internal Medicine, 23, 399-404. doi:10.1007/s11606-007-0482-z

68.

Wells

Williams

Treweek

Coyle

Taylor

(2012). Intervention description is not enough: Evidence from an in-depth multiple case study on the untold role and impact of context in randomised controlled trials of seven complex interventions. Trials Electronic Resource, 13, 95. doi:10.1186/1745-6215-13-95

69.

Whitehead

Wundke

Crotty

Finucane

(2003). Evidence-based clinical practice in falls prevention: A randomised controlled trial of a falls prevention service. Australian Health Review, 26, 88-97. doi:10.1071/ah030088

70.

Whitehead

C. H.

Wundke

Crotty

(2006). Attitudes to falls and injury prevention: What are the barriers to implementing falls prevention strategies? Clinical Rehabilitation, 20, 536-542. doi:10.1191/0269215506cr984oa

71.

World Health Organization . (2008). WHO global report on falls prevention in older age. WHO. https://apps.who.int/iris/handle/10665/43811

72.

Yee

Simpson-Young

Paton

Zuo

(2014). How do GPs want to learn in the digital era? Australian Family Physician, 43, 399-402.