Fall-Risk-Increasing Drugs: A Systematic Review and Meta-Analysis: II. Psychotropics

JAMDA 19 (2018) 371.e11e371.e17

JAMDA journal homepage: www.jamda.com

Original Study

Fall-Risk-Increasing Drugs: A Systematic Review and Meta-Analysis: II. Psychotropics Lotta J. Seppala MSc a, b, Anne M.A.T. Wermelink BM a, b, Max de Vries BM a, b, Kimberley J. Ploegmakers MD a, b, Esther M.M. van de Glind MD, PhD a, b, Joost G. Daams MA c, Nathalie van der Velde MD, PhD a, b, * , on behalf of the EUGMS task and Finish group on fall-risk-increasing drugs a b c

Department of Internal Medicine, Section of Geriatric Medicine, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands Amsterdam Public Health Research Institute, Amsterdam, The Netherlands Medical Library, Academic Medical Center, Amsterdam, The Netherlands

a b s t r a c t Keywords: Accidental falls psychotropic: antidepressant, benzodiazepine antipsychotic

Background and objective: Falls are a major public health problem in older adults. Earlier studies showed that psychotropic medication use increases the risk of falls. The aim of this study is to update the current knowledge by providing a comprehensive systematic review and meta-analysis on psychotropic medication use and falls in older adults. Methods and design: This study is a systematic review and meta-analysis. A search was conducted in Medline, PsycINFO, and Embase. Key search concepts were “falls,” “aged,” “medication,” and “causality.” Studies were included that investigated psychotropics (antipsychotics, antidepressants, anxiolytics, sedatives, and hypnotics) as risk factors for falls in participants
60 years of age or participants with a mean age of
70 years. Meta-analyses were performed using generic inverse variance method pooling unadjusted and adjusted odds ratio (OR) estimates separately. Results: In total, 248 studies met the inclusion criteria for qualitative synthesis. Meta-analyses using adjusted data showed the following pooled ORs: antipsychotics 1.54 [95% confidence interval (CI) 1.28e1.85], antidepressants 1.57 (95% Cl 1.43e1.74), tricyclic antidepressants 1.41 (95% CI 1.07e1.86), selective serotonin reuptake inhibitors 2.02 (95% CI 1.85e2.20), benzodiazepines 1.42 (95%, CI 1.22e1.65), long-acting benzodiazepines 1.81 (95%, CI 1.05e3.16), and short-acting benzodiazepines 1.27 (95%, CI 1.04e1.56) Most of the meta-analyses resulted in substantial heterogeneity that did not disappear after stratification for population and healthcare setting. Conclusions: Antipsychotics, antidepressants, and benzodiazepines are consistently associated with a higher risk of falls. It is unclear whether specific subgroups such as short-acting benzodiazepines and selective serotonin reuptake inhibitors are safer in terms of fall risk. Prescription bias could not be accounted for. Future studies need to address pharmacologic subgroups as fall risk may differ depending on specific medication properties. Precise and uniform classification of target medication (Anatomical Therapeutic Chemical Classification) is essential for valid comparisons between studies. Ó 2018 AMDA e The Society for Post-Acute and Long-Term Care Medicine.

This work was supported by the Clementine Brigitta Maria Dalderup fund, Amsterdam University fund. The sponsor played no part in the design, methods, data collection, analysis and preparation of this paper. The authors declare no conflicts of interest. * Address correspondence to Nathalie van der Velde, MD, PhD, Department of Internal Medicine, Section of Geriatric Medicine, Academic Medical Center, University of Amsterdam, F4 129, Meibergdreef 9, Amsterdam 1105AZ, The Netherlands. E-mail address: [email protected] (N. van der Velde). https://doi.org/10.1016/j.jamda.2017.12.098 1525-8610/Ó 2018 AMDA e The Society for Post-Acute and Long-Term Care Medicine.

Falls are the dominant cause of injury among older adults.1 Annually, one-third of the people over the age of 65 years fall at least once and 20% of these falls result in an injury.2 The high prevalence and injury rate co-occur with substantial healthcare costs. Fall injuries are among the 20 most expensive medical conditions among community-dwelling older adults.3 Moreover, older adults often experience a decline in functional status and social activities after a fall and report a reduced quality of life.4,5 All in all, falls form a great

371.e12

L.J. Seppala et al. / JAMDA 19 (2018) 371.e11e371.e17

public health concern, now even more in light of the world’s aging population. Psychotropic medication has consistently been reported to increase the risk of falls.6e10 However, earlier meta-analyses have predominantly used unadjusted data, thus, being susceptible to major bias. Only Leipzig et al6 and Woolcott et al7 performed a meta-analysis of limited medication groups using adjusted data. Furthermore, despite the well-established association between psychotropics and fall risk, knowledge on potential risk differences between specific pharmacologic subgroups is scarce. The objective of this systematic review and meta-analysis is to appraise the associations between falls (any falls, recurrent falls, injurious falls) and specific psychotropic medication groups, thus, providing a comprehensive update of current knowledge on this topic. Methods Identification of Studies A search was performed in Medline, PsycINFO, and Embase from onset until September 28, 2016. A customized search strategy was conducted for each database with the help of a clinical librarian (J.D.). The search contained the following key search concepts: “fall,” “aged,” “causality,” and “medication.” The key search term medication contained all known fall-risk-increasing drugs (FRIDs).6e8,11e13 Vosviewer was used to identify other possible medication classes to be added to the “medication” concept of the search.14 The full search is available as a supplement (Appendix 1). Selection of Studies All the titles and abstracts were screened by one author (L.S.). A 40% random subselection of articles was screened by a second reviewer (M.V.). After selection of potential articles, the latter were assessed by 2 reviewers (L.S. and M.V. or A.W.). In case of disagreement concerning the final selection of articles, a third author (N.V.) was consulted. Furthermore, reference checking was performed to search for potentially eligible studies from original and review articles. Randomized controlled trials (RCTs), cohort studies, case-control studies, cross-sectional studies, and case-crossover studies that investigated the association between psychotropic medications and any fall, recurrent falls (2 or more), or injurious falls were included. Psychotropics were defined as medications that are classified under N05 “psycholeptics” (antipsychotics, anxiolytics, hypnotics, and sedatives) or N06A “psychoanaleptics” (antidepressants) according to the Anatomical Therapeutic Chemical (ATC) classification15 system, and the general term psychotropics was also included. Only studies with participants aged
60 years, participants with a mean age of 70 years or over, and studies that reported results of suitable age subgroups were included. We included studies from all settings (populationbased, community-dwellers, hospital wards, long-term care institutions, and outpatient clinics). Articles were excluded if they were not written in English or if they were case reports, reviews, conference abstracts, letters, or articles concerning the withdrawal of medications. If an article used the same population and exposure for the second time, only 1 article was included giving priority to the one with adjusted analysis and a bigger sample size when both provided an adjusted analysis. Articles that used unclear medication definitions were excluded. Data Extraction A standardized form was designed to extract information systematically. Data were extracted by 2 reviewers (L.S. and M.V. or A.W. or K.P.) independently. In case of disagreement concerning the data

extraction, a third author (N.V.) was consulted. The following information was collected: study design; number and mean age of participants; setting; definition of falls; medication classification; type and method of medication and fall ascertainment; the effect of medication use duration and/or medication dose on fall risk; the outcome [odds rations (ORs), relative risk, incidence rate ratio, hazard ratios, 95% confidence interval, P values] and possible confounders. Definitions of a fall by Prevention of Falls Network Europe, Tinetti and Kellog group were used to compare with the definition in the articles.16e18 Quality Assessment The quality of the included studies was assessed by same reviewers (L.S. and M.V., or A.W., or K.P.). In case of disagreement concerning the quality assessment, a third author (N.V.) was consulted. The Newcastle Ottawa Scale for observational studies and the Cochrane tool for randomized controlled trials were used for quality assessment.19,20 The Newcastle Ottawa Scale was slightly adjusted to fit the review topic and optimize quality assessment of falls (Appendix 2). A score of 7 or higher was considered high quality, a score of 4e6 intermediate quality and a score of 0e3 low quality in case of observational studies. Data Synthesis and Analysis The included studies were categorized per medication class using the ATC classification system.15 Study characteristics and outcomes of all included studies can be found in tables (Appendix 3, Supplementary Tables S1eS6) and a descriptive approach was used to report the study results. In case of anxiolytics, sedatives, and hypnotics, the descriptive approach was used only for the studies using ATC classification. Meta-analyses were performed when there were 3 or more high or intermediate quality studies that provided OR. Adjusted and unadjusted estimates were pooled separately. The random-effects generic inverse variance method was used. Assessing heterogeneity between studies was done by using I2 statistics and by judging the overlap of confidence intervals of the studies. In accordance with the Cochrane handbook, an I2 value of 30%e60% represented moderate heterogeneity, 50%e90% substantial heterogeneity, and 75%e100% considerable heterogeneity.21 In case of substantial heterogeneity (>50%), the potential sources of heterogeneity (setting, fall type) were explored through subgroup analyses. The software Review Manager was used for all statistical tests (Review Manager, RevMan v 5, The Cochrane Collaboration, 2014; The Nordic Cochrane Center, Copenhagen, Denmark). Short-acting and long-acting benzodiazepines were subjected to meta-analyses. We used the definitions used in included studies. The definition of short-acting benzodiazepines in the studies was half-life shorter than 24 or 20 hours, or the exposure term included only benzodiazepines with a half-life shorter than 24 hours. The definition of long-acting benzodiazepines in the studies was half-life longer than 20 or 24 hours, or the exposure term included only benzodiazepines with a half-life longer than 24 hours. In addition, we performed metaanalyses adding the studies without the definition of short-or longacting benzodiazepines. Results The initial search provided 8927 articles after the removal of duplicates. Of these articles, 1156 full texts were assessed, and 248 articles were included in the qualitative synthesis (Figure 1). Supplementary Tables S1-S6 show the characteristics of included studies. One hundred thirty-eight studies were cohort studies, 57 were cross-sectional, 43 were case-control, 7 were randomized controlled trials, and 3 were case-crossover studies. Of the included studies, 101

L.J. Seppala et al. / JAMDA 19 (2018) 371.e11e371.e17

Records identified through database searching Medline (n = 5521), Embase(n = 5490) and PsycINFO (n = 517)

371.e13

Additional records identified through other sources (n = 48)

Records after duplicates removed (n = 8927) Records excluded (n = 7771) • •

Records screened (n = 8927)

• •

Full-text articles assessed for eligibility (n = 1156)

Full-text articles excluded, with reasons (n = 908) •

Studies included in qualitative synthesis (n = 248)

Not related to falls and/or medications Editorials, reviews, casereports Foreign language Did not meet one or more inclusion criteria

• • • • • • • • • •

Did not look at the association between medication of interest and falling or unclear psychotropic medication n = 361 Wrong publication type n = 259 Wrong population n = 117 Wrong study design n = 44 No access n = 39 Wrong outcome n = 31 Overlapping population n = 29 Duplicates n = 10 Foreign language n = 9 Insufficient data n = 8 Untrustworthy data n = 1

Fig. 1. Flow chart of study screening and inclusion.

were set in the community, 69 in long-term care institutions and 32 in hospital. The remaining 46 studies were conducted in other settings or included patients from various settings. Medication ascertainment at the time of fall for example by using pharmacy dispensing data was done in 14% of studies. Fifteen percent of observational studies used the ATC classification system for classifying medication, and 73 percent of observational studies did not mention how classification was done. In the 7 RCTs, falls were collected as adverse events. However, 3 RCTs did not give a definition for falls, and 5 studies were unclear about the method of fall ascertainment. A summary table of the quality assessment of included RCTs can be found in Appendix 3. Antipsychotics Seventy-five studies investigated the use of antipsychotics as a group (Supplementary Table S1) as a risk factor for falls. Fifty-two studies were scored as high or intermediate quality (52 of 75). Twenty-four studies22e45 showed an association with increased risk of falls (24 of 75). Three studies46e48 showed a protective association (3 of 75). Sixteen studies were included in the meta-analysis of antipsychotics using adjusted data, which resulted in a pooled OR of 1.54 (1.28e1.85) with substantial heterogeneity, I2 ¼ 67%. Table 1 shows the results of subgroup analysis. Thirty-three studies were included in

the meta-analysis using unadjusted data, which resulted in a pooled OR of 1.48 (1.23e1.79) with substantial to considerable heterogeneity, I2 ¼ 82%. Five studies25,33,42,49,50 investigated atypical antipsychotics as a risk factor for falls. Four studies were scored high or intermediate quality (4 of 5). Five studies found a positive association (5 of 5). Also, 5 studies investigated the use of typical antipsychotics as an exposure. All studies were intermediate to low quality. Two studies25,33 found a significant positive association (2 of 5). Eight studies investigated risperidone as an exposure. One intermediate quality study49 found a positive multivariably adjusted association (1 of 8). Results of the other individual antipsychotics are provided in Appendix 3. Antidepressants A total amount of 107 studies investigated antidepressant use (Supplementary Table S2) as an exposure for falls. Sixty-seven of those studies were scored as high or intermediate quality (67 of 107). Fortyeight studies23,25,26,28,30,40,42e44,50e88 showed an association with increased risk of falls (48 of 107). Twenty-two studies were included in the meta-analysis of antidepressants using adjusted data, which resulted in a pooled OR of 1.57 (1.43e1.74) with substantial to considerable heterogeneity I2 ¼ 76%. Table 1 shows the results of

371.e14

L.J. Seppala et al. / JAMDA 19 (2018) 371.e11e371.e17

Table 1 Summary of Meta-Analysis of Adjusted Data and Subgroup Analysis Antipsychotics

All studies Outcome Any fall Recurrent fall Injurious fall Population Community Long term care Hospital Other

Antidepressants

Benzodiazepines

No.

OR (95% Cl)

I2

No.

OR (95% Cl)

I2

No.

OR (95% Cl)

I2

16*

1.54 (1.28e1.85)

67%

22

1.57 (1.431.74)

76%

14

1.42 (1.22e1.65)

67 %

11 5* 1

1.43 (1.15e1.77) 1.70 (1.21e2.38) 1.66 (0.17e16.21)

54% 69% N/A

14 6 5*

1.35 (1.28e1.42) 1.90 (1.42e2.54) 1.72 (1.51e1.96)

0% 52% 72%

12 3 1*

1.38 (1.17e1.63) 1.45 (1.20e1.76) 1.70 (1.03e2.81)

66% 0% 67%

2.30 1.18 1.57 1.82

0% 88% 67% 86%

5 11 2 4*

1.48 1.46 1.57 1.75

63% 33% 76% 73%

6* 3 4 1

4 6 4 2*

(1.24e4.26) (0.97e1.43) (1.01e2.43) (1.10e3.00)

(1.24e1.77) (1.26e1.69) (1.43e1.74) (1.54e1.99)

1.40 1.11 1.69 1.93

(1.18e1.66) (0.84e1.47) (1.06e2.68) (1.24e1.65)

36% 0% 84% N/A

CI, confidence interval. *Study calculated only once in number of studies if results of men and women or different age groups separately reported. Both results pooled.

subgroup analysis. Forty-eight studies were included in the metaanalysis using unadjusted data, which resulted in a pooled OR of 1.69 (1.52e1.88) with substantial to considerable heterogeneity, I2 ¼ 76%. Twenty-three studies investigated selective serotonin reuptake inhibitors (SSRI) as an exposure. Fifteen studies25,30,42,58,65,74,75,79,89e94 found a positive association (15 of 23). Four studies were included in the meta-analysis of SSRI using adjusted data, which resulted in a pooled OR of 2.02 (1.85e2.20) with low heterogeneity, I2 ¼ 0%. Five studies were included in the meta-analysis of SSRI using unadjusted data, which resulted in a pooled OR of 1.98 (1.87e2.10) with low heterogeneity, I2 ¼ 0%. Twenty-four studies investigated tricyclic antidepressant (TCA) use as an exposure. Nine studies25,75,91,93e98 showed an association with increased risk of falls (9 of 24). Five studies were included in the meta-analysis of TCAs using adjusted data, which resulted in a pooled OR of 1.41 (1.07e1.86) with substantial heterogeneity, I2 ¼ 62%. Five studies were included in the meta-analysis of TCAs using unadjusted data, which resulted in a pooled OR of 1.95 (1.79e2.13) with low heterogeneity, I2 ¼ 0%. Results of the other antidepressant are provided in Appendix 3. In addition, results of the effect of the duration of the antidepressant therapy and dosage are provided in Appendix 3.

Thirteen studies investigated short-acting benzodiazepines as an exposure. Three studies33,50,72 showed an association with increased risk of falls (3 of 13). Four studies were included in the meta-analysis of short-acting benzodiazepines using adjusted data, which resulted in a pooled OR of 1.27 (1.04e1.56) with low heterogeneity I2 ¼ 0 %. When two studies without a definition of short-acting benzodiazepines were added in a meta-analysis a pooled OR of 1.27 (1.09e1.48), I2 ¼ 0% was found. Four studies were included in the meta-analysis of short-acting benzodiazepines using unadjusted data, which resulted in a pooled OR of 1.42 (1.22e1.65) with low heterogeneity, I2 ¼ 0 %. For individual benzodiazepines diazepam was most often investigated. Six studies assessed it as an exposure. Three studies38,113,114 showed a positive association (3 of 6). Results of the other individual benzodiazepines are provided in Appendix 3. Results of the effect of the duration of the benzodiazepine therapy and dosage are provided in Appendix 3.

Benzodiazepines

Results of both pharmacologic groups of anxiolytics, and hypnotics and sedatives and psychotropic medication are described in Appendix 3.

Sixty-seven studies investigated the use of benzodiazepines (Appendix, Supplementary Table S3) as a group as an exposure. Fortyfour studies were scored high or intermediate quality (44 of 67). Twenty-five studies30,33e36,38,57e59,85,95,98e111 showed an association with increased risk of falls (25 of 67). Fourteen studies were included in the meta-analysis of benzodiazepines using adjusted data, which resulted in a pooled OR of 1.42 (1.22e1.65) with substantial heterogeneity, I2 ¼ 67%. Table 1 shows the results of subgroup analysis. Twenty-six studies were included in a meta-analysis using unadjusted data, which resulted in a pooled OR of 1.73 (1.49e2.01) with substantial heterogeneity, I2 ¼ 64%. Seventeen studies assessed long-acting benzodiazepines as an exposure. Eight studies33,50,58,70,72,108,111,112 showed an association with increased risk of falls (8 of 17). Four studies were included in the meta-analysis of long-acting benzodiazepines using adjusted data, which resulted in a pooled OR of 1.81 (1.05e3.16) with substantial heterogeneity I2 ¼ 56 %. When 2 studies without a definition of longacting benzodiazepines were added in a meta-analysis a pooled OR 1.63 (1.20e2.23), I2 ¼ 32% was found. Four studies were included in the meta-analysis of long-acting benzodiazepines using unadjusted data, which resulted in a pooled OR of 1.54 (1.28e1.85) with low heterogeneity, I2 ¼ 0 %.

Benzodiazepine-Related Drugs Six studies assessed benzodiazepine-related drugs as a group as an exposure. One study115 reported an association (1 of 6). Anxiolytics, Hypnotics, and Sedatives and Psychotropic Medication

Discussion This systematic review comprehensively included 248 articles in a qualitative synthesis that evaluated the association between psychotropic medication use and the risk of falling in older adults. Seven medication classes (antipsychotics, antidepressants, TCAs, SSRIs, benzodiazepines, short-acting benzodiazepines, long-acting benzodiazepines) were associated with increased risk of falls in meta-analyses of adjusted data. In the qualitative synthesis, we observed inconsistent associations for antipsychotics (24 of 75), antidepressants (48 of 107), SSRIs (15 of 23), TCAs (9 of 24), benzodiazepines (25 of 67), long-acting benzodiazepines (8 of 17), short-acting benzodiazepines (3 of 13), hypnotics/sedatives (2 of 6), and anxiolytics (2 of 6). Initiation of antidepressants and benzodiazepines appear to increase fall-risk. Higher dosages of psychotropics also appear to result in a higher fall-risk. We found a significantly increased fall risk in meta-analyses among antidepressant users as well as both subgroups of TCA and SSRI users. Our findings on antidepressants are supported by the results of earlier meta-analyses.6,7 When first launched it was generally assumed that SSRIs would be safer to use in older adults than TCAs. However, our systematic review and meta-analysis, together

L.J. Seppala et al. / JAMDA 19 (2018) 371.e11e371.e17

with earlier systematic reviews,8,9 suggest that the use of SSRIs results in a comparable fall risk. Nevertheless, it has been shown that factors contributing to falls (eg, orthostatic hypotension) are less common among SSRI users than among TCA users.116 Possibly our outcomes are the result of preferential prescribing: prescribing SSRIs instead of TCAs to frail older adults with a higher fall risk. There are very few data on the use of SNRI and other antidepressants and fall risk. Thus, whether specific groups of antidepressants are safer in terms of fall risk needs to be assessed in prospective trials. Also, benzodiazepine use was associated with an increased risk of falls. This finding is in line with the meta-analysis of adjusted data by Woolcott et al.7 Both our qualitative synthesis and meta-analysis suggest that long-acting benzodiazepines might be more fall-riskincreasing than short-acting benzodiazepines, which is in contrast with the results of Leipzig et al,6 who found no difference between the benzodiazepines. Somewhat different studies were included in the meta-analyses due to different availability of the data and different quality criteria. We were able to conduct meta-analyses using adjusted data in contrast to Leipzig et al.6 Studies on benzodiazepine related drugs are limited and therefore further research is warranted. Antipsychotic use was associated with increased risk of falls in our meta-analysis. Our findings are in line with the meta-analysis by Woolcott et al.7 There were too few studies to perform a separate meta-analysis of typical or atypical antipsychotics. Our findings in the qualitative synthesis do not point toward a difference in fall risk between atypical and typical antipsychotics. However, more research is needed to evaluate whether typical and atypical antipsychotics differ concerning fall-risk-increasing properties.

371.e15

has been shown to be effective in lowering the fall risk.119 However, deprescribing is challenging with a high risk of noncompliance because of beneficial effects of medications and beliefs of patients and prescribers. In the future, falls as adverse events should be collected in clinical trials of new medications. To contribute better to current understanding, observational studies should use a precise definition and classification of their target medication. Medication classification systems, preferably ATC, are essential to enable valid comparisons between studies. Furthermore, pharmacologic subgroups and individual medications need more investigation (eg, benzodiazepine related drugs and antidepressants beyond TCAs and SSRIs). To accomplish safer prescribing, more studies should investigate the effect of dosage and duration of medication use in relation to fall-risk. Conclusions This systematic review and meta-analysis indicates consistent associations between use of antipsychotics, antidepressants, and benzodiazepines and an increased risk of falls in older adults. Because of limited data, evaluation of pharmacologic subgroups and chemical substances was limited to qualitative analysis. To contribute better to current findings henceforward, observational studies should use a clear definition of their target medication and use ATC classification. In addition, studies should investigate the effect of dosage and duration of medication use in relation to fall-risk, as well as pharmacologic subgroups. To raise the level of evidence, falls as adverse events should be included in clinical trials of new medication.

Strengths and Limitations Acknowledgments Our systematic review provides an extensive overview of the literature available on different pharmacologic groups, subgroups, and chemical substances belonging to psychotropics and fall risk. Moreover, we were able to perform several meta-analyses using adjusted data. To our knowledge, TCAs, SSRIs, and long- and short-acting benzodiazepines were subjected to meta-analysis using adjusted data for the first time. Several databases were used to perform a broad search. We might have missed some relevant articles by including only English language articles. Also, the results of cohorts of participants from age 50 years and over was beyond the scope of this review as well as evaluating the relationship between medications and syncope, though there is a considerable overlap between syncope and falls.117 By reporting pooled ORs of adjusted and unadjusted data, we present all available evidence on this topic. Adjusted ORs account for potential confounding and reporting unadjusted data as well prevents overestimation of the association. Substantial heterogeneity was apparent in most of the meta-analyses and did not mostly disappear after stratification for setting and population. The literature has significant limitations. The majority of observational studies collected data on medication only at baseline, which makes it likely that medication use by participants changed during follow-up.118 Furthermore, the validity of fall ascertainment is another major limitation of the literature. In addition, polypharmacy among older people is high. However, even if the studies included other FRIDs in their multivariate adjusted analysis, possible drug-drug interactions were not studied. Moreover, these observational studies are prone to confounding by indication. However, many of the studies are adjusted for medical disorders related to psychotropic medication use. Future Implications and Clinical Perspectives Psychotropics are established risk factors in fall-prevention guidelines, and this systematic review and meta-analysis supports their status as FRIDs. In addition, psychotropic medication withdrawal

The authors would like to acknowledge the contribution of EUGMS task and finish group fall-risk-increasing drugs. The group consists following members: Hubert Blain (Montpellier University Hospital), Jean Bousquet (Montpellier University Hospital), Gösta Bucht (Umeå University), Maria Angeles Caballero-Mora (Hospital Universitario De Getafe, Madrid),Tischa van der Cammen (Delft University of Technology), Patrik Eklund (Umeå University), Marielle Emmelot-Vonk (University Medical Centrum, Utrecht), Yngve Gustafson (Umeå University), Sirpa Hartikainen (University of Eastern Finland), Rose Anne Kenny (Trinity College Dublin), Lucie Laflamme (Karolinska Institutet), Francesco Landi (Catholic University of Rome), Tahir Masud (Nottingham University Hospitals NHS Trust), Irene O’Byrne-Maguire (State Claims Agency, Ireland), Mirko Petrovic (Ghent University), Leocadio Rodriguez (Hospital Universitario De Getafe, Madrid), Lotta Seppälä (Academic Medical Centre Amsterdam), Olle Svensson (Umeå  ska (Jagiellonian University Medical University), Katarzyna Szczerbin College), Heinrich Thaler (Trauma Hospital Meidling, Vienna), and Nathalie van der Velde (Academic Medical Center Amsterdam). Supplementary Data Supplementary data related to this article can be found online at https://doi.org/10.1016/j.jamda.2017.12.098. References 1. EuroSafe. Injuries in the European Union, Summary on injury statistics 2012e2014. Amsterdam (The Netherlands): European Association for Injury Prevention and Safety Promotion (EuroSafe); 2016. 2. Centers for Disease Control and Prevention. Home and Recreational Safety. Important facts about fallsAvailable at: http://www.cdc.gov/ homeandrecreationalsafety/falls/adultfalls.html. Accessed May 1, 2017. 3. Carroll NV, Slattum PW, Cox FM. The cost of falls among the communitydwelling elderly. J Manag Care Pharm 2005;11:307e316.

371.e16

L.J. Seppala et al. / JAMDA 19 (2018) 371.e11e371.e17

4. Stel VS, Smit JH, Pluijm SMF, et al. Consequences of falling in older men and women and risk factors for health service use and functional decline. Age Ageing 2003;33:58e65. 5. Hartholt KA, van Beeck EF, Polinder S, et al. Societal consequences of falls in the older population: Injuries, healthcare costs, and long-term reduced quality of life. J Trauma 2011;71:748e753. 6. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: A systematic review and meta-analysis: I. Psychotropic drugs. J Am Geriatr Soc 1999;47:30e39. 7. Woolcott JC, Richardson KJ, Wiens MO, et al. Meta-analysis of the impact of 9 medication classes on falls in elderly persons. Arch Intern Med 2009;169: 1952e1960. 8. Hartikainen S, Lonnroos E, Louhivuori K. Medication as a risk factor for falls: Critical systematic review. J Gerontol A Biol Sci Med Sci; 2007:62. 9. Park H, Satoh H, Miki A, et al. Medications associated with falls in older people: Systematic review of publications from a recent 5-year period. Eur J Clin Pharmacol 2015;71:1429e1440. 10. Bloch F, Thibaud M, Dugue B, et al. Psychotropic drugs and falls in the elderly people: Updated literature review and meta-analysis. J Aging Health 2011;23: 329e346. 11. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: A systematic review and meta-analysis: II. Cardiac and analgesic drugs. J Am Geriatr Soc 1999;47:40e50. 12. Boye ND, van der Velde N, de Vries OJ, et al. Effectiveness of medication withdrawal in older fallers: Results from the Improving Medication Prescribing to reduce Risk Of FALLs (IMPROveFALL) trial. Age Ageing 2017;46: 142e146. 13. Sjoberg C, Wallerstedt SM. Effects of medication reviews performed by a physician on treatment with fracture-preventing and fall-risk-increasing drugs in older adults with hip fractureeA randomized controlled study. J Am Geriatr Soc 2013;61:1464e1472. 14. van Eck NJ, Waltman L. Software survey: VOSviewer, a computer program for bibliometric mapping. Scientometrics 2010;84:523e538. 15. Methodology, WCCfDS. ATCAvailable at: http://www.whocc.no/atc/structu re_and_principles/. Accessed September 1, 2016. 16. Lamb SE, Jorstad-Stein EC, Hauer K, et al. Development of a common outcome data set for fall injury prevention trials: The Prevention of Falls Network Europe consensus. J Am Geriatr Soc 2005;53:1618e1622. 17. Kellogg International Work Group (1987). The prevention of falls in later life. A report of the Kellogg International Work Group on the Prevention of Falls by the Elderly. Dan Med Bull 1987;34:1e24. 18. Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med 1988;319:1701e1707. 19. Wells GA, Shea B, Oconnell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality if nonrandomized studies in meta-analysesAvailable at: http://www. ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed September 1, 2016. 20. Higgins JPT, Altman DG, Gøtzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343:d5928. 21. Higgins JPT, GSe. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration Available at: www.cochrane-handbook.org. Accessed March 1, 2017. 22. Abreu HC, Reiners AA, Azevedo RC, et al. Incidence and predicting factors of falls of older inpatients. Revista de Saude Publica 2015;49:37. 23. Askari M, Eslami S, Scheffer AC, et al. Different risk-increasing drugs in recurrent versus single fallers: Are recurrent fallers a distinct population? Drugs Aging 2013;30:845e851. 24. Flicker L, Mead K, MacInnis RJ, et al. Serum vitamin D and falls in older women in residential care in Australia. J Am Geriatr Soc 2003;51:1533e1538. 25. French DD, Campbell R, Spehar A, et al. Drugs and falls in communitydwelling older people: A national veterans study. Clin Therapeut 2006;28: 619e630. 26. French DD, Werner DC, Campbell RR, et al. A multivariate fall risk assessment model for VHA nursing homes using the minimum data set. J Am Med Dir Assoc 2007;8:115e122. 27. Horikawa E, Matsui T, Arai H, et al. Risk of falls in Alzheimer’s disease: A prospective study. Intern Med 2005;44:717e721. 28. Iinattiniemi S, Jokelainen J, Luukinen H. Falls risk among a very old homedwelling population. Scand J Prim Health Care 2009;27:25e30. 29. John de Carle A, Kohn R. Risk factors for falling in a psychogeriatric unit. Int J Geriatr Psychiatry 2001;16:762e767. 30. Kallin K, Gustafson Y, Sandman PO, et al. Drugs and falls in older people in geriatric care settings. Aging Clin Exp Res 2004;16:270e276. 31. Kiely DK, Kiel DP, Burrows AB, et al. Identifying nursing home residents at risk for falling. J Am Geriatr Soc 1998;46:551e555. 32. Kron M, Loy S, Sturm E, et al. Risk indicators for falls in institutionalized frail elderly. Am J Epidemiol 2003;158:645e653. 33. Landi F, Onder G, Cesari M, et al. Psychotropic medications and risk for falls among community-dwelling frail older people: An observational study. J Gerontol Ser A Biol Sci Med Sci 2005;60:622e626. 34. Liu SW, Sri-On J, Tirrell GP, et al. Serious conditions for ED elderly fall patients: A secondary analysis of the Basel Non-Specific Complaints study. Am J Emerg Med 2016;34:1394e1399. 35. Mecocci P, von Strauss E, Cherubini A, et al. Cognitive impairment is the major risk factor for development of geriatric syndromes during hospitalization: Results from the GIFA study. Dement Geriatr Cogn Disord 2005;20:262e269.

36. Moller U, Midlov P, Kristensson J, et al. Prevalence and predictors of falls and dizziness in people younger and older than 80 years of age-A longitudinal cohort study. Arch Gerontol Geriatr 2013;56:160e168. 37. Mustard CA, Mayer T. Case-control study of exposure to medication and the risk of injurious falls requiring hospitalization among nursing home residents. Am J Epidemiol 1997;145:738e745. 38. Neutel CI, Hirdes JP, Maxwell CJ, et al. New evidence on benzodiazepine use and falls: The time factor. Age Ageing 1996;25:273e278. 39. Pautex S, Herrmann FR, Zulian GB. Factors associated with falls in patients with cancer hospitalized for palliative care. J Palliat Med 2008;11:878e884. 40. Ryynanen OP, Kivela SL, Honkanen R, et al. Medications and chronic diseases as risk factors for falling injuries in the elderly. Scand J Soc Med 1993;21: 264e271. 41. Stenhagen M, Ekstrom H, Nordell E, et al. Falls in the general elderly population: A 3- and 6- year prospective study of risk factors using data from the longitudinal population study ‘good ageing in skane’. BMC Geriatr 2013;13:81. 42. Sterke CS, van Beeck EF, van der Velde N, et al. New insights: Dose-response relationship between psychotropic drugs and falls: A study in nursing home residents with dementia. J Clin Pharmacol 2012;52:947e955. 43. Van Doorn C, Gruber-Baldini AL, Zimmerman S, et al. Dementia as a risk factor for falls and fall injuries among nursing home residents. J Am Geriatr Soc 2003;51:1213e1218. 44. van Strien AM, Koek HL, van Marum RJ, et al. Psychotropic medications, including short acting benzodiazepines, strongly increase the frequency of falls in elderly. Maturitas 2013;74:357e362. 45. Yip YB, Cumming RG. The association between medications and falls in Australian nursing-home residents. Med J Aust 1994;160:14e18. 46. Brown SA, Edelstein BA. Falls among psychogeriatric patients. Clin Gerontol 2000;21:3e17. 47. Spar JE, la Rue A, Hewes C, et al. Multivariate prediction of falls in elderly inpatients. Int J Geriatr Psychiatry 1987;2:185e188. 48. Spoelstra SL, Given BA, Schutte DL, et al. Do older adults with cancer fall more often? A comparative analysis of falls in those with and without cancer. Oncol Nurs Forum 2013;40:E69e78. 49. Bozat-Emre S, Doupe M, Kozyrskyj AL, et al. Atypical antipsychotic drug use and falls among nursing home residents in Winnipeg, Canada. Int J Geriatr Psychiatry 2015;30:842e850. 50. Olazaran J, Valle D, Serra JA, et al. Psychotropic medications and falls in nursing homes: A cross-sectional study. J Am Med Dir Assoc 2013;14: 213e217. 51. Blake AJ, Morgan K, Bendall MJ, et al. Falls by elderly people at home: Prevalence and associated factors. Age Ageing 1988;17:365e372. 52. Bonnerup Vind A, Andersen HE, Pedersen KD, et al. Who will fall again? Predictors of further falls in one year following an injurious fall. Eur Geriatr Med 2011;2:145e149. 53. Boutin T, Kergoat M-J, Latour J, et al. Vision in the global evaluation of older individuals hospitalized following a fall. J Am Med Dir Assoc 2012;13:187. e115e187.e119. 54. Bruce D, Hunter M, Peters K, et al. Fear of falling is common in patients with type 2 diabetes and is associated with increased risk of falls. Age Ageing 2015; 44:687e690. 55. Callaly EL, Ni Chroinin D, Hannon N, et al. Falls and fractures 2 years after acute stroke: The North Dublin Population Stroke Study. Age Ageing 2015;44: 882e886. 56. Damian J, Pastor-Barriuso R, Valderrama-Gama E, et al. Factors associated with falls among older adults living in institutions. BMC Geriatr 2013;13:6. 57. Ebly EM, Hogan DB, Fung TS. Potential adverse outcomes of psychotropic and narcotic drug use in Canadian seniors. J Clin Epidemiol 1997;50:857e863. 58. Ensrud KE, Blackwell TL, Mangione CM, et al. Central nervous system-active medications and risk for falls in older women. J Am Geriatr Soc 2002;50. 59. Ham AC, Swart KM, Enneman AW, et al. Medication-related fall incidents in an older, ambulant population: The B-PROOF study. Drugs Aging 2014;31: 917e927. 60. Hayley AC, Williams LJ, Kennedy GA, et al. Excessive daytime sleepiness and falls among older men and women: Cross-sectional examination of a population-based sample. BMC Geriatr 2015;15:74. 61. Hien le TT, Cumming RG, Cameron ID, et al. Atypical antipsychotic medications and risk of falls in residents of aged care facilities. J Am Geriatr Soc 2005; 53:1290e1295. 62. Kallin K, Lundin-Olsson L, Jensen J, et al. Predisposing and precipitating factors for falls among older people in residential care. Public Health 2002;116: 263e271. 63. Kallin K, Jensen J, Olsson LL, et al. Why the elderly fall in residential care facilities, and suggested remedies. J Fam Pract 2004;53:41e52. 64. Kelly KD, Pickett W, Yiannakoulas N, et al. Medication use and falls in community-dwelling older persons. Age Ageing 2003;32:503e509. 65. Kerse N, Flicker L, Pfaff JJ, et al. Falls, depression and antidepressants in later life: A large primary care appraisal. PLoS One 2008;3;e2423. 66. Kerse N, Butler M, Robinson E, et al. Wearing slippers, falls and injury in residential care. Aust N Z J Public Health 2004;28:180e187. 67. Kuschel BM, Laflamme L, Moller J. The risk of fall injury in relation to commonly prescribed medications among older peopledA Swedish casecontrol study. Eur J Public Health 2015;25:527e532. 68. Lipsitz LA, Habtemariam D, Gagnon M, et al. Reexamining the effect of antihypertensive medications on falls in old age. Hypertension 2015;66:183e189.

L.J. Seppala et al. / JAMDA 19 (2018) 371.e11e371.e17 69. Liu BA, Topper AK, Reeves RA, et al. Falls among older people: Relationship to medication use and orthostatic hypotension. J Am Geriatr Soc 1995;43: 1141e1145. 70. Lord SR, Anstey KJ, Williams P, et al. Psychoactive medication use, sensorimotor function and falls in older women. Br J Clin Pharmacol 1995;39: 227e234. 71. Lord SR, March LM, Cameron ID, et al. Differing risk factors for falls in nursing home and intermediate-care residents who can and cannot stand unaided. J Am Geriatr Soc 2003;51:1645e1650. 72. Luukinen H, Koski K, Laippala P, et al. Predictors for recurrent falls among the home-dwelling elderly. Scand J Prim Health Care 1995;13:294e299. 73. Luukinen H, Koski K, Laippala P, et al. Risk factors for recurrent falls in the elderly in long-term institutional care. Public Health 1995;109:57e65. 74. Marcum ZA, Perera S, Thorpe JM, et al. Antidepressant use and recurrent falls in community-dwelling older adults: Findings from the Health ABC Study. Ann Pharmacother 2016;50:525e533. 75. Masud T, Frost M, Ryg J, et al. Central nervous system medications and falls risk in men aged 60e75 years: The Study on Male Osteoporosis and Aging (SOMA). Age Ageing 2013;42:121e124. 76. Mayo NE, Korner-Bitensky N, Becker R, et al. Predicting falls among patients in a rehabilitation hospital. Am J Phys Med Rehabil 1989;68:139e146. 77. Mitchell RJ, Lord SR, Harvey LA, et al. Obesity and falls in older people: Mediating effects of disease, sedentary behavior, mood, pain and medication use. Arch Gerontol Geriatr 2015;60:52e58. 78. Moden B, Merlo J, Ohlsson H, et al. Psychotropic drugs and falling accidents among the elderly: A nested case control study in the whole population of Scania. Sweden. J Epidemiol Community Health 2010;64:440e446. 79. Quach L, Yang FM, Berry SD, et al. Depression, antidepressants, and falls among community-dwelling elderly people: The MOBILIZE Boston study. J Gerontol Ser A Biol Sci Med Sci 2013;68:1575e1581. 80. Rosendahl E, Lundin-Olsson L, Kallin K, et al. Prediction of falls among older people in residential care facilities by the Downton index. Aging Clin Exp Res 2003;15:142e147. 81. Schonwetter RS, Kim S, Kirby J, et al. Etiology of falls among cognitively intact hospice patients. J Palliat Med 2010;13:1353e1363. 82. Stalenhoef PA, Diederiks JP, Knottnerus JA, et al. The construction of a patient record-based risk model for recurrent falls among elderly people living in the community. Fam Pract 2000;17:490e496. 83. Stone KL, Ancoli-Israel S, Blackwell T, et al. Actigraphy-measured sleep characteristics and risk of falls in older women. Arch Intern Med 2008;168: 1768e1775. 84. Svensson ML, Rundgren A, Landahl S. Falls in 84- to 85-year-old people living at home. Accid Anal Prev 1992;24:527e537. 85. Titler MG, Shever LL, Kanak MF, et al. Factors associated with falls during hospitalization in an older adult population. Res Theory Nurs Pract 2011;25: 127e148. 86. Verhoeven V, Lopez Hartmann M, Wens J, et al. Happy pills in nursing homes in Belgium: A cohort study to determine prescribing patterns and relation to fall risk. J Clin Gerontol Geriatr 2014;5:53e57. 87. von Heideken Wagert P, Gustafson Y, Kallin K, et al. Falls in very old people: The population-based Umea 85þ study in Sweden. Arch Gerontol Geriatr 2009;49:390e396. 88. Whitney J, Close JC, Lord SR, et al. Identification of high risk fallers among older people living in residential care facilities: A simple screen based on easily collectable measures. Arch Gerontol Geriatr 2012;55:690e695. 89. Arfken CL, Wilson JG, Aronson SM. Retrospective review of selective serotonin reuptake inhibitors and falling in older nursing home residents. Int Psychogeriatr 2001;13:85e91. 90. Carrière I, Farré A, Norton J, et al. Patterns of selective serotonin reuptake inhibitor use and risk of falls and fractures in community-dwelling elderly people: The Three-City cohort. Osteopor Int 2016;27:3187e3195. 91. Coupland C, Dhiman P, Morriss R, et al. Antidepressant use and risk of adverse outcomes in older people: Population based cohort study. BMJ 2011;343: d4551. 92. Fisher AA, Davis MW, McLean AJ, et al. Epidemiology of falls in elderly semiindependent residents in residential care. Aust J Ageing 2005;24:98e102. 93. Gribbin J, Hubbard R, Gladman J, et al. Serotonin-norepinephrine reuptake inhibitor antidepressants and the risk of falls in older people: Case-control

94. 95.

96.

97. 98. 99.

100.

101.

102.

103. 104.

105. 106.

107. 108. 109. 110. 111.

112.

113.

114. 115.

116. 117. 118.

119.

371.e17

and case-series analysis of a large UK primary care database. Drugs Aging 2011;28:895e902. Thapa PB, Gideon P, Cost TW, et al. Antidepressants and the risk of falls among nursing home residents. N Engl J Med 1998;339:875e882. Thapa PB, Gideon P, Fought RL, et al. Psychotropic drugs and risk of recurrent falls in ambulatory nursing home residents. Am J Epidemiol 1995;142: 202e211. Mahoney JE, Palta M, Johnson J, et al. Temporal association between hospitalization and rate of falls after discharge. Arch Intern Med 2000;160: 2788e2795. Lewis CL, Moutoux M, Slaughter M, et al. Characteristics of individuals who fell while receiving home health services. Phys Ther 2004;84:23e32. Chiu MH, Lee HD, Hwang HF, et al. Medication use and fall-risk assessment for falls in an acute care hospital. Geriatr Gerontol Int 2015;15:856e863. Aisen PS, Deluca T, Lawlor BA. Falls among geropsychiatry inpatients are associated with PRN medications for agitation. Int J Geriatr Psychiatry 1992;7: 709e712. Chang CM, Chen MJ, Tsai CY, et al. Medical conditions and medications as risk factors of falls in the inpatient older people: A case-control study. Int J Geriatr Psychiatry 2011;26:602e607. de Rekeneire N, Visser M, Peila R, et al. Is a fall just a fall: Correlates of falling in healthy older persons. The Health, Aging and Body Composition Study. J Am Geriatr Soc 2003;51:841e846. de Vries OJ, Peeters G, Elders P, et al. The elimination half-life of benzodiazepines and fall risk: Two prospective observational studies. Age Ageing 2013; 42:764e770. Frels C, Williams P, Narayanan S, et al. Iatrogenic causes of falls in hospitalised elderly patients: A case-control study. Postgrad Med J 2002;78:487e489. Gales BJ, Menard SM. Relationship between the administration of selected medications and falls in hospitalized elderly patients. Ann Pharmacother 1995;29:354e358. Hung CY, Wu TJ, Wang KY, et al. Falls and atrial fibrillation in elderly patients. Acta Cardiologica Sinica 2013;29:436e443. Panneman MJ, Goettsch WG, Kramarz P, et al. The costs of benzodiazepineassociated hospital-treated fall Injuries in the EU: A Pharmo study. Drugs Aging 2003;20:833e839. Pellfolk T, Gustafsson T, Gustafson Y, et al. Risk factors for falls among residents with dementia living in group dwellings. Int Psychogeriatr 2009;21:187e194. Ray WA, Thapa PB, Gideon P. Benzodiazepines and the risk of falls in nursing home residents. J Am Geriatr Soc 2000;48:682e685. Lamis RL, Kramer JS, Hale LS, et al. Fall risk associated with inpatient medications. Am J Health Syst Pharm 2012;69:1888e1894. Pariente A, Dartigues JF, Benichou J, et al. Benzodiazepines and injurious falls in community dwelling elders. Drugs Aging 2008;25:61e70. Tromp AM, Pluijm SM, Smit JH, et al. Fall-risk screening test: A prospective study on predictors for falls in community-dwelling elderly. J Clin Epidemiol 2001;54:837e844. Berdot S, Bertrand M, Dartigues J-F, et al. Inappropriate medication use and risk of falls e A prospective study in a large community-dwelling elderly cohort. BMC Geriatr 2009;9:30. Ballokova A, Peel NM, Fialova D, et al. Use of benzodiazepines and association with falls in older people admitted to hospital: A prospective cohort study. Drugs Aging 2014;31:299e310. Cumming RG, Miller JP, Kelsey JL, et al. Medications and multiple falls in elderly people: The St Louis OASIS study. Age Ageing 1991;20:455e461. Rhalimi M, Helou R, Jaecker P. Medication use and increased risk of falls in hospitalized elderly patients: A retrospective, case-control study. Drugs Aging 2009;26:847e852. Darowski A, Chambers SA, Chambers DJ. Antidepressants and falls in the elderly. Drugs Aging 2009;26:381e394. Shaw FE, Kenny RA. The overlap between syncope and falls in the elderly. Postgrad Med J 1997;73:635e639. Ray WA, Thapa PB, Gideon P. Misclassification of current benzodiazepine exposure by use of a single baseline measurement and its effects upon studies of injuries. Pharmacoepidemiol Drug Saf 2002;11:663e669. Campbell AJ, Robertson MC, Gardner MM, et al. Psychotropic medication withdrawal and a home-based exercise program to prevent falls: A randomized, controlled trial. J Am Geriatr Soc 1999;47:850e853.