TY - JOUR
AU - Hornberger,, Michael
AB - Abstract Background and Objectives People with dementia are at risk of exiting premises unsupervised, eloping, or getting lost, potentially leading to harmful or distressing consequences. This review aimed to estimate the effectiveness of interventions for preventing people with dementia from exiting or getting lost. Research Design and Methods A systematic review of English sources was undertaken. Health care (EMBASE, BNI, Medline, PubMed, CINAHL, PsycINFO, AMED, HTA, CENTRAL) and gray literature (OpenGrey) databases were searched using prespecified search terms. Additional studies were identified by hand-searching bibliographies of relevant reviews and included studies. Wide inclusion criteria were set to capture a range of intervention types. Data extraction and risk of bias assessment were completed independently by two reviewers. Methods were preregistered on PROSPERO. Results Individual and overall risk of bias was too high for statistical meta-analyses. A narrative synthesis was therefore performed. Twenty-five studies with 814 participants were included, investigating a range of nonpharmacological interventions aiming to prevent exiting, facilitate retrieval, educate participants, or a combination of these. Seventeen (68%) of the included studies had critical risks of internal bias to outcomes, providing no useful evidence for the effectiveness of their respective interventions. The remaining 8 (32%) studies had serious risks of bias. Narrative synthesis of results yielded no overall robust evidence for the effectiveness of any interventions. Discussion and Implications No evidence was found to justify the recommendation of any interventions included in this review. Future studies should focus on high-quality, controlled study designs. Alzheimer’s disease, Wandering, Missing incidents Spatial navigation symptoms (i.e., disorientation, getting lost) are core features of dementia, as the underlying brain systems are affected in the disease (Chiu et al., 2004; Coughlan, Laczó, Hort, Minihane, & Hornberger, 2018). Consequently, people with dementia are at risk of getting lost in unfamiliar and familiar environments without carers’ knowledge of their whereabouts (Yatawara et al., 2017). Reports estimate that 30%–70% of people with dementia become lost at least once during the course of the disease, often unpredictably during routine tasks and with few antecedents (Bowen, McKenzie, Steis, & Rowe, 2011; Kwok, Yuen, Ho, & Chan, 2010; McShane et al., 1998; Pai & Lee, 2016). Becoming lost can be highly distressing for people with dementia and their carers (Kwok et al., 2010). Extreme cases may result in injury or death (Woolford, Weller, & Ibrahim, 2017), the risk of which increases with age, length of time missing, and season (Bantry White & Montgomery, 2014). Informal carers may respond to the increased risk of people with dementia becoming lost by monitoring them more closely, which may result in a reduced sense of freedom of the care recipient (McShane et al., 1998). Moreover, multiple incidents of getting lost have been shown to increase the chances of informal carers institutionalizing the person with dementia (McShane et al., 1998), who may express resistance to this and a desire to stay in their own home (van der Roest et al., 2007). Lost people with dementia can also incur large costs to law enforcement and other community search and rescue services due to retrieval efforts (Bowen et al., 2011). As getting lost is a prevalent problem for people with dementia, their carers, and the wider community, there is a need to investigate effective interventions to safeguard against it. Recommendations from published literature and public health guidelines suggest a wide range of strategies and techniques to prevent people with dementia becoming lost including caregiver planning, out-of-sight door bolts, tracking devices, and more (see https://alz.org/help-support/caregiving/stages-behaviors/wandering; Bowen et al., 2011; Pai & Lee, 2016). However, the effectiveness of these interventions remains unclear without systematic evaluation. Previous reviews have reported on interventions aimed at reducing “wandering” (Hermans, Htay, & Cooley, 2007; Robinson et al., 2007). However, the term wandering is complex, with conceptual and operational definitions regarding it as a syndrome of locomotive behaviors (e.g., pacing, lapping) with possible associated outcomes such as exiting (also known as “eloping”) or getting lost (Algase, Moore, Vandeweerd, & Gavin-Dreschnack, 2007). Strategies and interventions aiming to mitigate harm or adverse consequences have often been tailored to one type of wandering-related behavior (such as pacing), but studies have often assessed their effectiveness with nonspecific outcome measures (see Robinson et al., 2007). Furthermore, interventions for reducing “behavioral and psychological symptoms of dementia (BPSD)” often include “wandering” among a list of diverse presentations. These are imprecise approaches for investigating specific wandering-related concepts, including the most potentially dangerous outcomes of wandering: the risk of people with dementia becoming lost, or leaving premises unattended (“exiting”) that may lead to them becoming lost (Rowe et al., 2012). Neither exiting nor getting lost, as specific and problematic outcomes, have been the subject of any systematic reviews of intervention studies. Therefore, evidence-based recommendations cannot currently be made for safeguarding against their associated risks. Filling this gap is important due to the prevalence and potential consequences of these behaviors, as mentioned above. A review of the evidence is also timely given that recent technological advancements have yielded a wide range of tracking and alarm devices claiming to alleviate incidents or consequences of people with dementia becoming lost (Pulido Herrera, 2016). We therefore systematically reviewed the literature to determine whether evidence does exist for the effectiveness of these or any other interventions in preventing people with dementia specifically from becoming lost, or exiting as a precursor to this. Review Question How effective are interventions which aim to prevent, reduce frequency, or decrease adverse consequences of people with dementia exiting or becoming lost? Methods Protocol and Registration Methods for this systematic review followed guidelines from the Centre for Reviews and Dissemination (2009), reporting standards from Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Liberati et al., 2009; Moher, Liberati, Tetzlaff, & Altman, 2009) and AMSTAR 2—a critical appraisal tool for systematic reviews of health care interventions (Shea et al., 2017). The review was preregistered on the International Prospective Register of Systematic Reviews (PROSPERO, CRD42018097229, https://www.crd.york.ac.uk/PROSPERO/) with details of the review question, search strategy, eligibility criteria, and methodological assessment. Details of data synthesis were not provided beforehand, except that they would follow methodology from CRD and that a meta-analysis was planned. Eligibility Criteria Primary research studies of any design were included, except for case studies due to their very low generalizability. This broad criterion was set to allow an overview of a diversity of intervention types from studies reporting on their effectiveness. However, effectiveness of interventions was assessed against high-quality designs for intervention studies. Studies with people who had a diagnosis of dementia of any age, gender, or disease severity were included. The study could have been undertaken in any care setting (i.e., hospital, care, or community). Studies looking exclusively at mild cognitive impairment and any other nondementia groups were excluded. Studies were included if they examined any intervention, treatment, or tool aimed at reducing or preventing wandering behaviors, exiting, elopement, getting lost, missing incidents, or adverse consequences of these. Studies of interventions for improving wayfinding in controlled environments were excluded because they were not directly related to exiting or getting lost. Studies without control groups were included. However, the risk of bias for uncontrolled studies was assessed against high-quality controlled trials. The main outcomes of interest were any measure of, or incidence of, exiting (including “eloping”) or getting lost (including “missing incidents”). Studies were excluded if they measured “wandering” without inclusion of the above measures, or conflated with agitation, pacing, or some other behavior. Studies could also have included assessment of the consequences of the intervention or lack thereof, such as: accidents; injuries; falls; fractures; deaths; activity in daily tasks; quality of life, anxiety, or distress of the person with dementia or their carer(s); carer burden; institutionalization. Only English records were included as resource limitations prevented translation of non-English records. Searching and Information Sources Search terms and databases are documented in Table 1. This included online databases of published and gray literature (from inception to March 2019: the date of the most recent search update), bibliographies from included studies, and bibliographies from relevant reviews and other secondary sources found from systematic searching of databases. A list of these latter papers can be found in Supplementary Table 3. Table 1. Search Strategies Databases/source(s) Search terms/details EMBASE, BNI, Medline, PubMed, CINAHL, PsycINFO, AMED, HTA dement* OR alzheimer* OR frontotemporal OR lewy OR corticobasal OR “primary progressive aphasia” OR “posterior cortical atrophy”  AND tag* OR track* OR alarm* OR device* OR technolog* OR electronic OR GPS OR restrain* OR lock* OR barrier* OR snoezelen OR aromatherapy OR music OR therap* OR manag* OR prevent* OR interven* OR treat* OR independence OR RFID OR radiofrequency OR “radio frequency” OR environment*  AND wander* OR walk* OR exit* OR elop* OR orientation OR disorientation OR navigat* OR lost OR wayfind* OR ambulat* OR “unexplained absence” OR abscond* CENTRAL, OpenGrey (SIGLE) Same as above but with the following in place of their respective truncations: Dementia; alzheimer; tag; track; alarm; device; technology; restrain; lock; barrier; therapy; manage; prevent; intervention; treat; environment; wander; walk; exit; elopement; navigation; wayfinding; ambulation; abscond. Bibliographies from included studies (n = 26) See Supplementary Table 4 for list and details of included studies. Bibliographies from reviews and secondary sources (n = 20) See Supplementary Table 3 for list of reviews and sources. Databases/source(s) Search terms/details EMBASE, BNI, Medline, PubMed, CINAHL, PsycINFO, AMED, HTA dement* OR alzheimer* OR frontotemporal OR lewy OR corticobasal OR “primary progressive aphasia” OR “posterior cortical atrophy”  AND tag* OR track* OR alarm* OR device* OR technolog* OR electronic OR GPS OR restrain* OR lock* OR barrier* OR snoezelen OR aromatherapy OR music OR therap* OR manag* OR prevent* OR interven* OR treat* OR independence OR RFID OR radiofrequency OR “radio frequency” OR environment*  AND wander* OR walk* OR exit* OR elop* OR orientation OR disorientation OR navigat* OR lost OR wayfind* OR ambulat* OR “unexplained absence” OR abscond* CENTRAL, OpenGrey (SIGLE) Same as above but with the following in place of their respective truncations: Dementia; alzheimer; tag; track; alarm; device; technology; restrain; lock; barrier; therapy; manage; prevent; intervention; treat; environment; wander; walk; exit; elopement; navigation; wayfinding; ambulation; abscond. Bibliographies from included studies (n = 26) See Supplementary Table 4 for list and details of included studies. Bibliographies from reviews and secondary sources (n = 20) See Supplementary Table 3 for list of reviews and sources. Note: * indicates truncated searched term. Open in new tab Table 1. Search Strategies Databases/source(s) Search terms/details EMBASE, BNI, Medline, PubMed, CINAHL, PsycINFO, AMED, HTA dement* OR alzheimer* OR frontotemporal OR lewy OR corticobasal OR “primary progressive aphasia” OR “posterior cortical atrophy”  AND tag* OR track* OR alarm* OR device* OR technolog* OR electronic OR GPS OR restrain* OR lock* OR barrier* OR snoezelen OR aromatherapy OR music OR therap* OR manag* OR prevent* OR interven* OR treat* OR independence OR RFID OR radiofrequency OR “radio frequency” OR environment*  AND wander* OR walk* OR exit* OR elop* OR orientation OR disorientation OR navigat* OR lost OR wayfind* OR ambulat* OR “unexplained absence” OR abscond* CENTRAL, OpenGrey (SIGLE) Same as above but with the following in place of their respective truncations: Dementia; alzheimer; tag; track; alarm; device; technology; restrain; lock; barrier; therapy; manage; prevent; intervention; treat; environment; wander; walk; exit; elopement; navigation; wayfinding; ambulation; abscond. Bibliographies from included studies (n = 26) See Supplementary Table 4 for list and details of included studies. Bibliographies from reviews and secondary sources (n = 20) See Supplementary Table 3 for list of reviews and sources. Databases/source(s) Search terms/details EMBASE, BNI, Medline, PubMed, CINAHL, PsycINFO, AMED, HTA dement* OR alzheimer* OR frontotemporal OR lewy OR corticobasal OR “primary progressive aphasia” OR “posterior cortical atrophy”  AND tag* OR track* OR alarm* OR device* OR technolog* OR electronic OR GPS OR restrain* OR lock* OR barrier* OR snoezelen OR aromatherapy OR music OR therap* OR manag* OR prevent* OR interven* OR treat* OR independence OR RFID OR radiofrequency OR “radio frequency” OR environment*  AND wander* OR walk* OR exit* OR elop* OR orientation OR disorientation OR navigat* OR lost OR wayfind* OR ambulat* OR “unexplained absence” OR abscond* CENTRAL, OpenGrey (SIGLE) Same as above but with the following in place of their respective truncations: Dementia; alzheimer; tag; track; alarm; device; technology; restrain; lock; barrier; therapy; manage; prevent; intervention; treat; environment; wander; walk; exit; elopement; navigation; wayfinding; ambulation; abscond. Bibliographies from included studies (n = 26) See Supplementary Table 4 for list and details of included studies. Bibliographies from reviews and secondary sources (n = 20) See Supplementary Table 3 for list of reviews and sources. Note: * indicates truncated searched term. Open in new tab Study Selection L. Emrich-Mills extracted all search results and performed an initial screening of titles and abstracts using the eligibility criteria above. A second screening of titles and abstracts was performed by L. Emrich-Mills and V. Puthusseryppady independently, with reasons for exclusions provided. This process was briefly piloted for consistency and refined accordingly. Potential inclusions from either L. Emrich-Mills or V. Puthusseryppady were eligible for full-text eligibility screening. Full texts were then assessed for final eligibility by L. Emrich-Mills and V. Puthusseryppady independently. This process was first piloted for consistency and refined accordingly. Disagreements during the full selection process were resolved through discussion and arbitration by M. Hornberger where necessary. Decisions on eligibility of full texts were recorded (Supplementary Spreadsheet, available for review as separate document). Authors were contacted if more information was required to assess eligibility of articles. Papers were excluded if no reply was received before the cutoff date for data extraction (March 2019). Data Collection Process Preregistered data items were included in a data extraction form. The form was piloted on two studies and refined accordingly. L. Emrich-Mills and V. Puthusseryppady extracted data independently before comparing for consistency with disagreements resolved by discussion. Study characteristics were collected alongside assessment of reporting quality, risk of internal bias, and risk of external bias (see Supplementary Spreadsheet for full data extraction and risk of bias assessment). Appraisal of Individual Studies An unblinded critical appraisal of each included study was undertaken by L. Emrich-Mills and V. Puthusseryppady in duplicate, with disagreements resolved by discussion. The assessment was split into three parts: internal validity (risk of bias), external validity (representativeness, or external selection bias), and reporting quality. Although some critical appraisal scales cover all three of these areas, an adapted combination of tools was used to emphasize risk of bias and to avoid conflating the three areas into an overall score or rating (Liberati et al., 2009). Internal validity Cochrane’s Risk Of Bias In Non-randomized Studies of Interventions tool (ROBINS-I; Sterne et al., 2016) was used as a basis for assessing internal validity. Cochrane’s revised risk of bias tool for randomized trials (RoB 2; Higgins et al., 2016) was also used for assessing risk of bias in included randomized controlled trials (RCTs). RoB 2 ratings were adapted for comparison with the ROBINS-I. For example, whereas ROBINS-I allows for risk of bias ratings from “low,” through “moderate,” to “serious” and “critical”; the RoB 2 uses “low,” “some concerns,” and “high” risks of bias. The RoB 2 “low” ratings were kept as “low”; “some concerns” was adapted to either “moderate” or “serious” risk of bias depending on details; and “high” was adapted to “serious” risk of bias in all domains except confounding, where there was potential for “critical” risk (Higgins et al., 2016). The RoB 2 tool has additional considerations for crossover trials (Higgins et al., 2016). These were used to inform risk of bias assessments of within-subject designs. For example, within-subject designs are susceptible to carryover effects of interventions and period effects from changes in study or background conditions over time. Additionally, the risk of bias due to confounding in single-group within-subject studies was partially assessed based on whether the study included just reversal (e.g., ABA) or also reintroduction (e.g., ABAB) of the intervention. The former accounts for confounding variables that may have influenced changes from A to B, but the latter is required to determine whether effects can be replicated (Cox, 2016). Single-group within-subject studies that reintroduced the intervention were deemed to have lower risk of bias due to confounding than reversal alone (Cox, 2016). Similarly, risk of bias due to confounding in these study designs was partially assessed based on potential effects of intervention ordering, which are usually controlled in crossover trials by design (Higgins et al., 2016). Finally, all within-subject designs were marked down for confounding due to period effects unless they accounted for underlying trends over time; this study design is considered inappropriate when investigating intervention effects on people who have progressive or unstable conditions such as dementia (Higgins et al., 2016). Uncontrolled before-after studies (e.g., AB) were deemed to automatically have a critical risk of bias for any intervention effect, as all study-related and background factors become confounding factors (Armstrong, Waters, & Doyle, 2011; Cox, 2016). However, these study designs were also automatically assigned a low risk of bias for many other domains because there could be no imbalance between intervention and control conditions. Supplementary Table 1 compares bias domains across tools and study designs. Studies with an overall critical risk of bias were excluded from any synthesis (Sterne et al., 2016). Risk of bias for remaining studies was taken into account in the evidence synthesis. External validity External validity was assessed based on risk of external selection bias, using the relevant subsection of the Downs and Black scale (1998). Each of the three items could be rated “yes,” “no,” or “unclear.” Studies with “no” or “unclear” for all three items were given a high risk of external selection bias. Conversely, studies that were rated yes for all three items were given a low risk. Those with a mixture of yes’s and no’s were either given a rating of moderate or high risk, depending on details. Reporting quality Reporting quality was assessed separately from internal and external validity using the relevant subsection of the Downs and Black scale (1998). This is an 8-point subscale consisting of seven questions covering clarity of descriptions of: aims/hypotheses, participants, confounders, interventions, outcome measures, withdrawals, and adverse events (see Supplementary Spreadsheet). Each study was given a total score between 1 and 8, with lower scores indicating lower reporting quality. Synthesis of Results One or more meta-analyses of intervention effects were planned, but studies were too heterogeneous in outcomes and intervention types, and risks of bias were too high. A narrative synthesis was undertaken instead, informed by published guidelines (CRD, 2009; Popay et al., 2006). This consisted of the following steps: 1. Theory development of intervention effects. 2. Tabulation and grouping of study characteristics. 3. Comparing direction and magnitude of effects of similar interventions, taking risk of bias into consideration alongside study characteristics. 4. Assessment of robustness of synthesis (overall level of evidence, critical reflection of methods of synthesis, comparison to other reviews). Results Study Selection Twenty-five studies met the eligibility criteria and were included in the review. Figure 1 details the inclusions and exclusions from the search strategy. A list of excluded full texts and reasons for exclusions can be found in the Supplementary Spreadsheet. Figure 1. Open in new tabDownload slide Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of inclusions and exclusions. Figure 1. Open in new tabDownload slide Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of inclusions and exclusions. Study Characteristics A summary of characteristics of included studies can be found in Supplementary Table 4. Study designs Figure 2B shows total included study designs. The most common design was the uncontrolled before-after study (n = 11). Two further studies (Bantry White, Montgomery, & McShane, 2010; Horvath, Harvey, & Trudeau, 2007) were classed as uncontrolled after-only (UAO, also known as “post-test-only”) designs. Five studies employed a within-subject design with reversal of the intervention (e.g., ABA). Two further studies included reintroduction of the intervention (e.g., ABAB). Two unblinded randomized controlled trials (URCTs; Rowe, Greenblum, Boltz, & Galvin, 2009; Shalek, Richeson & Buettner, 2004) and one nonrandomized controlled trial (NRCT) were included (Levy-Storms, Cherry, Lee, & Wolf, 2017). One study employed an observational cross-sectional design across groups of different intervention users (Chen & Leung, 2012). This is referred to as a controlled after-only (CAO) study to highlight comparison to other study designs. One study had an unclear design (Moore & Haley, 2014). Figure 2. Open in new tabDownload slide Summary charts from critical appraisal. (A) Total risk of bias ratings per domain and overall. (B) Overall risk of bias by study design: (C) external validity by individual question score and overall risk of bias. (D) Reporting quality total points per question: N = 0, Y = 1 except for the confounders question where partially = 1, Y = 2. Figure 2. Open in new tabDownload slide Summary charts from critical appraisal. (A) Total risk of bias ratings per domain and overall. (B) Overall risk of bias by study design: (C) external validity by individual question score and overall risk of bias. (D) Reporting quality total points per question: N = 0, Y = 1 except for the confounders question where partially = 1, Y = 2. Participants Eight-hundred and fourteen reported participants were involved in the 25 included studies (median = 20). Of these, six included carer-care recipient dyads as participants (total n = 299), one reported staff as participants (Cohen-Mansfield, Werner, Culpepper, & Barkley, 1997), and the remainder included people with dementia only as participants. Two studies reported total residents in care units without reporting the number of affected individuals (Chafetz, 1990; Sherman, 1999). Eight studies reported an unspecified diagnosis of dementia. Five did not report details on diagnosis but were included because the setting or context was dementia-specific. Twelve studies reported specific diagnoses of dementia, with totals of 232 Alzheimer’s disease, 15 vascular or multi-infarct dementia, 1 frontotemporal dementia, 3 Parkinson’s disease, 13 mixed dementia, 36 “other types of dementia” (or equivalently nonspecific), 1 “early onset,” and 1 no diagnosis. Four of these studies included only participants with Alzheimer’s disease (total n = 150). Disease severity was reported sparsely and heterogeneously (see Supplementary Spreadsheet). Interventions and theory of intervention effects Figure 3 shows a theoretical model of effects and outcomes of interventions included in this review. The model represents an as-usual pathway to exiting or becoming lost, with included interventions being linked to the section of the pathway in which they intervene. The effects of these interventions are also represented, discerned through stated and interpreted intervention actions from reading individual studies. This was used to form four superordinate groups of interventions for this review: Figure 3. Open in new tabDownload slide Model of intervention effects and actions. Green boxes represent ultimate aims of interventions. Orange boxes represent interventions. Orange arrows represent general or varying effects of educational interventions. Figure 3. Open in new tabDownload slide Model of intervention effects and actions. Green boxes represent ultimate aims of interventions. Orange boxes represent interventions. Orange arrows represent general or varying effects of educational interventions. • Most included studies investigated interventions for preventing exits from supervised locations. This is illustrated in Figure 3 with all interventions with rightward arrows toward the outcome “prevent exit.” This is the most preventative type of strategy to stopping people with dementia from becoming lost or coming to harm. • Another superordinate grouping was for interventions with rightward arrows toward the terminal outcome “retrieval; reduce negative consequences” in Figure 3. These are distinct aims from preventing exits, as they assume that the person with dementia is already outside supervised or safe premises. • Educational interventions may vary considerably but tend to target several types of strategies for the carer to employ. Therefore, they are represented in Figure 3 as affecting other interventions and intervention effects. • The final grouping of interventions was for those that used a combination of approaches for a combination of effects, referred to in this review as multicomponent and multi-aim interventions (not represented in Figure 3). These four superordinate groups were further subdivided by type of intervention, represented in Figure 3 as individual interventions within boxes. These groupings and categorizations were used to order studies in Supplementary Table 4 and the results and synthesis of intervention effects. Outcome measures Primary Twenty-four studies (92%) included measures consistent with eligibility for primary outcome measures for this review: measures of exiting, eloping, or getting lost. Substantial diversity in reported outcome measurements was found. The most commonly reported type of metric was an absolute measure (e.g., total, mean, frequency) of exits, exit attempts, or door approaches (n = 11). For other exit prevention interventions, authors reported proportional metrics of exit attempts (Hewawasam, 1996; Mayer & Darby, 1991), duration of exit-seeking behavior (Cohen-Mansfield & Werner, 1998), exits plus injuries (Rowe et al., 2009), or the Eloping Behavior subscale of the Algase Wandering Scale (Shalek, Richeson & Buettner, 2004; Traynor, Veerhuis, Johnson, Hazelton, & Gopalan, 2018). Two studies measured an outcome directly related to getting lost: Lau, Chan, and Szeto (2018) measured missing incidents per year and average searching time before and after their intervention. Levy-Storms and colleagues (2017) included a Likert-style self-report scale of frequency of getting lost. Secondary Two studies with interventions aiming to facilitate retrieval measured caregiver feelings and views only (Bantry White et al., 2010; Pot, Willemse, & Horjus, 2012). Settings Fifteen studies were undertaken exclusively in institutional settings, 12 of which were set in one specific nursing home, care unit, or other inpatient facility. Two studies were across two different care units (Cohen-Mansfield & Werner, 1998; Traynor et al., 2018); one was across 21 different nursing homes (Cohen-Mansfield et al., 1997). Seven studies were undertaken in domestic or community settings; two studies (Bass, Rowe, & Moreno, 2007; Roberts, 1999) were across a mixture of settings. Generally, interventions designed to facilitate retrieval were in community settings, as were multi-aim and multicomponent interventions. All studies with interventions designed to prevent exiting (n = 18) were in institutional settings except for two in community settings (Moore & Haley, 2014; Rowe et al., 2009). Publication status Twenty-three studies were published in peer-reviewed journals; two studies (Bass et al., 2007; Horvath et al., 2007) were published in the same book. Gray literature included one unpublished Masters dissertation (Hamilton, 1993) and one online registered trial report (Moore & Haley, 2014). Appraisal of Included Studies A summary table of critical appraisal results by study can be found in Figure 4. Reasons for risk of bias judgments and individual reporting quality scores can be found in the Supplementary Spreadsheet. Figure 4. Open in new tabDownload slide Summary of critical appraisal ratings of individual studies. Figure 4. Open in new tabDownload slide Summary of critical appraisal ratings of individual studies. Internal validity (risk of bias) Seventeen studies (68%) were judged to have an overall critical risk of bias, eight (32%) a serious risk of bias, and one (4%) did not have enough information to inform a risk of bias judgment (Figure 2A; percentages total over 100% because one study had different risk of bias across intervention effects [Namazi, Rosner, & Calkins, 1989]). No studies had an overall rating of moderate or low risk of bias. Thirteen studies had a critical risk of confounding due to an uncontrolled design. Twenty-two studies had designs that accounted for internal selection bias (UBA, RCT, WS). All study designs with independent or within-subject control conditions (n = 12) had a serious or critical risk of bias due to deviation from intended interventions. This was mainly due to the absence of blinding of participants and staff to intervention status without mitigation against risk of imbalanced co-interventions, as well as risks of carryover effects for within-subject designs. Additionally, there was no blinding of outcome assessors to intervention status across studies. This affected risk of bias differently depending on details of the outcome measurements. No studies had an available preregistered analysis and could not be rated low risk for selection of reported results (Higgins et al., 2016; Sterne et al., 2016). External validity Nineteen (76%) studies were rated high for risk of external selection bias, mainly due to small samples in singular institutional settings. Five (20%) studies were rated moderate for risk of external selection bias. All five of these were deemed to have representative settings, facilities, and staff, but a lack of clarity on the representativeness of included participants. One study was rated potentially low for external selection bias due to explicit comments on the representativeness of the sample for the area based on demographic characteristics (Horvath et al., 2007). However, this study had an overall critical risk of bias for internal validity, limiting its external validity. Reporting quality Reporting quality across items is represented in Figure 2D. The item most commonly marked down on was clarity of distribution of principal confounders. Reporting sufficiency of other items varied. Full details can be found in the Supplementary Spreadsheet. Results and Synthesis of Intervention Effects Studies with an overall critical risk of bias were not included in any analysis to avoid overemphasizing results that provide no useful evidence. For the remaining studies (n = 8), a statistical meta-analysis was avoided for the following reasons: • The risk of bias in included studies was too high (all serious); • The reported study outcome measures were too diverse; • Interventions were too diverse. Results of eligible studies were narratively synthesized, examining intervention effects and relevant features within and across studies where possible and appropriate. Results and syntheses are categorized by intervention type, informed by the theoretical model of intervention action (Figure 3). Graphical representations of effects across studies or outcomes were not attempted to avoid providing misleading results. Interventions Aiming to Prevent Exiting Visual barriers for preventing exiting Ten studies tested the effectiveness of interventions that modified the environment to disguise the exit door or deter people with dementia from interacting with it. These barriers were purely visual; they made no physical barrier to opening the door. Grid patterns The most common intervention overall was the use of grid patterns on or near the exit door. Four of the six studies (Hamilton, 1993; Hussian & Brown, 1987; Namazi et al., 1989; Roberts, 1999) were judged as having overall critical risks of bias and therefore provided no useful evidence for effectiveness. The results of the two remaining studies were inconsistent, with one study finding no intervention effect (Chafetz, 1990), and the other a large effect (Hewawasam, 1996; reviewers’ analysis, see Supplementary Spreadsheet). Although Hewawasam’s (1996) study reduced risk of confounding through a stronger study design than Chafetz (1990), both reports had serious risks of bias in multiple domains (Figure 4). Therefore, one result cannot clearly take precedence over the other and the effectiveness of grid patterns on exit-seeking behavior is inconclusive. Covering the exit door Five studies examining the effect of covering the entire exit door or features of it had critical risks of bias (Dickinson et al., 1995; Dickinson & McLain-Kark, 1998; Kincaid & Peacock, 2003; Namazi et al., 1989; Roberts, 1999), with one providing unclear results (Moore & Haley, 2014). Results from one study (Namazi et al., 1989) suggest covering the exit doorknob may reduce exiting compared to no-intervention baseline conditions. However, no statistical analysis was attempted, and no measures of spread were provided for reviewers to perform calculations themselves. Therefore, this study does not provide sufficient evidence for the effectiveness of this strategy. Mirror on the exit door The effectiveness of a mirror on the exit door for reducing exiting is unclear due to problematic studies: One study reported a reduction in percentage of exit-door approaches resulting in door contacts but did not provide a base rate of absolute exit-door approaches, obscuring results necessary to determine effectiveness (Mayer & Darby, 1991. One other study reported using a mirror on the exit door but had a critical risk of bias and very brief reporting of results (Roberts, 1999). Indoor alarms and tracking systems for preventing exiting Three papers reported using alarm and tracking/tagging systems to alert caregivers to exit attempts by people with dementia. Two of these were too problematic in both internal validity and reporting quality to provide any useful evidence (Altus, Mathews, Xaverius, Engelman, & Nolan, 2000; Connell & Sanford, 1998). Rowe and colleagues (2009) reported an URCT of the effect of a home security system and bed occupancy sensor on the likelihood of unattended exits and home injuries (“adverse events”). In their primary analyses, they found no significant intervention effect, although the study had serious risks of bias in all domains of the RoB 2 tool. They did find an intervention effect when analyzing based on intervention fidelity, but this is open to additional bias caused by per-protocol analysis (Ranganathan, Pramesh, & Aggarwal, 2016). Effectiveness of this intervention is therefore inconclusive. Distracting or occupying the person with dementia to prevent exiting Five studies evaluated interventions to prevent exiting through occupying or distracting people with dementia. Two studies (Connell & Sanford, 1998; Traynor et al., 2018) had an overall critical risk of bias. The interventions in the remaining three studies were substantially different from each other and therefore their effects were not synthesized. Corridor scenes Cohen-Mansfield and Werner (1998) examined the effect of two corridor “scenes” on the duration of a number of behaviors including exit-seeking. They found no significant difference in exit-seeking duration between conditions, but there was a serious risk of bias in multiple domains mainly due to possible contamination effects between the two corridors and risk of selective reporting of results. Service dog Results from Sherman (1999) suggest the effectiveness of a trained service dog for reducing exit attempts in an Alzheimer’s special care unit, despite risks of bias in measurement of outcomes and reporting of results. Air mat therapy Shalek, Richeson, and Buettner (2004) reported a significant reduction in the Eloping Behavior subscale of the Algase Wandering Scale before versus after the intervention for the intervention group in an URCT. However, no comparison was made between intervention and control groups’ scores and so the effectiveness of the intervention cannot be determined. Multicomponent Strategies to Prevent Exiting One study tested the effect of introducing multiple changes (camouflaged exit doors, a new wandering path, private bedrooms, and an outdoor patio) on exiting (Mazzei, Gillan, & Cloutier, 2013), rated as having an overall critical risk of bias. Interventions Aiming to Facilitate Retrieval Although preventing exits is a strategy for preventing people with dementia from becoming lost (see Figure 3), some interventions aim to facilitate safety and rapid location once a person with dementia has become lost. Two studies reported on carers’ views following use of global positioning system (GPS) devices for people with dementia (Bantry White et al., 2010; Pot et al., 2012). Both studies had overall critical risks of bias. No evidence of the effectiveness of GPS devices on any primary or secondary outcomes can therefore be concluded. Educational Interventions Both studies examining exclusively educational interventions (Cohen-Mansfield et al., 1997; Levy-Storms et al., 2017) had overall critical risks of bias. Multi-aim and Multicomponent Interventions Three studies were categorized as multi-aim studies, meaning they used a combination of exit prevention, retrieval facilitation, or education (Chen & Leung, 2012; Horvath et al., 2007; Lau et al., 2018). For example, Lau and colleagues (2018) investigated an individualized program involving, among other features, education, environmental modification for preventing exits, and GPS devices for facilitating retrieval. All three studies in this category were rated as having overall critical risks of bias. Discussion Summary of Evidence Overall, there is insufficient evidence to determine the effectiveness of any interventions for preventing people with dementia exiting or getting lost. Grading systems for overall level of evidence were deemed unnecessary as risk of bias was so high. Although we assessed external selection bias, its relevance to our conclusions is slight as internal validity was so low. For preventing exits, most studies had too high risk of bias to contribute to evidence synthesis. Only two studies investigating grid patterns were not at critical risk of bias. These studies had inconsistent results, possibly due to imbalanced co-interventions and selective reporting. Results from individual studies suggest potential support for a mirror on the door, a trained service dog, and covering features of the door for preventing exiting from institutional settings. However, due to the overall serious risks of internal bias and low external validity results should be considered preliminary and not robust evidence for effectiveness. No studies met minimum risk of bias requirements (i.e., not at critical risk of bias) for providing useful evidence for the effectiveness of interventions for facilitating the retrieval of people with dementia who had become lost, or preventing the negative consequences of these events (e.g., GPS tracking). Assumptions and Robustness of Synthesis This is the first systematic review to focus specifically on interventions for preventing people with dementia from exiting or becoming lost. Study designs, risks of bias, and reporting shortcomings mean that all outcomes were found to be highly problematic. However, some key assumptions that led to this conclusion are important to discuss. Firstly, the synthesis was partially based on groupings and categories determined by our theoretical model of intervention effects. This model represented the pathway to exiting or getting lost, and the effects of interventions for preventing these. For example, the first step along the pathway was a “desire to explore or exit,” with relevant interventions reducing this desire and ultimately preventing exit attempts. This understanding may, in fact, be imprecise or simplistic. However, for this review, the model is not meant to thoroughly represent all antecedent and consequent factors in getting lost (see Rowe et al., 2015 for a more comprehensive discussion on this), or all steps in logic models of included interventions. Rather, we aimed to summarize and highlight different routes to preventing exits and getting lost from included interventions. Although the model was used to structure the narrative synthesis, threats to internal validity were so great that different assumptions would not have yielded different conclusions. Nevertheless, the model may provide a useful basis for future work aiming to develop new interventions. As mentioned, the main factor preventing evidence synthesis in this review was risk of bias of individual studies. We included weaker study designs and assessed them against high-quality trials to highlight the problems with existing literature for concluding intervention effectiveness. However, it is possible that our bespoke assessment of risk of bias for these studies led to conservative conclusions. Indeed, other systematic reviews (Fleming & Purandare, 2010; Hodgkinson, Koch, Nay, & Lewis, 2007; Jensen & Padilla, 2017; Letts et al., 2011; Padilla, 2011) have concluded moderate evidence for the effectiveness of environmental modification strategies on reducing exiting based on several of the same included studies as this review. However, these reviews included the results of other systematic reviews (i.e., each other) as the highest “levels of evidence,” often without methodological appraisal of primary research. This runs the risk of compounding and propagating biased findings. Indeed, when methodological appraisal was undertaken (Fleming & Purandare, 2010) authors used a checklist by Forbes (1998) that may have been interpreted as giving a “moderate” rating for uncontrolled before-after studies. In contrast, we assumed that studies without a control group or comparator condition were at critical risk of bias due to confounding. It is well-established that an uncontrolled study cannot isolate the effect of the intervention under investigation (Armstrong et al., 2011). We also assumed that potential carryover effects and intervention order effects may have existed for within-subject investigations. These effects may seem intuitively unlikely for environmental modification interventions for people with memory impairments, but their absence cannot be assumed without further evidence. Moreover, no studies suffered from risk of carryover or intervention order effects alone—all controlled studies had at least serious risk of bias in two or more domains. Conclusions and Future Directions This review highlights a mismatch between the scale of the issue of getting lost and the evidence for strategies to mitigate against it. Indeed, most included studies are small-scale or preliminary and not appropriate for estimating intervention effectiveness. Therefore, high-quality RCTs, NRCTs, and crossover trials are urgently needed for further investigation of any intervention for people with dementia exiting, eloping, or getting lost. For institutional interventions that affect an entire inpatient unit (e.g., environmental modification, service dog), cluster randomized or crossover trials may be most appropriate. However, the practical and resource-related barriers to multi-site studies may be off-putting for many potential investigators working at specific clinical sites. For research looking to maximize internal validity without the need for generalizability (e.g., looking to reduce exiting on a specific ward), the use of within-subject designs with repeated reversal and reintroduction of interventions would reduce many common risks of confounding. However, investigators must also consider mitigation of carryover effects and bias due to unblinded outcome measurement. An additional strategy might be the use of an interrupted time series analysis to account for trends over time due to, for example, changes in disease progression (Sterne et al., 2016). Despite several papers on the use of GPS for people with dementia who may become lost (see Pulido Herrera, 2016 for a review), only two studies with critical risks of bias came near to quantifying any intervention effects, neither of which examined an eligible primary outcome measure for this review. Therefore, perhaps surprisingly, there is no evidence for the effectiveness of GPS tracking devices for helping with retrieval of a lost person with dementia. Although the use and utility of GPS technology is so widespread (Pulido Herrera, 2016), its effectiveness for supporting people with dementia cannot be assumed: Devices often rely on a minimum level of technological aptitude that people with dementia or their carers may not have. This has implications for ethical considerations surrounding the use of tracking technology: Evidence of the effectiveness of tracking devices is necessary to counterbalance concerns of stigma and breaching privacy (Hughes & Louw, 2002; McShane et al., 1998; Nicolle, 1998). Observational studies may be a good starting point for building evidence of the effectiveness of GPS as many dyads already use them (Pulido Herrera, 2016). For experimental studies, ensuring intervention fidelity and usability is imperative before starting expensive RCTs. One key strategy not included in this review is that of retrieval and missing incident initiatives or programs. For example, in the United States, the Alzheimer’s Association’s Safe Return program (https://alz.org/help-support/caregiving/safety/medicalert-safe-return) facilitates the engagement of law enforcement and community services to help speed the retrieval of people with dementia. One report stated that the program had facilitated the recovery of 11,200 people at the time of writing (2007), with a 99% success in safely returning those enrolled (Bass, Rowe, & Moreno, 2007). In addition, the United States has Project Lifesaver (https://projectlifesaver.org/) and Silver Alerts for missing persons with dementia. Other countries have similar programs, such as Dementia Australia’s version of Safe Return (https://www.dementia.org.au/resources/safe-return) and the currently trialed Purple Alert mobile application from Alzheimer Scotland (http://purplealert.org.uk/). Although studies of the effectiveness of these programs could not be found for inclusion in this review, they may play a crucial role in efforts to reduce the impact of people with dementia getting lost in the community. Detailed and systematic evaluations of these initiatives could reveal promising avenues for their implementation elsewhere. Further Implications The absence of evidence for the effectiveness of interventions in this review implies an inability of practitioners and policy-makers to form evidence-based decisions or guidelines regarding strategies for preventing people with dementia becoming lost or exiting. Hesitancy in recommending strategies or technologies to individuals, dyads, or institutions should be taken, particularly when large financial costs may be incurred. With the increasing prevalence of dementia, the issue and negative consequences of people with dementia getting lost will only continue to grow without effective mitigation. We hope this consideration will galvanize practitioners and researchers into thorough investigations of promising interventions to help safeguard people with dementia against becoming lost. Acknowledgments We would like to thank all staff at the Frank Curtis Library at Norfolk and Suffolk NHS Foundation Trust for their help finding papers for this review. Funding This work was supported by the Earle & Stuart Charitable Trust and the Faculty of Medicine and Health Sciences, University of East Anglia (Grant number R205319). Conflict of Interest None reported. References Algase , D. L. , Moore , D. H. , Vandeweerd , C. , & Gavin-Dreschnack , D. J . ( 2007 ). Mapping the maze of terms and definitions in dementia-related wandering . Aging & Mental Health , 11 , 686 – 698 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Altus , D. E. , Mathews , R. M. , Xaverius , P. K. , Engelman , K. K. , & Nolan , B. A. D . ( 2000 ). Evaluating an electronic monitoring system for people who wander . American Journal of Alzheimer’s Disease , 15 , 121 – 125 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Armstrong , R. , Waters , E. , & Doyle , J . ( 2011 ). Chapter 21: Reviews in health promotion and public health. In J. P. Higgins & S. Green (Eds.), Cochrane handbook for systematic reviews of interventions version 5.1.0 . Retrieved from www.handbook.cochrane.org Google Preview WorldCat COPAC Bantry White , E. , & Montgomery , P . ( 2014 ). Dementia, walking outdoors and getting lost: Incidence, risk factors and consequences from dementia-related police missing-person reports . Aging & Mental Health , 19 , 224 – 230 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Bantry White , E. , Montgomery , P. , & McShane , R . ( 2010 ). Electronic tracking for people with dementia who get lost outside the home: A study of the experience of familial carers . British Journal of Occupational Therapy , 73 , 152 – 159 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Bass , E. , Rowe , M. A. , & Moreno , M . ( 2007 ). The Alzheimer’s Association’s Safe Return program for persons who wander. In A. L. Nelson & D. L. Algase (Eds.), Evidence-based protocols for managing wandering behaviors (pp. 251 – 258 ). New York : Springer . Google Preview WorldCat COPAC Bowen , M. E. , McKenzie , B. , Steis , M. , & Rowe , M . ( 2011 ). Prevalence of and antecedents to dementia-related missing incidents in the community . Dementia and Geriatric Cognitive Disorders , 31 , 406 – 412 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Centre for Reviews and Dissemination . ( 2009 ). CRD’s guidance for undertaking reviews in healthcare . York, UK : York Publishing Services . Google Preview WorldCat COPAC Chafetz , P. K . ( 1990 ). Two-dimensional grid is ineffective against demented patients’ exiting through glass doors . Psychology and Aging , 5 , 146 – 147 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Chen , Y.-C. , & Leung , C.-Y . ( 2012 ). Exploring functions of the lost seeking devices for people with dementia . Work , 41 ( Suppl. 1 ), 3093 – 3100 . doi: Google Scholar PubMed WorldCat Crossref Chiu , Y. C. , Algase , D. , Whall , A. , Liang , J. , Liu , H. C. , Lin , K. N. , & Wang , P. N . ( 2004 ). Getting lost: Directed attention and executive functions in early Alzheimer’s disease patients . Dementia and Geriatric Cognitive Disorders , 17 , 174 – 180 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Cohen-Mansfield , J. , & Werner , P . ( 1998 ). The effects of an enhanced environment on nursing home residents who pace . The Gerontologist , 38 , 199 – 208 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Cohen-Mansfield , J. , Werner , P. , Culpepper , W. J. , & Barkley , D . ( 1997 ). Evaluation of an inservice training program on dementia and wandering . Journal of Gerontological Nursing , 23 , 40 – 47 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Connell , B. R. , & Sanford , J. A . ( 1998 ). Evaluation of interventions to prevent elopement among nursing home patients . Rehabilitation R&D Progress Reports , 35 , 105 . WorldCat Coughlan , G. , Laczó , J. , Hort , J. , Minihane , A. M. , & Hornberger , M . ( 2018 ). Spatial navigation deficits - overlooked cognitive marker for preclinical Alzheimer disease? Nature Reviews Neurology , 14 , 496 – 506 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Cox , D. J . ( 2016 ). A brief overview of within-subject experimental design logic for individuals with ASD . Austin Journal of Autism & Related Disabilities , 2 , 1025 . WorldCat Dickinson , J. I. , & McLain-Kark , J . ( 1998 ). Wandering behavior and attempted exits among residents diagnosed with dementia-related illnesses: A qualitative approach . Journal of Women & Aging , 10 , 23 – 34 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Dickinson , J. I. , McLain-Kark , J. , & Marshall-Baker , A . ( 1995 ). The effects of visual barriers on exiting behaviour in a dementia care unit . The Gerontologist , 35 , 127 – 130 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Downs , S. H. , & Black , N . ( 1998 ). The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions . Journal of Epidemiology & Community Health , 52 , 377 – 384 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Fleming , R. , & Purandare , N . ( 2010 ). Long-term care for people with dementia: Environmental design guidelines . International Psychogeriatrics , 22 , 1084 – 1096 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Forbes , D. A . ( 1998 ). Strategies for managing behavioural symptomatology associated with dementia of the Alzheimer type: A systematic overview . The Canadian Journal of Nursing Research , 30 , 67 – 86 . Google Scholar PubMed WorldCat Hamilton , C. L . ( 1993 ). The use of tape patterns as an alternative method for controlling wanderers’ exiting behavior in a dementia care unit . Masters dissertation , Virginia Polytechnic Institute and State University , Blacksburg . Google Preview WorldCat COPAC Hermans , D. , Htay , U. H. , & Cooley , S. J . ( 2007 ). Non-pharmacological interventions for wandering of people with dementia in the domestic setting . Cochrane Database of Systematic Reviews , 1 , Art. No.: CD005994 . doi: WorldCat Crossref Hewawasam , L. C . ( 1996 ). The use of two-dimensional grid patterns to limit hazardous ambulation in elderly patients with Alzheimer’s disease . NT Research , 1 , 217 – 227 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Higgins , J. , Sterne , J. , Savović , J. , Page , M. , Hróbjartsson , A. , Boutron , I. , … Eldridge , S . ( 2016 ). A revised tool for assessing risk of bias in randomized trials . Cochrane Database of Systematic Reviews , 10 ( Suppl. 1 ), 29 – 31 . WorldCat Hodgkinson , B. , Koch , S. , Nay , R. , & Lewis , M . ( 2007 ). Managing the wandering behaviour of people living in a residential aged care facility . International Journal of Evidence-Based Healthcare , 5 , 406 – 436 . doi: Google Scholar PubMed WorldCat Crossref Horvath , K. , Harvey , R. , & Trudeau , S . ( 2007 ). A home safety program for community based wanderers: Outcomes from the veterans home safety project. In A. L Nelson & D. L. Algase (Eds.), Evidence-based protocols for managing wandering behaviors (pp. 259 – 276 ). New York : Springer . Google Preview WorldCat COPAC Hughes , J. C. , & Louw , S. J . ( 2002 ). Electronic tagging of people with dementia who wander: Ethical considerations are possibly more important than practical benefits . BMJ , 325 , 847 – 848 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Hussian , R. A. , & Brown , D. C . ( 1987 ). Use of two-dimensional grid patterns to limit hazardous ambulation in demented patients . Journal of Gerontology , 42 , 558 – 560 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Jensen , L. , & Padilla , R . ( 2017 ). Effectiveness of environment-based interventions that address behavior, perception, and falls in people with Alzheimer’s disease and related major neurocognitive disorders: A systematic review . American Journal of Occupational Therapy , 71 , 7105180030p1 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Kincaid , C. , & Peacock , J. R . ( 2003 ). The effect of a wall mural on decreasing four types of door-testing behaviors . Journal of Applied Gerontology , 22 , 76 – 88 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Kwok , T. C. , Yuen , K. S. , Ho , F. K. , & Chan , W. M . ( 2010 ). Getting lost in the community: A phone survey on the community-dwelling demented people in Hong Kong . International Journal of Geriatric Psychiatry , 25 , 427 – 432 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Lau , W. M. , Chan , T. Y. , & Szeto , S. L . ( 2018 ). Effectiveness of a home-based missing incident prevention program for community-dwelling elderly patients with dementia . International Psychogeriatrics , 31 , 91 – 99 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Letts , L. , Minezes , J. , Edwards , M. , Berenyi , J. , Moros , K. , O’Neill , C. , & O’Toole , C . ( 2011 ). Effectiveness of interventions designed to modify and maintain perceptual abilities in people with Alzheimer’s disease and related dementias . The American Journal of Occupational Therapy , 65 , 505 – 513 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Levy-Storms , L. , Cherry , D. L. , Lee , L. J. , & Wolf , S. M . ( 2017 ). Reducing safety risk among underserved caregivers with an Alzheimer’s home safety program . Aging & Mental Health , 21 , 902 – 909 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Liberati , A. , Altman , D. G. , Tetzlaff , J. , Mulrow , C. , Gøtzsche , P. C. , Ioannidis , J. P. , … Moher , D . ( 2009 ). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration . Annals of Internal Medicine , 151 , W65 – W94 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Mayer , R. , & Darby , S. J . ( 1991 ). Does a mirror deter wandering in demented older people? International Journal of Geriatric Psychiatry , 6 , 607 – 609 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Mazzei , F. , Gillan , R. , & Cloutier , D . ( 2013 ). Exploring the influence of environment on the spatial behavior of older adults in a purpose-built acute care dementia unit . American Journal of Alzheimer’s Disease & Other Dementias , 29 , 311 – 319 . doi: Google Scholar Crossref Search ADS WorldCat Crossref McShane , R. , Gedling , K. , Keene , J. , Fairburn , C. , Jacoby , R. , & Hope , T . ( 1998 ). Getting lost in dementia: A longitudinal study of a behavioral symptom . International Psychogeriatrics , 10 , 253 – 260 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Moher , D. , Liberati , A. , Tetzlaff , J. , & Altman , D. G . ( 2009 ). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement . Annals of Internal Medicine , 151 , 264 – 269 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Moore , D. , & Haley , J . ( 2014 ). Home evaluation of visual exit barriers in dementia-related wandering . Retrieved April 16, 2019 , from clinicaltrials.gov website: https://clinicaltrials.gov/ct2/show/study/NCT00997425 Namazi , K. H. , Rosner , T. T. , & Calkins , M. P . ( 1989 ). Visual barriers to prevent ambulatory alzheimer’s patients from exiting through an emergency door . Gerontologist , 29 , 699 – 702 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Nicolle , C . ( 1998 ). Issues in the use of tagging for people who wander: A European perspective . Personal Social Services in Northern Ireland , 58 , 10 – 22 . WorldCat Padilla , R . ( 2011 ). Effectiveness of environment-based interventions for people with Alzheimer’s disease and related dementias . American Journal of Occupational Therapy , 65 , 514 – 522 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Pai , M. C. , & Lee , C. C . ( 2016 ). The incidence and recurrence of getting lost in community-dwelling people with Alzheimer’s disease: A two and a half-year follow-up . PLoS One , 11 , e0155480 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Popay , J. , Roberts , H. , Sowden , A. , Petticrew , M. , Arai , L. , Rodgers , M. , … Duffy , S . ( 2006 ). Guidance on the conduct of narrative synthesis in systematic reviews: A product from the ESRC Methods Programme (Vol. 1 ). Lancaster, UK : Institute for Health Research, Lancaster University . doi: Google Preview WorldCat COPAC Crossref Pot , A. M. , Willemse , B. M. , & Horjus , S . ( 2012 ). A pilot study on the use of tracking technology: Feasibility, acceptability, and benefits for people in early stages of dementia and their informal caregivers . Aging & Mental Health , 16 , 127 – 134 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Pulido Herrera , E . ( 2016 ). Location-based technologies for supporting elderly pedestrian in “getting lost” events . Disability and Rehabilitation: Assistive Technology , 12 , 315 – 323 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Ranganathan , P. , Pramesh , C. S. , & Aggarwal , R . ( 2016 ). Common pitfalls in statistical analysis: Intention-to-treat versus per-protocol analysis . Perspectives in Clinical Research , 7 , 144 – 146 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Roberts , C . ( 1999 ). The management of wandering in older people with dementia . Journal of Clinical Nursing , 8 , 322 – 323 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Robinson , L. , Hutchings , D. , Dickinson , H. O. , Corner , L. , Beyer , F. , Finch , T. , … Bond , J . ( 2007 ). Effectiveness and acceptability of non-pharmacological interventions to reduce wandering in dementia: A systematic review . International Journal of Geriatric Psychiatry , 22 , 9 – 22 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref van der Roest , H. G. , Meiland , F. J. M. , Maroccini , R. , Comijs , H. C. , Jonker , C. , & Dröes , R.-M . ( 2007 ). Subjective needs of people with dementia: A review of the literature . International Psychogeriatrics , 19 , 559 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Rowe , M. A. , Greenblum , C. A. , & D’Aoust , R. F . ( 2012 ). Missing incidents in community-dwelling people with dementia . American Journal of Nursing , 112 , 30 – 35 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Rowe , M. , Houston , A. , Molinari , V. , Bulat , T. , Bowen , M. E. , Spring , H. , … McKenzie , ... ( 2015 ). The concept of missing incidents in persons with dementia . Healthcare , 3 . doi: WorldCat Crossref Rowe , M. A. , Kelly , A. , Horne , C. , Lane , S. , Campbell , J. , Lehman , B. , … Benito , A. P . ( 2009 ). Reducing dangerous nighttime events in persons with dementia by using a nighttime monitoring system . Alzheimer’s & Dementia , 5 , 419 – 426 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Shalek , M. , Richeson , N. E. , & Buettner , L. L . ( 2004 ). Air mat therapy for treatment of agitated wandering: An evidence-based recreational therapy intervention . American Journal of Recreation Therapy , 3 , 18 – 26 . WorldCat Shea , B. J. , Reeves , B. C. , Wells , G. , Thuku , M. , Hamel , C. , Moran , J. , … Henry , D. A . ( 2017 ). AMSTAR 2: A critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both . BMJ , 358 , j4008 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Sherman , K . ( 1999 ). A touch of “magic”: Using a service dog for dementia care . Nursing Homes , 1999 , 29 – 34 . WorldCat Sterne , J. A. , Hernán , M. A. , Reeves , B. C. , Savović , J. , Berkman , N. D. , Viswanathan , M. , … Higgins , J. P . ( 2016 ). ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions . BMJ , 355 , i4919 . doi: Google Scholar Crossref Search ADS PubMed WorldCat Crossref Traynor , V. , Veerhuis , N. , Johnson , K. , Hazelton , J. , & Gopalan , S . ( 2018 ). Evaluating the effects of a physical activity on agitation and wandering (PAAW) experienced by individuals living with a dementia in care homes . Journal of Research in Nursing , 23 , 125 – 138 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Woolford , M. H. , Weller , C. , & Ibrahim , J. E . ( 2017 ). Unexplained absences and risk of death and injury among nursing home residents: A systematic review . Journal of the American Medical Directors Association , 18 , 366.e1 – 366.e15 . doi: Google Scholar Crossref Search ADS WorldCat Crossref Yatawara , C. , Lee , D. R. , Lim , L. , Zhou , J. , & Kandiah , N . ( 2017 ). Getting lost behavior in patients with mild alzheimer’s disease: a cognitive and anatomical model . Frontiers in Medicine , 4 . doi: WorldCat Crossref © The Author(s) 2019. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Effectiveness of Interventions for Preventing People With Dementia Exiting or Getting Lost
JF - The Gerontologist
DO - 10.1093/geront/gnz133
DA - 1997-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/effectiveness-of-interventions-for-preventing-people-with-dementia-xa4EvvJ0Jc
SP - 1
VL - Advance Article
IS - 
DP - DeepDyve
ER -