An exploratory study on municipal measures for preventing and addressing incidents of missing older persons with dementia living alone in Japan
Abstract
Background
It is predicted that the number of older persons with dementia (PwD) living alone in Japan will increase in the future. Consequently, it is also expected that the number of missing older PwD living alone will rise. However, there is little research on missing PwD who live alone. The aim of this study was to clarify the framework of measures that are deemed necessary by municipalities when older PwD who are living alone go missing.
Methods
A mail survey was conducted among 1741 municipalities, and we asked about the necessity of 22 measures to address the issue of missing PwD who are living alone. A total of 478 municipalities with no missing values in the analyzed variables were included in the analysis, and factor analysis was conducted.
Results
The following five factors were extracted: collaboration (Factor 1), prior agreement (Factor 2), the SOS network (Factor 3), information and communication technology-based solutions (Factor 4), and policy guidelines or protocol frameworks (Factor 5).
Conclusion
Each of the five factors was unique. The strategies for dealing with missing persons include both relatively easy and challenging measures. Municipalities that have not yet made significant progress should begin with the easier measures and gradually implement the more challenging measures.
INTRODUCTION
In Japan, according to the National Police Agency, the number of missing persons among persons with dementia (PwD) (including suspected cases) has been annually increasing, reaching 19 039 individuals in 2023.1 Additionally, it is predicted that the number of older PwD living alone (PwDLA) will reach 1.47 million in 2025 and 1.81 million in 2040.2 Although this suggests that the prevalence of missing incidents among older PwDLA will increase in the future, few studies have focused on this topic.3-6
Measures for preventing and addressing missing incidents vary by country, with the implementation of systems such as the Silver Alert in the United States,7 Canada8 and Australia,9 the Herbert Protocol in the United Kingdom,10 and the Dementia Go-to Points in Singapore.11 In Japan, municipalities have the responsibility to develop municipal welfare plans in compliance with the Social Welfare Act.12 Therefore, both the police and municipalities must address the issues surrounding missing incidents among older PwD. While the police must respond in accordance with regulations,13 the measures of municipalities are independently determined. Some municipalities proactively deal with missing incidents, while others take more passive approaches. Previous studies have revealed that living alone at the time of disappearance is significantly associated with death at the time of discovery.14, 15 Therefore, it is crucial for municipalities to appropriately handle missing incidents when older PwDLA go missing.
However, more than half of the municipalities reported they did not effectively respond to missing incidents among PwDLA.16 If measures for preventing and addressing missing incidents among PwDLA have not been developed in municipalities, measures such as building local networks, developing social resources, and providing support in individual cases must be developed. Nevertheless, it remains unclear what specific measures should be taken regarding the disappearance of PwDLA. The aim of this study was to clarify the framework for measures that are deemed necessary by municipalities when PwDLA go missing.
METHODS
Study design and sample size
A mail survey of all 1741 municipalities was conducted in Japan in September 2020. A total of 562 municipalities responded, and of them, 487 had no missing values for the study variables and were included in the analysis.
Measured variables
As noted above, there are few studies on older missing PwDLA. Therefore, there are few studies on measures used by municipalities for preventing and addressing missing incidents among PwDLA. Consequently, we developed 22 items (Table 1) based on the 20 items regarding how municipalities respond when PwD go missing.17 Incidentally, the validity and reliability of these 20 items have not been verified. In considering the survey items, we also referred to the measures to be used for missing PwD that were posted on some municipalities' official websites. Through discussions, we decided on the 22 survey items.
| Factor 1 | Factor 2 | Factor 3 | Factor 4 | Factor 5 | |||
|---|---|---|---|---|---|---|---|
| Avg. | SD | Collaboration | Prior agreement | SOS network | ICT-based solutions | Policy guidelines or protocol frameworks | |
| Training for local residents on speaking accordingly to older persons with dementia who may be missing | 3.79 | 0.97 | 0.594 | 0.116 | 0.201 | 0.076 | 0.223 |
| Holding a case conference for older individuals with dementia at risk of going missing | 3.65 | 0.85 | 0.571 | 0.212 | 0.179 | 0.049 | 0.239 |
| Collaboration with condominium associations | 3.57 | 1.17 | 0.530 | 0.154 | 0.193 | 0.117 | 0.211 |
| Collaboration with neighbourhood associations and/or community councils | 4.46 | 0.90 | 0.505 | 0.473 | 0.250 | 0.089 | −0.013 |
| Ensuring the involvement of older individuals with dementia in measures related to a missing incident | 3.65 | 0.94 | 0.470 | 0.213 | 0.163 | 0.154 | 0.302 |
| Creating a community that is easy to live in (and go out in) for people with dementia | 4.73 | 0.91 | 0.461 | 0.326 | 0.341 | 0.102 | −0.013 |
| Broad collaboration with other municipalities | 3.97 | 1.02 | 0.457 | 0.183 | 0.226 | 0.149 | 0.214 |
| Determining in advance who will file a missing person report with the police | 3.57 | 1.02 | 0.435 | 0.275 | 0.088 | 0.078 | 0.429 |
| Appointment of a legal guardian for adults | 3.72 | 0.89 | 0.434 | 0.234 | 0.047 | 0.101 | 0.221 |
| Developing a framework for the seamless sharing of personal information among relevant organisations | 4.30 | 0.88 | 0.162 | 0.694 | 0.211 | 0.148 | 0.253 |
| Identifying in advance older individuals with dementia who are at risk of going missing | 4.32 | 0.86 | 0.234 | 0.606 | 0.215 | 0.084 | 0.200 |
| Prearranging a way to communicate with family members or relatives | 4.53 | 0.96 | 0.354 | 0.531 | 0.153 | 0.112 | 0.097 |
| Determining in advance the response protocol for when a missing individual with dementia is located | 4.00 | 0.94 | 0.247 | 0.529 | 0.170 | 0.057 | 0.452 |
| Information sharing and collaboration with the police | 5.01 | 0.86 | 0.184 | 0.482 | 0.396 | 0.079 | −0.008 |
| Monitoring through SOS network | 4.36 | 0.96 | 0.238 | 0.199 | 0.838 | 0.104 | 0.185 |
| Activating the SOS network | 4.44 | 0.96 | 0.208 | 0.256 | 0.781 | 0.135 | 0.193 |
| Registration of older individuals with dementia in an SOS network | 4.17 | 1.04 | 0.219 | 0.219 | 0.624 | 0.061 | 0.251 |
| Monitoring by community residents | 4.90 | 0.87 | 0.322 | 0.396 | 0.416 | 0.064 | −0.177 |
| Development of ICT-based search devices for ascertaining the location of older individuals who have gone missing | 3.42 | 1.18 | 0.150 | 0.156 | 0.115 | 0.934 | 0.117 |
| Development of ICT-based monitoring devices for early awareness regarding missing individuals | 3.35 | 1.17 | 0.176 | 0.111 | 0.119 | 0.915 | 0.109 |
| Establishment of implementation outlines and protocols for cases involving missing individuals with dementia | 3.15 | 1.00 | 0.188 | 0.081 | 0.187 | 0.109 | 0.627 |
| Enactment of ordinances related to missing individuals with dementia | 2.76 | 0.94 | 0.263 | 0.078 | 0.067 | 0.065 | 0.597 |
| Eigen value | 8.275 | 1.699 | 1.536 | 1.220 | 1.094 | ||
| Explained variance (%) | 13.390 | 11.928 | 11.901 | 8.702 | 8.088 | ||
| Cumulative (%) | 13.390 | 25.318 | 37.219 | 45.921 | 54.009 | ||
- Note: Bartlett's test of sphericity: p < 0.001 (in grey shade).
- 1: Not necessary; 2: Rather unnecessary; 3: Somewhat necessary; 4: Necessary; 5: Very necessary; 6: Absolutely necessary. SOS network: an organisation generally established by municipalities and aims to monitor persons with dementia (PwD) and search for missing PwD, with the participation of the municipality, police, various local agencies, community residents, etc. Factor extraction method: principal factor method. Rotation method: Varimax rotation with Kaiser normalisation.
- Avg, average score; SD, Standard deviation; ICT, information and communication technology.
However, there is no clear distinction between PwDLA and PwD living with others in the measures for preventing and addressing missing incidents. If PwDLA go missing, measures are taken based on the individual circumstances. Hence, it was difficult to evaluate measures targeting only PwDLA. Thus, for each of the 22 items, respondents were asked to rate the extent to which they felt that a measure was necessary if PwDLA went missing. Responses were collected using a six-point scale, from one (not necessary) to six (absolutely necessary).
The analysis involved calculating descriptive statistics for each variable. Next, exploratory factor analysis was conducted for the 22 items. The analysis was performed by using SPSS version 29 (Japanese version).
Ethics consideration
This study was approved by the Ethics Committee of the Tokyo Metropolitan Institute for Geriatrics and Gerontology (R21-39).
RESULTS
Descriptive statistics
The average scores and standard deviations of the 22 items are shown in Table 1. The item with the highest average score (5.01 points) was ‘information sharing and collaboration with the police’, followed by ‘monitoring by community residents’ (4.90 points), ‘creating a community that is easy to live in (and go out in) for people with dementia’ (4.73 points) and ‘prearranging a way to communicate with family members or relatives’ (4.53 points).
The item ‘enactment of ordinances related to missing individuals due to dementia’ had the lowest score (2.76 points). The next lowest scores were as follows: 3.15 points for the item ‘establishment of implementation outlines and protocols for cases involving missing individuals with dementia’, 3.35 points for the item ‘development of information and communication technology (ICT)-based monitoring devices for early awareness regarding missing individuals’, and 3.42 points for the item ‘development of ICT-based search devices for ascertaining the location of older individuals who have gone missing’.
Factor analysis
Since there are few studies on measures taken by municipalities for preventing and addressing missing incidents among PwDLA, an exploratory factor analysis was conducted to clarify the framework for these measures (Table 1). Eigenvalues exceeding 1.0 were observed for five factors. The measure of sampling adequacy according to the Kaiser-Meyer-Olkin test was 0.892, and Bartlett's test of sphericity yielded a P-value of <0.001, indicating no issues with the validity of the factor analysis. Furthermore, all item factor loadings were above 0.40, resulting in a five-factor structure effectively utilising all 22 items. The Cronbach's α coefficients were calculated and determined to be 0.917 for the 22 items, 0.848 for the first factor, 0.814 for the second factor, 0.844 for the third factor, 0.961 for the fourth factor, and 0.756 for the fifth factor. The Cronbach's α coefficients were greater than 0.700 for the 22 items and five factors, so there was no problem with the reliability of the analysis.
The factors were named as follows: collaboration (Factor 1), prior agreement (Factor 2), the SOS network (Factor 3), information and communication technology-based solutions (Factor 4), and policy guidelines or protocol frameworks (Factor 5).
DISCUSSION
The factor analysis resulted in the extraction of five factors, each of which was unique. This study was conducted on an exploratory basis and targeted Japanese municipalities; therefore, it is unclear whether the results are applicable to other countries. However, in the case of individuals living alone, although there may be differences among countries, appropriate measures taken by municipalities might contribute to the prevention of missing incidents and the location of missing persons. For example, in Japan, the use of long-term care insurance services is significantly associated with the early detection of missing PwDLA.4
In this study, for each of the 22 items, representatives of each municipality were asked to what extent they felt that a measure was necessary if PwDLA go missing. Of the 22 items, the item with the highest average score was ‘information sharing and collaboration with the police’. This is probably a reasonable result since cooperation with the police is essential in the case of missing persons. In Japan, if a missing person report has been filed with the police, 72.3% of missing PwD (including cases of death) are found within the same day.1 Neubauer et al. noted that delays in calling the police are the most concerning issue for police. Initial investigations by police are very important and conclusive.18 Respondents on behalf of the municipalities indicated the importance of cooperation with the police.
The item ‘monitoring by community residents’ had the next highest score. Since approximately 60% of people who report missing incidents are family members living with PwD,19 respondents on behalf of the municipalities probably thought that if there were no family members living with a PwD, it would be necessary for them to be watched by neighbours. The third item was ‘creating a community that is easy to live in (and go out in) for people with dementia’. This refers to the creation of a dementia-friendly community20 and the need for the community as a whole to address missing incidents among PwDLA. The fourth item was ‘prearranging a way to communicate with family members or relatives’. When PwDLA go missing, prior communication arrangements with family and relatives are needed. In some cases, these individuals may be required to file a missing person report with the police.
Conversely, the average scores for the enactment of ordinances, the establishment of implementation outlines and protocols, ICT-based monitoring devices and ICT-based search device items were lower. It is assumed that the first and second items, especially the second item, are necessary measures for municipalities to handle missing incidents. However, the opposite results were obtained. For municipalities, the documentation of implementation outlines and protocols are necessary measures to implement in order to handle missing incidents. However, documentation is required regardless of whether a missing person lives alone or with others. Therefore, it is possible that the need for this was evaluated as less important. The third and fourth items referred to ICT devices. The respondents may have thought PwDLA would not be able to use ICT devices.
Next, as a result of factor analysis, five factors were extracted. Factor 1 was named ‘collaboration’. This factor included collaboration that was considered challenging, such as training local residents and ensuring the involvement of older individuals with dementia; however, collaboration could also be relatively easy, such as broad collaboration with other municipalities and collaboration with neighbourhoods. Factor 1 included various levels of collaboration. Training for local residents can only be carried out after a long period of practice regarding measures for preventing and addressing missing incidents in the community and with the understanding of local residents. Therefore, municipalities may not immediately implement challenging measures. However, it will be necessary to implement challenging measures in the future.
Factor 2 was ‘prior agreement’. When PwD go missing, it is necessary to have prior agreements in place regarding how different agencies should collaborate to promptly initiate effective search activities. This includes prior agreement regarding interpersonal cooperation to address missing incidents. These measures are considered relatively feasible in the current situation. Since the items in Factor 2 are basic measures, municipalities that have not yet implemented them should do so as soon as possible.
Factor 3 was ‘the SOS network’. A network that consists of various social resources in the community—sometimes more extensive areas—is essential for preventing and addressing missing incidents. The SOS network is thought to play a similar role, although not exactly the same, as the Silver Alert in the United States and other countries. The network is positioned at the core of measures for preventing and addressing missing incidents. However, in Japan, approximately 40% of municipalities believe that their SOS networks are not functioning because of reasons such as the small number of pre-registrants, difficulties in sharing personal information, and the lack of collaborating organisations.16 It is necessary to make the SOS networks be effective in functioning.
Factor 4 was ‘ICT-based solutions’. This factor consists of two items. Japan is experiencing a declining birth rate along with increased population ageing.21 In the foreseeable future, it is anticipated that community residents who are engaged in monitoring PwD and searching for missing PwD will also be ageing. Additionally, previous studies have noted that the use of ICT devices is effective for preventing and addressing missing incidents.22, 23 Considering this situation, the use of ICT devices would seem to be effective, but the respondents probably thought that PwDLA would not be able to successfully use the devices. However, soon, if ICT devices progress and become usable even for PwDLA, the utilisation of ICT devices is expected to become indispensable.
Factor 5 was ‘policy guidelines or protocol frameworks’. Two items were included in this factor. If a municipality takes measures to handle missing incidents, those measures must be documented. The documentation of measures provides municipalities with responsibilities and specific methods for preventing and addressing missing incidents.
The results of the factor analysis showed that each of the five factors was unique. There may be an order of priority in which measures for preventing and addressing missing incidents among PwDLA are implemented. In other words, basic measures should be taken first, and once these are implemented, more challenging measures should be implemented. The findings obtained from this study provide strategies to address the issue of missing PwDLA. Municipalities that are behind on implementing measures for missing PwDLA should begin with simpler steps and progressively develop more comprehensive strategies.
This study has several limitations. First, this study targeted municipalities in Japan, and the actual situations in other countries may differ. Therefore, it is unclear whether the findings of this study can be applied to other countries. Second, this study only analyzed 487 of the 1741 municipalities (28.0%). Therefore, representativeness is not ensured. However, the average response rate of mail surveys in Japan is approximately 30%.24 Thus, the response rate of 28% in this study was not especially low. The low response rate is considered due to the use of a mail survey. The municipalities that responded are considered interested in this issue, even if to varying degrees. According to a previous study that compared the measures of five issues—elderly abuse, self-neglect, consumer harm, promotion of the use of adult guardianship and missing incidents that municipalities have conducted—the measures of missing incidents were the lowest to be developed. Taking this into account, the interest of municipalities in this area was not high. Nevertheless, some municipalities were actively working on this issue, suggesting that the measures varied considerably by municipality.17 Third, it was difficult to consider measures targeting only missing PwDLA; therefore, the answers of the respondents were subjective. Finally, few studies have focused on municipal measures for preventing and addressing missing incidents among PwDLA. Therefore, sufficient discussion based on previous research was not possible.
Although this study was exploratory, we were able to clarify the framework of measures that are deemed necessary by municipalities when PwDLA go missing. We hope that this research will contribute to promoting the development of measures by municipalities and providing a road map to implement these measures.
AUTHOR CONTRIBUTIONS
KK had full access to all the data in this study. KK was responsible for the accuracy of data analysis. Concept and design: KK, TO and TI. Data acquisition: KK and TI. Analysis or interpretation of the data: KK, TI and SA. Manuscript drafting: SA, TO and KK. Critical revisions: all authors. Statistical analysis: KK.
ACKNOWLEDGMENTS
This study was supported by a grant from the Pfizer Health Research Foundation (Tokyo, Japan). The funder had no role in this study. We appreciate the cooperation of the municipalities.
DISCLOSURE
The authors declare no conflicts of interest.
Open Research
DATA AVAILABILITY STATEMENT
The data are not available because we did not obtain consent from the survey participants for data provision.
