Helping behaviours of community members towards older adults and the related factors: a cross‐sectional study

Community members can play important roles in helping older adults in their community. This study aimed to clarify the actual situation of community members' helping behaviours towards older adults and examine the related factors.


INTRODUCTION
In ageing societies worldwide, efforts have been made to realise 'ageing in place', which aims to enable older adults to live in their own communities as long as possible.As an international initiative, the development of age-friendly communities/cities (AFCCs) 'encourages active ageing by optimising opportunities for health, participation and security in order to enhance quality of life as people age' 1 and has been promoted to realise ageing in place. 2 In Japan, the government has promoted a communitybased integrated care system 3 aimed at managing AFCCs.In this system, it is expected that various community members, including those other than health/social care professionals, will support older adults in the community.Therefore, individual community members should play a role in providing help to older adults.
In previous literature, the helping behaviours of community members have been classified as 'formal planned helping' (e.g., donating money, serving as a volunteer), 'informal planned helping' (e.g., looking after a sick friend, giving a ride to a friend without transportation), or 'spontaneous helping', defined as 'helping that you provide immediately, with little or no prior thought' (e.g., giving directions to a stranger, providing assistance to a stranger who has fallen over). 4Under the notion of a 'community-based integrated care system', municipalities and community health/social care professionals have made efforts to build community networks to support older adults in the community (e.g., a system for monitoring older adults living alone, and mutual help among neighbours 5,6 ).This aims to promote 'formal' or 'informal' planned helping among community members.
At the same time, community members might also be compelled to provide spontaneous as well as planned help, since older adults with physical and/or cognitive disabilities can face sudden difficulties in their daily lives, such as falling or getting lost.If, in addition to planned behaviours, people develop the ability to spontaneously help older adults, it could further strengthen community-based integrated care systems.To consider strategies for promoting spontaneous helping behaviours among community members, it is necessary to clarify the nature of these helping behaviours for older adults and identify the related factors.
Few studies have investigated spontaneous helping behaviours for older adults in the community.Most previous studies have focused specifically on planned helping behaviours for people living with dementia (PLWD).[9][10] 10 Experiences of caring for PLWD were associated with positive attitudes towards PLWD in nursing students 11 and a person-centred care approach in healthcare professionals. 12Interventions have also been provided to improve attitudes and helping behaviours towards PLWD.Attending a dementia supporter training program, which is a standardised Japanese dementia education program, improved confidence in nursing students and nurses in caring for PLWD. 13After a training program including interaction with PLWD, volunteers' knowledge of dementia and attitudes towards PLWD improved, 14 and after a service-learning program that involved interaction with older adults, university students' comfort in working with adults with cognitive impairment significantly improved. 15,16Furthermore, the psychological sense of community-the extent to which people feel connected to a communitypredicted the concurrent and future volunteerism of community members. 17hus, community members' knowledge of support for PLWD, experience in caring for or being involved with PLWD, and sense of community could affect also spontaneous helping behaviours towards older adults, including those without dementia.Therefore, this study aimed to clarify the actual situations of spontaneous helping behaviours of community members towards older adults and examine related factors such as knowledge of support, involvement with PLWD, and community perceptions.

Sample and design
A survey was conducted to understand the need to develop educational programs for PLWD.The survey participants were recruited from the registered members of an internet research company (Rakuten Research).They were aged 15-69 years and were distributed across nine cities in the Tokyo Metropolis.The quota-sampling method was based on 11 participants for each stratum based on city (nine cities), age (six age groups: teens to sixties), and sex (two strata).If a stratum contained fewer than 11 participants, it was supplemented with participants from an adjacent age group; in particular, there was a limited number of registered members aged 15-19 years, meaning that only 10 participants were recruited from that age group.Finally, 1000 participants were recruited for this study.Survey responses were voluntary, and respondents were given rewards by the survey company.

Measurements
In the questionnaire, participants were asked about their demographics and their experience with helping older adults.To understand the factors related to helping behaviours, we also asked about their involvement with PLWD, knowledge of care, and perceptions of the community.

Demographics
The characteristics of the participants included age, sex, living arrangements, duration of living in the current community, and whether they worked as health or social care professionals.
Experiences of support and inability to support older adults First, the participants were asked about their experience with helping or not being able to help an older adult other than their family in the community who faced difficulties.Those who had such experiences were asked to describe them as open-ended responses.Participants with the experience of helping an older adult were asked to describe (i) what difficulty the older adult faced and (ii) how they helped.Those with experience of not being able to help were asked to describe (i) what difficulty the older adult faced and (ii) why they could not help.If they had more than one helping or non-helping experience, they were asked to describe the one most impressive to them.

Involvement with a person with dementia
To measure involvement with a PLWD, participants were asked whether they had any family members with dementia and whether they had any experience interacting with a PLWD other than a family member.
Knowledge of care: dementia and community resources for support Knowledge of care was measured by experience with attending a dementia supporter program 18 and knowledge of dementia and community general support centres.
Knowledge of dementia was measured using a scale developed and validated by Mikami et al. 19 This scale comprises 10 statements; for example, 'Dementia is mainly managed through inpatient treatment', 'Some forms of dementia are associated with shivering and stiffness of the limbs', and 'Appropriate interaction with PLWD can relieve the symptoms of dementia'.In this study, the participants responded to each question with 'yes' or 'no', and the number of correct answers was summed (range: 1-10) for each participant.Higher scores indicated greater knowledge.

Perceptions of the community
To understand the participants' perception of the community, we measured 'a sense of community'.This includes the feeling that members have of belonging-that members matter to one another and to the group-and a shared faith that members' needs will be met through their commitment to being together. 20Sense of community was measured using the short version of the Sense of Community scale, 21,22 which was rated using a 5-point Likerttype scale ranging from 1 (disagree) to 5 (agree).The scale consisted of the following subscales: 'solidarity and proactiveness' (three items), 'self-determination' (three items), 'sense of attachment' (three items), and 'reliance on others' (three items).The internal consistency reliability of the scale was found to be acceptable (Cronbach's alpha coefficient = 0.837). 22

Analysis
After obtaining the descriptive statistics of the participants' characteristics, content analysis and multinomial logistic regression analysis were used to clarify their helping behaviours and examine the related factors, respectively.

Content analysis
We analysed the free-answered data regarding experiences of helping or not being able to help older adults in the community using the content analysis method described by Funashima, 23 based on Berelson's method. 24The questions for the analysis concerned (i) what difficulty the older adult faced (helping situations), (ii) how the participants helped older adults in situations in which they could help (helping behaviour), and (iii) why they could not help the older adult in the situations in which they could not help (reasons for not helping).We excluded cases of 'informal planned helping behaviours' by a family member or friend and 'formal planned helping behaviours' by a care professional or volunteer from the analysis with the intention of understanding spontaneous helping behaviours.
The following procedure was used to analyse the data for each question.First, the first author described the contents corresponding to the answers to the questions from each case (making a recording unit).Second, each recording unit was categorised based on its similarity in terms of content, and each category was named.Third, to examine the reliability of the categorisation, the second author categorised data on a sample of record units, which were randomly extracted from 20% of all record units, into the defined categories.The concordance rate between the two researchers was calculated using the method of Scott et al., 25 and a criterion of 70% or above was set to ensure reliability. 23

Logistic regression analysis
To understand the characteristics of community members who helped older adults, we performed multinominal logistic regression analysis to examine factors in the participants' characteristics related to helping behaviours.For this analysis, we classified each category of helping behaviour based on the content analysis into two types-a little helping and advanced helping-and used this as an outcome variable (i.e., no helping, a little helping, and advanced helping), and examined factors related to the implementation of a little helping and advanced helping, with reference to no helping.
The participants' ages were entered into the model using the forced entry method.We also entered sex, years of residence in the community, and whether they worked as health or social care professionals, had a family member with dementia, had experience with a PLWD other than a family member, knew the community general support centre, and had experience attending a dementia supporter program.Scores for dementia knowledge and the sense of community scale were entered using the forward stepwise selection method.
The significance level was set at <0.05 (two-tailed).Statistical analysis was performed using SPSS Statistics version 22.0 for Windows.

Ethical considerations
Survey responses were voluntary, and the respondents were given rewards by the survey company.Registrants under 18 years of age registered with their parents' consent.All data were anonymous, and informed consent was obtained from the online survey system.The study was conducted in accordance with the ethics guidelines of the Declaration of Helsinki.This study was approved by the Research Ethics Committee of the Graduate School of Medicine, University of Tokyo (#2019206NI).

Characteristics of participants
Table 1 shows the descriptive statistics of the participants' characteristics (N = 1000).The mean age (AE standard deviation) was 44.6 (AE 13.8) years, 50% were men, and approximately 10% were working as health/social care professionals.Only 6% of the participants had experience attending a dementia supporter program.
Experience with helping or not helping older adults Among the participants, 481 had had helping experiences while 241 had encountered an older adult in trouble but could not help.Among these 481 and 241 participants, 40 and 34 were excluded, respectively, because of a lack of details about the situations (eight and 12) or they described helping a family member/friend or as a care professional/volunteer (22 and 32).Therefore, the helping and not-helping experiences of 441 (44.1%) and 207 (20.7%) participants, respectively, were analysed.

Situations in which participants encountered older adults in trouble
Table 2 shows the situations in which the participants encountered older adults in trouble in cases in which they helped (n = 441) and could not help (n = 207).The situations were classified into seven categories, comprising 18 subcategories.If multiple categories were applicable for each situation, each was counted.
The most common situations in which older adults faced difficulties included being lost in the city or on public transportation, regardless of whether they had dementia (34.7% and 30.9% for cases in which participants helped and could not help, respectively).About 20% of participants encountered an older adult who had trouble walking or moving (19.1% and 16.4%, respectively) or had an accident or was in poor physical condition (18.1% and 22.7%, respectively).
'Situations indicating the possibility of dementia' frequently occurred in the same cases as older adults who were lost in the city/on public transportation or had had an accident, as mentioned above.More participants could not help (13.6%)than could help (5.5%) in situations with unusual behaviour or appearance, such as shouting or talking to themselves or being confused because they did not understand their own situations (Table 2).

Helping behaviours provided
Table 3 shows the helping behaviours provided by the participants in the cases where they helped older adults (n = 441).The most common helping behaviours were helping them walk or move, including carrying their baggage (20.0%), and taking care of them in accidents, such as checking their condition by talking to them or helping them get up (16.8%).More than 10% of the participants reported that they supported an older adult with difficulties in daily life (10.4%),gave directions to or accompanied an older adult to a destination (11.6% and 12.9%, respectively), and contacted police or an ambulance (11.8%).Some participants contacted store/station staff or family members (7.9%), while others offered their seats on transportation or their turn in line (7.3%;Table 3).

Reasons for not being able to help
Table 4 shows the reasons why participants could not help older adults (n = 207).The most common reason was that they were in a situation that prevented them from helping (e.g., in a hurry to get somewhere, physically unable to help because they were driving or far away; 42.5%).About 20% did not feel a sense of responsibility to help the older adult (19.3%), either because someone else had already helped or they assumed someone else would help.Some said they did not know how to help older adults (17.4%) and hesitated to help (14.4%)because they did not want to get involved or were afraid of the older adult's unusual appearance (Table 4).

Factors related to helping behaviours
We classified the helping behaviours extracted from the content analysis into 'a little help' (e.g., giving up a seat on public transit, letting them move ahead in line, giving directions to a destination) and 'advanced help' (e.g., accompanying them to a destination, helping them walk/move, providing accident care, contacting police or an ambulance, contacting staff or a family member, providing support for daily living, talking with them or watching over them).
Table 5 shows the results of the multinomial logistic regression analysis, which was used to examine the associations between helping experience and participants' characteristics.The characteristics related to a little helping experience were fewer years of residence in the current community (adjusted odds ratio (AOR) = 0.97, P = 0.007) and higher scores for dementia knowledge (AOR = 1.28,P = 0.010).The characteristics related to advanced helping experience were older age (AOR = 1.01,P = 0.044), female (AOR = 1.52,P = 0.005), having a family member with dementia (AOR = 1.73, p = 0.001), having experience interacting with a PLWD other than a family member, continuous involvement/involvement for a short period of time (AOR = 4.99 and 2.52, respectively; P < 0.001), knowing about the community general support centre (AOR = 1.59,P = 0.004), higher scores for dementia knowledge (AOR = 1.17,P = 0.003), and higher scores for sense of community (AOR = 1.04,P = 0.001).

DISCUSSION
This study investigated community members' experience of spontaneous helping behaviours by asking whether they had helped an older adult in trouble.The results showed that about 40% and 20% of participants had experience with providing help or not being able to do so.The multivariate analyses demonstrated that participants with helping experience had more experience involving PLWD, knowledge of dementia and community resources for supporting older adults, and a stronger sense of community.To our knowledge, this is the first study to clarify actual situations of community members across diverse age groups spontaneously helping older adults in the community.The results could provide insightful knowledge to promote community members' helping behaviours for older adults, contributing to the realisation of AFCCs.
The results of the content analysis showed that community members experienced a variety of The boldface is the total in each category.
spontaneous helping behaviours, from a little help on the spot (e.g., yielding seats or giving directions to a destination) to further in-depth help (e.g., taking care of older adults in an accident, accompanying the older adult to their destination, contacting public and private agencies/family, support in daily life).Since little is known about community members' spontaneous helping behaviours towards older adults, the categories developed in our content analysis could be useful as a draft framework for future research on community members' helping behaviours.
The results of the multivariate analyses showed that there were differences in related factors among the types of helping behaviours.In the category of little help without specific knowledge or skills (e.g., yielding seats/ turns, giving directions to a destination), fewer years of residence in the current community and higher dementia knowledge scores were the only associated factors.However, for in-depth advanced helping behaviours, the associated factors were older age, being female, experience with PLWD, knowledge of dementia and general community support centres, and sense of community.There could be different strategies for promoting various helping behaviours.The various factors identified in this study could be useful for considering specific strategies, especially for in-depth advanced helping behaviours.
The results showed that experience with interacting with PLWD or having a family member with dementia was associated with advanced helping behaviours.Meanwhile, more knowledge of dementia was associated with both a little helping behaviour and advanced helping behaviour.These results are consistent with previous studies that revealed the associations between knowledge/attitudes and the helping behaviours of community members towards PLWD. 7,8,10Our results also suggest that dementia knowledge and interactions with PLWD could promote in-depth advanced helping behaviours in addition to attitude improvement by reducing dementia-related stigma.While this study did not specifically investigate ageist attitudes or interactions with older adults without  dementia, these could also influence helping behaviours and should therefore be investigated in future research.
In addition, a sense of community and knowledge of community resources (e.g., community general support centres) were also associated with advanced helping behaviours.This is consistent with previous studies that showed an association between people's sense of community and participation in volunteering, 17,26 a type of planned helping behaviour.Higher perceptions of community and knowledge of community resources might lead to a high awareness of support and willingness to provide advanced support for older adults.
The results for the associated factors could provide useful suggestions for promoting community helping behaviours for older adults.Providing educational programs that include elements of involvement with older adults (including PLWD) and approaches that foster perceptions of community could be effective.In a previous study, we developed evaluated a dementia educational program for community members that included a short film illustrating the daily life of an older woman living with dementia and virtual reality movies from her perspective.The results showed an improvement in attitudes towards PLWD and the intention to help PLWD. 27Simulated experience involving PLWD could be effective for increasing community members' awareness of support.Another study indicated that small-group workshops with discussions improved participants' psychological sense of community and their intention to engage in social actions such as volunteering. 28his suggests that including small-group discussions in future educational programs could help foster people's awareness of community and its resources, thus leading to advanced helping behaviours.
Our content analyses of reasons for not being able to help an older adult could also provide insightful suggestions.Aside from being physically unable to help, common reasons for not being able to help were that they could not determine the need for support, did not feel responsible for helping, felt hesitant to help, or did not know how to help (6%-19%).These elements are partially consistent with Latane and Darley's 29 decision model of helping.Overcoming such barriers requires training on responsibility for helping and how to provide help.Gaesser et al. revealed that when presented with a situation depicting another person's plight, participants who imagined an event in which they helped the person (episodic simulation 30 ) showed increased prosocial intentions. 31In our previous study, a dementia educational program using a simulation card game was effective for improving attitudes towards PLWD. 32uch simulation training designed to evoke an image of helping could be effective for motivating and teaching helping behaviours.
This study has several limitations.First, since the web survey was conducted for community members registered with the survey company and living in certain municipalities of the Tokyo Metropolis, the generalisability of the results is unknown.For example, it is possible that the subjects who voluntarily responded to the web survey performed advanced helping behaviours more frequently than community members as a whole.Also, community members in rural areas might provide advanced helping that is unique, but it might not be clarified in this study.Second, this study might not have identified all cases in which the participants could or could not help.For the cases in which they could help, they were asked to select the most memorable cases.However, there might have been other cases in which they provided help.For those not able to help, there might be difficult situations for older adults that the participants were unaware of.Third, since this was a crosssectional study, the causal relationships were unclear.In the future, longitudinal studies should be conducted to examine whether specific training for community members can promote helping behaviours towards older adults.

CONCLUSION
Community members provided older adults with various types of spontaneous help, including help with walking, accident care, giving directions to or accompanying them to a destination, and support in daily life.Having a family member with dementia, experience interacting with a PLWD, knowledge of dementia and community general support centres, and a stronger sense of community were associated with advanced helping behaviours.Educational interventions for community members, including interactions with older adults, including PLWD, a focus on community perceptions, and training in actual helping actions could effectively promote helping behaviours in the community, leading to the construction of effective AFCCs.

Table 1
Participant characteristics (n = 1000) © 2023 The Authors Psychogeriatrics published by John Wiley & Sons Australia, Ltd on behalf of Japanese Psychogeriatric Society.

Table 2
Situations in which the participants encountered older adults in trouble Able to support (n = 441) Not able to support (n = 207)

Table 4
Reasons for being unable to help (N = 207)