Perceived reasons of unsafety among independently living older adults in Sweden

ABSTRACT

Feeling safe is important for quality of life in advanced age. The current study aimed to investigate whether different perceived reasons for unsafety uniquely contribute to the feelings of unsafety in the neighbourhood and at home while ageing. Data from a cross-sectional survey study on older adults living independently in Sweden were analysed (N = 622, age range 64–106 years, 60.6% female). Binary logistic regressions revealed unique associations between fear of crime, unattractive social climate in the neighbourhood, and inconvenient infrastructure at home with experienced feelings of unsafety, while controlling for socio-demographic factors. When distinguishing between reasons for feeling unsafe in the neighbourhood and at home, different associations with socio-demographic factors emerged. Higher age was positively associated with health-related unsafety both in the neighbourhood and at home and was not related to fear of crime neither in the neighbourhood nor at home. Female gender was associated with both health-related unsafety and fear of crime in the neighbourhood and with health-related unsafety at home. Overall, the findings highlight the presence of a range of perceived reasons of unsafety relevant for older adults and the importance to consider these subjective unsafety reasons in further research and practice.

KEYWORDS:

• Feelings of unsafety
• fear of crime
• neighbourhood
• ageing
• older adults

Introduction

Feeling safe in a place of residence is important for quality of life in advanced age (Bowling et al., 2003). Perceived unsafety underlies stress and anxiety responses (Brosschot et al., 2017) and may alter older people’s mobility and participation in their community lives (Bowling et al., 2003; Farrall et al., 2009). Most research on feelings of unsafety experienced by older adults has focused on fear of crime (Ceccato & Bamzar, 2016; Hale, 1996; Tulloch, 2000), while research on various perceived reasons of unsafety in late life is very limited (Won et al., 2016). However, subjective unsafety can be explained by a broader range of insecurities: from the perspective of symbolic paradigm, potential non-criminal threats or vulnerabilities should also be considered (De Donder et al., 2012). For instance, health and mobility limitations constitute another important reason for feeling unsafe in advanced age (e.g., fear of falling, Fonad et al., 2006; vision and hearing problems; Kappes et al., 2013). Closely related to health limitations are needs for adapted infrastructure at home and in the neighbourhood. Home hazards that threaten safe movement or wayfinding can be crucial for an older adult (Wahl & Gitlin, 2019). Moreover, social connectedness is an important element of safety while ageing in place (Wiles et al., 2012). Although these different perceived reasons for feeling unsafe have been identified within previous studies, they have mostly been studied in different lines of research, hence masking their potential unique contribution to feelings of unsafety in advanced age. Thus, research on feelings of safety in advanced age should seek to address reasons for such feelings from a broader perspective.

Although various reasons may explain feelings of unsafety in advanced age, empirical research investigating several perceived reasons of unsafety simultaneously is scarce (Won et al., 2016). Hence, there is a need for a theoretically based analytic approach that takes into consideration multiple reasons of unsafety and allows to determine whether these perceived reasons have a unique contribution to the overall feelings of unsafety. The current study addresses this research gap by applying the Generalized Unsafety Theory of Stress (GUTS) (Brosschot et al., 2017, 2018) and explores fear of crime, health limitations, infrastructure problems, and social climate as independent perceived reasons of unsafety experienced by older adults, while controlling for a range of socio-demographic factors. Gaining such understanding of different reasons for perceived unsafety in advanced age is important because it can guide preventive efforts and initiatives aimed at promoting safety.

Generalized unsafety theory of stress

According to GUTS, perceived generalized unsafety may result from one or several compromised life domains and does not require a specific stressor or a threat (Brosschot et al., 2017, 2018). Examples of such compromised domains include the body (e.g., physical illnesses or functional limitations), social networks (feeling lonely or disconnected from others), and the environmental context (e.g., unfavourable perception of one’s housing or neighbourhood). Perceived generalized unsafety thus reflects deficits in one or several of these important life domains and can develop in ‘an objectively safe world’ (Brosschot et al., 2017, p. 288). Such theoretical approach allows to adopt a broader perspective on feelings of safety in advanced age and interpret such feelings as being potentially explained by many different reasons. This is of great importance since past studies might have exaggerated the contribution of single types of reasons for unsafety, while other reasons could potentially merit priority attention.

Regarding older adults, GUTS considers older age in general as a vulnerability factor for perceived unsafety. Brosschot et al. (2018) argue that because an older individual might have a slower reaction to an external threat, his or her subjective world might be perceived as less safe. However, older age is a period of high heterogeneity with functional, psychosocial, and lifestyle aspects differing substantially from one individual to another (Diehl & Wahl, 2020; Ferraro, 2018). These differences in ageing might explain the presence of distinct perceived reasons for feeling unsafe among older adults. Moreover, from the environmental point of view, perceived unsafety differs in the outside and inside home contexts. Older adults tend to feel more unsafe in their neighbourhoods compared to at home (Allik & Kearns, 2017), and feeling fearful in the neighbourhood and at home is associated with different socio-demographic, lifestyle, and other personal characteristics of older adults (Bazargan, 1994). Hence, these contexts must be considered when examining different reasons for feeling unsafe as the neighbourhood and home contexts of unsafety might be also sources of different perceived reasons of unsafety.

Reasons for feeling unsafe

Overall, attention to various factors explaining feelings of unsafety is increasing in recent literature (Valente & Valera Pertegas, 2018). Multiple domains of unsafety in the neighbourhood context and their differential associations with health and physical activity indicators in older adults were previously outlined (Won et al., 2016). In the current study, following the propositions of GUTS (Brosschot et al., 2017, 2018), we approach feelings of unsafety as a subjective experience which can be triggered by various sources of unsafety and which could differ depending on the home or neighbourhood context. Based on the description of the key life domains addressed by this theory, we hypothesize the presence of the following perceived reasons of unsafety relevant among older adults: Fear of crime, health limitations, infrastructure problems, and social climate. Namely, the fear of crime factor corresponds to the compromised context domain, health limitations factor corresponds to the compromised body domain, and the social climate factor corresponds to the compromised social networks domain. Moreover, we considered the ageing context in unsafety experience (Wahl & Gitlin, 2019) and included Infrastructure problems (e.g., inconvenient staircases, difficult thresholds, lack of lighting, etc.) as an additional factor which reflects both compromised context and body domains.

Fear of crime

Perceived unsafety in the physical environment may result from a perceived threat of being victimized. If an older adult perceives the living environment as unsafe, this might reflect the compromised context (Brosschot et al., 2018) in which a real or presumed criminal threat creates the feelings of unsafety. Fear of crime in older adults has been researched extensively while considering its affective, cognitive, and behavioural facets (Greve et al., 2018); gender aspects (Lee et al., 2020; Pain, 1995); negative mental health consequences (Beaulieu et al., 2004; Olofsson et al., 2012), etc. In this line of research, older adults’ feelings on unsafety have been explained mainly as a consequence of greater vulnerability for victimization (Farrall et al., 2009; Hale, 1996). Although fear of crime might be one important explanation for feeling unsafe in advanced age, the lack of studies including other relevant factors might lead to overestimating its importance. In line with current models on unsafety (Valente & Valera Pertegas, 2018; Valente et al., 2020), this study addresses other potential reasons for feelings of unsafety over and above fear of crime.

Health limitations

Another source of perceived unsafety can develop due to ageing related bodily changes and illnesses (Brosschot et al., 2018). When researched from the physical vulnerability perspective in fear of crime (Henson & Reyns, 2015), health limitations have been linked to fear of crime experience (Chadee et al., 2017; Cossman & Rader, 2011; Stiles et al., 2003) and are considered especially important for older adults (Ceccato & Bamzar, 2016). However, diminished health of older adults is known to increase vulnerability to overall environmental risks (Oswald et al., 2011). In advanced age, impaired vision (Kappes et al., 2013; Wahl & Gitlin, 2019), instability and involuntary falls (Peel et al., 2000) or diminished mobility might lead to perception of the daily environment as less safe. Therefore, in this study, we conceptualize health limitations as a source of unsafety which might be independent from fear of crime.

Infrastructure problems

Being a reflection of a compromised physical environment (Brosschot et al., 2018), such inconvenient infrastructure elements as lack of light in the neighbourhood, difficult staircases in the building, or problematic thresholds at home might be an external source of unsafety for older adults. In a qualitative study, older adults described their loss of safety following a fall accident and their need for a safer environment at home (design, furniture, doorsteps, etc.) as a frequent reason for relocation to a retirement home (Fonad et al., 2006). In both outside and inside home situations, when physical environment is perceived as inconvenient, feelings of unsafety might be triggered.

Social climate (for feelings of unsafety in the neighbourhood only)

When social climate in the neighbourhood is perceived as unattractive due to experienced unfriendliness of others, perceived social exclusion or lack of feeling of belonging to a community, social context is at risk of being compromised resulting in experienced unsafety (Brosschot et al., 2018). On the one hand, perceived social environment and antisocial behaviour of others is important in explaining fear of crime (Farrall et al., 2007; Lorenc et al., 2012) and feelings of unsafety (De Donder et al., 2012; Valente et al., 2020). On the other hand, social connectedness in their neighbourhoods is very important for older adults’ feelings of safety while ageing in place (Wiles et al., 2012). Therefore, lack of these social elements is hypothesized as a perceived reason for unsafety while ageing.

Current study

The current study addresses the important knowledge gap in research on feelings of unsafety in older adults by studying the unique contribution of various perceived reasons of unsafety to the feelings of unsafety in the neighbourhood and at home simultaneously. Because higher age, female gender, lower education and financial status, and living alone are considered vulnerability factors for feelings of unsafety (Henson & Reyns, 2015; Killias & Clerici, 2000), these socio-demographic characteristics are controlled for in the study. Additionally, we control for mobility outside of the apartment because this life-style characteristic was previously linked to feelings of unsafety (Hale, 1996; Stjernborg, 2017). We also control for the amount of years lived in the area as this characteristic might reflect one’s rootedness in the area which was negatively associated with fear of crime in previous research (McGarrell et al., 1997). This study aims to answer the following research questions:

1. Are there associations between feelings of unsafety and perceived reasons of unsafety with socio-demographic variables? While we expect the associations between feelings of unsafety and socio-demographic variables to reflect the vulnerability hypotheses assumptions (described above), the perceived reasons of unsafety might be differentially related to different socio-demographic variables. Specifically, we expect fear of crime, unsafety due to health limitations and due to infrastructure problems to be higher for older participants (Olofsson et al., 2012; Wahl & Gitlin, 2019) and those living alone (Fonad et al., 2006; Killias & Clerici, 2000), and women to be more fearful of crime compared to men (Collins, 2016). Lower levels of education and lower financial status are hypothesized to be associated with more fear of crime (Henson & Reyns, 2015) and more health-related unsafety (Tesch-Römer & Wahl, 2017). We expect all perceived reasons for unsafety to serve as barriers for going out and thus to be associated with lower mobility outside of the apartment. However, this hypothesizing is rather tentative because of the lack of previous research on multiple reasons of unsafety in advanced age (Won et al., 2016).

2. What is the unique contribution of various perceived reasons of unsafety to the feelings of unsafety in the neighbourhood and at home, while controlling for socio-demographic variables? In line with GUTS (Brosschot et al., 2017), threats of crime and health limitations may be central to potentially compromised life domains of an older person. Thus, tentatively, fear of crime and health limitations as perceived reasons of unsafety are expected to have stronger unique associations with feelings of unsafety, compared to the inconvenient infrastructure and unattractive social climate.

Method

Participants and procedure

Data for the current study come from a cross-sectional survey study, the 65+ and Safe Study, which examined safety, fear of crime, and health among independently living older people in Sweden. Data collection was carried out in May – September 2019. Inclusion criteria for the study participants were formulated as 1) turning 65 years old in the year of 2019, or older; 2) community dwelling in senior apartments owned by a Swedish municipal housing company, and 3) being absent from severe cognitive impairment. The senior apartments are regular rental apartments (not equivalent to a service or nursing home) restricted to people 65 years or older. Most of them are equipped with an elevator, automatic door openers in the entrance, and similar structures adopted for older age. The apartments are spread throughout the municipally, located in rural and non-rural areas. Apartments are diverse in size and located in neighbourhoods with varying socioeconomic status.

The target sample for the 65+ and Safe Study was identified through the Swedish Population Register provided by the Swedish Tax Agency. From the list of registered residents received from the Agency, 1299 residents fulfilled the inclusion criteria of the study. However, the final target sample was reduced to 1256 due to reasons such as residents had moved or deceased. To announce the study, postcards were sent out inviting these older adults who meet the inclusion criteria to participate in the study. The postcards were followed by posted paper questionnaires along with the information letter containing detailed description of the study, central ethical considerations, and the research group contacts. It also contained a pre-paid response envelope. The information letter included instructions and individual log-in information for answering a web-version of the questionnaire. These letters were followed by reminder letters and phone calls made within the following weeks. The study received approval from the Swedish Ethical Review Agency (Dnr: 2019–02248).

The final study sample comprised 622 responses (587 paper questionnaire responses and 35 online web-questionnaire responses; two participants responded both online and by filling in the paper questionnaire; only their online responses were processed), resulting in a response rate of 49.5%. Study participants were on average 77.6 years old (range from 64 to 106) and 60.6% were women. On average, they lived for 10.22 years in their neighbourhoods (ranging from 0 to 70 years) and 56.6% lived alone. Regarding financial status, 85% indicated that if they were in a situation in which they had to get a hold of 15 000 SEK in one week, they would be able to do so. Sixty-eight per cent had a high school level of education or lower. When asked to indicate their mobility outside of the apartment, 83.9% indicated that they went outside every day. The participating group did not significantly differ from the non-participating group concerning gender, χ2 (2, N = 1237) = 1.17, p = .56. However, the non-participating group had a slightly significantly higher mean age (M = 79.09, SD = 8.12) compared to the participating group (M = 77.6, SD = 7.22), t(1220) = −3.41, p = .001.

Measurements

Feelings of unsafety in the neighbourhood and at home

Feelings of unsafety in the neighbourhood and at home were assessed through one item each. Feelings of unsafety in the neighbourhood were assessed with the question ‘During the last year, did you ever feel unsafe in the area where you live?’ Feelings of unsafety at home were assessed with the question ‘During the last year, did you ever feel unsafe in the apartment in which you live?’ Response options for both items included ‘Never’, ‘Very rarely’, ‘Quite rarely’, ‘Quite often’, and ‘Very often’. Such assessment of feelings of unsafety is more suitable for older adults compared to measures that emphasize being outside alone at night, considering that this may not be a part of a natural lifestyle for many older adults (De Donder et al., 2015).

Perceived reasons of unsafety

After each unsafety question (i.e., in neighbourhood and at home), respondents were presented with the following statement: ‘Please indicate to what extent the following factors make you feel unsafe in your neighbourhood/in your apartment’. The factors included: Fear of crime (e.g., fear of being robbed on a street/fear of a stranger breaking in or having access to the apartment), health limitations (e.g., vision or hearing problems, fear of falling), and infrastructure (e.g., shortcomings in the design of the neighbourhood/apartment, difficult stairs, insufficient lighting, or thresholds). For unsafety in the neighbourhood, social climate (e.g., not attractive social climate in neighbourhood, unfriendly neighbours) was added. The questions thus referred to the actual current situation and not to a hypothetical effect of each of the factors. Also, following the rationale of the GUTS, each of the reasons of unsafety was constructed in such a way that they referred to the overall current situation and did not require a specific stressor or episode of feeling unsafe. Each factor was rated with response options ‘Not at all’, ‘To a minor extent’, ‘To some extent’, or ‘To a great extent’.

Sociodemographic factors

In total, the following seven factors were included as control variables: Age (assessed in years); Gender (1 = women, 2 = men); Education (1 = high school or lower, 2 = education above high school); Living alone (0 = yes, 1 = living with someone); Financial status (assessed with the question: ‘If you were to suddenly find yourself in a situation in which you had to get a hold of 15 000 SEK, would you manage to do so?’, 1 = no; 2 = yes); Years in neighbourhood (assessed in years), mobility outside apartment (assessed with the question: ‘How often do you usually go out of your apartment?’, 1 = less than once a week, 2 = once or twice a week, 3 = 3–4 times per week, 4 = every day).

Analytic strategy

The two variables representing feelings of unsafety in the neighbourhood and at home were not normally distributed with skewness of 1.56 and 2.24 (SE = 0.1), kurtosis of 1.91 and 5.27 (SE = 0.2) respectively. Therefore, non-parametric methods of analysis were applied (Field, 2013). New categorical variables for feelings of unsafety in the neighbourhood and at home were created. Those who responded ‘Never’ or ‘Very rarely’ were classified as feeling safe, and those who responded ‘Quite rarely’, ‘Quite often’, and ‘Very often’ were classified as feeling unsafe.¹

To answer the first research question, Spearman’s correlation established the relationship between feelings of unsafety in the neighbourhood and at home, perceived reasons of unsafety, and socio-demographic factors. To answer the second research question, two binary logistic regression analyses estimated the unique, independent relations of perceived reasons of unsafety (i.e., fear of crime, health limitation, infrastructure, and social climate [only for neighbourhood]) with feelings of unsafety in the neighbourhood and at home, separately, while controlling for seven socio-demographic factors (i.e., age, gender, education, living alone, financial status, years in neighbourhood, and Mobility outside apartment). All analyses were carried out in IBM SPSS Statistics 25.

Results

Results showed that most respondents feel predominantly safe (never or very rarely unsafe) in their neighbourhoods (81.5%) and at home (89.7%) (Table 1). For those who reported unsafety in their neighbourhood, fear of crime (51.9%) followed by health limitations (42.1%) was stated as contributing to their feeling of unsafety to some extent. Fewer respondents stated that social climate in the neighbourhood (20.4%) or infrastructure (17.4%) was a reason for feeling unsafe in the neighbourhood. For those feeling unsafe at home, health limitations (40.7%) followed by fear of crime (33.3%) and inconvenient infrastructure (18.2%) were reported as reasons for feeling unsafe.

Table 1. Descriptive statistics of feelings of unsafety in the neighbourhood and at home

			N	%
Feeling unsafe in the neighbourhood		Never Very rarely Quite rarely Quite often Very often	347 160 48 33 10	55.8 25.7 7.7 5.3 1.6
Feeling unsafe at home		Never Very rarely Quite rarely Quite often Very often	452 106 33 10 2	72.7 17.0 5.3 1.6 0.3

Feelings of unsafety, perceived reasons of unsafety, and socio-demographic factors

Regarding the first research question, results of the Spearman’s correlations (Table 2) show that feelings of unsafety in the neighbourhood are positively associated with years lived in the neighbourhood. Feelings of unsafety at home are not associated with any of the examined socio-demographic variables (i.e., age, gender, education, living alone, financial status, years lived in neighbourhood, and mobility outside apartment). However, the relationship between perceived reasons of feelings of unsafety and socio-demographic factors is more nuanced. In the neighbourhood, older respondents report that they experience more feelings of unsafety due to health limitations and less feelings of unsafety due to unattractive social climate. Regarding gender, women feel more unsafe because of their health limitations and are more fearful of crime than men. Respondents with lower levels of education report feeling more unsafe due to their health limitations and those living alone experience more feelings of unsafety associated with health and with fear of crime. Older adults with higher financial status report less feelings of unsafety due to health and due to unattractive social climate. Respondents who go outside of their apartments more frequently experience less health-related feelings of unsafety.

Table 2. Spearman’s correlations among feelings of unsafety and study variables

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.	13.	14.	15.	16.
1. Feelings of unsafety in the neighbourhood	1
2. Feelings of unsafety at home	,512**	1
3. Age	-,066	-,036	1
4. Gender	-,052	-,065	-,047	1
5. Education	,045	-,064	-,181**	-,009	1
6. Living alone	,077	,060	,191**	-,211**	,032	1
7. Financial status	-,041	-,064	,047	,061	,043	-,092*	1
8. Years lived in the neighbourhood	,132**	,057	,103*	-,074	-,112*	,061	-,051	1
9. Mobility outside of apartment	-,038	-,081	-,199**	,168**	,110*	-,168**	,053	-,036	1
Perceived reasons of unsafety in the neighbourhood
10. Fear of crime	,514**	,341**	-,025	-,101*	-,021	,102*	-,061	,080	-,049	1
11. Health limitations	,095*	,167**	,215**	-,107*	-,115*	,163**	-,105*	,044	-,219**	,294**	1
12. Infrastructure	,120**	,235**	,014	-,021	-,001	,014	-,076	,001	-,073	,202**	,392**	1
13. Social Climate	,412**	,338**	-,129**	-,042	,008	,066	-,089*	,071	,004	,336**	,167**	,302**	1
Perceived reasons of unsafety at home
14. Fear of crime	,475**	,433**	-,008	-,066	-,023	,136**	-,066	,084	-,153**	,593**	,292**	,160**	,348**	1
15. Health limitations	,083	,138**	,228**	-,114*	-,068	,151**	-,104*	,056	-,193**	,285**	,785**	,332**	,149**	,356**	1
16. Infrastructure	,113*	,221**	,011	-,013	,020	,000	-,078	,012	-,044	,145**	,402**	,696**	,282**	,158**	,352**	1

At home, higher age is associated with more feelings of unsafety only due to own health limitations. Similarly, women feel less safe than men only due to their health. In contrast, older adults with higher financial status report less feelings of unsafety due to their health. Older adults living alone experience more health-related unsafety and more fear of crime-related unsafety. Respondents reporting more frequent mobility outside of apartment experience less health-related and fear of crime-related unsafety.

Perceived reasons of unsafety

The second research question examined the unique contribution of various perceived reasons of unsafety to the feelings of unsafety in the neighbourhood and at home while controlling for socio-demographic factors. Two binary logistic regression analyses were performed with feelings of unsafety in the neighbourhood and at home as dependent variables, respectively (Table 3.). The first model in which feelings of unsafety in neighbourhood served as outcome shows that fear of crime and social climate but neither health limitations nor infrastructure significantly and independently predict feelings of unsafety in the neighbourhood. The second model in which feelings of unsafety at home served as outcome shows that fear of crime and inconvenient infrastructure but not health limitations significantly predict feelings of unsafety at home, while controlling for socio-demographic factors.

Table 3. Logistic regression predicting feelings of unsafety in the neighbourhood and at home (controlling for socio-demographic factors)

	B	Wald	P	Odds Ratio	95% C.I. for Odds Ratio
					Lower	Upper
In the neighbourhood ^a
Age	,003	,012	,913	1,003	,953	1,056
Gender	,138	,151	,698	1,148	,573	2,298
Education	-,662	3,535	,060	,516	,259	1,029
Living alone	-,185	,272	,602	,831	,415	1,665
Financial status	,189	,179	,673	1,208	,502	2,908
Years lived in the neighbourhood	,019	1,672	,196	1,019	,990	1,049
Mobility outside of apartment	-,145	,196	,658	,865	,455	1,644
Reasons for unsafety:
Fear of crime	1,742	71,272	<,001	5,707	3,809	8,551
Health	-,363	2,610	,106	,696	,448	1,080
Infrastructure	-,033	,010	,919	,967	,513	1,826
Social climate	1,246	23,543	<,001	3,476	2,101	5,750
Constant	−6,057	5,581	,018	,002
At home ^b
Age	-,037	1,356	,244	,964	,906	1,025
Gender	,355	,553	,457	1,427	,559	3,639
Education	1,079	3,851	,050	2,943	1,001	8,648
Living alone	-,066	,017	,895	,936	,352	2,494
Financial status	,009	,000	,986	1,009	,345	2,951
Years lived in the neighbourhood	,019	1,184	,276	1,019	,985	1,054
Mobility outside of apartment	-,041	,015	,903	,959	,495	1,860
Reasons for unsafety:
Fear of crime	1,893	55,231	<,001	6,637	4,029	10,934
Health	-,270	,813	,367	,763	,424	1,373
Infrastructure	1,597	19,337	<,001	4,938	2,423	10,061
Constant	−6,051	4,645	,031	,002

^aOmnibus test = χ² (11) = 170.98, p < 0.001. Cox & Snell R square = 0.29. Nagelkerke R square = 0.50

^bOmnibus test = χ² (10) = 113.63, p < 0.001. Cox & Snell R square = 0.20. Nagelkerke R square = 0.48

Discussion

The current study investigated the feelings of unsafety, the perceived reasons of unsafety in advanced age, and their associations with the socio-demographic factors. Regarding the first research question, in contrast to our predictions based on the vulnerability approach (Henson & Reyns, 2015), we found no associations between feelings of unsafety at home and socio-demographic variables. The only significant association with feelings of unsafety in the neighbourhood was years lived in the neighbourhood: in contrast to our initial conceptualization, those living longer in a neighbourhood reported more feelings of unsafety. A possible explanation for this, however, could be that those living longer in their neighbourhoods are likely to observe more changes that could provoke the fear of unknown which can be closely linked to feelings of unsafety in advanced age (Ceccato & Bamzar, 2016). An alternative explanation could be in that the years lived in the neighbourhood are also positively associated with age and negatively associated with the level of education, thus the constellation of these factors might explain that those living longer in their neighbourhoods experience more feelings of unsafety.

Further, exploring the associations of socio-demographic factors with perceived reasons of unsafety revealed several nuanced tendencies. Namely, regarding the feelings of unsafety in the neighbourhood, more significant associations were observed between health limitations and socio-demographic variables, compared to other perceived reasons of unsafety. Interestingly, age, education, financial status, and mobility outside of the apartment were related to feelings of unsafety due to health limitations and were not related to fear of crime. Thus, in our study, the factors that were previously linked to fear of crime were related only to health-related unsafety in the neighbourhood. Female gender and living alone were associated with both health limitations and fear of crime as perceived reasons of unsafety in the neighbourhood. This result corroborates previous findings on the role of female gender (Collins, 2016) and living alone (Killias & Clerici, 2000) in fear of crime experience and extends the importance of these factors to the health-related feelings of unsafety. Regarding the situation at home, health limitations were associated with higher age, female gender, living alone, lower financial status, and less mobility outside of the apartment. Fear of crime at home was only associated with living alone and lower mobility outside of the apartment. These results add nuances to earlier research in which different unsafety reasons were seldom explored.

The second research question examined the unique contribution of perceived reasons of unsafety to experienced feelings of unsafety in the neighbourhood and at home while controlling for a range of sociodemographic factors. Results showed that perceived unsafety for older adults in our sample originated mainly from external to the older adult threats – crime, unattractive social climate in the neighbourhood, and inconvenient infrastructure at home. Thus, these specific risks in the socio-physical environment which is especially important in advanced age (Oswald et al., 2011; Wahl & Gitlin, 2019) were shown to contribute to perceived unsafety.

In line with our prediction, fear of crime was significantly uniquely associated with feeling of unsafety for older adults. A perceived threat of crime links together threats to safety in the physical, social, and environmental life domains simultaneously (Brosschot et al., 2018) thus triggering feelings of unsafety. Moreover, unattractive social climate in the neighbourhood played a more central role in feelings of unsafety in the neighbourhood than it was originally hypothesized. However, this finding is in line with previous conceptualizations of unsafety experience. Socially unattractive neighbourhood is a threat to the social context domain, and when it is compromised, perceived unsafety arises (Brosschot et al., 2018). Perceiving other people as potentially threatening has been previously described in the ‘fear of “strangers”’ (Heber, 2011, p. 64) and the ‘fear as the “other”’ (Walklate, 1998, p. 404) discourses. Further, community disaffection contributes significantly to the subjective feelings of unsafety among adults (Valente et al., 2020). These descriptions of social climate-related unsafety differ from fear of crime in that they may include a range of non-crime-related unsafety experiences for older adults. For instance, in a qualitative study, older adults described safety in their residential areas as comprising familiarity, attachment, and social connection (Wiles et al., 2012). Given that most of senior accommodations offer the residents of the apartments some special arrangements for maintaining social community with other residents (e.g., a shared living room in each building, organized activities), evaluation of the social climate as unfriendly may reflect a lack of such connection or belonging together with other residents in our sample. Our study showed that low quality of such social connection on the level of a neighbourhood is an important perceived reason of unsafety independently of fear of crime.

Another independent perceived reason of feelings of unsafety that emerged in the study is inconvenient infrastructure at home. The results showed that home hazards (Wahl & Gitlin, 2019) constitute an important perceived reason of feeling unsafe at home for older adults, thus emphasizing the importance of the support function of the immediate environment for feeling safe while ageing (Wahl & Weisman, 2003). As shown by the study, this compromised context element (Brosschot et al., 2018) triggers feelings of unsafety at home independently and over and above fear of crime.

In contrast to our prediction, feeling unsafe because of health limitations at home and in the neighbourhood, as well as infrastructure-related unsafety in the neighbourhood, did not have a unique association with feelings of unsafety. We believe that one explanation for this could be the sample specifics: most of senior accommodations in which the participants of our study live have improved infrastructure of the buildings and contain some design elements important for older adults. Their needs for health-related safety are more likely to be met by infrastructure developments compared to those older adults living in regular apartments or houses. Moreover, considering the high importance of infrastructure-related unsafety at home and the fact that design limitations might be especially detrimental for those with diminished health (Wahl & Gitlin, 2019), results might implicitly indicate that it is experienced unsafe because of health limitations. Moreover, the correlational analyses revealed that older study participants experience more health-related unsafety. Therefore, feelings of unsafety due to health limitations remain relevant for future investigations of perceived unsafety with a special focus on the oldest old.

Overall, the results of the study showed that most of the respondents felt safe in their neighbourhoods and in their apartments. These results endorse the discourse of older adults as competent actors rather than potentially vulnerable victims’ discourse (Rypi, 2012) and contribute to research that challenged the victimization-fear paradox assumptions regarding older adults (Greve et al., 2018; Pain, 1995; Ziegler & Mitchell, 2003). From the GUTS (Brosschot et al., 2018) point of view, our results might indicate that for most older adults in our sample the crucial life domains remain uncompromised and that advanced age cannot be regarded as an a priori factor for perceived generalized unsafety. Various emotional and behavioural compensation mechanisms may account for optimal functioning in the most important life domains despite possible declines of ageing (Carstensen et al., 2003; Diehl & Wahl, 2020).

The key conclusion of this study is that it identifies the presence of several unique factors associated with feelings of unsafety for older adults. However, for further research, it is important to consider that combinations of specific perceived reasons contributing uniquely to feeling unsafe might be different for different subgroups of older adults (i.e., residing in regular accommodations, in special care settings, etc.). Moreover, the differences in perceived reasons for feelings of unsafety described in our study have important practical implications for creating safe opportunities for ageing in place (Wahl & Gitlin, 2019; Wiles et al., 2012). It is likely that older people experiencing unsafety for different reasons would benefit from different interventions promoting safety. Such interventions would be more targeted when based on profound understanding of leading sources of unsafety for specific individuals and differences in these sources of unsafety in at home and in the neighbourhood situations. For instance, it is likely that older adults experiencing fear of crime would require different interventions compared to those experiencing unsafety due to noncriminal factors. This is important to consider when working on creating safe conditions for residents of senior apartments.

Among the strengths of the present study is the broad measurement of perceived reasons of unsafety and how these reasons contribute to overall feelings of unsafety in the neighbourhood and at home. Including such theoretical and analytic approach is of great importance. Previous research on fear of crime tends to emphasize how fear of being a victim of crime influences people’s sense of safety, whereas health literature might stress the impaired health and infrastructure design as the main vulnerability factors for unsafety in advanced age. All such arguments have empirical support, but none is likely to be a sufficient explanation alone, given that unsafety in late life is a multifaceted concept. Important to notice is that the sample includes a large age range with a high oldest age (i.e., 64–106 years of age). The study also included a range of the socio-demographic factors and showed differential associations of these factors with the perceived reasons of unsafety. The study complements the previously reported findings on differences in fear of crime experience of older adults and extends this heterogeneity to broader feelings of unsafety. In contrast to research aimed at confirming or disconfirming the victimization-fear paradox assumptions by comparing older adults to younger adults, the current study addressed differences within the older adult sample and highlighted the differences among older adults in their perceived reasons of unsafety.

The findings of this study must also be interpreted in the context of its limitations. The study sample includes older adults living in rented senior apartments, therefore, there should be caution in generalizing the results to all community-dwelling older adults due to possible differences in feelings of unsafety between these older adults and those owning their property. Additionally, the study excluded non-Swedish speaking residents because the questionnaire was available only in Swedish and English languages. Further, the survey questionnaire was introduced to the prospective participants as a scientific study on safety and fear of crime. From a symbolic paradigm perspective on unsafety, representations and communication play an important role in perceived unsafety (Elchardus et al., 2008). Hence, our presentation of the study, namely, describing it as a study on safety and fear of crime, could possibly prime respondents with crime-related associations, even when responding to questions not explicitly mentioning crime or fear of crime. Finally, the question on perceived reasons of unsafety contained a fixed number of statements constructed by researchers, referring to the neighbourhood and at home situations. This list could be further extended with unsafety due to traffic which was shown to be related to physical activity outcomes for older adults (Won et al., 2016). Alternatively, posing open questions on potential perceived reasons of unsafety within a qualitative approach could further our understanding of phenomenology of feeling unsafe in advanced age.

Conclusion

The current study investigated the feelings of unsafety and perceived reasons of unsafety in advanced age and revealed the presence of distinct perceived reasons of unsafety and their differential associations with socio-demographic variables and with general feelings of unsafety. Overall, the study demonstrated that fear of crime, social climate in the neighbourhood, and the quality of infrastructure at home are important factors associated with feelings of unsafety in advanced age. The findings allow to conclude that perceived unsafety is a complex phenomenon and that various sources of unsafety, including those not related to criminal threat, contribute to feeling unsafe both at home and in the neighbourhood for older adults. Hence, the results imply that greater attention should be given to multiple reasons for unsafety in advanced age. Further research might take into account the presence and importance of these different perceived reasons of unsafety for older adults.

Acknowledgments

We would like to thank Örebro bostäder AB (ÖBO) for their assistance with forming the target sample. We would also like to thank Dr. Jessie Dezutter and the Meaning and Existence research group at KU Leuven (Belgium) for the valuable methodological contribution while designing this paper. This study was accomplished within the context of the Swedish National Graduate School for Competitive Science on Ageing and Health (SWEAH) funded by the Swedish Research Council.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The study was completed within the Newbreed doctoral program – the project that has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement No 754285.

Notes

1. The decision for dichotomizing the dependent variable was guided by the content of the response options as well as by aiming to retain the statistical power. When considering an alternative – classifying only those who feel ‘Often’ or ‘Very often’ unsafe as the unsafe group – the size of this unsafe subgroup reduces considerably. However, we performed the analyses with this dichotomizing option and the results showed no changes in the direction and in patterns of significance. Given these considerations, we report the results of the analyses with the dependent variable dichotomized as presented in the Analytic Strategy above.