Discussion and conclusions

We examined the correlates of food insecurity among older people in Canada using data from the 2013–2014 CCHS. Generally, our findings show that seniors in lower age categories were more likely to be food insecure compared to those in higher age categories. Some socio-economic variables including age, visible minority status, province of residence, income, living arrangement, sense of community belonging, and self-rated physical and mental health were significant predictors of food insecurity among elderly people. These findings suggest that food insecurity among elderly people is complex and requires an in-depth appreciation of a range of socio-economic and demographic factors. Recent research on the population-level determinants of food insecurity in Canada revealed the important role of province of residence, income, housing tenure, education, Aboriginal status and household structure (see Tarasuk et al., Reference Tarasuk, St-Germain and Mitchell2019). While population-level factors such as income, province of residence and visible minority status are consistent determinants of food insecurity among older adults, this study also demonstrates that living arrangement, age cohort, sense of community belonging, and self-rated physical and mental health are uniquely associated with food insecurity among elderly people.

Generally, the literature on food insecurity demonstrates a link between old age and food insecurity across several contexts (Lee and Frongillo, Reference Lee and Frongillo2001; Temple, Reference Temple2006; Fernandes et al., Reference Fernandes, Rodrigues, Nunes, Santos, Gregório, de Sousa, Dias and Canhão2018). However, a key gap that remains less understood is how food insecurity may vary across age categories among elderly people. By disaggregating elderly people into sub-categories, our study contributes to the literature by demonstrating that seniors in lower age groups, i.e. between 64 and 75, were more likely to be food insecure compared to those in relatively higher age categories. Thus, our study builds on earlier evidence on the association between old age and food insecurity by demonstrating that even among elderly people, certain age categories may be more susceptible to food insecurity. This finding corroborates earlier findings that suggest a protective age-cohort effect in food insecurity among elderly people, whereby seniors in older age categories tend to receive better attention from the state, such as being prioritised in food-assistance initiatives (Duerr, Reference Duerr2007; Dean et al., Reference Dean, Sharkey and Johnson2011). Ensuring a holistic reduction in food insecurity would entail paying more attention to relatively younger seniors who are more likely to be at risk of food insecurity.

Regarding living arrangement, the finding that elderly persons living alone are more likely to be food insecure is consistent with existing literature (Woo, Reference Woo2000; Lee and Frongillo, Reference Lee and Frongillo2001; Temple, Reference Temple2006; Dean et al., Reference Dean, Sharkey and Johnson2011; Kim et al., Reference Kim, Park and Kim2019; Park et al., Reference Park, Saint Ville, Schwinghamer and Melgar-Quiñonez2019). Familial social capital plays a crucial role in food accessibility and utilisation among elderly people given that the presence of a spouse at home provides a safety net against food insecurity in several ways (Park et al., Reference Park, Saint Ville, Schwinghamer and Melgar-Quiñonez2019). Aside from providing daily assistance in grocery shopping and food preparation (Ishikawa et al., Reference Ishikawa, Yokoyama, Nakaya, Fukuda, Takemi, Kusama, Yoshiike, Nozue, Yoshiba and Murayama2016), spousal support has also been found to be crucial in improving access to resources for obtaining food in the right quantity and quality, and providing the companionship and emotional support necessary for regular eating (Wolfe et al. Reference Wolfe, Olson, Kendall and Frongillo1996; Dean et al., Reference Dean, Sharkey and Johnson2011). The buffering role of spousal support on food insecurity for elderly people is even pronounced in situations of functional impairment (Unger et al., Reference Unger, McAvay, Bruce, Berkman and Seeman1999).

Our finding that elderly people in lower-income categories are more likely to be food insecure is also consistent with existing literature (Lee and Frongillo, Reference Lee and Frongillo2001; Guthrie and Lin, Reference Guthrie and Lin2002; Russell et al., Reference Russell, Flood, Yeatman and Mitchell2014; McIntyre et al., Reference McIntyre, Dutton, Kwok and Emery2016; Fernandes et al., Reference Fernandes, Rodrigues, Nunes, Santos, Gregório, de Sousa, Dias and Canhão2018). Income plays a direct role in ensuring both the accessibility and utilisation of food. Apart from improving purchasing power, income sufficiency also affects the quality and desirability of food. Thus, seniors with sufficient income can choose between meals compared to those in lower-income categories who may have to eat low-cost meals or meals they do not desire (Guthrie and Lin, Reference Guthrie and Lin2002; Russell et al., Reference Russell, Flood, Yeatman and Mitchell2014; Park et al., Reference Park, Saint Ville, Schwinghamer and Melgar-Quiñonez2019). For instance, Guthrie and Lin (Reference Guthrie and Lin2002) find that in the United States of America (USA), lower-income elderly people tend to consume significantly fewer calories and servings from major Food Guide Pyramid food groups compared to elderly people in higher-income categories. Moreover, for elderly people who are living alone, those with sufficient income can employ someone to help in their general up-keep, including cooking. Closely related to income is the province of residence. Seniors in Prince Edward Island, Nova Scotia, New Brunswick, Ontario, Saskatchewan and Alberta had higher odds of being food insecure compared to their counterparts in Quebec. This finding may be explained by provincial differences in economic conditions and social support for seniors. As observed by previous studies, seniors living in Quebec may have lower odds of reporting food insecurity due to the relatively generous social programmes offered by the province (Tarasuk et al., Reference Tarasuk, St-Germain and Mitchell2019).

The association between having a weak sense of community belonging and food insecurity among elderly people is also supported by previous research (Locher et al., Reference Locher, Ritchie, Roth, Baker, Bodner and Allman2005; Ogg, Reference Ogg2005; Dean et al., Reference Dean, Sharkey and Johnson2011). Aside from familial social capital as expressed in living arrangement, the role of collective social functioning at the community level in providing fall back for elderly people who are food insecure has been acknowledged. For instance, Dean et al. (Reference Dean, Sharkey and Johnson2011) found that elderly people with low community social capital were more likely to report being severely food insecure compared to those with strong community ties. This highlights the importance of creating avenues for community-level engagement among elderly people.

Visible minority status was also found to be positively associated with food insecurity among elderly people. This finding is consistent with the literature on food insecurity (Sharkey and Haines, Reference Sharkey and Haines2001; Bengle et al., Reference Bengle, Sinnett, Johnson, Johnson, Brown and Lee2010; Leroux et al., Reference Leroux, Morrison and Rosenberg2018) and may be explained by experiences of discrimination, especially with food assistance programmes. Indeed, Locher et al. (Reference Locher, Ritchie, Roth, Baker, Bodner and Allman2005) found a significant relationship between experience of discrimination and food insecurity among older adults who were black. The association between food insecurity and visible minority status may also be explained by the relatively low-income status of most visible minorities in Canada (Maroto, Reference Maroto2016), given the role income plays in ensuring adequate access to food. In an analysis of black–white differences in nutritional risk in the USA, Sharkey and Haines (Reference Sidenvall, Nydahl and Fjellström2001) find that older adults who are black were four times more likely to be at the highest level of nutritional risk, with the odds increasing further when economic need was taken into consideration.

The association between self-rated physical and mental health and food insecurity is largely consistent with the literature (Choi et al., Reference Choi, Park, Kim and Jang2004; Quine and Morrell, Reference Quine and Morrell2006; Green-LaPierre et al., Reference Green-LaPierre, Williams, Glanville, Norris, Hunter and Watt2012). The role of physical health on food insecurity among older people may be explained by the limiting role of poor physical functioning on their food access and utilisation, particularly the ability to go grocery shopping and prepare appropriate meals, especiaqlly those living alone (Wylie et al. Reference Wylie, Copeman and Kirk1999; Sidenvall et al., Reference Sidenvall, Nydahl and Fjellström2001; Burns et al., Reference Burns, Bentley, Thornton and Kavanagh2011). For instance, in an assessment of the dietary intake of older adults, Wylie et al. (Reference Wylie, Copeman and Kirk1999) found that seniors living with restricted mobility consumed vegetables and cooked meals less frequently. Health care may compete with food budgets, especially for elderly people who are more prone to age-related health complications. However, using the 2012 round of the CCHS, Leroux et al. (Reference Leroux, Morrison and Rosenberg2018) found no significant association between self-rated health and food insecurity among older Canadians. This variation in our findings further reinforces the fact that food insecurity and its associated determinants are not static. This reinforces the need to explore continually the determinants of food insecurity among older populations. Further, it is argued that poor mental health conditions such as depression, cognitive decline and anxiety decrease people's ability to access and utilise food resources, and this is particularly important for older adults who are more at risk of poor mental health conditions (Pierce et al., Reference Pierce, Sheehan and Ferris2002; Brostow et al., Reference Brostow, Gunzburger, Abbate, Brenner and Thomas2019). In the USA, for instance, Brostow et al. (Reference Brostow, Gunzburger, Abbate, Brenner and Thomas2019) found that seniors with mental illness had higher odds of being food insecure compared to their counterparts with no mental health conditions. The relationship between health and food insecurity can also be bidirectional. Indeed, there is evidence that food insecurity can shape physical and mental health (Darling et al., Reference Darling, Fahrenkamp, Wilson, D'Auria and Sato2017; Tuthill et al., Reference Tuthill, Sheira, Palar, Frongillo, Wilson, Adedimeji, Merenstein, Cohen, Wentz and Adimora2019; Leung et al., Reference Leung, Kullgren, Malani, Singer, Kirch, Solway and Wolfson2020). In a recent study among older adults in the USA, Leung et al. (Reference Leung, Kullgren, Malani, Singer, Kirch, Solway and Wolfson2020) found that food insecurity was positively associated with multiple chronic conditions and lower self-reported health status.

Although these findings provide useful insights for understanding food insecurity among elderly people, there are some noteworthy limitations. The cross-sectional nature of the data limits our findings to associations, for which reason causal inferences cannot be made. Moreover, given that the experience of food security among elderly people may be temporal, a longitudinal analysis of food insecurity, particularly using a monthly interval, may be useful. Also, the experience of food insecurity is subjective. As a result, future research could benefit from using a qualitative approach to uncover the lived experiences of elderly people. The dataset for this analysis is not representative of all provinces in Canada, given that some provinces did not implement the module on food security in the 2013–2014 round of the CCHS. Moreover, although Canadians living on reservations tend to experience higher rates of food insecurity, they were not included in the CCHS. It is important to mention that some of the provinces that did not implement the food security module are known to have some of the highest food insecurity rates in Canada (Tarasuk et al., Reference Tarasuk, St-Germain and Mitchell2019)

Notwithstanding these limitations, this study provides useful insights for addressing food insecurity among elderly people in Canada and similar contexts. Our findings suggest that food insecurity among elderly people is associated with a multiplicity of factors. As a result, approaches that go beyond poverty alleviation should be prioritised. In Canada, and many other Global North countries, poverty reduction strategies have featured prominently in policy responses for reducing food insecurity among elderly people. Canada's main approach for ensuring the welfare of elderly people has largely been an income improvement approach through the income support programs such as the OASP and the CPP (Statistics Canada, 2014). While these strategies are timely, especially for low-income elderly people, our findings suggest a need to pay attention to other socio-economic factors, such as sense of belonging, age, minority status, state of physical health and living arrangement. Indeed, the impact of these income-related programmes can be enhanced if attention is paid to age-specific nuances and other socio-economic predictors of food insecurity.