Household Food Insecurity Indicators in Rural Aotearoa New Zealand: Patterns, Protections, and Persistent Inequities

Abstract

Food insecurity is a complex phenomenon in high income countries. While there is food security information about Māori, Pasifika, and children in Aotearoa New Zealand (Aotearoa), there is little known about food security by geographic area. Using data from the 2019/2020 New Zealand Health Survey, we report on a set of eight statements relating to having sufficient resources to provide food for all members in the household, by rurality. We further investigate these measures stratified by socio‐demographic variables and by rurality. Results showed there are households experiencing food insecurity in rural Aotearoa. Results further showed that food resource insufficiency was less prevalent in rural households than in nonrural households across almost every variable of age, gender, and deprivation. A lower proportion of rural Māori households faced concerns of resource insufficiency than nonrural Māori households, yet rural Māori still experience rates of resource insufficiency higher than nonrural non‐Māori for each metric measured. Our results highlight ethnicity as an important marker of food insecurity and support views expressed by public health experts calling for policies and initiatives to address food resource insufficiency that results from structural racism.

1. Introduction

Aotearoa New Zealand (Aotearoa) is a net food producer, with more than enough food produced to feed everyone who lives here (Rush and Obolonkin 2020). Yet, accessing sufficient healthy, nutritious, affordable, and culturally appropriate food is a challenge faced by many people in Aotearoa. The New Zealand Health Survey reports that in 2023/24, 27% of children lived in houses where food often or sometimes runs out; this was 34% for tamariki Māori (Māori children) and 55% for Pacific children (Ministry of Health 2024). Similar inequities were reported for eating less because of cost; 26% for the total population of children, 33% for tamariki Māori and 53% for Pacific children.

Issues of resource insufficiency and food insecurity for Māori in Aotearoa are rooted in the systematic dispossession of land, subsequent economic marginalization, and the deliberate acts by successive state governments to deny Māori the equity promised in the Treaty of Waitangi | Te Tiriti o Waitangi (Beavis et al. 2019; McKerchar et al. 2024; Walker 1990). This expropriation of lands and freshwater and marine waterways severed Māori communities from mahinga kai | traditional food sources and systems and associated practices and values (McKerchar et al. 2015; Moeke‐Pickering et al. 2015; Revington 2017). Along with severed access, the intergenerational transmission of mātauranga Māori | Indigenous knowledge around natural systems, resource use, and food security, was also subsequently disrupted (Kawharu et al. 2024; Smith and Hutchings 2024). The legacies of dispossession and disruption continue to drive food insecurity for Māori.

Rural communities, vital for food production, ironically also face food insecurity challenges. Geographic isolation, limited retail and transport options, and longer distances to larger, more affordable supermarkets or discount stores contribute to these issues (Minister of Health 2023). Limited food availability affects access to culturally appropriate foods, particularly impacting migrant groups in rural communities looking to preserve and maintain their cultural food traditions (Minister of Health 2023). Food and nutrition security are issues for those on modest or fixed incomes in rural communities, such as families with children and the elderly (Minister of Health 2023). High fuel costs and fewer charitable food assistance options may exacerbate financial pressures. Additionally, those without discretionary income cannot take advantage of cost‐saving measures like buying in bulk. These factors combined create a complex web of food insecurity challenges unique to rural communities, impacting both their economic stability and overall well‐being.

1.1. Framing Rurality

A problematic deficit framing around rurality ‘implies a rural deficit and leads to unhelpful assumptions and generalizations about rural populations’ (Eggleton 2024, p. 230). This framing operates through discourse and rhetoric that positions urban as the norm and rural as ‘other’ (ibid, p. 230) with often ‘racialised, colonial, ableist, and gendered aspects’ (ibid).

Binary rural–urban categories homogenize diverse rural people and communities, overlooking the complexity and internal diversity within all communities. According to Gaventa's theory of power, rural institutions are constrained through lack of control over establishment and funding, exclusion from agenda setting, and inability to shape discourse and influence social forces (Eggleton 2024). Reframing rural inequity as a critique of urban privilege shifts attention to where power is held and enables disruptive opportunities for rural‐led solutions. Accepting the criticism that food insecurity research often has an urban‐centric focus (Eggleton 2024), we acknowledge our own prior geographic narcissism (Fors 2018) and note the approach taken in this study centers rural communities, particularly rural Māori in Aotearoa.

1.2. Rural Health in Aotearoa

People in rural areas have poorer health outcomes than urban areas (Walker et al. 2024) and consistently higher overall mortality rates than nonrural populations (Crengle et al. 2022; Nixon et al. 2023). This may be due in part to socioeconomic challenges (Minister of Health 2023) or difficulties with access to health services (Minister of Health 2023). The excess all‐cause and amenable mortality experienced by rural Māori, compared to urban Māori, makes visible the additional challenges associated with living rurally in Aotearoa for Māori (Crengle et al. 2022). However, the lack of research and data on rural food access contributes to uncertainty around best ways to prioritize and utilize resources to improve rural health (Minister of Health 2023).

1.3. Rural Food Insecurity

Community‐based food systems are systems that integrate food production, processing, distribution, retail, consumption, and waste in ways that enhance the environmental, economic, social, and nutritional health of a particular place. To be well, whānau in rural communities tell us they need ‘healthy food, places to exercise and socialize, a healthy home and positive relationships’ (Participant voice, Ministry of Health 2023, p. 15). Around two‐thirds of rural‐based Māori students receive school lunches daily through the Ka Ora Ka Ako program because they attend schools in the least advantaged communities (Minister of Health 2023).

Food insecurity and the resulting health inequities faced by Māori is a relatively recent phenomenon and has arisen primarily as a consequence of colonization; a recent study identified displacement from whenua, loss of rangatiratanga, and suppression of mātauranga Māori as key factors which have influenced food security recently (Shelling and Te Morenga 2026). Rural communities benefit from other collective initiatives, such as iwi and hapū indigenous gardening practices (Minister of Health 2023). Although there is a lack of research around rural food insecurity, for rural communities there may be other protective factors that draw from mātauranga Māori and mahinga kai | traditional food sources and systems, and associated practices and values. Self‐reliance activities such as hunting, fishing, and growing fruit trees and vegetables, and resource stretching are common strategies used by rural dwellers to provide food security (Buck‐McFadyen 2015; Urlich et al. 2024). Thus, food security in rural communities is complex, with many interrelated factors.

The relationship between rurality and food insecurity is complex. In their synthesis of heterogeneous studies, Carter et al. (2014) found an inverse association between rural living and food insecurity in seven of eleven studies, uncovering a ‘potential protective effect of rural living on food insecurity’ (p. 97). While three studies found no association, one study reported a positive association between rural living and food insecurity, highlighting the context of rural food insecurity and suggesting the influence of many intersecting factors.

Rural communities often face a paradoxical situation—lands of plenty with agricultural abundance alongside food insecurity. As Hossfeld and Mendez (2018) observed in rural Mississippi, being an agricultural state does not guarantee food security when production focuses on commodities rather than local food systems. This paradox manifests globally, including in Aotearoa, where food insecurity disproportionately impacts whānau Māori despite the country's agricultural leadership (Ministry of Health 2023). Alongside systemic barriers to food security in rural areas, such as high transportation costs and limited competition and few local options, whānau Māori in rural areas may experience food insecurity due to disconnection from whenua | lands and wai | freshwater and marine waterways and disrupted access to mahinga kai | traditional food sources and systems.

Several studies report on food access for rural and small‐town Māori (Moeke‐Pickering et al. 2015; Smith and Hutchings 2024; Stein et al. 2018; Urlich et al. 2024). A study of food insecure mothers demonstrates the need for resilience and resourcefulness in their survival strategies to meet food access needs and reduce the severity of food insecurity experienced by their whānau (Urlich et al. 2024). Time consuming food access activities, reacting to unexpected costs and unstable housing contributes to the stress and anxiety of food insecurity for rural Māori māmā | mothers in Aotearoa (Urlich et al. 2024).

Given this background, the aim of our study is to describe levels of food insecurity in rural Aotearoa and investigate socio‐demographic determinants of food insecurity for rural communities.

2. Methods

Data on food insecurity were obtained from the 2019/2020 New Zealand Health Survey (Ministry of Health 2020c). This is a national survey that has been continually in the field since 2011, with annual updates released periodically. However, questions about food insecurity are not asked each year, and these are often asked in the children's not the adult's survey, with a considerably smaller sample size. Analysis was therefore restricted to the adult 2019/20 wave. The NZHS employs a complex survey design; full details for the 2019/20 wave are available (Ministry of Health 2020a). The response rate in 2019/20 was 75% (Ministry of Health 2020b), but the overall sample size was lower than previous and subsequent years due to the first Covid‐19 related lockdown, which occurred in Aotearoa from April to May 2020, during which recruitment was paused. Statistics New Zealand (StatsNZ) provided access to the survey confidentialised Unit Record Files (CURFs), that is, individual level data that have been modified to protect confidentiality. The 2019/20 NZHS was approved by the multiregion Ethics Committee (MEC/10/10/103), and our analysis of the anonymised data was approved by the Victoria University of Wellington Human Ethics Committee (2024/HE000310).

Food insecurity is measured using a set of eight statements relating to having sufficient resources to provide food for all members in the household (Ministry of Health 2020a; Parnell and Gray 2014), see Table 1. Each statement uses a three‐point response scale – ‘often’, ‘sometimes’, ‘never’. The responses relate to the respondents’ experience of household food insecurity in the past year. For the purposes of this analysis, we have categorized responses of items 2–8 as ‘often’ or ‘sometimes’ as contributing to a degree of food insecurity, and “never” as indicating a food secure household. The first question is phrased in reverse (‘we can afford to eat properly’); food insecurity is defined as people who answered ‘sometimes’ or ‘never’ to this question. These were analyzed as a single combined unweighted index, and as individual items.

TABLE 1.

Measurement of food insecurity in the New Zealand Health Survey.

Item	Description
1	We can afford to eat properly
2	Food runs out in our household due to lack of money
3	We eat less because of lack of money
4	The variety of foods we are able to eat is limited by a lack of money
5	We rely on others to provide food and/or money for food, for our household, when we don’t have enough money
6	We make use of special food grants or food banks when we do not have enough money for food
7	I feel stressed because of not having enough money for food
8	I feel stressed because I can’t provide the food I want for social occasions

Rurality was measured using the Statistical Standard for Geographic Areas 2018 (Statistics NZ, 2017), based on home address mapping to one of Rural settlement/Rural Other/Other. The other categories (Major Urban Area, Large Urban Area, Medium Urban Area, and Small Urban Area) were together considered a nonrural. Area‐level deprivation was measured using the New Zealand Index of Deprivation 2018 (NZDep2018) and analyzed in quintiles. Other socio‐demographic variables were recorded in the NZHS. Self‐identified ethnicity was analyzed as Māori or non‐Māori. Age was analyzed in four categories: 15–24 years, 25–44 years, 45–64 years, and 65 years and over.

Due to the complex survey design, all analyses used the survey weights provided by StatsNZ. Weighted methods were used to calculate proportions of food insecurity. Logistic regression, with survey weights applied, was used to estimate crude odds ratios (OR) and associated 95% confidence intervals (CIs), as well as adjusted ORs (aOR), adjusting for socio‐demographic measures. Tests of statistical interaction were used to determine differential effects of rurality on food insecurity by Māori ethnicity. p‐values from Wald tests of the interaction term are reported.

3. Results

A total of 9 699 people were included in the analysis, of whom 1 344 (14%) lived rurally. People living rurally tended to be older, were more likely to be Māori, and less likely to live in areas of deprivation (Table 2). For the total population, 19% of households reported being food insecure on at least one metric. This varied from 4.2% of people who utilized formal support such as special food grants or food banks to 17.5% of people who reported financially compromised diet variety (Table 3).

TABLE 2.

Weighted distribution of participants by rurality.

	Rural		p‐value
	Yes	No
Gender
Female	51.2%	50.7%	0.800
Male	48.8%	49.3%
Age Group
15–24	10.4%	17.1%	<0.001
25–44	28.8%	34.7%
45–64	40.1%	29.5%
65+	20.6%	18.8%
Māori
Yes	16.7%	13.8%	0.023
No	83.3%	86.2%
Deprivation quintile
Least deprived	25.0%	19.3%	<0.001
2	25.4%	19.5%
3	24.3%	19.5%
4	14.6%	21.0%
Most deprived	10.6%	20.7%

Note: since all analyses were conducted using survey weights, the distribution of participants reflects the general population, not the survey respondents.

TABLE 3.

Food insecurity indicators in the 2019/20 New Zealand Health Survey, stratified by rurality.

	Total	Rural	Nonrural
We cannot afford to eat properlya	12.2 (11.3 to 13.1)	8.1 (6.5 to 10.2)	12.8 (11.8 to 13.9)
Food runs out in our household due to lack of moneyb	10.9 (10.0 to 11.8)	6.4 (4.9 to 8.4)	11.6 (10.6 to 12.7)
We eat less because of lack of moneyb	10.8 (10.0 to 11.8)	6.7 (5.1 to 8.9)	11.5 (10.5 to 12.6)
The variety of foods we are able to eat is limited by a lack of moneyb	17.5 (16.4 to 18.7)	11.3 (9.2 to 13.8)	18.6 (17.3 to 19.9)
We rely on others to provide food and/or money for food, for our household, when we don’t have enough moneyb	6.2 (5.6 to 7.0)	2.8 (1.9 to 4.3)	6.8 (6.0 to 7.7)
We make use of special food grants or food banks when we do not have enough money for foodb	4.2 (3.7 to 4.8)	1.7 (1.0 to 2.8)	4.6 (4.1 to 5.3)
I feel stressed because of not having enough money for foodb	9.7 (8.9 to 10.6)	7.8 (6.0 to 10.0)	10.0 (9.1 to 11.0)
I feel stressed because I can’t provide the food I want for social occasionsb	8.3 (7.5 to 9.1)	5.8 (4.3 to 8.0)	8.7 (7.8 to 9.6)

Note: Figures show the weighted proportion (%) and 95% confidence interval (95% CI).

^a For reporting purposes item 1 has been phrased in reverse; food insecurity is defined as an answer of “sometimes” or “never” in response to the statement “we can afford to eat properly”.

^b Food insecurity is people who answer “often” or “sometimes”.

Food insecurity manifested across rural households in varying degrees of severity (Table 3). Among the most concerning indicators, rural households reported food shortages due to lack of money (6.4%) and financial barriers to food accessibility (8.1%). Additionally, rural households reduced consumption (6.7%) and restricted nutrition (11.3%) because of financial constraints. The psychological impact of food insecurity was notable; money‐related food provision stress affected nearly one in thirteen (7.8%) rural people. One in eighteen rural people reported social occasion‐related food provision stress (5.8%). Interestingly, similar levels of money‐related food provision stress were reported by people in both rural (7.8%) and nonrural (10.0%) households, suggesting that while the frequency of food insecurity differed, its psychological impact is comparable across geographic areas.

3.1. Ethnicity

For each of the eight metrics, Māori were two to four times more likely to report food insecurity than non‐Māori (Table 4). In the total population, there is an ethnicity food accessibility gap comprising: food affordability challenges, food shortages, financially forced food rationing, and economically driven restricted nutrition. These consistent patterns highlight the pervasive nature of ethnic inequities in measures of food security.

TABLE 4.

Food insecurity indicators in the 2019/20 New Zealand Health Survey, stratified by socio‐demographic variables and by rurality.

	Total	Rural	Nonrural
We cannot afford to eat properly a
Māori
Yes	23.6 (20.6 to 26.8)	14.2 (9.6 to 20.5)	25.42 (22.1 to 29.0)
No	10.4 (9.5 to 11.4)	7.0 (5.3 to 9.3)	10.91 (9.9 to 12.0)
Gender
Female	13.1 (11.8 to 14.4)	8.8 (6.4 to 11.9)	13.8 (12.4 to 15.3)
Male	11.3 (10.1 to 12.7)	7.6 (5.4 to 10.6)	12.0 (10.6 to 13.5)
Age Group
15–24	15.1 (12.4 to 18.3)	5.5 (2.4 to 11.9)	16.1 (13.2 to 19.6)
25–44	13.7 (11.8 to 15.7)	12.1 (7.6 to 18.8)	13.8 (11.9 to 16.1)
45–64	12.9 (11.5 to 14.5)	7.8 (5.5 to 10.8)	14.1 (12.5 to 15.9)
65+	7.3 (6.2 to 8.6)	7.2 (4.4 to 11.7)	7.3 (6.1 to 8.7)
Deprivation quintile
Least deprived	5.5 (3.9 to 7.6)	5.4 (2.8 to 10.4)	5.5 (3.7 to 7.9)
2	6.0 (4.6 to 7.8)	5.1 (3.0 to 8.7)	6.2 (4.6 to 8.3)
3	10.4 (8.5 to 12.8)	4.7 (2.8 to 7.8)	11.6 (9.3 to 14.3)
4	15.9 (13.8 to 18.2)	19.9 (13.2 to 28.8)	15.4 (13.2 to 17.8)
Most deprived	24.2 (21.9 to 26.7)	14.0 (9.2 to 20.5)	25.1 (22.6 to 27.8)
Food runs out in our household due to lack of money b
Māori
Yes	24.1 (21.1 to 27.4)	13.04 (8.0 to 20.7)	26.34 (23.0 to 30.0)
No	8.9 (7.9 to 9.7)	5.17 (3.8 to 7.1)	9.39 (8.4 to 10.5)
Gender
Female	11.2 (10.0 to 12.6)	6.3 (4.2 to 9.2)	12.0 (10.7 to 13.5)
Male	10.6 (9.4 to 11.9)	6.5 (4.5 to 9.4)	11.3 (9.9 to 12.8)
Age Group
15–24	16.6 (13.7 to 20.0)	8.2 (3.5 to 18.1)	17.5 (14.3 to 21.1)
25–44	11.0 (9.4 to 12.9)	8.4 (5.0 to 13.8)	11.3 (9.6 to 13.3)
45–64	12.1 (10.7 to 13.7)	6.4 (4.3 to 9.5)	13.4 (11.8 to 15.3)
65+	4.7 (3.8 to 5.7)	4.2 (2.3 to 7.5)	4.8 (3.9 to 5.8)
Deprivation quintile
Least deprived	4.4 (3.0 to 6.5)	4.7 (2.1 to 10.0)	4.4 (2.8 to 6.8)
2	5.2 (3.8 to 7.0)	5.9 (3.4 to 10.0)	5.0 (3.5 to 7.1)
3	8.4 (6.6 to 10.6)	4.7 (2.8 to 7.6)	9.2 (7.0 to 11.8)
4	14.0 (12.0 to 16.3)	11.4 (6.4 to 19.4)	14.3 (12.2 to 16.8)
Most deprived	23.5 (21.2 to 26.0)	9.0 (5.3 to 14.9)	24.8 (22.3 to 27.5)
We eat less because of lack of money b
Māori
Yes	21.5 (18.5 to 24.7)	13.7 (8.1 to 22.2)	23.0 (19.8 to 26.6)
No	9.2 (8.3 to 10.2)	5.5 (4.0 to 7.5)	9.8 (8.8 to 10.9)
Gender
Female	11.4 (10.2 to 12.7)	7.8 (5.5 to 11.0)	12.0 (10.6 to 13.4)
Male	10.4 (9.1 to 11.8)	5.9 (3.8 to 9.0)	11.1 (9.7 to 12.7)
Age Group
15–24	15.2 (12.3 to 18.7)	10.2 (4.3 to 22.3)	15.7 (12.7 to 19.4)
25–44	12.6 (10.8 to 14.7)	8.8 (5.2 to 14.5)	13.1 (11.1 to 15.4)
45–64	11.0 (9.7 to 12.5)	6.7 (4.5 to 10.0)	12.0 (10.5 to 13.6)
65+	5.4 (4.5 to 6.6)	3.9 (2.1 to 7.0)	5.7 (4.7 to 7.0)
Deprivation quintile
Least deprived	4.6 (3.1 to 6.7)	4.4 (1.7 to 11.0)	4.6 (3.0 to 7.0)
2	6.6 (4.9 to 8.8)	4.4 (2.5 to 7.8)	7.0 (5.1 to 9.6)
3	8.4 (6.6 to 10.6)	5.3 (2.9 to 9.5)	9.0 (6.9 to 11.6)
4	13.5 (11.5 to 15.7)	14.1 (8.4 to 22.7)	13.4 (11.3 to 15.8)
Most deprived	22.3 (20.0 to 24.7)	11.4 (7.3 to 17.4)	23.2 (20.8 to 25.8)
The variety of foods we are able to eat is limited by a lack of money b
Māori
Yes	29.6 (26.4 to 33.1)	20.8 (14.4 to 29.1)	31.4 (27.8 to 35.2)
No	17.8 (16.4 to 18.7)	9.5 (7.4 to 12.1)	16.6 (15.4 to 18.0)
Gender
Female	18.3 (16.8 to 19.9)	11.1 (8.4 to 14.5)	19.5 (17.8 to 21.3)
Male	16.9 (15.3 to 18.5)	11.5 (8.5 to 15.3)	17.8 (16.0 to 19.6)
Age Group
15–24	23.8 (20.3 to 27.6)	11.2 (4.8 to 23.9)	25.0 (21.3 to 29.2)
25–44	18.7 (16.5 to 21.1)	15.7 (10.6 to 22.7)	19.1 (16.7 to 21.7)
45–64	18.5 (16.7 to 20.4)	11.9 (8.9 to 15.8)	20.0 (18.0 to 22.1)
65+	10.3 (9.0 to 11.8)	7.3 (4.7 to 11.0)	10.9 (9.4 to 12.5)
Deprivation quintile
Least deprived	9.6 (7.5 to 12.1)	8.2 (4.8 to 13.7)	9.9 (7.5 to 12.8)
2	11.4 (9.2 to 14.0)	9.8 (6.0 to 15.6)	11.7 (9.3 to 14.7)
3	14.7 (12.4 to 17.3)	8.8 (5.7 to 13.2)	15.9 (13.2 to 18.9)
4	20.8 (18.4 to 23.4)	20.3 (13.6 to 29.3)	20.8 (18.3 to 23.6)
Most deprived	32.7 (30.1 to 35.5)	15.8 (10.8 to 22.7)	34.1 (31.3 to 37.1)
We rely on others to provide food and/or money for food, for our household, when we don’t have enough money b
Māori
Yes	13.4 (11.1 to 16.1)	6.7 (3.4 to 12.7)	14.7 (12.1 to 17.8)
No	5.1 (4.5 to 5.9)	2.12 (1.3 to 3.5)	5.6 (4.9 to 6.5)
Gender
Female	6.0 (5.2 to 7.0)	2.2 (1.3 to 3.8)	6.6 (5.7 to 7.8)
Male	6.4 (5.4 to 7.6)	3.4 (1.9 to 5.8)	7.0 (5.8 to 8.3)
Age Group
15–24	11.0 (8.6 to 14.0)	4.1 (1.6 to 9.7)	11.7 (9.1 to 15.0)
25–44	7.0 (5.7 to 8.5)	6.1 (2.9 to 12.5)	7.1 (5.7 to 8.7)
45–64	5.9 (5.0 to 7.1)	2.4 (1.3 to 4.5)	6.7 (5.6 to 8.1)
65+	2.3 (1.7 to 3.1)	0.7 (0.2 to 2.3)	2.6 (1.9 to 3.5)
Deprivation quintile
Least deprived	2.9 (1.8 to 4.7)	0.7 (0.2 to 2.8)	3.4 (2.0 to 5.5)
2	3.0 (2.0 to 4.5)	4.9 (2.4 to 9.7)	2.6 (1.6 to 4.3)
3	4.5 (3.1 to 6.4)	1.1 (0.4 to 3.2)	5.2 (3.5 to 7.5)
4	7.5 (5.9 to 9.3)	3.7 (1.7 to 8.2)	7.9 (6.2 to 9.9)
Most deprived	14.1 (12.3 to 16.3)	5.9 (2.9 to 11.6)	14.9 (12.8 to 17.1)
We make use of special food grants or food banks when we do not have enough money for food b
Māori
Yes	12.1 (10.0 to 14.6)	6.15 (3.2 to 11.6)	13.3 (10.9 to 16.2)
No	3.0 (2.5 to 3.5)	0.87 (0.4 to 1.8)	3.32 (2.8 to 3.9)
Gender
Female	4.2 (3.5 to 4.9)	1.8 (0.9 to 3.5)	4.5 (3.8 to 5.4)
Male	4.3 (3.5 to 5.1)	1.6 (0.8 to 3.3)	4.7 (3.9 to 5.7)
Age Group
15–24	5.5 (3.9 to 7.6)	2.3 (0.8 to 6.2)	5.8 (4.1 to 8.1)
25–44	5.5 (4.4 to 6.9)	4.3 (1.8 to 9.9)	5.6 (4.5 to 7.1)
45–64	4.3 (3.6 to 5.2)	1.2 (0.6 to 2.6)	5.0 (4.2 to 6.1)
65+	1.8 (1.3 to 2.4)	0.4 (0.1 to 1.2)	2.0 (1.5 to 2.8)
Deprivation quintile
Least deprived	0.8 (0.3 to 1.8)	No obs	0.9 (0.4 to 2.2)
2	0.9 (0.5 to 1.6)	0.8 (0.2 to 2.3)	0.9 (0.5 to 1.7)
3	3.4 (2.3 to 4.9)	1.3 (0.6 to 2.8)	3.8 (2.6 to 5.7)
4	4.8 (3.6 to 6.4)	4.6 (1.8 to 11.5)	4.9 (3.6 to 6.5)
Most deprived	11.9 (10.2 to 13.7)	5.0 (2.2 to 11.1)	12.5 (10.7 to 14.4)
I feel stressed because of not having enough money for food b
Māori
Yes	20.4 (17.5 to 23.6)	14.1 (8.3 to 23.1)	21.6 (18.5 to 25.2)
No	8.0 (7.2 to 8.9)	6.6 (4.9 to 8.9)	8.2 (7.3 to 9.3)
Gender
Female	10.7 (9.5 to 12.0)	6.2 (4.3 to 9.0)	11.4 (10.1 to 12.9)
Male	8.8 (7.7 to 10.1)	9.1 (6.4 to 12.9)	8.7 (7.5 to 10.1)
Age Group
15–24	12.7 (10.1 to 15.8)	7.5 (3.1 to 17.1)	13.2 (10.5 to 16.6)
25–44	11.0 (9.3 to 13.0)	15.2 (9.5 to 23.5)	10.5 (8.8 to 12.5)
45–64	11.2 (9.8 to 12.7)	7.3 (4.9 to 10.7)	12.1 (10.6 to 13.8)
65+	4.0 (3.2 to 5.1)	3.7 (2.0 to 6.8)	4.1 (3.2 to 5.2)
Deprivation quintile
Least deprived	4.6 (3.1 to 6.6)	7.0 (3.5 to 13.4)	4.1 (2.6 to 6.3)
2	6.3 (4.7 to 8.4)	7.9 (4.6 to 13.4)	6.0 (4.2 to 8.3)
3	8.1 (6.3 to 10.4)	4.2 (2.5 to 7.2)	8.9 (6.8 to 11.5)
4	10.1 (8.5 to 12.0)	11.9 (6.7 to 20.2)	9.9 (8.2 to 11.9)
Most deprived	20.2 (18.0 to 22.6)	11.9 (7.6 to 18.2)	20.9 (18.6 to 23.5)
I feel stressed because I can’t provide the food I want for social occasions b
Māori
Yes	16.4 (13.9 to 19.4)	12.4 (7.3 to 20.4)	17.2 (14.4 to 20.5)
No	7.0 (6.2 to 7.9)	4.6 (3.1 to 6.9)	7.4 (6.5 to 8.4)
Gender
Female	9.8 (8.6 to 11.1)	6.9 (4.5 to 10.4)	10.3 (9.0 to 11.7)
Male	7.0 (6.0 to 8.1)	5.0 (3.1 to 8.0)	7.3 (6.2 to 8.6)
Age Group
15–24	10.4 (8.1 to 13.4)	3.4 (1.5 to 7.7)	11.2 (8.6 to 14.4)
25–44	9.9 (8.2 to 11.8)	11.7 (6.1 to 21.4)	9.7 (8.0 to 11.6)
45–64	9.1 (7.9 to 10.5)	5.2 (3.4 to 8.0)	10.0 (8.6 to 11.6)
65+	3.9 (3.1 to 4.9)	4.0 (2.2 to 7.2)	3.9 (3.1 to 5.0)
Deprivation quintile
Least deprived	4.6 (3.0 to 6.9)	7.0 (3.2 to 14.4)	4.0 (2.4 to 6.6)
2	4.3 (3.1 to 6.1)	4.2 (2.1 to 8.4)	4.3 (2.9 to 6.4)
3	6.5 (5.0 to 8.5)	3.6 (2.0 to 6.5)	7.1 (5.3 to 9.4)
4	9.5 (7.9 to 11.4)	8.5 (4.4 to 15.6)	9.6 (8.0 to 11.6)
Most deprived	17.4 (15.3 to 19.7)	9.0 (5.6 to 14.1)	18.1 (15.9 to 20.6)

Note: All numbers in the table are percentages with estimated 95% confidence intervals.

^a For reporting purposes item 1 has been phrased in reverse; the answer “sometimes” or “never” to the statement “we can afford to eat properly” is included in the definition of food insecurity in the table above.

^b Food insecurity is people who answer “often” or “sometimes”.

The metrics show a protective effect of rurality, but ethnic disparities persisted regardless of rurality. Among Māori, responses from rural people were consistently lower than those from nonrural areas across all measures. Similarly, rural non‐Māori rates are lower than rates for nonrural non‐Māori across all measures.

Ethnic disparities are evident in coping strategies such as seeking informal food or financial support from others, or in utilizing formal support such as food grants or food banks. Māori (13.4%) in the total population were much more likely than non‐Māori (5.1%) to rely on others when financially constrained. This pattern was demonstrated in both rural and nonrural areas for informal support. In the total population, Māori (12.1%) accessed formal support at higher rates than non‐Māori (3.0%). Nonrural Māori were the highest users of both informal (14.7%) and formal coping strategies (13.3%) when seeking food or financial assistance for food.

Inequities continue, with difference in psychological impacts evident across all settings. In the total population, Māori (20.4%) reported finance‐related food stress at higher rates than non‐Māori (8.0%). This ethnic disparity persisted across geographic subgroups, with more rural Māori (14.1%) experiencing finance‐related food stress than rural non‐Māori (6.6%) and more nonrural Māori (21.6%) than nonrural, non‐Māori (8.2%). A similar pattern emerged regarding social occasion‐related food stress.

For each measure individually, Māori living rurally were protected to some extent compared to Māori living in nonrural areas. Yet, for every indicator, the level of inequity between Māori and non‐Māori was so pervasive that rural Māori, who are ‘protected’ by rural living still experience higher rates of food insecurity than nonrural, nonMāori on most indicators.

3.2. Gender

Overall, small but consistent differences suggest women experience slightly higher rates of direct food insecurity. There was a geographic variation in coping strategies. Rural females (2.2%) sought informal support at less than half the rate of nonrural females (6.6%) and utilized formal supports such as food grants and food banks (1.8%) at much lower rates than nonrural females (4.5%). Similarly, rural males (3.4%) sought informal support at lower rates than nonrural males (7.0%) and utilized formal supports such as food grants and food banks (1.6%) at much lower rates than nonrural males (4.7%).

Gender differences continue around psychological dimensions, but with an important rural variation. Across the total population, females (10.7%) reported higher rates of finance‐related food stress than males (8.8%) and higher rates (9.8%) of social occasion‐related food stress than males (7.0%). Nonrural females were more likely (11.4%) than nonrural males (8.7%) to report finance‐related food stress. Interestingly, this pattern reversed in rural settings, where rural males reported higher rates of finance‐related food stress (9.1%) compared to rural females (6.2%). This exception alongside the reported low use of informal support (3.4%) and low utilization rates of formal support (1.6%) by rural males may reflect economic pressures and highlight the need for further research around gender and geography in food insecurity experiences.

3.3. Age

Age‐related patterns in the food insecurity indicators show distinct arrangements that challenge common assumptions and illustrate complexity in rural settings. The data show a consistent pattern with rates decreasing steadily from the youngest age group to the oldest in the total population but differential effects of age in rural and nonrural adults. In the total population, young adults (15–24 years) consistently experienced the highest levels of food insecurity, lower in the 25–44 years and 45–64 years groups, and was lowest among those aged over 65 years.

When the data were examined by rurality, a different pattern was seen. Young adults (15–24 years) living rurally were less affected by food insecurity than their nonrural counterparts on all indicators. For example, they were less likely to experience food affordability challenges (5.5% compared to 16.1% nonrural), and they have much lower odds of feeling stressed about providing food for social occasions (3.4% compared to 11.2% for nonrural).

Mid‐life (25–44 years) rural adults present as a particularly vulnerable group. Among those living in rural areas, this age group reported the highest rate of food affordability challenges (12.1%), of food shortages (8.4%), and of restricted nutrition (15.7%) due to cost. They were the highest rural seekers of informal (6.7%) and formal (4.3%) support. People in this age group reported subjective psychological impacts more than double those of other rural age groups and higher than their nonrural counterparts. The high rates of food‐related financial stress (15.2%, compared to 10.5% for nonrural, OR 1.53) and social occasion‐related food provision stress (11.7%, compared to 9.7% for nonrural, OR 1.23) are a pattern reversal associating rurality with unique stressors for mid‐life rural adults. The responses for this rural age group suggest a distinct set of social and economic pressures, expectations, or responsibilities across the rural mid‐life course that heighten their vulnerability to household food insecurity.

The elderly (65+) reported consistently low levels across all food insecurity indicators. Rural and nonrural elderly experienced similarly low rates across all metrics, suggesting age‐related protective factors against food insecurity potentially including stable income such as superannuation, or assets and skills accumulated over the life‐course. The very low support seeking from rural elderly suggests potential differences in service availability, accessibility, attitudes toward support and assistance, and/or possible alternative methods of support and assistance in rural communities. Low rates of stress caused by not having enough money for food among the elderly were similar across rural and nonrural settings. These low rates potentially reflect differences in financial stability and/or coping mechanisms developed over the life course.

3.4. Deprivation

There was a clear socioeconomic gradient in food insecurity, with rates rising steadily from least deprived (Quintile 1) to most deprived (Quintile 5) areas. Using all metrics combined, food insecurity indicators ranged from 11.3% in the least deprived area to 32.0% in the most deprived area. The data show a consistent socioeconomic gradient in food insecurity in the total population, a rural deviation evidencing a nuanced rather than consistent protective effect of rurality, and a clear critical threshold effect between Quintiles 1–3 and Quintiles 4–5.

Rural households exhibited a nonlinear socioeconomic gradient with Quintile 4 higher than the most deprived quintile for food affordability (19.9% vs. 14.0%), food shortages (11.4% vs. 9.0%), reduced consumption due to financial constraints (14.1% vs. 11.4%), and limited food variety due to lack of money (20.3% vs. 15.8%) for example.

Rurality's protective effect against food insecurity varied across deprivation levels, showing a nuanced pattern rather than consistent protection. There are mixed effects across Quintiles 1–4, where some rural areas show slightly worse outcomes than nonrural areas. In the least deprived Quintiles (1–2), rurality has a lesser effect where overall households are more food secure. However, for quintile 5, where food insecurity is most severe, there is strong rural protection.

3.5. Multivariable Analysis

Logistic regression modeling confirmed the protective effect of rurality on household food insecurity (Table 5). In a crude model, the odds ratio comparing rural to nonrural areas for the overall measure of food insecurity was 0.60 (95%CI 0.49–0.75), which was partially attenuated having adjusted for Māori ethnicity, age group, gender, and area‐level deprivation (aOR 0.70, 95%CI 0.56–0.88). There was no evidence of a differential effect of rurality among Māori compared to non‐Māori (p = 0.80).

TABLE 5.

The effect of rurality on food insecurity among 1,906 Māori and 7,793 non‐Māori adults in 2019/20 New Zealand Health Survey.

	Crude OR, 95% CI	Adjusted OR, 95% CI
We cannot afford to eat properly a	0.60 (0.46 to 0.78)	0.72 (0.54 to 0.95)
Food runs out in our household due to lack of money b	0.52 (0.38 to 0.70)	0.62 (0.45 to 0.86)
We eat less because of lack of money b	0.56 (0.41 to 0.76)	0.68 (0.49 to 0.94)
The variety of foods we are able to eat is limited by a lack of money b	0.56 (0.44 to 0.71)	0.66 (0.51 to 0.85)
We rely on others to provide food and/or money for food, for our household, when we don’t have enough money b	0.40 (0.26 to 0.62)	0.51 (0.32 to 0.80)
We make use of special food grants or food banks when we do not have enough money for food b	0.36 (0.21 to 0.60)	0.45 (0.26 to 0.77)
I feel stressed because of not having enough money for food b	0.76 (0.56 to 1.03)	0.89 (0.64 to 1.24)
I feel stressed because I can’t provide the food I want for social occasions b	0.66 (0.46 to 0.93)	0.79 (0.54 to 1.15)

Note: Odds ratios (ORs) show the association between food insecurity in rural compared to non‐rural areas. Adjusted Ors adjust for Māori ethnicity, age group, deprivation group and gender.

^a For reporting purposes item 1 has been phrased in reverse; the answer “sometimes” or “never” to the statement “we can afford to eat properly” is included in the definition of food insecurity in the table above.

^b Food insecurity is people who answer “often” or “sometimes”.

When each of the dietary (non‐stress) metrics was analyzed individually, there was a similar pattern: a protective effect of rurality on food insecurity, with attenuation following adjustment for socio‐demographic measures. For each of the six dietary (nonstress) related measures, rural households had a statistically significant lower odds of experiencing food insecurity compared to nonrural households.

For the two stress‐related measures, the protective effect of rurality is attenuated and no longer statistically significant following adjustment for Māori ethnicity, age group, gender, and area‐level deprivation. This indicates that although rural living was protective for dietary (nonstress) related food provision, both rural and nonrural households experienced comparative levels of stress around food provision and social occasions.

There was no evidence of the effect of rurality on food insecurity being different among Māori compared to non‐Māori for any of the metrics studied, p‐values ranged from 0.15 to 0.87 (results not shown). This is particularly important because it highlights the protective effect of rurality benefits both Māori and non‐Māori similarly.

4. Discussion

We sought to analyze NZHS data to identify any rural and nonrural differences in metrics of food insecurity. Investigation of the food insecurity indicators individually provides for a nuanced understanding of the different dimensions of food insecurity in Aotearoa. We analyzed responses by ethnicity, gender, age, and socioeconomic deprivation level. A greater understanding of rural food insecurity would help inform the development of effective policies and design of appropriate interventions. We believe that we are the first to document food insecurity by rurality on a national scale. We discuss our insights from six key findings below.

4.1. Protective Effect of Rurality

Overall, our findings describe an inverse association between rural living and food insecurity. Rural living appears to have a protective effect on food insecurity in Aotearoa. Confirming a synthesis of research in other countries that identified rurality as protective, but not across all contexts (Carter et al. 2014; Gorton et al. 2010), there are exceptions, which we discuss below. Our findings support the theoretical conceptualization of rural food security as a mixture of compounding stressors interacting with supportive factors (Piaskoski et al. 2020).

4.2. Persistent Ethnic Inequities and Intersecting Disadvantage

Our analyses identified the high prevalence of food insecurity in Aotearoa and provided evidence of persistent ethnic disparities across food insecurity indicators, with nuance between rural and nonrural settings. Interaction analyses demonstrated that the effect of living rurally on food insecurity was similar for Māori and non‐Māori, but the prevalence of food insecurity among Māori was considerably higher than non‐Māori. These findings confirm ethnic inequities in food insecurity documented elsewhere (Beavis et al. 2019; McKerchar et al. 2015; Ministry of Health 2012, 2019, 2024), although they are limited in their provision of a complete understanding of food insecurity. One study with a chronically food insecure urban population in Aotearoa found no significant differences in severity based on ethnicity or gender (Robinson et al. 2021). In our study, rural Māori experience higher rates of food insecurity than nonrural non‐Māori across most indicators. Nonrural Māori experience the highest rates across nearly all of the food insecurity indicators, facing a double burden of urbanicity and ethnicity. Ethnic disparities persist highlighting an ethnicity food accessibility gap including disparities in food shortages and diet quality, coping strategies and impacts of stress. Rurality moderates but does not fully eliminate ethnic disparities.

4.3. Reversed Gender Pattern for Finance‐Related Stress

Previous research identifies high vulnerability and disproportionate burden of food insecurity for women (Buck‐McFadyen 2015; Buckelew 2022; Gorton et al. 2010; Piaskoski et al. 2020). Our findings describe a numerically small, yet consistent, pattern of higher rates across the food insecurity indicators for women. Our findings of a reversed gender pattern in rural settings for finance‐related food stress echoes a similarly gendered experience in the United States (Bradley et al. 2024). The higher rates of finance‐related stress may reflect gendered provider norms, coping strategies, and/or psychological responses to economic hardship and highlight the need for further research to understand the intersectionality between rural gender dynamics and food insecurity.

4.4. Pattern of Mid‐Life Age Vulnerability

An important insight from this study is the illumination of several age‐related patterns. Aligning with the literature, there is a general gradient in the total population, where the youngest group (15–24 years) are the most vulnerable to food insecurity (Lê et al. 2015; Pool and Dooris 2022; Pullen et al. 2021), and this vulnerability decreases with age, with consistent protection for the elderly (65+), who had low levels across all food insecurity indicators (Garasky et al. 2006; Pool and Dooris 2022; Pullen et al. 2021). However, our study shows a geographic variation where rural young adults (15–24 years) fare much better than their nonrural counterparts on all measures. This finding warrants further research. Rather than age, rural settings are protective for young adults, suggesting future research should be undertaken to identify the underlying factors that differentiate rural and nonrural environments and drive food insecurity for young adults. Another age deviation in rural settings from our study is that mid‐life rural adults (25–44 years) emerge as the most vulnerable group. This adds to the ethnographic work of Buck‐McFadyen (2015) that highlights challenges for mid‐life rural adults. The vulnerability warrants further investigation and suggests age‐related food insecurity support should be tailored differently for rural and nonrural populations.

4.5. Socioeconomic Gradient in Diet

Our analyses identified a clear socioeconomic gradient in access to nutrition in Aotearoa across the total population, with nuance between rural and nonrural settings. A socioeconomic gradient in diet quality exists where more nutrient‐dense diets are associated with higher socioeconomic conditions and nutrient‐poor, energy‐dense diets are associated with lower socioeconmic conditions (Darmon and Drewnowski 2008, 2015; Drewnowski and Darmon 2005). Food prices affect affordability, accessibility, and diet quality for lower income groups as nutrient‐dense diets are more expensive than nutrient‐poor, energy‐dense diets (Darmon and Drewnowski 2015). Similar to other research (Gorton et al. 2010; Lê et al. 2015), the highest rates of food insecurity are evidenced in the most deprived quintile and lowest rates in the least deprived quintile. There is a rural deviation in the socioeconomic gradient in diet; a nonlinear irregular progression where, for multiple measures Quintile 4 has the highest rates, higher than Quintile 5, the most deprived quintile. Thus, high deprivation has a more severe impact in nonrural settings, and rurality is, to a degree, protective for Quintile 5 households. Protective community factors in rural communities such as informal support systems, community‐provisioning practices, and different social expectations warrant further investigation. Not yet noted in other research, there appears to be a critical threshold between Quintiles 1–3 and Quintiles 4–5, suggesting a breaking point where socioeconomic pressures outweigh abilities to afford optimal food and nutrition.

4.6. Pervasive Stress

An important insight from this study is the contrasting effect of rurality on the six‐food accessibility and coping strategy dimensions versus the psychological stress dimensions of food insecurity. The effect of rurality on the two measures of stress (food provision‐related and social occasion‐related) became nonsignificant following adjustment, suggesting that while rurality is protective for food accessibility and coping strategies, it is not protective for psychological stress dimensions. Rural households experience comparable levels of stress to that of nonrural households about food insecurity. This finding is consistent with other research identifying various ways stress pervades rural communities, through depression and anxiety (Garasky et al. 2006; Pool and Dooris 2022), social isolation, stigma and shame (Bradley et al. 2024; Buck‐McFadyen 2015; Piaskoski et al. 2020) and compounding stressors (Piaskoski et al. 2020). The pervasiveness of stress illuminates the need for interventions to address both food provision and psychological dimensions of food insecurity and/or the need to understand how psychological dimensions contribute to experiences of food insecurity and how psychological components can contribute to food security programs.

4.7. Limitations

There are several strengths and limitations of the study. Although we used the standard food security questions in Aotearoa, we did not operationalize them as some publications do (Ministry of Health 2019; Parnell and Gray 2014). This was an explicit decision to allow exploration of each measure independently, given the lack of previous evidence of rural food insecurity in Aotearoa. Importantly, the food security questions used do not align with or represent Māori food systems, values, and traditions. The 2018 Te Pukenga survey (Statistics NZ, 2020) found that 41% of respondents reported ‘gathering traditional Māori food’, a practice that was more common in rural (55%) than urban (38%) areas. Since our analyses were based on previously collected data, we were also limited by the sample size, particularly in some socio‐demographic groupings. Furthermore, as the NZHS is not explicitly designed to include a representative sample of people from rural areas, the distribution of demographic factors (see Table 2), may not be representative of the population. For example, deprivation is distributed unevenly across rural/nonrural areas in our data: only 11% of participants from rural areas lived in the most deprived areas of deprivation, compared to 21% of people in nonrural areas.

We were limited in our analyses by the data collected in the NZHS. Although this is continuous in the field, the included questions change each year, and between 2011/12 and 2021/22 (the years for which CURF data were available), 2019/20 was the only year in which food insecurity was asked in the adult survey and rurality was determined. The data in our analysis thus precedes the impact of the Covid‐19 pandemic and the subsequent cost of living crisis, so provides an important pre‐Covid‐19 snapshot of food insecurity in Aotearoa. We speculate that the protective effect of rural living on food insecurity, which is likely to include self‐reliance on home grown fruit and vegetables, and protein sources from hunting and fishing, are likely to have continued to benefit rural compared with nonrural households.

The cross‐sectional nature of the survey data does not allow exploration of the direction of association between food insecurity and rural living. For example, are urban‐living Māori who are exposed to food insecurity selectively migrating to rural areas? Likewise, it may be that retired people who choose to move to rural areas are more affluent, so less likely to experience food insecurity.

It is important to note that, although lower than in urban areas, the prevalence of food insecurity remains a challenge in rural communities. This is particularly the case for rural Māori, who experience levels of food insecurity higher than nonrural non‐Māori for each metric measured. Charitable and noncommercial initiatives such as co‐ops, food banks, pātaka kai, and community‐led projects such as sharing economies and koha kai function in urban communities (Davies et al. 2022). Further research could explore the extent to which these types of activities also provide food and nutrition security for people in rural communities.

The measurement of rurality in this study was limited to the Statistical Standard for Geographic Areas (2018). More recent work has identified that the geographic classification for health (GCH) may be a more appropriate measure. It sub‐classifies rural areas into three groups, based on distance to urban centres as well as population density. Although the Ministry of Health identifies that further exploratory work on this measure is required (Minister of Health 2023), it is likely to be useful in understanding rural barriers to health and other services. The Ministry of Health (2025) has recently released aggregated data from the NZHS using GCH. For the 3‐year period 2021/22 to 2023/24, 19.6% of children reported living in households where food sometimes or often runs out. Among those living rurally, this was 17.9%, compared to 18.8% in those living in larger urban areas and 21.9% in smaller urban areas. Similarly for other metrics, children in rural households were less likely to report eating less due to money or using food grants. A similar pattern is seen for Māori households, with less food insecurity in rural compared to urban areas, but greater food insecurity among Māori compared to non‐Māori households.

5. Conclusion

The consistent protective effect of rurality across settings and place warrants future research to identify and leverage any specific factors that could translate into informing policy and interventions to improve food insecurity in nonrural settings. Although it has been suggested that food insecurity may particularly affect older people (Minister of Health 2023), as well as families with young children, in rural areas, our data shows the opposite. Older people (aged 65 years and older) reported lower levels of food insecurity than adults at younger ages. It is possible that this could be due to selective migration of people in retirement; those able to afford to move to rural areas are unlikely to face financial food insecurity. Further work to explore this is warranted.

This analysis illustrates the layered way food insecurity exists in households in Aotearoa New Zealand, with a lower prevalence in rural compared to nonrural households across all metrics. The protective effect of rurality persists after adjusting for ethnicity, age, gender, and deprivation. Despite this protective rural effect, substantial inequities persist for Māori. Māori experience disproportionately higher rates across all food insecurity metrics regardless of geographic area.

Funding

This work was supported by the Health Research Council of New Zealand 24/996.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from Statistics New Zealand. Restrictions apply to the availability of these data, which were used under license for this study.