Gender Differences in Food Insecurity, Anxiety, and Nutrition in Israel: A Cross-Sectional Study

Abstract

Background:

Food insecurity and psychological distress, especially anxiety, are often interconnected, with notable gender-based differences. Understanding these relationships is crucial for informing targeted interventions during crises such as war.

Objectives:

To estimate food insecurity prevalence by gender and examine its association with anxiety and diet quality among Israeli adults in 2023–2024.

Methods:

A cross-sectional online survey was conducted among a representative random sample of Israeli adults (≥18 years). Food insecurity was assessed using the two-item Hunger Vital Sign, anxiety symptoms via the General Anxiety Disorder (GAD)-7, and diet quality with the Israel Mediterranean Diet Scale (I-MEDAS). The survey was fielded November 7–18, 2023, 4–6 weeks after the 7 October war broke, a period marked by nationwide school closures or partial remote learning, reserve-duty mobilization of ∼360,000 individuals, frequent air-raid sirens in multiple regions, temporary business disruptions, and government relocation/hoteling of some communities, even outside the immediate Gaza border area.

Results:

Among 530 participants, food insecurity prevalence was 8.2%. Males reported higher smoking rates (31% vs. 22.2%, p = 0.022) and physical activity levels both before (120 vs. 60 minutes, p = 0.001) and during the war (70 vs. 15 minutes, p < 0.001), while females exhibited more dietary structure changes (25.8% vs. 16.8%, p = 0.023) and supplement use (30.5% vs. 20.8%, p = 0.011). No gender differences were found in weight change or Mediterranean Diet adherence. Females had significantly higher anxiety scores (median GAD-7: 10 vs. 5, p < 0.001). Female gender, food insecurity (OR 2.436, p = 0.011), current smoking, and worsening living conditions were associated with elevated anxiety; older age was protective.

Discussion:

These findings highlight critical gender-based disparities in mental health and lifestyle behaviors. Higher anxiety and distinct coping strategies among women align with feminist theories of emotional labor and “crisis patriarchy,” suggesting disproportionate burdens during systemic disruptions. Public health and primary care strategies should incorporate screening and tailored interventions. Future research should apply an intersectional lens to explore how ethnicity, religion, and socioeconomic status intersect with gender to influence vulnerability.

Introduction

Food insecurity refers to a condition in which individuals lack reliable, safe, and adequate access to nutritious foods that meet their health needs and cultural preferences, enabling an active and healthy life. This situation can arise when one or more essential components, physical availability of food, economic means to obtain it, and social or cultural accessibility, are missing. ¹

The drivers of food insecurity are multifaceted, encompassing economic, social, environmental, and geographic dimensions. ² Economic challenges include unemployment, low income, and single-parent family structures. ³ Social determinants such as low educational attainment, discrimination, physical or mental disability, and ethnic background may increase vulnerability.^4,5 Geographic barriers, including long distances or difficulty reaching food outlets, a lack of accessible public transportation, and high travel costs, can further restrict access to quality foods. Environmental stressors, such as natural disasters, floods, or droughts can disrupt food production and supply chains, leading to price volatility and reduced availability. ⁶ The war’s early weeks disrupted daily life across Israel beyond the border region, for example, widespread school and childcare interruptions, mobilization of reservists, sporadic sheltering due to sirens, supply-chain perturbations, and temporary relocation of select communities. These acute stressors plausibly affected anxiety, eating patterns, and perceived living conditions.

The consequences of food insecurity are not distributed equally across populations. Women, especially single mothers, elderly women, and those in low-income or rural areas, may face heightened vulnerability due to caregiving responsibilities, limited mobility, and gender-based disparities in income and resource access. ⁷ Environmental and societal disruptions, such as economic downturns or supply chain instability, may further exacerbate food insecurity, often with disproportionate effects on women. ⁸

Food insecurity can negatively influence both physical and mental health. Limited and inconsistent access to nutritious food may lead individuals to alter their eating habits in ways that increase the risk of obesity and complicate the management of chronic illnesses such as type 2 diabetes. ⁹ Beyond its nutritional impact, food insecurity is frequently linked with heightened psychological distress, including anxiety.^10,11 Prolonged stress can activate the hypothalamic–pituitary–adrenal axis, triggering the release of hormones and neuropeptides such as cortisol and neuropeptide Y. ¹² These physiological responses can increase appetite and drive the consumption of calorie-dense, highly processed foods, further affecting diet quality and overall well-being.^13,14

Anxiety is a common psychological condition defined by pervasive feelings of worry, fear, and apprehension, often focused on future events without a specific or immediate cause. Although it affects individuals across all demographics, anxiety can lead to significant impairment and, in some cases, long-term disability if left unaddressed. ¹⁵ While it affects individuals across the lifespan, its expression and impact differ by gender. Research has consistently shown that women report higher levels of anxiety than men, and these differences tend to widen during periods of social or environmental stress.^16–19 Biochemical pathways, such as hormonal fluctuations, interact with sociocultural pressures, including caregiving roles, economic insecurity, and exposure to gender-based violence, to increase women’s susceptibility to anxiety.^19,20 Notably, dietary patterns and psychological well-being are often interrelated; studies during the COVID-19 pandemic, for instance, found that adherence to the Mediterranean diet (MedDiet) was associated with lower odds of anxiety, especially among women.^17,18

Research consistently demonstrates significant gender-based differences in both the prevalence and experience of anxiety. ¹⁸ Women are more likely than men to report elevated anxiety levels, and they may be more vulnerable to its effects during periods of crisis. ¹⁹ Global data indicate that during periods of widespread disruption, such as an emergency event, women not only experienced higher levels of anxiety but also more significant declines in psychological well-being, especially when compounded by challenges like food insecurity and unemployment. ²⁰

Given the increased pressures on daily life during national periods of uncertainty, there is a growing need to better understand how food insecurity and mental health interact, particularly through a gender-based lens. The present study investigates the associations between food insecurity, anxiety symptoms, and diet quality among Israeli adults during a time of war (2023–2024), with attention to potential gender differences in these relationships.

Materials and Methods

Study design

The study was conducted online in a representative sample of Israeli adults. The present cross-sectional survey, the data of which have been previously collected and reported, ²¹ simultaneously assessed the degree to which food insecurity was associated with changes in dietary habits compared with prearmed conflict patterns; anxiety measured using the General Anxiety Disorder (GAD) 7 questionnaire; food quality, assessed using the Israel-Mediterranean diet screener (I-MEDAS); and demographic characteristics, with the current article addressing a distinct research question.

Ethics

The study was approved by the Institutional Ethics Board of Ariel University, Israel (approval number AU-HEA-MB-20231105, November 5, 2023).

Study location

The survey was distributed online from November 7, 2023, to November 18, 2023, through the services of a survey company (“SekerNet”).

Study population

Participants were drawn by SekerNet from a standing national online panel of Israeli residents aged ≥18 years. The panel recruits via probability and mixed-mode outreach with phone verification; one submission per invitee was allowed. To approximate national sociodemographics, SekerNet applied stratified invitations by sex, age group, and region; we then compared achieved distributions to Central Bureau of Statistics (CBS) benchmarks. The study population was a representative random sample of Israeli adults who provided informed consent by clicking the appropriate button.

Inclusion criteria

All Israeli adults who indicated their age as 18 years or older and who provided informed consent were included. Survey recruitment and data collection were halted when >500 responses were accrued.

Exclusion criteria

Survey respondents who did not provide informed consent by clicking on the appropriate button and those who indicated their age was younger than 18 years could not proceed with the survey and thus were excluded.

Study procedures

Participant flow: A total of 530 eligible Israeli adults (≥18 years) provided consent and were included in the final analytic sample (no exclusions), a representative, random sample of Israeli adults. All data were de-identified all data prior to transferring the survey-generated dataset to investigators.

Sample size and study power

The present study was powered to include 500 participants, providing a 95% confidence level and a 4.4% confidence limit for the survey.

Data acquisition and survey characteristics

Collected in the framework of the survey were the following data: (1) demographic information: age; gender; country of residence; level of education; ethnicity (religion); religiosity; family status; (2) food insecurity (Hunger Vital Sign 2-question food insecurity screening tool) ²² ; (3) anxiety questionnaire (GAD−7) ²³ ; and (4) nutritional habits questionnaire (change in the present vs. prewar diet questions; I-MEDAS ²⁴ ).

Variables and operational definitions

Food insecurity was assessed with the 2-item Hunger Vital Sign; respondents were classified as food insecure if they answered “often true” or “sometimes true” (coded “yes”) to ≥1 item. The Hunger Vital Sign screens food insecurity over the prior 12 months; therefore, it may capture prewar economic hardship in addition to acute crisis-related effects.

Anxiety was measured with the GAD-7 (range 0–21). Severity groups used conventional cut-points: 0–4 minimal, 5–9 mild, 10–14 moderate, ≥15 severe; for multivariable analyses, we modeled anxiety as (≥10 vs. <10) (moderate/severe vs. other). Diet quality was assessed using the I-MEDAS (17 items; total 0–17; higher score indicates greater adherence). Physical activity was self-reported in minutes/week “before the war” and “in the past week.” ΔPA was calculated as (past-week minutes−pre-war minutes). Smoking was “present smoker” if ≥1 cigarette/day. Weight change captured any gain or loss since the war’s start and, if present, the magnitude (kg). Dietary pattern changes included: less structured eating, more snacks, more total food, fewer fruits/vegetables, healthier now (all yes/no). Sociodemographics included age (years), gender, family status, children <18 at home (count), region, religion (Jewish/Muslim/Christian/Druze/Other), religiosity (secular/traditional/religious/ultra-religious), and education. “Worse now” indicates respondents who reported that their current living conditions were worse than before the war.

Bias and quality control

To limit selection bias, the survey company (SekerNet) sampled a random panel subset of Israeli adults ≥18 years and restricted one submission per invitee; participation required consent before survey access.

To reduce information bias, we used validated instruments (Hunger Vital Sign, GAD-7, I-MEDAS) and standardized questions with fixed response options. The GAD-7 recall window is 2 weeks by design, minimizing long-recall effects; “pre-war” questions anchored behaviors to the commonly understood period preceding October 2023.

To limit confounding, covariates were selected a priori based on published evidence and conceptual relevance (age, gender, smoking, food insecurity, and “worse now”). Model assumptions were checked.

Statistical analysis

Data were stored on an Excel spreadsheet and analyzed on SPSS Statistical Analysis Software v29. The Kolmogorov–Smirnov test was used to assess the normality of continuous variables. Because none of the continuous variables were normally distributed, they were described as median (interquartile range [IQR]). Nominal variables were presented as n (%). Food insecurity by gender was considered present if the subject answered “yes” to either of the two Hunger Vital Sign questions. Continuous variables were compared by gender using the Mann–Whitney U test. Nominal variables were compared by food insecurity using the chi-square test. Food insecurity by gender was modeled using binary logistic regression analysis. All analyses were intention-to-treat, and all tests were two-sided and considered significant at p < 0.05.

Because continuous variables were non-normally distributed (Kolmogorov-Smirnov), we report medians (IQR) and used Mann–Whitney U tests. For GAD-7, we report both the continuous score and severity categories using established thresholds (5/10/15) to improve interpretability and comparability across studies. I-MEDAS totals (0–17) were analyzed continuously; item-level binary scoring followed instrument definitions. The primary regression modeled moderate–to–severe anxiety (GAD-7 ≥10) versus lower levels; we chose ≥10 because it balances sensitivity/specificity for clinically meaningful anxiety in general populations. For I-MEDAS item-level comparisons (17 tests), we controlled the false discovery rate using Benjamini–Hochberg (q = 0.05).

Missing data were minimal and handled by complete-case analysis; denominators are shown where applicable. No imputation was performed.

All analyses were stratified by gender, and multivariable models were constructed to evaluate gender-specific associations among food insecurity, anxiety, and lifestyle behaviors that were not examined in the earlier report.

Results

The study population consisted of 530 Israeli adults aged 18 years or older. Of these, 44 (8.2%) answered “yes” to one or both of the food insecurity screening questions on the Hunger Vital Signs questionnaire, indicating they are food insecure.

Displayed in Table 1 are characteristics of the study population by gender. Age and religiosity did not differ by gender. Food insecurity, education, and religious affiliation did not differ by gender (p = 0.526, p = 0.071, and p = 0.440, respectively). In terms of residence since the onset of the war, the vast majority of participants in both groups reported continuing to reside at home (92.5% of females vs. 90.6% of males). A small proportion of participants reported relocation by the army or government (1.1% females vs. 2.3% males, p = 0.800), staying with relatives or friends, or living in other arrangements.

Table 1.

The Demographic Characteristics of the Study Population by Gender

Characteristic	Female (n = 275)	Male (n = 255)	p value
Age (years, median (IQR))	35 (22)	44 (25)	0.001
family status n (%)
Married/Domestic partnership	174 (63.4)	166 (65.1)	0.945
Single	73 (26.5)	67 (26.3)
Divorced	25 (9)	20 (7.8)
Widowed	3 (1.1)	2 (0.8)
Children in the household <18 years of age (median [interquartile range])	0.00 (2)	0.00 (2)	0.319
Region of residence because onset of war n (%)
Central Region (excluding Tel Aviv)	60 (21.5)	64 (25.1)	0.205
Northern Region	54 (19.4)	42 (16.5)
Tel Aviv	47 (16.8)	36 (14.1)
Southern Region	36 (12.9)	27 (10.6)
Jerusalem	28 (10)	27 (10.6)
Carmel Region (excluding Haifa)
Judea and Samaria	21 (7.5)	10 (3.9)
Mixed city (i.e., Haifa, Lod, Ramleh)	8 (2.9)	10 (3.9)
Southern Triangle (i.e., Jaljulia, Taibeh)	5 (1.8)	4 (1.6)
Northern Triangle (i.e., Um al Fahem, Kfar Kara)	3 (1.1)	5 (2)
Other	1 (0.4)	0 (0)
Residence since onset of war n (%)	(n = 248)	(n = 219)
Home	258 (92.5)	231 (90.6)	0.800
Relocated by army/government (i.e., hotel, field, school)	3 (1.1)	5 (2.0)
Staying with relatives/friends	13 (4.7)	13 (5.1)
Other	5 (1.8)	6 (2.4)
Religion n (%)
Jewish	220 (80.3)	209 (82)	0.440
Muslim	29 (10.4)	25 (9.8)
Christian	14 (5)	12 (4.7)
Druze	10 (3.6)	4 (1.6)
Other	2 (0.7)	5 (2)
Religiosity n (%)
Secular	125 (46.2)	108 (42.4)	0.484
Traditional	91 (32.6)	82 (32.2)
Religious	37 (13.3)	46 (18)
Ultra-religious	22 (7.9)	19 (7.5)
Education n (%)
Fewer than 12 years	1 (0.4)	3 (1.2)	0.071
High school diploma/matriculation	20 (7.2)	27 (10.6)
Professional license (technician, tradesperson, etc.)	66 (23.7)	39 (15.3)
Bachelor’s degree	66 (23.7)	77 (30.2)
Master’s degree or higher	77 (27.6)	64 (25.1)
Other	49 (17.6)	45 (17.6)
Food Insecurity (%)	25 (9)	19 (7.5)	0.526

The bold data highlights the median age of participants, indicating significant demographic differences between genders, which may influence health and social outcomes.

IQR, Interquartile Range.

The lifestyle and dietary characteristics of the study population, stratified by gender, are presented in Table 2. Before the onset of the war, males reported a significantly higher median number of weekly exercise minutes than females (120 vs. 60 minutes, p = 0.001). This pattern persisted during the week preceding the survey, with males engaging in significantly more exercise (70 vs. 15 minutes, p < 0.001). Although females exhibited a greater reduction in physical activity (delta physical activity [−23.5] ± 99.5 minutes) compared with males (delta physical activity 9.3 ± 138.0), this difference did not reach statistical significance (p = 0.053).

Table 2.

Lifestyle and Dietary Characteristics of the Study Population by Gender

Characteristic	Female (n = 275)	Male (n = 255)	p value
Minutes of exercise/week prior to war (median [IQR])	60 (115)	120 (170)	0.001
Minutes of exercise/week in the past week (median [IQR])	15 (60)	70 (100)	<0.001
Delta PA (median [IQR])	0 (60)	0 (50)	0.229
Present smoker (at least one cigarette smoked per day) n (%)	62 (22.2)	79 (31)	0.022
Weight change since the start of war n (%)
No weight change
Yes, weight gain	96 (34.4)	84 (32.9)	0.72
Yes, weight loss	38 (13.6)	31 (12.2)	0.615
Don’t know
Quantity of weight change among those reporting change (in kg)
Weight gained (median [IQR])	0.00 (1)	0.00 (1)	0.72
Weight lost (median [IQR])	0.00 (0)	0.00 (0)	0.615
Has cooking facilities as present location (n [%])
Change in diet quality n (%)
Healthier prior to the war	60 (21.5)	57 (22.4)	0.813
Less structure now	72 (25.8)	45 (16.8)	0.023
More snacks now	63 (22.6)	42 (16.5)	0.076
More food now	51 (18.3)	41 (16.1)	0.501
Fewer fruits/vegetables now	34 (12.2)	30 (11.8)	0.881
Healthier now during the war	12 (4.4)	10 (3.9)	0.762
Nutrition supplements	85 (30.5)	53 (20.8)	0.011
MedDiet Score (median [IQR])	8.0 (3)	8.0 (4)	0.597

Bold figures represent significant differences in physical activity levels between genders. Specifically, males engaged in more exercise compared to females before and during the war.

IQR, Interquartile Range; PA, Physical Activity; MedDiet, Mediterranean Diet.

Smoking prevalence differed significantly by gender, with a higher proportion of males (31%) reporting current smoking compared with females (22.2%, p = 0.022). No significant differences were observed between males and females regarding weight changes since the start of the war. Specifically, the proportions of participants reporting weight gain (p = 0.720) or weight loss (p = 0.615) and the median of weight change among those reporting a change did not differ between genders.

Access to cooking facilities was similarly reported among males and females. Regarding dietary quality changes during the war, females were significantly more likely than men to report less structured eating patterns (25.8% vs. 16.8%, p = 0.023). No significant gender differences were noted in the proportions reporting healthier diets before the war, increased snack consumption, increased food intake, reduced fruit and vegetable consumption, or healthier diets during the war.

Finally, the use of nutritional supplements differed significantly by gender, with a higher proportion of females reporting use than males (30.5% vs. 20.8%, p = 0.011). The median MedDiet Score was identical in females and males, with a median score of 8.0 for both groups. The IQR was slightly narrower among females (IQR = 3) compared with males (IQR = 4). There was no statistically significant difference in MedDiet Score between females and males (p = 0.597, Mann–Whitney U test).

Anxiety measures related to food insecurity, stratified by gender, are presented in Table 3. Females demonstrated significantly higher levels of anxiety compared with males across all measures. The median total GAD-7 score was significantly higher among females (10 [IQR: 9]) than males (5 [IQR: 7]; p < 0.001).

Table 3.

Anxiety Measures GAD-7 by Gender

Characteristic	Female (n = 275)	Male (n = 255)	p value
Total GAD-7 Score (median [IQR])	10 (9)	5 (7)	<0.001
GAD-7 score (median [IQR]) a
Feeling nervous, anxious, or on edge	2 (2)	1 (1)	<0.001
Not being able to stop or control worrying	1 (2)	1 (1)	<0.001
Worrying too much about different things	2 (2)	1 (1)	<0.001
Trouble relaxing	1 (1)	0 (1)	<0.001
Being so restless that it is hard to sit still	1 (2)	0 (1)	<0.001
Becoming easily annoyed or irritable	1 (2)	1 (1)	<0.001
Feeling afraid as if something terrible might happen	2 (3)	1 (1)	<0.001
Categorized GAD-7 score n (%)
Minimal anxiety (GAD-7 score 0–4)	43 (15.4)	112 (43.9)	<0.01
Mild anxiety (GAD-7 score 5–9)	84 (30.1)	83 (32.5)
Moderate anxiety (GAD-7 score 10–14)	73 (26.2)	31 (12.2)
Severe anxiety (GAD-7 score ≥15)	75 (28.3)	29 (11.4)

^aThe GAD-7 scale asks the respondent to refer to the two weeks prior to the survey. Each item on the scale can receive is scored as follows: 0 (not at all), 1 (several days), 2 (more than half of the days), and 3 (nearly every day); thus, the total score can receive a value from 0 to 21.

Each individual GAD-7 item score was significantly higher in females compared with males, including feelings of nervousness, inability to control worrying, excessive worrying about different things, trouble relaxing, restlessness, irritability, and feelings of fear as if something terrible might happen (all p < 0.001).

When categorizing GAD-7 scores, a greater proportion of females experienced moderate (GAD-7 score 10–14; 26.2% vs. 12.2%) and severe anxiety (GAD-7 score ≥15; 28.3% vs. 11.4%) compared with males, and this difference was statistically significant (p < 0.01). Conversely, minimal anxiety (GAD-7 score 0–4) was more frequently reported among males (43.9%) than females (15.4%). The prevalence of mild anxiety (GAD-7 score 5–9) was similar between genders. Adherence to individual components of the I-MEDAS, stratified by gender, is presented in Table 4. Overall, most I-MEDAS items did not differ significantly between females and males. Both genders reported similar adherence rates for the use of olive oil as the main culinary fat (62% females vs. 60.8% males, p = 0.772), preference for poultry/white meat over red meat (71% vs. 74.5%, p = 0.399), consumption of at least two vegetable servings per day (45.4% vs. 44.7%, p = 0.899), and several other core items including butter/cream consumption, whole grain intake, dairy intake, legume consumption, fish intake, nut intake, hummus/tahina consumption, and sweet baked goods consumption (all comparisons p > 0.05).

Table 4.

Adherence to I-MEDAS Items by Gender

	% receiving a point for the item
I-MEDAS items n (%)	Female (n = 275)	Male (n = 255)	p value
Uses Olive Oil as Main Culinary Fat	173 (62)	155 (60.8)	0.781
Eats poultry/white meat more than red meat	198 (71)	190 (74.5)	0.298
At least two vegetable servings/day	123 (45.4)	115 (45.8)	0.798
Three or more fruit servings/day	29 (10.4)	39 (15.3)	0.091
Less than one butter/margarine/cream servings/day	152 (54.5)	154 (60.4)	0.166
Less than one sugar-sweetened beverages/day	170 (60.9)	133 (52.2)	0.043 †
Three or more whole grain servings/day	54 (19.4)	56 (22)	0.457
Two or more unsweetened dairy servings/day	151 (55.1)	128 (50.6)	0.219
Fewer than seven red/processed meat servings/week	268 (96.1)	241 (94.5)	0.374
Seven or more alcoholic beverages/week	3 (1.1)	9 (3.5)	0.051
Three or more legume servings/week	69 (24.8)	77 (30.2)	0.164
Three or more fish servings/week	36 (12.9)	31 (12.2)	0.808
Three or more nut servings/week	80 (28.7)	75 (29.4)	0.851
Three or more hummus/tahina servings/week	98 (35.1)	100 (39.2)	0.382
Fewer than three sweet baked goods servings/week	108 (38.7)	118 (46.3)	0.077
Fewer than two savory baked goods servings/week	235 (84.2)	197 (77.3)	0.041 †
Three or fewer salty snacks servings/week	235 (84.2)	210 (82.4)	0.561

The bolded data indicate that females reported consuming less than one sugar-sweetened beverage per day, compared to males, with a statistically significant difference (p = 0.043). Additionally, more females reported consuming fewer than 2 servings of savory baked goods per week than males (p = 0.041), highlighting dietary preferences and health-conscious behaviors.

P-values are FDR-adjusted (Benjamini–Hochberg). Items remaining significant after adjustment are indicated with †.

I-MEDAS, Israeli Mediterranean Diet Score.

Significant differences were observed for a few specific dietary behaviors. A greater proportion of females reported consuming fewer than one sugar-sweetened beverage per day compared with males (60.9% vs. 52.2%, p = 0.041). Additionally, females were more likely than males to consume fewer than three savory baked goods servings per week (84.2% vs. 77.3%, p = 0.041). No significant differences were found between genders for consumption of fewer than three salty snacks per week.

Table 5 presents the results of a multivariable logistic regression analysis predicting anxiety symptoms, as measured by the GAD-7 scale.

Table 5.

Multivariable Logistic Regression Model of GAD-7 by Gender

Variable	Odds ratio	95% confidence interval	p value
Age (years)	0.982	0.970–0.995	0.006
Gender (Male = 1)	0.240	0.161–0.358	<0.001
Reported dietary intake worse than before (Worse now) (1 = yes)	1.869	1.186–2.943	0.007
Food insecurity (1 = yes)	2.436	1.225–4.846	0.011
Current smoker (1 = yes)	1.782	1.155–2.749	0.009
Constant	1.770		0.06

Generalized Anxiety Disorder score, GAD-7.

Indicators for variables received a value of 1. Indicators were: male gender (vs. female); Jewish (vs. any other religion); consumes <1 serving per day of butter/margarine/cream. The model was significant (p < 0.001) and correctly classified 91.9% of participants with respect to anxiety status (GAD-7 ≥ 10).

Several variables were significantly associated with higher GAD-7 scores. Increasing age was associated with a slight but significant decrease in the odds of higher anxiety scores (odds ratio [OR]: 0.982, 95% confidence interval [CI]: 0.970–0.995, p = 0.006). Male gender was strongly associated with lower odds of higher anxiety scores compared with female gender (OR: 0.240, 95% CI: 0.161–0.358, p < 0.001).

Participants who reported that their situation was worse at the time of the survey (“worse now”) had significantly higher odds of elevated anxiety (OR: 1.869, 95% CI: 1.186–2.943, p = 0.007). Food insecurity was a strong predictor of higher GAD-7 scores (OR: 2.436, 95% CI: 1.225–4.846, p = 0.011). Additionally, current smokers had significantly higher odds of elevated anxiety symptoms (OR: 1.782, 95% CI: 1.155–2.749, p = 0.009).

The overall model was statistically significant (p < 0.001) and demonstrated strong classification performance, correctly classifying 91.9% of participants with respect to anxiety status (GAD-7 ≥10).

Discussion

This analysis builds on prior publications based on the same nationally representative survey, which examined food insecurity, diet quality, and anxiety during the early phase of the war. The present study extends this work by focusing explicitly on sex/gender differences in anxiety, lifestyle behaviors, and dietary responses to crisis.

Women’s higher anxiety in our data is consistent with prior work demonstrating sex/gender differences in stress responses. While we did not measure caregiving or domestic labor, prior literature suggests that during societal disruptions, women may experience disproportionate emotional and logistical burdens. We therefore present this as context rather than a tested mechanism in our data, and we have accordingly tempered our speculative language.

This cross-sectional study explored the associations between food insecurity, anxiety symptoms, and lifestyle behaviors among Israeli adults during war, with a focus on gender differences. While the overall prevalence of food insecurity was relatively low and showed no significant variation by gender, distinct gender-specific patterns were observed in mental health and lifestyle behaviors. Females reported significantly higher anxiety levels across all GAD-7 measures, whereas males exhibited higher levels of physical activity and a greater prevalence of smoking. Dietary patterns were largely comparable between genders in terms of overall MedDiet adherence, although females reported greater changes in dietary structure and higher rates of nutritional supplement use. Multivariable analysis identified female gender, food insecurity, current smoking, and worsening living conditions as independent predictors of elevated anxiety, whereas older age appeared protective.

The elevated anxiety levels observed among women in our study resonate with longstanding feminist analyses that conceptualize caregiving not merely as a set of domestic tasks but as a deeply gendered form of labor embedded in emotional, moral, and social expectations. ²⁵ Scholars such as Hochschild ²⁶ have highlighted how women disproportionately bear the burden of “emotional labor”—the work of managing others’ feelings, needs, and emotional stability—particularly during periods of collective crisis.^27,28 In the context of war, such as the October 2023 conflict in Israel, these responsibilities become more acute, as societal disruptions heighten demands for caregiving, domestic coordination, and emotional regulation within households. ²⁹ These forms of labor, while often invisible to dominant policy frameworks, exact a cumulative psychological toll that extends beyond material deprivation. Empirical evidence from both the COVID-19 lockdowns and the recent war demonstrates how moments of societal rupture tend to reinforce traditional gender roles, assigning women the primary responsibility for sustaining household functionality amid uncertainty. Feminist political economists have termed this dynamic “crisis patriarchy,^30,31” a system in which women serve as shock absorbers for institutional breakdown. In times of systemic disruption, they are positioned as buffers for state retrenchment and fractured infrastructures. This gendered risk absorption is structurally produced: women’s unpaid labor, both emotional and material, becomes the hidden scaffolding of social resilience. ³² Managing food insecurity, maintaining emotional stability for children, and preserving daily routines under duress are all forms of unpaid crisis labor that remain undervalued in official policy, yet are essential for household survival. The heightened anxiety reported by women in this study thus reflects not only the psychological strain of crisis, but also the embodied consequences of their structural positioning. The expectation that women will “hold the household together” during national emergencies constitutes a feminized form of crisis governance-one reliant on moralized ideals of care and duty rather than on institutional support.

The finding that females experienced higher anxiety symptoms is also consistent with extensive literature demonstrating a greater psychological burden among women during periods of societal stress.^33–36 The observed association between food insecurity and elevated anxiety symptoms supports previous evidence suggesting that concerns about access to sufficient food act as a chronic stressor, with wide-ranging psychological impacts. ³⁷ During the COVID-19 pandemic, very low food security was associated with markedly increased odds of psychological distress, including over sevenfold greater odds of depression and nearly elevenfold greater odds of high stress. A dose-response relationship has been observed, with worsening food insecurity linked to higher levels of distress. ³⁸ A narrative review highlights concerns about health, income, and the ability to feed one’s family as key contributors to this burden. ³⁹ These findings support the view of food insecurity as a significant social determinant of mental health. Interestingly, despite differences in lifestyle behaviors, diet quality as assessed by adherence to the MedDiet did not differ significantly between males and females, suggesting that dietary structure changes reported by females may have been more related to eating patterns and meal regularity than overall food choice quality.

In contrast to some earlier studies conducted during periods of crisis, the prevalence of food insecurity reported in our sample was lower than expected. For instance, surveys conducted during the COVID-19 pandemic in Israel and internationally reported food insecurity rates ranging from 14% to 25%, depending on population vulnerability.^5,6 This discrepancy may partly reflect underreporting due to the anonymous, self-reported nature of the survey or differences in the acute phase of societal disruption when the study was conducted.

Lifestyle findings in this study echo prior observations that while men are engaged in significantly more physical activity, both men and women decreased their level of physical activity in response to stressful events.^40,41 In line with the literature, men also reported a higher prevalence of current smoking compared with women. ⁴² However, it seems that smoking reflects a usual habit because we have not checked the change in smoking habits during stressful events. The maintenance of similar Mediterranean diet adherence scores across genders is noteworthy, particularly given previous research suggesting that psychological stress can negatively impact diet quality. This finding aligns with previous evidence linking stress to changes in eating behaviors, such as increased consumption of unhealthy foods and reduced intake of nutrient-dense options.^17,43 However, these qualitative changes did not significantly alter diet quality scores, suggesting that structured assessments may not fully capture stress-related dietary shifts. Further research is warranted to explore the long-term impact of such disruptions on overall diet quality.

Although our findings revealed no significant gender differences in overall adherence to the MedDiet, the greater use of nutritional supplements and the more pronounced changes in dietary structure reported by women may reflect distinct gendered coping mechanisms. Recent studies confirm that women are more likely than men to engage in food-related responses to stress, including increased monitoring of food intake, supplement use, and meal structuring, especially during times of uncertainty and crisis.^44–46 These behaviors are not solely functional or health-driven; rather, they are deeply shaped by internalized gender norms that frame women as the primary custodians of family health and bodily regulation. ⁴⁷ As such, women’s dietary adaptations under stress can be understood as performances of normative femininity, where control over food, health, and the body serves both emotional and moral purposes. ⁴⁸ These patterns are reinforced by broader social scripts that position nutritional vigilance and self-care as ethical obligations for women, particularly in the context of caregiving roles. ⁴⁸ Applying a gender lens to these coping strategies reveals how food practices during crises, such as war or food insecurity, are not neutral but culturally and politically encoded.

Overall, the present study extends the existing literature by providing gender-stratified evidence on the interplay between food insecurity, mental health, and lifestyle during a period of war in Israel.

Although the survey’s anonymous format likely minimized social desirability bias, it may not have eliminated it. This residual bias may have influenced responses to the two-item Hunger Vital Sign food insecurity screener, potentially resulting in an underreporting of food insecurity within the study population, an observation supported by the lower prevalence found compared with national estimates. Similarly, participants may have underreported weight changes and other diet-related behaviors. Collectively, these underestimations may have attenuated the observed association between food insecurity and anxiety during periods of armed conflict.

Food insecurity was reported by only 44 participants, of whom just 7.5% were female. This unexpectedly low prevalence limited the ability to conduct subgroup analyses by ethnicity, geographic region, or other demographic factors within the food-insecure group. Although the small number of food-insecure women in this study limited subgroup analyses by ethnicity, geography, or socioeconomic status, future research must address the intersectional nature of vulnerability. Not all women experience food insecurity and anxiety in the same way. In Israel, Arab women, ultra-Orthodox mothers, and low-income single mothers often face compounded risks due to structural exclusions, cultural expectations, and limited access to resources.^49,50 These intersecting factors-ethnicity, religion, class, and family structure- shape both exposure to hardship and coping capacity. An intersectional lens ⁵¹ is essential to avoid universalizing women’s experiences and to develop more targeted and equitable public health interventions.

Limitations

Prior research had suggested that the prevalence of food insecurity would be nearly double that observed in the present study, highlighting a discrepancy that may reflect underreporting or sampling limitations. Another limitation lies in the study’s cross-sectional design, which measures exposure and outcome simultaneously, preventing conclusions about causality. As a result, it remains unclear whether food insecurity contributed to elevated anxiety or whether pre-existing anxiety, through its impact on cognitive function and decision-making, may have led to behaviors that increased the risk of food insecurity.^52,53

Although invitations were stratified to be nationally balanced, online panel membership can under- or over-represent specific subgroups (e.g., very low-income, non-Hebrew speakers). Our achieved distributions for sex, age, and region were broadly similar to CBS profiles; however, residual deviations may remain.

Because Hunger Vital Sign spans 12 months, our food-insecurity prevalence does not solely reflect the first post-attack month. This likely attenuates contemporaneous associations between the war period and food insecurity.

Because this analysis is based on a previously collected dataset, the scope of inference is constrained to the variables originally measured; however, the gender-stratified focus of the present study provides additional interpretive depth beyond report and prior analyses.

This study was conducted during a period of regional conflict and instability, which likely affected multiple populations in diverse ways. Our analysis focuses solely on Israeli residents and should be interpreted within this defined context.

Implications for Practice and/or Policy

The findings of this study underscore the urgent need for public health strategies and policies that are sensitive to gender-based and gendered experiences of food insecurity and psychological distress, particularly during periods of societal crisis such as war. Given that women in this study exhibited significantly higher levels of anxiety, alongside distinct patterns of dietary disruption and supplement use, health systems should prioritize integrated interventions that address both mental health and nutritional resilience. Primary care providers and community health programs should incorporate routine screening for food insecurity and anxiety, especially among women, and offer referrals to psychological support services and nutrition assistance programs. Policymakers should also recognize the disproportionate emotional and logistical burdens placed on women during emergencies, investing in structural supports such as targeted food subsidies, flexible childcare resources, and community-based psychosocial programs. Moreover, adopting an intersectional framework is essential to ensure that interventions reach women who face compounded vulnerabilities due to factors like ethnicity, socioeconomic status, and family structure. Finally, embedding these insights into preparedness plans can strengthen societal resilience, ensuring that responses to future crises do not inadvertently deepen existing gender and health inequities.

Conclusions

Distinct gender-based differences emerged in lifestyle and psychological responses to societal strain. While food insecurity and most sociodemographic factors did not differ by gender, males reported higher physical activity, smoking rates, and personal exposure to injury within their social network. Females were more likely to experience dietary disruptions and reported significantly higher anxiety levels across all measures. Multivariable analysis identified female gender, food insecurity, current smoking, and worsening living conditions as independent predictors of elevated anxiety, while older age was protective. Despite similar MedDiet adherence, behavioral and emotional responses to the crisis diverged markedly. These findings underscore the importance of gender-specific strategies in public health preparedness, particularly in addressing mental health vulnerabilities and supporting adaptive health behaviors during emergencies.

Declaration of Generative AI and AI-Assisted Technologies in the Writing Process

During the preparation of this work, I, Daniela Abigail Navarro, used OpenAI’s ChatGPT in order to refine the English language, primarily to shorten paragraphs, enhance clarity, and make the writing more polished and professional. After using this tool, I carefully reviewed and edited the content to ensure it accurately represents my intended meaning and meets scholarly standards. I take full responsibility for the content of this publication.

Authors’ Contributions

Conceptualization: D.A.N., N.D.Y., and V.K.-S. Data curation: D.A.N. and V.K.S. Formal analysis: D.A.N. Data interpretation: D.A.N., N.D.Y., M.M., and V.K.-S. Funding acquisition: None. Methodology: D.A.N. and V.K.-S. Project administration: V.K.-S. Writing—original draft: DAN. Writing—review and editing: D.A.N., N.D.Y., M.M., and V.K.-S. All authors approved the final version.

Abbreviations Used

ΔPA

Delta Physical Activity

AI

Artificial Intelligence

CBS

Central Bureau of Statistics

CI

Confidence Interval

COVID-19

Coronavirus disease 19

FDR

False Discovery Rate

GAD-7

General Anxiety Disorder

I-MEDAS

Israel Mediterranean Diet Scale

IQR

Interquartile Range

MedDiet

Mediterranean Diet

OR

Odd ratio

PA

Physical Activity

STROBE

Strengthening the Reporting of Observational Studies in Epidemiology

Ethics Statement

The study was approved by the Institutional Ethics Board of Ariel University, Israel (approval number AU-HEA-MB-20231105, November 5, 2023). No survey respondent participated in the study without providing informed consent. Informed consent was indicated by clicking on the appropriate button at the start of the survey, and individuals who did not do so could not proceed with the survey.

Data Availability Statement

The dataset analyzed during the current study is not publicly available due to participant confidentiality and ethical restrictions but may be made available from the corresponding author on reasonable request, contingent upon approval by the Ariel University Institutional Ethics Committee.

Cite this article as: Navarro DA, Yaroslaviz ND, Maor M, Kaufman-Shriqui V (2026) Gender Differences in food insecurity, anxiety, and nutrition in Israel: A cross-sectional study, Women’s Health Reports, 2026, 7, 26884844261436547, DOI: 10.1177/26884844261436547.