ABSTRACT
Background
Household food insecurity has been associated with pregnancy complications and poorer birth outcomes in the United States and with maternal mental disorders in the United Kingdom, but there has been little investigation of the effects of food insecurity during this life stage in Canada.
Objectives
Our objective was to examine the relationship between the food insecurity status of women during pregnancy and maternal and birth outcomes and health in infancy in Canada.
Methods
We drew on data from 1998 women in Ontario, Canada, whose food insecurity was assessed using the Household Food Security Survey Module on the Canadian Community Health Survey, cycles 2005 to 2011–2012. These records were linked to multiple health administrative databases to identify indications of adverse health outcomes during pregnancy, at birth, and during children's first year of life. We included women who gave birth between 9 months prior and 6 months after their interview date, and for whom infant outcome data were available. Multivariable Poisson regression models were used to compare outcomes by maternal food security status, expressed as adjusted relative risks (aRR) with 95% CIs.
Results
While pregnant, 5.6% of women were marginally food insecure and 10.0% were moderately or severely food insecure. Food insecurity was unrelated to pregnancy complications and adverse birth outcomes, but 26.8% of women with moderate or severe food insecurity had treatment for postpartum mental disorders in the 6-month postpartum period, compared to 13.9% of food-secure women (aRR, 1.86; 95% CI, 1.40–2.46). Children born to food-insecure mothers were at elevated risk of being treated in an emergency department in the first year of life (aRR, 1.18; 95% CI, 1.01–1.38).
Conclusions
Maternal food insecurity during pregnancy in Ontario, Canada, is associated with postpartum mental disorders and a greater likelihood of infants being treated in an emergency department.
Keywords: food insecurity, mental disorders, infants, pregnancy, Canada
Introduction
Household food insecurity—the inadequate or insecure access to food due to financial constraints—is a serious public health concern in high-income countries. It is also a problem that disproportionately affects households with children (1, 2). In Canada, 12.7% of households experienced marginal, moderate, or severe food insecurity in 2017–18, but the prevalence was higher (16.2%) among households with a child under 18 years of age (3). While there has been considerable research in Canada to examine the relationships between household food insecurity and adults’ health and health-care costs, the implications of food insecurity for maternal and infant health have received less attention. Yet, household food insecurity has been associated with significantly poorer dietary intakes among women of childbearing age in Canada (4, 5), potentially predisposing food-insecure women who become pregnant to higher risks of poor birth outcomes. Additionally, food insecurity among women in Canada has been associated with higher rates of perceived stress and social isolation (6), cost-related nonadherence to prescription medications (7), and higher rates of depression and depressive symptoms (6, 8, 9). Given that pregnancy and childbirth place additional pressures on women's mental health and their resources, women who are food insecure at this critical life stage may be at increased risk of complications during pregnancy, adverse birth outcomes, and mental illness postpartum (10, 11).
In the United States, food insecurity in households with children has been associated with increased risks of pregnancy complications (12), hospitalization following discharge after birth (13), and adverse birth outcomes (14–16). Additionally, findings from a cohort study from the United Kingdom recently suggested that food insecurity during pregnancy increases the risk of common mental disorders (17). Extrapolations to the Canadian context are limited, however, given between-country differences in health-care systems and the presence of large-scale government interventions for at-risk families in the United States, including the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children. No similar nutrition interventions exist on this scale in Canada, a country for which there is very little data on the effects of food insecurity on maternal and infant health. In Canada, apart from studies of the relationship between household food insecurity and infant feeding practices (18–20), the only insight into the effects of food insecurity on health in early life comes from an ecological study in the province of Quebec, which documented higher rates of small-for-gestational-age births in areas with higher rates of food insecurity (21).
Drawing on linked survey and administrative health-care data from a population-based sample of women living in the province of Ontario, we aimed to examine the relationship between the food insecurity status of women during pregnancy and maternal and birth outcomes and health in infancy.
Methods
Data sources and sample
The Canadian Community Health Survey (CCHS) is a cross-sectional survey of approximately 130,000 respondents per cycle. It is designed to be representative of 98% of the Canadian population aged 12 years and over, omitting individuals living on First Nation reserves, in institutions, in the Canadian Armed forces, and in some remote areas in Quebec. Since 2005, household food insecurity over the past year has been assessed on this survey using an adapted version of the 18-item Household Food Security Survey Module, developed by the US Department of Agriculture (22). The questions in this module range in severity from experiences of anxiety that food will run out before household members have money to buy more, to skipping or cutting the size of meals, to being hungry without eating, and, in the extreme, going whole days without eating, all because of a lack of money for food. Of the 18 items, 10 questions refer to the experiences of adults and 8 refer specifically to the experiences of any children under 18 years of age in the household.
This study drew on data from a linked data set comprising all respondents to CCHS cycles 2005, 2007–2008, 2009–2010, and 2011–2012 who were living in the province of Ontario and who, when surveyed, agreed to allow their responses to be linked to health administrative data for research purposes. These are the years for which standardized food insecurity measurements are available on CCHS and the survey data have been linked to Ontario provincial health administrative data. The data sets were linked using unique, encoded identifiers and were analyzed at the Institute for Clinical and Evaluative Sciences. Approximately 83% of women respondents to these survey cycles consented to this linkage. Delivery/birth information for them was identified using deterministically linked hospital discharge records of the mother and the newborn from the Canadian Institute for Health Information Discharge Abstract Database (23).
We included women living in Ontario who responded to any of the 4 linked cycles of the survey, had a valid Ontario Health Insurance Plan number, and delivered a live infant between 9 months prior and 6 months after the interview date. This time window was selected to ensure that the 12-month measurement of household food insecurity administered at the time of the interview would encompass at least 1 trimester of pregnancy. We excluded women with multiple pregnancies, those with missing responses to the food security module, and those with missing information on education or home ownership. The final analytic sample comprised 1988 women with infants who were followed to their first birthday.
Food insecurity measurement and classification
Maternal food insecurity status was determined from the 10-item adult subscale in the food security module. Using this subscale ensured comparability in the determination of food insecurity among women with and without children at the time of the interview. Applying the classification scheme developed by Health Canada (22), women with 2 or more affirmative responses were considered to be moderately or severely food insecure, referred to here as “food insecure.” Women with a single affirmative response were considered marginally food insecure, given prior research indicating the vulnerability of women and children in households that report any indication of food insecurity on this module (8, 12, 24–29). Health Canada's definition is more inclusive than that of the US Department of Agriculture, which requires 3 or more affirmative responses on the scale to classify a household as food insecure (30). In addition, the term “marginal food insecurity” that is used in Canada is generally called “marginal food security” in the United States.
Health outcome measurements and classifications
We used the Canadian Institute for Health Information Discharge Abstract Database, Ontario Health Insurance Plan claims, Ontario Mental Health Reporting System, and National Ambulatory Care Reporting System to identify indications of illness and adverse health outcomes during pregnancy, at birth, and during the infant's first year of life. The outcomes derived from these administrative databases are based on standardized, diagnostic coding that accompanies physician billings and emergency department or hospital discharge. Maternal health outcomes included admission to hospital during pregnancy, gestational hypertension or diabetes, and postpartum mental disorders. The latter variable described women who had been seen by a primary care physician or psychiatrist or had been hospitalized or treated in an emergency department for depression or other diagnosed mental health problems in the year following delivery. Birth outcomes of interest included birth before 37 completed gestation weeks, admission to a neonatal intensive care unit for congenital anomalies, being small for gestational age, and being large for gestational age, based on a Canadian population-based reference (31). Infant health outcomes included hospitalization and emergency department visits during the first year of life. The administrative health-care databases and diagnostic codes used are detailed in Supplemental Table 1.
Analyses
We compared sociodemographic characteristics and outcome rates among food-secure, marginally food-insecure, and food-insecure groups. All univariable descriptive analyses were performed using the Kruskal-Wallis test for continuous variables and chi-square tests for categorical variables.
Multivariable Poisson regression models with robust standard errors (32) were used to estimate the adjusted relative risks (aRR) for each binary outcome, with 95% CIs, comparing women with versus without food insecurity. The mean length of stay in hospital for childbirth was compared between groups using linear regression. Models were adjusted for factors associated with food insecurity and health outcomes that could confound the relationship between these variables. These included the woman's age group (<20, 20–24, 25–34, or ≥35 y), partnership status (partnered or not partnered), history of a previous live birth (yes/no), housing tenure (own or rent dwelling), residence (urban or rural), main source of household income (wages, social assistance, or other), immigration status (Canadian born or not stated versus immigrant), race (White or non-White), and education level (less than secondary school graduation, secondary school graduate, some post-secondary schooling, or post-secondary school graduate) (28, 33–35). Given other research suggesting that maternal depression can impact the relationship between food insecurity and children's health (36–39), we also ran the multivariable analyses of infant health outcomes, including postpartum mental disorders in the 12 months following birth (yes/no) and any health-care utilization for mental health reasons in the 12 months prior to the birth (yes/no) as covariates (40). Additionally, we reran the multivariable analysis of food insecurity in relation to postpartum mental disorders, adjusted for any health-care utilization for mental health reasons in the 12 months prior to the birth (yes/no).
Statistical significance was determined as a 2-tailed P value < 0.05. Analyses were performed using SAS version 9.4 (SAS Institute, Inc.). Ethics approval for this study was obtained from the Health Sciences Research Ethics Board at the University of Toronto.
Results
The sociodemographic characteristics of the sample, by household food insecurity status, are presented in Table 1. While pregnant, 5.6% of women were marginally food insecure and 10.0% were food insecure. Compared to the food secure, food-insecure and marginally food-insecure women tended to be younger, to be less well educated, and to rent rather than own their dwelling. A significantly higher proportion of food-insecure women did not have a partner and were reliant on income sources other than employment.
TABLE 1.
Sociodemographic characteristics of the sample, by household food insecurity status
| Characteristic | Total, n = 1988 | Food secure,1n = 1678 | Marginal insecurity,2n = 112 | Moderate or severe insecurity,3n = 198 | P value4 |
|---|---|---|---|---|---|
| Age at the index delivery, y | |||||
| Mean ± SD | 29.59 ± 5.38 | 29.96 ± 5.20 | 27.75 ± 5.80 | 27.46 ± 5.94 | <0.001 |
| Median [IQR] | 30 [26–33] | 30 [27–34] | 27 [24–32] | 27 [23–31] | <0.001 |
| Age at the index delivery, n (%) | — | — | — | — | <0.001 |
| <20 y | 75 (3.8) | 50 (3.0) | 12 (10.7) | 13 (6.6) | |
| 20–24 y | 279 (14.0) | 194 (11.6) | 21 (18.8) | 64 (32.3) | |
| 25–29 y | 621 (31.2) | 530 (31.6) | 39 (34.8) | 52 (26.3) | |
| 30–34 y | 649 (32.6) | 579 (34.5) | 29 (25.9) | 41 (20.7) | |
| ≥35 y | 364 (18.3) | 325 (19.4) | 11 (9.8) | 28 (14.1) | |
| Previous live birth | — | — | — | — | 0.16 |
| None | 937 (47.1) | 806 (48.0) | 49 (43.8) | 82 (41.4) | |
| ≥1 | 1051 (52.9) | 872 (52.0) | 63 (56.3) | 116 (58.6) | |
| Educational attainment, n (%) | — | — | — | — | <0.001 |
| <Secondary school graduation | 172 (8.7) | 119 (7.1) | 17 (15.2) | 36 (18.2) | |
| Secondary graduate | 297 (14.9) | 232 (13.8) | 21 (18.8) | 44 (22.2) | |
| Some post–secondary school | 112 (5.6) | 79 (4.7) | 10 (8.9) | 23 (11.6) | |
| Post–secondary school graduate | 1407 (70.8) | 1248 (74.4) | 64 (57.1) | 95 (48.0) | |
| Number of children <18 y in household | |||||
| Mean ± SD | 1.40 ± 1.05 | 1.36 ± 1.03 | 1.46 ± 1.08 | 1.70 ± 1.19 | <0.001 |
| Median [IQR] | 1 [1–2] | 1 [1–2] | 1 [1–2] | 2 [1–2] | <0.001 |
| Partnership status, n (%) | — | — | — | — | <0.001 |
| Married or common law | 1714 (86.2) | 1502 (89.5) | 86 (76.8) | 126 (63.6) | |
| Single, divorced, separated, or widowed | 274 (13.8) | 176 (10.5) | 26 (23.2) | 72 (36.4) | |
| White, n (%) | — | — | — | — | <0.001 |
| No | 410 (20.6) | 320 (19.1) | 32 (28.6) | 58 (29.3) | |
| Yes | 1578 (79.4) | 1358 (80.9) | 80 (71.4) | 140 (70.7) | |
| Immigrant status, n (%) | — | — | — | — | 0.20 |
| Non-immigrant or not stated | 1587 (79.8) | 1346 (80.2) | 82 (73.2) | 159 (80.3) | |
| Immigrant | 401 (20.2) | 332 (19.8) | 30 (26.8) | 39 (19.7) | |
| Main source of household income, n (%) | — | — | — | — | <0.001 |
| Wages, salaries, self-employment | 1768 (88.9) | 1554 (92.6) | 92 (82.1) | 122 (61.6) | |
| Social assistance | 92 (4.6) | 44 (2.6) | 7 (6.3) | 41 (20.7) | |
| Other | 128 (6.4) | 80 (4.8) | 13 (11.6) | 35 (17.7) | |
| Housing tenure, n (%) | — | — | — | — | <0.001 |
| Owner | 1402 (70.5) | 1282 (76.4) | 59 (52.7) | 61 (30.8) | |
| Renter | 586 (29.5) | 396 (23.6) | 53 (47.3) | 137 (69.2) | |
| Rural residence, n (%) | — | — | — | — | 0.17 |
| Urban | 1575 (79.2) | 1319 (78.6) | 89 (79.5) | 167 (84.3) | |
| Rural | 413 (20.8) | 359 (21.4) | 23 (20.5) | 31 (15.7) | |
| Any health-care utilization for mental health reasons in the 12 mo prior to the index birth, n (%) | — | — | — | — | <0.001 |
| Yes | 326 (16.4) | 254 (15.1) | 16 (14.3) | 56 (28.3) | |
| No | 1662 (83.6) | 1424 (84.9) | 96 (85.7) | 142 (71.7) | |
Values are mean ± SD, median [IQR], or frequency (%).
No affirmative response on the 10-item adult subscale of the Household Food Security Survey Module.
1 affirmative response to the 10-item adult subscale of the Household Food Security Survey Module.
≥2 affirmative responses to the 10-item adult subscale of the Household Food Security Survey Module.
P values were derived from the Kruskal-Wallis test for continuous variables and chi-square tests for categorical variables.
Pregnancy and maternal health
Although there was little difference in rates of hospitalization during pregnancy and gestational hypertension or diabetes by food insecurity status, the prevalence of postpartum mental disorders was higher among food-insecure women (34.8% versus 15.2% among marginally food-insecure women and 20.6% among the food secure; Table 2).
TABLE 2.
Maternal, birth, and infant health outcome rates, by food insecurity status
| Characteristic | Total | Food secure | Marginal insecurity | Moderate or severe insecurity | P value1 |
|---|---|---|---|---|---|
| Maternal outcomes | |||||
| Admission to hospital during pregnancy, n (%) | 135 (6.8) | 109 (6.5) | 9 (8.0) | 17 (8.6) | 0.47 |
| Length of stay in hospital for childbirth episode, d, mean ± SD | 2.32 ± 1.58 | 2.31 ± 1.61 | 2.24 ± 1.01 | 2.38 ± 1.59 | 0.73 |
| Pregnancy complication: hypertension or diabetes, n (%) | 157 (7.9) | 130 (7.7) | 8 (7.1) | 19 (9.6) | 0.63 |
| Postpartum mental disorders | |||||
| 1–179 d after the index delivery, n (%) | 297 (14.9) | 234 (13.9) | 10 (8.9) | 53 (26.8) | <0.001 |
| 180–364 d after the index delivery, n (%) | 267 (13.4) | 215 (12.8) | 10 (8.9) | 42 (21.2) | 0.002 |
| 0–364 d after the index delivery, n (%) | 431 (21.7) | 345 (20.6) | 17 (15.2) | 69 (34.8) | <0.001 |
| Birth outcomes | |||||
| Newborn with clinical gestation <37 wk, n (%) | 107 (5.4) | 89 (5.3) | 6 (5.4) | 12 (6.1) | 0.91 |
| Admission to neonatal intensive care unit, n (%) | 231 (11.6) | 193 (11.5) | 10 (8.9) | 28 (14.1) | 0.36 |
| Small for gestational age, n (%) | 152 (7.6) | 131 (7.8) | 6 (5.4) | 15 (7.6) | 0.64 |
| Large for gestational age, n (%) | 245 (12.3) | 205 (12.2) | 14 (12.5) | 26 (13.1) | 0.93 |
| Congenital anomalies, n (%) | 91 (4.6) | 73 (4.4) | 9 (8.0) | 9 (4.5) | 0.20 |
| Infant outcomes | |||||
| At least 1 hospital admission in first year of life, n (%) | 166 (8.4) | 132 (7.9) | 10 (8.9) | 24 (12.1) | 0.12 |
| At least 1 emergency department visit in first year of life, n (%) | 877 (44.1) | 708 (42.2) | 59 (52.7) | 110 (55.6) | <0.001 |
Values are mean ± SD, median [IQR], or frequency (%).
P values were derived from the chi-square test for categorical variables and Kruskal-Wallis test for continuous variable.
In multivariable analyses, maternal food insecurity was not associated with hospitalization during pregnancy, length of stay in hospital for childbirth, or the risk of complications due to gestational hypertension or diabetes (Table 3). Compared to food-secure women, food-insecure women had elevated risks of being treated for postpartum mental disorders in the first 6 months following delivery (aRR, 1.86; 95% CI, 1.40–2.46) and in the subsequent 6 months (aRR, 1.38; 95% CI, 1.00–1.90; Table 3). When health-care utilization for mental health reasons in the 12 months prior to the birth was taken into account, only the risk of food-insecure women being treated for postpartum mental disorders in the first 6 months remained significantly elevated (aRR, 1.61; 95% CI, 1.21–2.13; Table 4). No difference in risk was observed between marginally food-insecure and food-secure women.
TABLE 3.
Associations between food insecurity status and maternal, birth, and infant health outcomes
| Marginal food insecurity | Moderate/severe food insecurity | |||
|---|---|---|---|---|
| Characteristic | Crude RR1 (95% CI) | Adjusted RR2 (95% CI) | Crude RR1 (95% CI) | Adjusted RR2 (95% CI) |
| Maternal outcomes | ||||
| Admission to hospital during pregnancy | 1.24 (0.64–2.38) | 1.09 (0.57–2.09) | 1.32 (0.81–2.16) | 0.94 (0.55–1.60) |
| Mean length of stay in hospital for childbirth episode, d | 0.97 (0.86–1.09)3 | 1.00 (0.89–1.11)4 | 1.03 (0.95–1.12)3 | 1.03 (0.94–1.13)4 |
| Pregnancy complication: hypertension or diabetes | 0.92 (0.46–1.83) | 0.99 (0.51–1.92) | 1.24 (0.78–1.96) | 1.44 (0.88–2.35) |
| Postpartum mental disorders | ||||
| 1–179 d after the index delivery | 0.64 (0.35–1.17) | 0.66 (0.36–1.22) | 1.92 (1.48–2.49)** | 1.86 (1.40–2.46)** |
| 180–364 d after the index delivery | 0.70 (0.38–1.28) | 0.65 (0.35–1.19) | 1.66 (1.23–2.23)** | 1.38 (1.00–1.90)* |
| 0–364 d after the index delivery | 0.74 (0.47–1.16) | 0.72 (0.46–1.13) | 1.69 (1.37–2.10)** | 1.53 (1.21–1.93)** |
| Birth outcomes | ||||
| Newborn with clinical gestation <37 wk | 1.01 (0.45–2.26) | 1.03 (0.45–2.35) | 1.14 (0.64–2.05) | 0.98 (0.51–1.90) |
| Admission to neonatal intensive care unit | 0.78 (0.42–1.42) | 0.75 (0.41–1.38) | 1.23 (0.85–1.78) | 1.09 (0.73–1.62) |
| Congenital anomalies | 1.85 (0.95–3.59) | 1.96 (0.99–3.86) | 1.04 (0.53–2.06) | 1.13 (0.52–2.45) |
| Small for gestational age | 0.69 (0.31–1.52) | 0.57 (0.26–1.27) | 0.97 (0.58–1.62) | 0.80 (0.47–1.37) |
| Large for gestational age | 1.02 (0.62–1.70) | 1.13 (0.68–1.87) | 1.07 (0.73–1.57) | 1.14 (0.75–1.72) |
| Infant outcomes | ||||
| At least 1 hospital admission in first year of life | 1.14 (0.61–2.10) | 1.08 (0.59–1.95) | 1.54 (1.02–2.32)* | 1.44 (0.89–2.32) |
| At least 1 emergency department visit in first year of life | 1.25 (1.04–1.50)* | 1.17 (0.97–1.40) | 1.32 (1.15–1.51)** | 1.18 (1.01–1.38)* |
*P < 0.05, **P < 0.001.
RRs are reported unless otherwise specified.
Unless otherwise specified, values are RRs derived from Poisson regression analyses, adjusted for age group, partnership status, previous live birth, housing tenure, rural residence, main income source of the household, immigration status, race, and education level.
Mean difference in length of stay (95% CI).
Mean difference in length of stay (95% CI), derived from linear regression model adjusted for age, partnership status, previous live birth, housing tenure, rural residence, main income source of the household, immigration status, race, and education level.
TABLE 4.
Associations between food insecurity status and maternal and infant outcomes, adjusting for maternal mental health
| Adjusted1 RRs (95% CI) | ||
|---|---|---|
| Characteristic | Marginal food insecurity | Moderate/severe food insecurity |
| Maternal outcomes, postpartum mental disorders | ||
| 1–179 d after the index delivery | 0.68 (0.38–1.22) | 1.61 (1.21–2.13)** |
| 180–364 d after the index delivery | 0.65 (0.37–1.16) | 1.18 (0.86–1.63) |
| 0–364 d after the index delivery | 0.73 (0.47–1.12) | 1.36 (1.07–1.71)* |
| Infant outcomes | ||
| At least 1 hospital admission in first year of life | 1.10 (0.61–2.00) | 1.41 (0.87–2.31) |
| At least 1 emergency department visit in first year of life | 1.18 (0.98–1.41) | 1.17 (1.00–1.36) |
| Small for gestational age | 0.57 (0.26–1.28) | 0.80 (0.47–1.37) |
| Large for gestational age | 1.13 (0.68–1.88) | 1.16 (0.77–1.76) |
*P < 0.05, **P < 0.01.
Relative risk (95% CI) derived from Poisson regression models, adjusted for age group, partnership status, previous live birth, housing tenure, rural residence, main income source of the household, immigration status, race, education level, and any health-care utilization by the mother for mental health reasons in the 12 months prior to the birth. Models predicting infant outcomes were additionally adjusted for evidence of postpartum mental disorders in the 12 months following birth.
Birth outcomes and health in infancy
There were no significant differences in the risks of a preterm birth, congenital anomalies, or infants who were small or large for gestational age by food insecurity status (Table 3). Similarly, maternal food insecurity was unrelated to the risk of infants being admitted to a neonatal intensive care unit.
The risk of being hospitalized in the first year of life did not differ by maternal food insecurity status, once maternal and household sociodemographic characteristics were taken into account (Table 3). Children born to food-insecure mothers were at elevated risk of being treated in an emergency department in the first year of life (aRR, 1.18; 95% CI, 1.01–1.38). This association was similar upon adjustment for the mother's prior history of treatment for mental health reasons (aRR, 1.17; 95% CI, 1.00–1.36; Table 4). There was no relationship between marginal food insecurity and emergency department visits (Table 3).
Discussion
In a population-based sample of women in the Canadian province of Ontario (2005 to 2012), those who reported moderate or severe food insecurity while pregnant were more likely than other women to be treated for postpartum mental disorders. However, maternal food insecurity was not associated with complications during pregnancy or adverse birth outcomes. Children born to moderately or severely food-insecure mothers were at greater risk of being treated in an emergency department during infancy. After accounting for confounding factors, marginal food insecurity was not associated with the risk of maternal or infant health outcomes.
Our finding of an increased risk of postpartum mental disorders among food-insecure women is consistent with research in the United Kingdom (17), and with the much broader body of literature linking food insecurity to mental illness in Canada (6, 8, 9, 40) and the United States (38, 41–44). Our results also add to prior research charting the elevated health-care utilization among food-insecure adults in Ontario (5, 40, 45), suggesting that household food insecurity is linked to health and health-care utilization postpartum and in infancy.
Although research in the United States has documented significant associations between food insecurity and pregnancy complications (12), hospitalization following discharge after birth (13), and adverse birth outcomes (14–16), we did not find evidence of similar associations in this Canadian sample. The absence of differences in preterm births and the prevalence of infants who were small or large for their gestational age could perhaps in part be attributable to the supports available to vulnerable pregnant women through the Canada Prenatal Nutrition Program (CPNP) (46). This federal program was launched with the explicit goal of increasing the number of babies born with healthy birth weights. Delivered in partnership with the provinces and territories, CPNP funds community programs to deliver a range of services and supports, including some food and nutrition supplements, to at-risk women. A 2012 evaluation indicated better birth outcomes among program participants (47), but no population data on participation rates or program impacts among women in food-insecure households are available. Without information on CPNP program participation among the women in our sample, it is impossible to determine whether this program affected their birth outcomes.
Although we did not find higher rates of pregnancy complications or birth outcomes among food-insecure women, this group had marked differences in their mental health, which may relate to the particular nature of the publicly funded Canadian health-care system. This system appears to mitigate socioeconomic disparities in access to physical health care to a greater extent than mental health services (35). Although the mental health care provided by primary care physicians and psychiatrists is fully funded, care by clinical psychologists or other nonphysician providers is generally not funded by the public system. We speculate that food-insecure women may be more likely to experience prolonged or untreated mental health challenges due to their reduced access to services and social supports that are not publicly funded; as a result, they may be at increased risk of eventually meeting the physician-ascertained outcomes that were measured in this study (i.e., assessment or treatment by psychiatrist; emergency department visit for a mental health issue).
Strengths of this study include our use of a population-based sample, with a standardized, well-validated measure of food insecurity and objective measures of health and health-care use based on physician diagnoses and administrative data on hospitalizations and emergency department visits in a universal health-care system. Our study pooled data for women in Ontario from 2005 to 2012: a period when the provincial prevalence of moderate/severe household food insecurity ranged from 7.7% to 9.2% (48). The provincial prevalence has remained relatively stable since then, with 9.0% of households in Ontario classified as moderately or severely food insecure in 2017–18 (3). Although there are regional and socioeconomic differences in health-care access in Ontario (34), such biases are likely to lessen the probability of food-insecure mothers using the health-care system, thus dampening our ability to see associations between maternal food insecurity and the health outcomes considered here. Our study may also have been affected by nonresponse bias. Although 83% of women living in Ontario who participated in the CCHS agreed to have their data linked to their health records, the representativeness of the sample cannot be ascertained from the available data.
Our study has 3 limitations that can be addressed in future research. First, although we had data on women's histories of health-care use for mental health problems, we were reliant on a single, cross-sectional assessment of food insecurity and household material resources. Our 12-month measure of food insecurity encompassed at least 1 trimester of pregnancy, but knowing the extent to which food insecurity also occurred over other time periods would advance our understanding of the pathways through which maternal food insecurity influenced the outcomes charted here. Second, missing data for household income in CCHS precluded an adjustment for this variable in our models. Although we included adjustments for housing tenure and main source of household income to capture broad differences in household resources, future Canadian research should aim to differentiate the effects of food insecurity in relation to maternal and infant health from the effects of low income more generally. Third, the low number of women with moderate/severe food insecurity limited our power to assess the relationships between food insecurity and relatively rare birth outcomes. US studies documenting associations between maternal food insecurity and birth defects, preterm births, and low birth weights have been much larger and/or focused on deprived samples (14–16), suggesting that more extensive investigation of these conditions in association with food insecurity is warranted in Canada.
Prior Canadian research has documented food-insecure women's lower likelihood of sustaining exclusive breastfeeding (18, 20) and their struggles to access needed formula (20, 49–51). Our results importantly extend this understanding of vulnerability, revealing the elevated risk of maternal mental disorders and infants’ treatment in emergency departments in the context of moderate/severe food insecurity. Although our data on food insecurity in these families were limited to a single 12-month period, food insecurity disproportionately affects families with children in Canada (3), and the sociodemographic characteristics of the women who were food insecure during pregnancy suggest that many are at elevated risk of food insecurity going forward (28). This is particularly concerning in light of US research linking food insecurity to impaired mental health and to development and behavioral problems among young children (36–38, 52–54), and evidence that the combination of maternal depression and food insecurity negatively impacts infant feeding practices, heightening the risk of overweight among toddlers (39). Our study does not furnish proof of a causal relationship between maternal food insecurity and maternal and infant health, but considering the present findings in the context of the broader body of research on food insecurity and health, it seems reasonable that interventions to reduce food insecurity among families in Canada would lay a better foundation for maternal and child health. Whereas public food and nutrition assistance programs such as SNAP are central to efforts to address food insecurity in the United States (55–57), cash transfer programs define the social safety nets of many other high-income countries, including Canada (58). In Canada, federal and provincial child benefits have been shown to have positive effects on the food security of low-income families (33, 59–61), but the potential of these benefit programs to address family food insecurity is yet to be realized. The prevalence of severe food insecurity among low-income families fell by one-third following the recent introduction of the Canada Child Benefit, but the benefit was not designed to reduce food insecurity, and the overall prevalence of food insecurity among Canadian families remains unchanged (59). However, the analysis suggested that improving the benefit for low-income families would be an effective means to reduce their risk of food insecurity. Our findings highlight the potential to improve maternal mental health and infant health through interventions that reduce food insecurity.
Supplementary Material
Acknowledgments
The author's responsibilities were as follows—VT, MLU, CG, and DER: conceptualized the study and designed the analysis; XW, VT, and MLU: had access to the data; XW: performed the statistical analyses; VT and MLU: oversaw all aspects of the data analysis and had primary responsibility for the final content; VT: led the writing of the manuscript; and all authors: participated in the interpretation of the results and manuscript development, and read and approved the final manuscript.
Notes
This study was supported by the Canadian Institutes of Health Research (PJT 153260 to VT and MLU) and by the Institute for Clinical and Evaluative Sciences, which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of data; or preparation, review, decision to submit for publication, or approval of the manuscript. The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by the Institute for Clinical and Evaluative Sciences or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred. Parts of this material are based on data and/or information compiled and provided by the Canadian Institute for Health Information. However, the analyses, conclusions, opinions and statements expressed in the material are those of the authors, and not necessarily those of Canadian Institute for Health Information.
Author disclosures: MLU is supported by a Canada Research Chair in Applied Population Health. VT, CG, XW, and DER, no conflicts of interests.
Supplemental Table 1 is available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at http://jn.nutrition.org.
Abbreviations used: aRR, adjusted relative risk; CCHS, Canadian Community Health Survey; CPNP, Canada Prenatal Nutrition Program; SNAP, Supplemental Nutrition Assistance Program.
Contributor Information
Valerie Tarasuk, Department of Nutritional Sciences, University of Toronto, Toronto, Canada.
Craig Gundersen, Department of Agricultural and Consumer Economics, University of Illinois, Urbana, IL, USA.
Xuesong Wang, Institute for Clinical and Evaluative Sciences Central, Toronto, Canada.
Daniel E Roth, Department of Nutritional Sciences, University of Toronto, Toronto, Canada; The Hospital for Sick Children, Toronto, Canada.
Marcelo L Urquia, Institute for Clinical and Evaluative Sciences Central, Toronto, Canada; Manitoba Centre for Health Policy, University of Manitoba, Manitoba, Canada; Dalla Lana School of Public Health, University of Toronto, Toronto, Canada.
References
- 1. Smith M, Rabbitt M, Coleman-Jensen A. Who are the world's food insecure? New evidence from the Food and Agriculture Organization's Food Insecurity Experience Scale. World Dev. 2017;93:402–12. [Google Scholar]
- 2. Coleman-Jensen A, Rabbitt M, Gregory C, Singh A. Household food security in the United States in 2017. United States Department of Agriculture, Economic Research Service;2018. [Google Scholar]
- 3. Tarasuk V, Mitchell A. Household food insecurity in Canada, 2017–18. Toronto (Canada): Research to Identify Policy Options to Reduce Food Insecurity (PROOF); 2020. [Google Scholar]
- 4. Kirkpatrick S, Tarasuk V.. Food insecurity is associated with nutrient inadequacies among Canadian adults and adolescents. J Nutr. 2008;138:604–12. [DOI] [PubMed] [Google Scholar]
- 5. Kirkpatrick S, Dodd KW, Parsons R, Ng C, Garriguet D, Tarasuk V. Household food insecurity is a stronger marker of adequacy of nutrient intakes among Canadian compared to American youth and adults. J Nutr. 2015;145:1596–603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Martin M, Maddocks E, Chen Y, Gilman S, Colman I. Food insecurity and mental illness: disproportionate impacts in the context of perceived stress and social isolation. Public Health. 2016;132:86–91. [DOI] [PubMed] [Google Scholar]
- 7. Men F, Gundersen C, Urquia M, Tarasuk V. Prescription medication nonadherence associated with food insecurity: a population-based cross-sectional study. CMAJ Open. 2019;7:E590–E7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jessiman-Perreault G, McIntyre L.. The household food insecurity gradient and potential reductions in adverse population mental health outcomes in Canadian adults. SSM Popul Health. 2017;3:464–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Muldoon KA, Duff PK, Fielden S, Anema A. Food insufficiency is associated with psychiatric morbidity in a nationally representative study of mental illness among food insecure Canadians. Soc Psychiatry Psychiatr Epidemiol. 2013;48:795–803. [DOI] [PubMed] [Google Scholar]
- 10. Blumenshine P, Egerter S, Barclay C, Cubbin C, Braveman P. Socioeconomic disparities in adverse birth outcomes: a systematic review. Prev Med. 2010;39:263–72. [DOI] [PubMed] [Google Scholar]
- 11. Ramakrishnan U, Grant F, Goldenberg T, Zongrone A, Martorell R. Effect of women's nutrition before and during early pregnancy on maternal and infant outcomes: a systematic review. Paediatr Perinat Epidemiol. 2012;26:285–301. [DOI] [PubMed] [Google Scholar]
- 12. Laraia BA, Siega-Riz AM, Gundersen C. Household food insecurity is associated with self-reported pregravid weight status, gestational weight gain and pregnancy complications. J Am Diet Assoc. 2010;110:692–701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Cook J, Frank DA, Levenson SM, Neault NB, Heeren TC, Black MM, Berkowitz C, Casey PH, Meyers AF, Cutts DB et al. . Child food insecurity increases risks posed by household food insecurity to young children's health. J Nutr. 2006;136:1073–6. [DOI] [PubMed] [Google Scholar]
- 14. Bryant Borders A, Grobman W, Amsden L, Holl J. Chronic stress and low birth weight in a low income population of women. Obstet Gynecol. 2007;109:331–8. [DOI] [PubMed] [Google Scholar]
- 15. Carmichael SL, Yang W, Herring A, Abrams B, Shaw GM. Maternal food insecurity is associated with increased risk of certain birth defects. J Nutr. 2007;137:2087–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Tucker C, Berrien K, Menard M, Herring A, Daniels J, Rowley D, Tucker Halpern C. Predicting preterm birth among women screened by North Carolina's Pregnancy Medical Home Program. Matern Child Health J. 2015;19:2438–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Power M, Uphoff E, Kelly B, Pickett K. Food insecurity and mental health: an analysis of routine primary care data of pregnant women in the Born in Bradford cohort. J Epidemiol Community Health. 2017;71:324–8. [DOI] [PubMed] [Google Scholar]
- 18. Orr S, Dachner N, Frank L, Tarasuk V. The relationship between household food insecurity and infant feeding practices among Canadian mothers. Can Med Assoc J. 2018;190:E312–E9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Frank L. Exploring infant feeding practices in food insecure households: what is the real issue? Food Foodways. 2015;23:186–209. [Google Scholar]
- 20. Frank L. Out of milk: infant food insecurity in a rich nation. Vancouver (Canada): UBC Press; 2020. [Google Scholar]
- 21. Savard N, Levallois P, Rivest L, Gingras S. A study of the association between characteristics of CLSCs and the risk of small for gestational age births among term and preterm births in Quebec, Canada. Can J Public Health. 2012;103:152–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Health Canada. Canadian Community Health Survey, cycle 2.2, nutrition (2004)—income-related household food security in Canada. Ottawa (Canada): Office of Nutrition Policy and Promotion, Health Products and Food Branch, Health Canada; 2007. Report No. 4696. [Google Scholar]
- 23. Joseph K, Fahey J. Validation of perinatal data in the discharge abstract database of the Canadian Institute of Health Information. Chronic Dis Can. 2009;29:96–100. [PubMed] [Google Scholar]
- 24. Coleman-Jensen AJ. U.S. food insecurity status: toward a refined definition. Soc Indic Res. 2010;95:215–30. [Google Scholar]
- 25. Cook JT, Black M, Chilton M, Cutts D, Ettinger de Cuba S, Heeren T, Rose-Jacobs R, Sandel M, Casey PH, Coleman S et al. . Are food insecurity's health impacts underestimated in the US population? Marginal food security also predicts adverse health outcomes in young US children and mothers. Adv Nutr. 2013;4:51–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Loopstra R, Tarasuk V. Severity of household food insecurity is sensitive to change in household income and employment status among low-income families. J Nutr. 2013;143:1316–23. [DOI] [PubMed] [Google Scholar]
- 27. Burke M, Martini L, Cayir E, Hartline-Grafton H, Meade R. Severity of household food insecurity is positively associated with mental disorders among children and adolescents in the United States. J Nutr. 2016;146:2019–26. [DOI] [PubMed] [Google Scholar]
- 28. Tarasuk VS, Fafard St-Germain AA, Mitchell A. Geographic and socio-demographic predictors of household food insecurity in Canada, 2011–12. BMC Public Health. 2019;19:1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Tarasuk V, Cheng J, de Oliveira C, Dachner N, Gundersen C, Kurdyak P. Association between household food insecurity and annual health care costs. Can Med Assoc J. 2015;187:E429–E36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Coleman-Jensen A, Rabbitt M, Gregory C, Singh A. Household food security in the United States in 2018. U.S. Department of Agriculture, Economic Research Service;2019. [Google Scholar]
- 31. Kramer M, Platt R, Wen S, Joseph K, Allen A, Abrahamowicz M, Blondel B, Breart G, Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System . A new and improved population based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108:1–7. [DOI] [PubMed] [Google Scholar]
- 32. Zou G. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702–6. [DOI] [PubMed] [Google Scholar]
- 33. Tarasuk V, Li N, Dachner N, Mitchell A. Household food insecurity in Ontario during a period of poverty reduction, 2005–2014. Can Public Policy. 2019;45:93–104. [Google Scholar]
- 34. Brien S, Grenier L, Kapral M, Kurdyak P, Vigod S. Taking stock: a report on the quality of mental health and addiction services in Ontario. Toronto (Canada): Health Quality Ontario and Institute for Clinical Evaluative Sciences; 2015. [Google Scholar]
- 35. Bartram M. Income-based inequities in access to mental health services in Canada. Can J Public Health. 2019;110:395–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Rose-Jacobs R, Black M, Casey P, Cook J, Cutts D, Chilton M, Heeren T, Levenson SM, Meyers AF, Frank D. Household food insecurity: associations with at-risk infant and toddler development. Pediatrics. 2008;121:65–72. [DOI] [PubMed] [Google Scholar]
- 37. Zaslow M, Bronte-Tinkew J, Capps R, Horowitz M, Moore K, Weinstein D. Food security during infancy: implications for attachment and mental proficiency in toddlerhood. Matern Child Health J. 2009;1366–80. [DOI] [PubMed] [Google Scholar]
- 38. Whitaker RC, Phillips SM, Orzol SM. Food insecurity and the risks of depression and anxiety in mothers and behavior problems in their preschool-aged children. Pediatrics. 2006;118:e859–e68. [DOI] [PubMed] [Google Scholar]
- 39. Bronte-Tinkew J, Zaslow M, Capps R, Horowitz A, McNamara M. Food insecurity works through depression, parenting, and infant feeding to influence overweight and health in toddlers. J Nutr. 2007;137:2160–5. [DOI] [PubMed] [Google Scholar]
- 40. Tarasuk V, Cheng J, Gundersen C, De Oliveira C, Kurdyak P. The relation between food insecurity and mental health service utilization in Ontario. Can J Psychiatry. 2018;63:557–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Stuff JE, Casey PH, Szeto KL, Gossett JM, Robbins JM, Simpson PM, Connell C, Bogle ML. Household food insecurity is associated with adult health status. J Nutr. 2004;134:2330–5. [DOI] [PubMed] [Google Scholar]
- 42. Leung C, Epel E, Willett WC, Rimm EB, Laraia BA. Household food insecurity is positively associated with depression among low-income Supplemental Nutrition Assistance Program participants and income-ineligible nonparticipants. J Nutr. 2015;145:622–7. [DOI] [PubMed] [Google Scholar]
- 43. Heflin CM, Siefert K, Williams DR. Food insufficiency and women's mental health: findings from a 3-year panel of welfare recipients. Soc Sci Med. 2005;61:1971–82. [DOI] [PubMed] [Google Scholar]
- 44. Maynard M, Andrade L, Packull-McCormick S, Perlman CM, Leos-Toro C, Kirkpatrick S. Food insecurity and mental health among females in high-income countries. Int J Environ Res Public Health. 2018;15:1–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Fitzpatrick T, Rosella L, Calzavara A, Petch J, Pinto A, Manson H, Goel V, Wodchis W. Looking beyond income and education: socioeconomic status gradients among future high-cost users of health care. Am J Prev Med. 2015;49:161–71. [DOI] [PubMed] [Google Scholar]
- 46. Government of Canada. Canada Prenatal Nutrition Program (CPNP). 2020; [Internet]. [Accessed 2020 Jun 18]. Available from: https://www.canada.ca/en/public-health/services/health-promotion/childhoodadolescence/programs-initiatives/canada-prenatal-nutrition-program-cpnp.html [Google Scholar]
- 47. Muhajarine N, Ng J, Bowen A, Cushon J, Johnson S. Understanding the impact of the Canada Prenatal Nutrition Program: a quantitative evaluation. Can J Public Health. 2012;103:S26–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Tarasuk V, Mitchell A, Dachner N. Household food insecurity in Canada, 2012. Toronto (Canada): Research to Identify Policy Options to Reduce Food Insecurity (PROOF); 2014. [Google Scholar]
- 49. Frank L. The breastfeeding paradox: a critique of policy related to infant food insecurity in Canada. Food Cult Soc. 2015;18:107–29. [Google Scholar]
- 50. Frank L. Finding formula: community-based organizational responses to infant formula needs due to household food insecurity. Can Food Stud. 2018;5:90–112. [Google Scholar]
- 51. Partyka B, Whiting S, Grunerud D, Archibald K, Quennell K. Infant nutrition in Saskatoon: barriers to infant food security. Can J Diet Pract Res. 2010;71:79–84. [DOI] [PubMed] [Google Scholar]
- 52. Althoff R, Ametti M, Bertmann F. The role of food insecurity in developmental psychopathology. Prev Med. 2016;92:106–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Council on Community Pediatrics and Committee on Nutrition. Promoting food security for all children. Pediatrics. 2015;136:e1431–8. [DOI] [PubMed] [Google Scholar]
- 54. Shankar P, Chung R, Frank D. Association of food insecurity with children's behavioral, emotional, and academic outcomes: a systematic review. J Dev Behav Pediatr. 2017;38:135–50. [DOI] [PubMed] [Google Scholar]
- 55. Gundersen C, Kreider B, Pepper J. Partial identification methods for evaluating food assistance programs: a case study of the causal impact of SNAP on food insecurity. Am J Agr Econ. 2017;99:875–94. [Google Scholar]
- 56. Bartfeld J, Gundersen C, Smeeding TM, Ziliak JP (eds). SNAP matters: how food stamps affect health and well-being. Redwood City (CA): Stanford University Press; 2015. [Google Scholar]
- 57. Gundersen C, Kreider B, Pepper J, Tarasuk V. Food assistance programs and food insecurity: implications for Canada in light of the mixing problem. Empir Econ. 2017;52:1065–87. [Google Scholar]
- 58. Fernald L, Gosliner W. Alternatives to SNAP: global approaches to addressing childhood poverty and food insecurity. Am J Public Health. 2019;109:1668–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Brown E, Tarasuk V. Money speaks: reductions in severe food insecurity follow the Canada Child Benefit. Prev Med. 2019;129:1–8. [DOI] [PubMed] [Google Scholar]
- 60. Ionescu-Ittu R, Glymour M, Kaufman J. A difference-in-difference approach to estimate the effect of income-supplementation on food insecurity. Prev Med. 2015;70:108–16. [DOI] [PubMed] [Google Scholar]
- 61. Milligan K, Stabile M. Do child tax benefits affect the well-being of children? Evidence from Canadian child benefit expansions. Am Econ J Econ Policy. 2011;3:175–205. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.