Household food insecurity in Ontario.

Tarasuk, Valerie ; Vogt, Janet

Although questions about food insecurity have been included on several national surveys, (1-4) the 2004 Canadian Community Health Survey (CCHS 2.2) marks the first time that food insecurity has been assessed in the Canadian population using a standardized instrument with known measurement properties. (5,6) The assessment identified 9.2% of Canadian households as food insecure in the previous 12 months. (5) This must be an underestimate, however, because CCHS 2.2 did not include the territories, on-reserve First Nations people, or homeless people--three groups in which high levels of food insecurity have been previously documented. (1,7-11) Within provinces, the observed prevalence ranged from 8.1% in Saskatchewan to 14.6% in Nova Scotia. (5) Ontario, with a prevalence of 8.4%, was home to an estimated 379,100 food-insecure households, one third of all food-insecure households in the country. (12)

An examination of the dietary intake data collected in CCHS 2.2 revealed that household food insecurity is associated with inadequate nutrient intakes among adults and adolescents. (13) These results, coupled with the evidence that individuals in food-insecure households in Canada have poorer health than others across a variety of measures, (1-4) present a clear population health imperative to address household food insecurity. Building on the descriptive statistics released by Health Canada, (12) we undertook further examination of the data from CCHS 2.2 to identify socio-demographic factors associated with food insecurity in the Ontario population, to help inform provincial responses to this problem. Province-specific analyses are needed because provincial governments are responsible for setting minimum wages, social assistance benefit levels, and other policies of particular relevance to problems of household food insecurity.

METHODS

All analyses were conducted using data from the Ontario Share File of the CCHS 2.2 (2004). This sample of 10,517 individuals comprises the 96.3% of Ontario respondents who gave permission to share their information with the province. Because food security was assessed at the level of the household rather than the individual, weights specifically designed for household-level analyses were applied in this study to estimate the number of households experiencing food insecurity.

Household food security status was determined from the 18-item Household Food Security Survey Module (HFSSM), following the classifications developed by Health Canada. (5) A household was considered food insecure if there were indications of compromise in the quality and/or quantity of food consumed among either adults or children in the household because of financial constraints. This was denoted by two or more affirmative responses on either the 10-item adult subscale or 8-item child subscale on the HFSSM.

[FIGURE 1 OMITTED]

Logistic regression was undertaken to identify the socio-demographic characteristics of households most likely to report food insecurity. Because food insecurity was assessed at the household level, we limited our examination to socio-demographic variables also measured at the household level: health region (Figure 1), household type, household income adequacy (Table 1), main source of income, and home ownership. The reference category for each variable was set as the category with the largest prevalence in our sample. Unadjusted odds ratios were generated from bivariate logistic regressions and adjusted odds ratios were generated from a multivariate model including all five variables. A bootstrap variance estimation method and household weights, both supplied by Statistics Canada, were used to calculate 95% confidence intervals and coefficients of variation. The analyses included 10,082 respondents with complete data for all variables. Comparison of our results with the reported prevalence of food insecurity for all respondents in Ontario (12) suggests that the exclusion of respondents with missing data did not bias the estimation of this parameter.

RESULTS

The socio-demographic profile of food-insecure households differed considerably from the profile of food-secure households (Table 2). Within the province, the prevalence of household food insecurity ranged from 6.9% in the South West and Central East health regions to 10.1% in the Toronto health region (Table 2), although an examination of the crude odds ratio for health region (Table 3) indicates that these differences are not statistically significant. Nonetheless, the regional variation in prevalence suggests that this problem poses a greater burden for some parts of the province than others. When other socio-demographic variables were taken into account, the only significant difference by region was a lower odds of food insecurity in the South West compared to the Central West region.

The prevalence of food insecurity increased markedly as income adequacy declined, rising to 47% in the lowest category of income adequacy (Table 2). Even after other socio-demographic factors were taken into account, the odds of experiencing household food insecurity rose in a stepwise fashion as the adequacy of household income declined (Table 3). Despite the much greater odds of food insecurity among those in the lowest income adequacy category, most food-insecure households in the province have incomes in the middle and lower-middle adequacy ranges, reflecting the greater number of households with incomes at these levels.

The highest prevalence of food insecurity occurred among households in which the main source of income was social assistance (Table 2), a categorization that includes both Ontario Works and the Ontario Disability Support Program. Sixty-one percent of households reliant on social assistance were food insecure, compared with 6.5% of households reliant on salaries and wages and 5% reliant on pensions or seniors' benefits. Even after controlling for other household socio-demographic characteristics, households reliant on social assistance had almost four times the odds of experiencing food insecurity compared to households that relied on salary or wages (Table 3). In contrast, the odds of experiencing food insecurity for households where the main source of income was pension or seniors' benefits was less than half that of a household that relied on salary or wages as its main source of household income.

Considering the distribution of food-secure and food-insecure households in Ontario by income source provides another window into this problem (Figure 2). Twenty-three percent of food-insecure households were reliant on social assistance, compared to just 1% of food-secure households. However, 55% of food-insecure households were reliant on salaries or wages for their incomes.

Although food insecurity was more prevalent among single-parent households and single-person households compared to couples with or without children (Table 2), there was no difference in the odds of experiencing food insecurity by household type, once the other variables were taken into consideration (Table 3).

Households who owned their dwelling had a much lower rate of food insecurity than those who reported not owning a dwelling (Table 2). After taking other socio-demographic factors into account, not owning a dwelling was associated with almost three times the odds of household food insecurity (Table 3).

DISCUSSION AND CONCLUSIONS

Three potent socio-demographic correlates of household food insecurity in Ontario can be delineated from our analysis: low income adequacy, social assistance as the main source of income, and not owning one's dwelling. These three variables have been repeatedly associated with increased odds of food insecurity in published examinations of national survey data. (2-4) Together, they chart the extreme financial vulnerability that underpins household food insecurity. The crude understanding of household financial resources provided by the income adequacy variable is extended by information on home ownership and income source. The greater vulnerability of tenant households to food insecurity in part reflects the impact of housing costs on the income left for food, (14) but home ownership is also an indicator of household 'wealth' not captured by income alone. Similarly, reliance on social assistance is independently associated with food insecurity because the vulnerability of recipient households extends beyond income. They lack savings, property, or other financial assets because this is typically a requirement for eligibility and benefit levels are too low for households to subsequently acquire assets.

Our finding that lone-parenthood was not associated with increased vulnerability to food insecurity once other socio-demographic factors were controlled for, stands in contrast to analyses of earlier national surveys reporting increased odds of food insecurity and related conditions among single-parent (3) or lone-mother households. (2,4) Given that only 1.5% of households in our sample were headed by male single parents, it is unlikely that the discrepancy between our findings and those of earlier studies reflects our failure to differentiate lone-parenthood by gender. The fact that the odds ratio for household food insecurity was significantly higher among single-parent families in Ontario prior to adjustment for other socio-demographic variables suggests that the greater vulnerability of single-parent families is largely a function of their greater poverty.

The high prevalence of food insecurity among social assistance recipients in Ontario comes as no surprise. The inadequacy of welfare incomes to cover estimates of basic living costs is routinely documented in conjunction with the release of costing estimates for the Nutritious Food Basket in Ontario, (15-18) and our results lend support to the advocacy efforts prompted by these comparisons. Welfare incomes for those not receiving disability support payments fell by more than 30% in Ontario between 1992 and 2005. (19) This drop reflects the 21.6% reduction to welfare incomes in 1995 and the non-indexation of benefits to inflation. The small increases in welfare incomes announced in recent years are insufficient to reverse this trend.

Further indication of the inadequacy of social assistance income levels in Ontario comes from an examination of the prevalence of severe food insecurity--a condition marked by indications of reduced food intake and disrupted eating patterns among adults and/or children in the household. (5) Although sample size limitations precluded a multivariate analysis of socio-demographic characteristics in relation to the severity of household food insecurity in Ontario, the descriptive statistics on this condition compiled by Health Canada indicate that fewer than one third of food-insecure households in Canada and in Ontario were classified as experiencing severe food insecurity. (5,12) However, 53% of food-insecure households reliant on social assistance in Ontario experienced severe food insecurity, compared with just 23% of food-insecure households reliant on employment incomes. (12)

Despite the extraordinary vulnerability of households on social assistance to food insecurity, it is important to recognize that most food-insecure households in Ontario are reliant on salaries or wages. This is also true nationally; employment earnings were insufficient to protect more than 600,000 Canadian households from food insecurity in 2004. (5) These findings lend support to calls for a restructuring of income security programs for working-age adults, highlighting the need to look beyond welfare to the full spectrum of programs that impact adults' incomes. (20,21)

Given the strong associations observed between household food security status and the adequacy of adults' and adolescents' nutrient intakes in this survey, (13) there can be no question that household food insecurity is a serious public health concern. The fact that this problem is so tightly linked to problems of income inadequacy highlights the need for a closer alignment of federal, provincial, and territorial income support programs with basic living costs. It is beyond the scope of our work to provide specific recommendations for policy changes that would ensure household food security for all Ontarians. A recent analysis of policy options for improving food security in British Columbia sheds some light on this question, (22) but inter-provincial differences in such factors as employment rates, wages, and the structure and delivery of income support programs necessitate province-specific analyses of policy options. Recent examinations of social policies and programs in Ontario point to some directions for policy reform, (21,23) but additional evaluation is needed to determine whether the proposed changes will enable households to achieve and maintain food security.

[FIGURE 2 OMITTED]

We recommend that Ontario undertake a systematic examination of the interrelationship between household food security and a broad spectrum of current policies, considering those related to minimum wages, Ontario Works, the Ontario Disability Support Program, and other income support programs, as well as policies that relate to the affordability of housing, child care, public transportation, higher education, dental care, prescription medications, special dietary needs, and other expenditures that can impact the food budgets of low-income households. We also recommend that the impact of proposed policy changes on household food security be routinely assessed as part of the process of evaluating the merit of policy options at the provincial, territorial, and federal levels. Only when food security becomes a deliberate policy objective will problems of household food insecurity diminish.

Received: July 2, 2008

Accepted: January 9, 2009

REFERENCES

(1.) Ledrou I, Gervais J. Food insecurity. Health Rep 2005;16(3):47-50.

(2.) Che J, Chen J. Food insecurity in Canadian households. Health Rep 2001;12(4):11-22.

(3.) Vozoris N, Tarasuk V. Household food insufficiency is associated with poorer health. J Nutr 2003;133:120-26.

(4.) McIntyre L, Connor SK, Warren J. Child hunger in Canada: Results of the 1994 National Longitudinal Survey of Children and Youth. CMAJ 2000;163(8):961-65.

(5.) Health Canada. Canadian Community Health Survey, Cycle 2.2, Nutrition (2004)--Income-Related Household Food Security in Canada. Ottawa, ON: Office of Nutrition Policy and Promotion, Health Products and Food Branch, Health Canada, 2007, Cat. No. H164-42/2007E.

(6.) Kirkpatrick S, Tarasuk V. Food insecurity in Canada. Can J Public Health 2008;99(4):324-27.

(7.) Dachner N, Tarasuk V. Homeless "squeegee kids": Food insecurity and daily survival. Soc Sci Med 2002;54(7):1039-49.

(8.) Lawn J, Harvey D. Nutrition and food security in Kangiqsujuaq, Nunavik. R2341/2004E. Ottawa, ON: Minister of Indian Affairs and Northern Development, 2004.

(9.) Lawn J, Harvey D. Nutrition and food security in Kugaaruk, Nunavut. R2265/2003E. Ottawa: Minister of Indian Affairs and Northern Development, 2003.

(10.) Lawn J, Harvey D. Nutrition and food security in Fort Severn, Ontario. R2350/2004E. Ottawa: Minister of Indian Affairs and Northern Development, 2004.

(11.) Tarasuk V, Dachner N, Li J. Homeless youth in Toronto are nutritionally vulnerable. J Nutr 2005;135:1926-33.

(12.) Health Canada. Canadian Community Health Survey Cycle 2.2, Nutrition (2004)--Income-Related Household Food Security in Canada. Supplementary Data Tables. Ottawa: Office of Nutrition Policy and Promotion, Health Products and Food Branch, Health Canada, 2007.

(13.) Kirkpatrick S, Tarasuk V. Food insecurity is associated with nutrient inadequacies among Canadian adults and adolescents. J Nutr 2008;138:604-12.

(14.) Kirkpatrick S, Tarasuk V. The relationship between income and household food expenditure in Canada. Public Health Nutr 2003;6(6):589-97.

(15.) Association of Local Public Health Agencies. Nutritious Food Basket Survey. Final Report. Toronto, ON: Association of Local Public Health Agencies, 2007.

(16.) Peterborough County-City Health Unit. Nutrition Matters. Limited Incomes: A Recipe for Hunger. Peterborough, ON: Peterborough County-City Health Unit, 2006.

(17.) Sudbury & District Health Unit. Nutritious Food Basket. The Cost of Eating Well Report 2006. Sudbury, ON: Sudbury & District Health Unit, 2006.

(18.) Toronto Public Health. The Cost of the Nutritious Food Basket in Toronto 2005. Toronto, ON: Toronto Public Health, 2005.

(19.) National Council of Welfare. Welfare Incomes 2005. Ottawa: Minister of Public Works and Government Services, 2006, Cat. No. SD25-2/2005E-PDF.

(20.) Battle K, Mendelson M, Torjman S. The modernization mantra: Toward a new architecture for Canada's adult benefits. Can Public Policy 2005;31(4):431-37.

(21.) Task Force on Modernizing Income Security for Working-Age Adults. Time for a fair deal. 2006. Available online at: http://www.torontoalliance.ca/MISWAA_ Report.pdf (Accessed May 8, 2006).

(22.) Kerstetter S, Goldberg M. A review of policy options for increasing food security and income security in British Columbia--A discussion paper. PHSA, 2007.

(23.) Drummond D, Manning G. From Welfare to Work in Ontario: Still the Road Less Travelled. TD Bank Financial Group, 2005. Available online at: http://www.td.com/economics/special/welfare05.jsp (Accessed September 9, 2005).

Authors' Affiliation

Department of Nutritional Sciences, University of Toronto, Toronto, ON

Correspondence and reprint requests: Valerie Tarasuk, Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, 150 College Street, Toronto, ON M5S 3E2, Tel: 416-978-0618, Fax: 416-978-5882, E-mail: valerie.tarasuk@utoronto.ca

Acknowledgements and Disclaimer: The authors are indebted to Elizabeth Rael for her thorough and thoughtful critiques of earlier versions of this work and to the Sudbury & District Health Unit (SDHU) Public Health Research, Education and Development (PHRED) Program for their preparation of the health region maps.

This research was funded by the Ministry of Health Promotion in Ontario. It was conducted with guidance from the Public Health Advisory Group, chaired by the SDHU PHRED Program. The views expressed in the manuscript are those of the authors and do not necessarily reflect those of the Government of Ontario or the Ministry of Health Promotion, of the SDHU, or of the Ontario PHRED Program.

Table 1. Category Definitions for Household Income
Adequacy Variable

Income Adequacy       Income Thresholds by Household Size

Lowest income         <$10,000 if 1 to 4 persons
                      <$15,000 if 5 or more persons

Lower-middle income   $10,000 to $14,999 if 1 or 2 persons
                      $10,000 to $19,999 if 3 or 4 persons
                      $15,000 to $29,999 if 5 or more persons

Middle income         $15,000 to $29,999 if 1 or 2 persons
                      $20,000 to $39,999 if 3 or 4 persons
                      $30,000 to $59,999 if 5 or more persons

Upper-middle income   $30,000 to $59,999 if 1 or 2 persons
                      $40,000 to $79,999 if 3 or 4 persons
                      $60,000 to $79,999 if 5 or more persons

Highest income        [greater than or equal to] $60,000 if
                      1 or 2 persons
                      [greater than or equal to] $80,000 if
                      3 or more persons

Table 2. Selected Socio-demographic Characteristics of
Ontario Households by Food Security Status, 2004

                                  N           % Food          95% CI
                                               Secure

Ontario                       4,553,500         91.7        90.8, 92.5

Health region
  South West                   540,800          93.1        91.7, 94.6
  Central South                430,600          92.1        90.1, 94.1
  Central West                 685,200          92.2        90.4, 94.1
  Central East                 715,900          93.1        91.0, 95.2
  Toronto                      903,000          89.9        87.5, 92.2
  East                         584,000          91.3        89.0, 93.6
  North                        314,700          90.3        88.0, 92.6

Main source of income
  Salary/Wages                3,150,300         93.5        92.7, 94.2
  Social assistance            143,200          38.9        30.5, 47.2
  Worker's compensation or      33,000          74.1        59.7, 88.4
    Employment Insurance
  Pension or seniors'          923,200          94.9        93.3, 96.4
  benefits
  Other                        159,100          88.1        83.6, 92.6

Household income adequacy
  Lowest                       133,500          52.8        43.3, 62.2
  Lower-middle                 278,000          69.6        64.4, 74.7
  Middle                       777,000          85.6        83.3, 87.8
  Upper-middle                1,404,900         94.8        93.7, 96.0
  Highest                     1,513,900         98.2        97.5, 98.8
  Not stated                   446,200          95.7        94.4, 97.1

Household type
  Couple with children        1,464,600         93.9        92.8, 95.0
  <25 years old, with or
  without others
  Single                      1,304,600         88.5        86.4, 90.5
  Single parent, children      368,100          83.7        80.7, 86.6
  <25 years old
  Couple alone                1,111,500         96.2        95.0, 97.3
  Other household types        285,300          88.4        84.3, 92.5

Home ownership
  Own dwelling                3,222,000         96.5        95.9, 97.1
  Do not own dwelling         1,329,900         80.1        77.8, 82.3

                               % Food          95% CI
                               Insecure

Ontario                          8.3          7.5, 9.2

Health region
  South West                     6.9          5.4, 8.3
  Central South                  8.0         6.0, 10.0
  Central West                   7.8          5.9, 9.6
  Central East                   6.9          4.8, 9.0
  Toronto                        10.1        7.8, 12.5
  East                           8.7         6.4, 11.0
  North                          9.7         7.4, 12.0

Main source of income
  Salary/Wages                   6.5          5.8, 7.3
  Social assistance              61.1        52.8, 69.5
  Worker's compensation or
    Employment Insurance          *              *
  Pension or seniors'            5.1          3.6, 6.7
  benefits
  Other                         11.9 E       7.5, 16.4

Household income adequacy
  Lowest                         47.2        37.8, 56.7
  Lower-middle                   30.4        25.3, 35.6
  Middle                         14.4        12.2, 16.7
  Upper-middle                   5.2          4.0, 6.3
  Highest                       1.8 E         1.2, 2.5
  Not stated                     4.3          2.9, 5.6

Household type
  Couple with children           6.1          5.0, 7.2
  <25 years old, with or
  without others
  Single                         11.5        9.5, 13.6
  Single parent, children        16.3        13.4, 19.3
  <25 years old
  Couple alone                   3.8          2.7, 5.0
  Other household types         11.6 E       7.5, 15.7

Home ownership
  Own dwelling                   3.5          3.0, 4.1
  Do not own dwelling            19.9        17.7, 22.2

Notes: N = weighted sample size, rounded to nearest 100;
CI = confidence interval. Estimates accompanied by an E are
considered of marginal quality. The reader is advised to
'use with caution'.

* Data with a cell size less than 25 have been suppressed.

Table 3. Household Food Insecurity in Relation to Selected
Household Socio-demographic Characteristics in
Ontario, 2004 (n = 10,082)

                                  Crude OR for          Adjusted OR for
                                 Reporting Food          Reporting Food
                                   Insecurity              Insecurity
                                    (95% CI)            [double dagger]
                                                            (95% CI)

Health region
  South West                0.87 (0.63, 1.21)      0.67 * (0.45, 0.99)
  Central South             1.02 (0.69, 1.51)      0.93 (0.59, 1.45)
  Central West ([dagger])              1.00                  1.00
  Central East               0.88 (0.59, 1.32)     1.07 (0.67, 1.70)
  Toronto                    1.34 (0.93, 1.93)     0.67 (0.45, 1.02)
  East                       1.13 (0.77, 1.66)     0.96 (0.62, 1.49)
  North                      1.28 (0.89, 1.83)     0.96 (0.60, 1.54)

Main source of
household income
  Salary/Wages [dagger]      1.00                  1.00
  Social assistance         22.43 * (15.46, 32.54) 3.69 * (2.33, 5.84)
  Worker's compensation or
    Employment Insurance     4.99 * (2.18, 11.40)  1.76 (0.63, 4.98)
  Pension or seniors'       0.77 (0.56, 1.08)      0.44 * (0.29, 0.67)
  benefits
  Other                     1.93 * (1.23, 3.02)    0.94 (0.55, 1.60)

Household income adequacy
  Lowest                   16.46 * (10.41, 26.04)  6.90 * (3.98, 11.960
  Lower-middle              8.04 * (5.73, 11.27)   5.00 * (3.35, 7.44)
  Middle                    3.10 * (2.30, 4.18)    2.57 * (1.89, 3.50)
  Upper-middle [dagger]     1.00                   1.00
  Highest                   0.34 * (0.22, 0.54)    0.40 * (0.25, 0.63)
  Not stated                0.82 (0.55, 1.24)      0.81 (0.49, 1.34)

Household type
  Couple with children
  <25 years old, with or   1.00                    1.00
    without others [dagger]
  Single                   2.01 * (1.53, 2.65)     0.98 (0.68, 1.40)
  Single parent, children
    <25 years old          3.01 * (2.24, 4.04)     0.99 (0.68, 1.44)
  Couple alone             0.61 * (0.44, 0.86)     0.81 (0.54, 1.19)
  Other household types    2.03 * (1.28, 3.20)     1.62 (0.96, 2.71)
Home ownership
  Own dwelling [dagger]    1.00                    1.00
  Do not own dwelling      6.79 * (5.48, 8.42)     2.94 * (2.28, 3.79)

Notes: n = number of respondents included in analysis,
OR = odds ratio, CI = confidence interval.

* Significantly different from estimate for reference
category (p<0.05).

([dagger]) Reference category.

([double dagger]) Adjusted for other variables listed in table.