Sage Journals: Discover world-class research

Abstract

Aim

The aim of this study was to examine the prevalence of household food insecurity in individuals reporting migraine within a large population-based sample of Canadians.

Methods

The Canadian Community Health Survey (CCHS) uses a stratified cluster sample design to obtain information on Canadians ≥12 years of age. Data on household food insecurity were assessed for individuals who reported having migraine or not, providing a current point prevalence. This was assessed for stability in two CCHS datasets from four and eight years earlier. Factors associated with food insecurity among those reporting migraine were examined and a logistic regression model of food insecurity was developed. We also examined whether food insecurity was associated with other reported chronic health conditions.

Results

Of 48,645 eligible survey respondents, 4614 reported having migraine (weighted point prevalence 10.2%). Food insecurity was reported by 14.8% who reported migraine compared with 6.8% of those not reporting migraine, giving an odds ratio of 2.4 (95% confidence interval 2.0–2.8%). This risk estimate was stable over the previous eight years. The higher risk for food insecurity was not unique to migraine and was seen with some, but not all, chronic health conditions reported in the CCHS.

Conclusions

Food insecurity is more frequent among individuals reporting migraine in Canada.

Keywords

Migraine food insecurity population-based survey Canada

Introduction

Food insecurity and obesity are the most significant food-related public health issues. There has been less awareness of food insecurity, which is defined as limited or intermittent access to nutritionally adequate, safe and acceptable foods, accessed in socially acceptable ways (1). Approximately 30% of Canadians in low-income households are food insecure (2) and almost 13% of Canadian households experienced some level of food insecurity in 2012 (3). Food-insecure households are more likely to include children with acute infections, chronic illness, and both developmental and mental health problems (4,5) and adults who experience stress, depression, and anxiety (6). People with threatened food security are more likely to omit or reduce the size of meals, to have less variety in their diets, and to eat less fruit, vegetables, dairy products, and vitamins (7). Food insecurity is associated with numerous health issues, including hyperlipidemia, hypertension, and obesity (8).

Migraine affects approximately one in six individuals (9) and its prevalence is inversely related to household income (10). We hypothesized an association between migraine and food insecurity within a Canadian population-based health survey.

The objectives of this study were to:

examine whether individuals reporting migraine are more likely to report household food insecurity;

determine whether this association is stable;

explore which factors are associated with reported household food insecurity within the population of individuals reporting migraine;

determine whether reported migraine continues to be associated with food insecurity within a parameterized model of food insecurity for the entire population (migraine and freedom from reported migraine); and

whether this effect is unique to reported migraine or is seen within other health conditions.

Methods

The Canadian Community Health Survey (CCHS) is a cross-sectional survey that collects information related to health status, health care utilization, and health determinants for the Canadian population. A stratified cluster sample design is used to obtain information on Canadians ≥12 years of age living in private dwellings in the ten provinces and three territories, exclusive of people living on First Nations Reserves or Crown lands, residents of institutions, full-time members of the Canadian Armed Forces, and residents of certain remote regions. Information was collected every two years until 2007 and annually since then (11). The response rate for the most recent iteration of the survey (2013) was 76.6% at the household level and 66.7% at the person level (12).

Our analysis was primarily based on the 2013 CCHS (data collection January–December 2013), but also made use of the 2009 and 2005 datasets to examine the prevalence of food security over time. Within these surveys there are common question sequences addressed to all respondents and sections that are addressed to specific subpopulations, generally to residents of specific provinces or health districts. All respondents were given the question: “I’d like to ask about certain chronic health conditions which you may have. We are interested in ‘long-term conditions’ which are expected to last or have already lasted 6 months or more and that have been diagnosed by a health professional. Do you have migraine headaches?” (13). Household food insecurity was subsequently assessed using the Canadian Food Security Status Module (14), which was adopted from the US model of food security status published by the US Department of Agriculture in 2000 (15). A summative assessment of household food security was derived from 18 questions indicating whether household members were able to afford the food they needed in the previous 12 months. The questions ranged in severity from worrying about running out of food, through the unavailability of nutritious food, to children within the household not eating for the whole day. Ten of the 18 questions were specific to the experiences of adults within the household or the household in general, while the remaining eight were specific to the experiences of children in the household (16). The module yielded a score based on which households could be classified as food secure, moderately food insecure and having severe food insecurity. For this study, we collapsed the two food insecurity classifications to give a binary scale: food secure or food insecure.

Eligible respondents were ≥12 years of age and resided in the Canadian provinces and territories, with the exception of British Columbia, Manitoba, Newfoundland and Labrador, and Yukon Territory, which did not participate in the food security question in 2013 (78.5% of all survey respondents). Our analysis was limited to individuals with non-missing responses for reported migraine (99.9% of all survey respondents) and household food security (96.4% of all survey respondents).

The primary analysis was to describe the prevalence of food security within the population of people reporting migraine compared with the population not reporting migraine. We assessed whether the prevalence of food insecurity remained stable by examining prevalence in four-year intervals culminating with the current dataset.

We examined variables associated with food security within the population of people reporting migraine (Table 1). These variables were derived from a previous analysis by Statistics Canada on food security in Canadian households (2) and a US study using NHANES data (17), reviewed within the 2009 dataset initially to determine the nature of their relationship with food insecurity before recoding any variables (Table 1). We also chose to include a single question on self-perceived health and reported mental health conditions of mood disorders and anxiety, as both may be associated with both food insecurity and migraine.

Table 1.

Factors assessed for association with food insecurity.

Factor	Coding
Sex	Male; female
Age (years)	12–49; 50–59; 60–69; ≥70
Household education	Not a secondary school graduate; higher level of education than secondary school graduate
Marital status	Married or cohabiting; single; widowed; separated; or divorced
Race	White; visible minority
Aboriginal	Yes; no
Immigrant	Yes; no
Total household income (CDN$)	<20,000; 20,000–29,000; 30,000–39,000; 40,000–59,000; ≥60,000
Family size	Single; multiple
Own dwelling	Yes; no
Self-perceived health	Excellent; very good; good; fair; poor
Reported mood disorder	Yes; no
Reported anxiety disorder	Yes; no

Variables in the univariate analysis that appeared to be related to the description of food insecurity by individuals reporting migraine were used to develop a multivariable logistic regression model of food insecurity within this population. The multivariable model was developed interactively using backward elimination, with testing of all eliminated variables simultaneously. In alignment with the requirements of Statistics Canada, variables with coefficients of variation >16% were also removed (18). Goodness of fit statistics and the area under the curve for the receiver-operating curve were used to assess the adequacy of the model.

To determine whether food insecurity was unique to people reporting migraine or was also associated with other chronic medical conditions, we identified a number of reported medical conditions within the CCHS that were persistent, highly prevalent within the population, and affected the entire age spectrum. Using these criteria, we elected to examine asthma, arthritis, bowel disorders, and diabetes. We developed a multivariable logistic regression model of food security within the complete population using identical methodology, then assessed the impact of adding reported migraine and then the other reported conditions (asthma, arthritis, bowel disorders, and diabetes).

The data were analyzed using Stata/SE 12 (19). Weights were recoded to reflect the sample size (an average weight of 1). Measures of associations are reported as odds ratios (OR) with 95% confidence intervals. All statistics given are for non-missing data. Estimates were obtained using sampling weights provided by Statistics Canada to account for design effects and non-response bias; standard errors were estimated using a bootstrapping procedure.

Results

Within the 2013 CCHS, a total of 48,645 respondents answered affirmatively or not to whether they had migraine diagnosed by a health professional and had evaluable responses as to whether they resided in a food secure household or not. Household food security was available across all respondent ages. The rates of missing data by the variables examined (Table 1) were low (range 0–3.2%).

Having migraine was reported by 4614 individuals, accounting for a weighted point prevalence of 10.2% (95% confidence interval (CI) 9.8–10.7%), representing 2,399,300 young people and adults reporting migraine nationwide. This point prevalence appears relatively stable over the past eight years (10.7% in 2005 and 10.5% in 2009). A total of 7.6% of the sample was assessed as experiencing household food insecurity (95% CI 7.1–8.1%). Household food insecurity was considerably more likely to be reported among those reporting migraine compared with respondents not reporting migraine (14.8 versus 6.8%, respectively; OR = 2.4, 95% CI 2.0–2.8). The rates of reporting food insecurity within the Canadian population and specifically food insecurity within those reporting migraine does not appear to have substantively changed over the past eight years (Figure 1).

Figure 1.

Reported food insecurity (%) for the Canadian population and those reporting migraine over the past eight years in the Canadian Community Health Survey (CCHS) 2005 (Canada, excluding Saskatchewan, Manitoba, New Brunswick, Newfoundland and Yukon Territory), CCHS 2009 (Canada, excluding New Brunswick and Prince Edward Island), and CCHS 2013 (Canada, excluding British Columbia, Manitoba, Newfoundland and Labrador, and Yukon Territory). Vertical bars reflect 95% confidence intervals for the population estimates.

We examined variables previously shown to be associated with food insecurity within the population of individuals reporting migraine. Several factors seemed to be associated with an increased likelihood of reporting food insecurity within a multivariable model (age, sex, income, family size, home ownership, self-perceived health, and reported mood disorder) (Table 2). The probability associated with the goodness of fit statistic was 0.29. As an assessment of the discriminative ability of the model, the area under the receiver-operating curve was 0.83 (95% CI 0.81–0.85) (20).

Table 2.

Multivariable logistic regression model of factors associated with household food insecurity among individuals reporting migraine (n = 4,614) in the Canadian Community Health Survey 2013 (excluding respondents from British Columbia, Manitoba, Newfoundland and Labrador, and Yukon Territory).

Factor	Odds ratio (95% confidence interval)	Proportion	p
Age (years)
12–49	1.9 (1.2–3.1)	0.67	0.008
≥50	1.0 (reference)	0.33
Sex
Female	1.6 (1.1–2.3)	0.70	0.009
Male	1.0 (reference)	0.30
Total household income (CDN$) ^a
<20,000	8.3 (5.4–12.7)	0.11	<0.001
20,000–29,000	4.6 (2.9–7.4)	0.10	<0.001
≥30,000	1.0 (reference)	0.80
Family size
Multiple	1.8 (1.2–2.7)	0.88	0.005
Single	1.0 (reference)	0.12
Own dwelling
No	2.9 (1.9–4.3)	0.31	<0.001
Yes	1.0 (reference)	0.69
Self-perceived health
Fair or poor	2.5 (1.6–3.7)	0.18	<0.001
Excellent, very good, or good	1.0 (reference)	0.82
Reported mood disorder
Yes	3.1 (2.1–4.5)	0.16	<0.001
No	1.0 (reference)	0.84

Notes: ^a Total household income, cumulative <0.001. Area under the curve for the receiver-operating curve was 0.83 (95% CI 0.81–0.85). Model developed on 4590/4614 eligible cases (99.5%).

We examined the variables that had previously been associated with household food insecurity in a univariate analysis within the entire sample. A multivariable logistic regression model was developed retaining age, marital status, aboriginal status, income, family size, home ownership, self-perceived health, reported mood disorder, and reported anxiety disorder. We then independently entered the reported health condition (migraine, asthma, arthritis, bowel disorders, or diabetes). The model goodness of fit statistic was χ²= 12.9, 9df, p = 0.16. When reported migraine was entered into this model, its association with household food insecurity persisted (OR 1.5; 95% CI 1.2–1.8%). The results for the other reported medical conditions are presented in Table 3.

Table 3.

Association of reported health conditions with food insecurity in a multivariable logistic regression model (n = 48,645) in the Canadian Community Health Survey 2013 (excluding respondents from British Columbia, Manitoba, Newfoundland and Labrador, and Yukon Territory).

Reported health condition	Adjusted odds ratio (95% confidence interval)	p
Migraine	1.5 (1.2–1.8)	<0.001
Asthma	1.3 (1.1–1.6)	<0.012
Arthritis	1.6 (1.3–1.9)	<0.001
Bowel disorders	1.0 (0.7–1.3)	0.90
Diabetes	1.2 (0.9–1.5)	0.21

Notes: Adjusted odds ratios after adjustment for age, marital status, aboriginal status, total household income, family size, own dwelling, self-perceived health, reported mood disorder, and reported anxiety disorder.

Discussion

Food insecurity is reported by approximately 15% of US (21) and 13% of Canadian households (3). The factors that underlie food insecurity include poverty and difficulty in accessing appropriate grocery stores. Travel to stores with healthy affordable food can be challenging for those who live in rural or urban food deserts, where shopping is often dependent on being able to access a vehicle. Even when the challenge of accessing stores is overcome, budgeting for appropriate food must compete with escalating housing and medical costs. A lack of education, unawareness of appropriate food choices, or mental illness may also be related to food insecurity.

Using a representative sample of Canadians, we demonstrated an association between migraine and food insecurity. Among individuals with migraine, the risk was greater in those who were younger, female, lived with others, earned less than CDN$30,000 per annum, and did not own their own dwelling. Self-reporting a mood disorder and self-perceiving health as only fair or poor were also associated with food insecurity. At a population level, even after correcting for the known associations, migraine was still associated with food insecurity. This association is unlikely to be merely a reflection of having a chronic illness, as neither diabetes nor bowel-related disorders was related to food insecurity, but migraine, asthma, and arthritis were related to food insecurity. This may reflect the greater impact of these conditions on mobility and the ability to access healthy food.

Migraine and food insecurity share a number of associations. Both migraine (22) and food insecurity (23) are associated with anxiety and depression. Psychological stress and dietary habits, including missed meals and hunger, are among the most frequent precipitants of migraine (24,25). Migraine may contribute to a higher incidence of food insecurity by resulting in days of inability or reduced ability to perform normal household activities (26,27), including shopping and cooking. For many individuals with migraine, the occurrence of headaches may also preclude attendance at work (28), which results in costs (29,30) including lost income.

Stewart et al. (31) found a higher incidence of migraine as household income decreased, which they proposed supported a social causation hypothesis, as opposed to a social selection hypothesis. In the former, lower household income is associated with stress and other factors that increase the incidence and duration of an illness. Social selection proposes that a disease or illness can result in a decline in social status because the afflicted individual may underperform in their educational or occupational endeavors. We have demonstrated an association between food insecurity and migraine that persists after correction for income, in addition to other factors known to be associated with food insecurity. The directionality of this relationship appears to be as challenging as the relationship between income and migraine formulated by both the social causation and selection hypotheses. The cross-sectional design of our study does not allow a conclusion on the temporal order of food insecurity and migraine.

The CCHS is a nationally representative, population-based survey and is more likely to be truly representative of those living with migraine than clinic-based studies. Nationally representative health surveys are primarily designed to guide health policy. Their secondary analyses, as in this study, have the potential to pose and answer important clinically relevant questions.

The diagnosis of migraine in our patients was based on the reporting of a diagnosis by a health practitioner rather than on specific criteria. To our knowledge, the question about migraine used in the CCHS has not been validated. However, a similar question, “Have you ever had migraine?”, has been validated against a diagnostic telephone interview administered by physicians trained in headache diagnosis, with a sensitivity of 75.8% (95% CI 72.4–79.0%) and specificity of 86.4% (95% CI 84.3–88.3%) (32) and, when assessed within a clinical interview, with an associated κ of 0.77 (33). In an internet-based study of patients reporting migraine, after reviewing the patients’ narrative description of headache qualities, a neurologist assigned a high or moderate probability of migraine to 97% of patients, and also to 97% of patients using International Headache Society criteria (34). These studies suggest that the question used in the CCHS may be valid. In addition, our reported migraine prevalence was comparable with other similar Canadian studies that found a prevalence of 7.9% (35) and 8.3% (36).

The assessment of experience-based food insecurity has been measured by various questionnaires, most notably the 18-item Household Food Security Survey Module (14) and its precursor the Core Food Security Measurement (37). The questions asked within the CCHS are adopted from this US model of food security (38). The favorable psychometric properties of the Household Food Security Survey Module (14) have been reviewed (37,39).

The experience of food insecurity appears to be more frequent among individuals reporting migraine and some, but not all, chronic diseases. Clinicians who treat migraine should be aware that their patients are more likely to experience food insecurity.

Qualifier

The analysis presented in this paper was conducted at the Atlantic Research Data Centre, which is part of the Canadian Research Data Centre Network. The services and activities provided by the Atlantic Research Data Centre are made possible by the financial or in-kind support of the Social Sciences and Humanities Research Council, the Canadian Institutes of Health Research, the Canada Foundation for Innovation, Statistics Canada, and Dalhousie University. The views expressed in this paper do not necessarily represent the Canadian Research Data Centre Network or those of its partners.

Public health relevance

This is the first description of increased household food insecurity by individuals reporting migraine derived from a large national health dataset of Canadians.

Food insecurity was reported by 14.8% of individuals reporting migraine versus 6.8% of those not reporting migraine (OR 2.4).

In a multivariable model of food insecurity at the population level, after correcting for a number of known associations with food insecurity, migraine continued to have a modest association with food insecurity (OR 1.5).

Footnotes

Acknowledgments

The authors acknowledge the assistance of the staff at the Atlantic Research Data Centre.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by the Social Sciences and Humanities Research Council of Canada RDC Access Granting Committee (grant number 14-SSHDAL-3854).

References

Economic and Social Development Department, Food and Agricultural Organization of the United Nations. Trade reforms and food security: conceptualizing the linkages, Rome 2003, ftp://ftp.fao.org/docrep/fao/005/y4671e/y4671e00.pdf (2003, accessed 3 October 2011).

Che

Chen

. Food insecurity in Canadian households. Health Rep 2001; 12: 11–22.

Tarasuk V, Mitchell A and Dachner N. Household food insecurity in Canada, 2012, http://nutritionalsciences.lamp.utoronto.ca/wp-content/uploads/2014/05/Household_Food_Insecurity_in_Canada-2012_ENG.pdf (2014, accessed 16 March 2015).

Kursmark

Weitzman

. Recent findings concerning childhood food insecurity. Curr Opin Clin Nutr Metab Care 2009; 12: 310–316.

Gundersen

Kreider

. Bounding the effects of food insecurity on children’s health outcomes. J Health Econ 2009; 28: 971–983.

Whitaker

Phillips

Orzol

. Food insecurity and the risks of depression and anxiety in mothers and behavior problems in their preschool-aged children. Pediatrics 2006; 118: e859–e868.

Seligman

Laraia

Kushel

. Food insecurity is associated with chronic disease among low-income NHANES participants. J Nutr 2010; 140: 304–310.

Vozoris

Tarasuk

. Household food insufficiency is associated with poorer health. J Nutr 2003; 133: 120–126.

Smitherman

Burch

Sheikh

. The prevalence, impact, and treatment of migraine and severe headaches in the United States: a review of statistics from national surveillance studies. Headache 2013; 53: 427–436.

10.

Stewart

Roy

Lipton

. Migraine prevalence, socioeconomic status, and social causation. Neurology 2013; 81: 948–955.

11.

Statistics Canada. Canadian Community Health Survey 2002, http://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&SDDS=3226 (2002, accessed 2 December 2014).

12.

Statistics Canada, Health Statistics Division. Canadian Community Health Survey 2013 User Guide, Ottawa: Statistics Canada, 2013.

13.

Statistics Canada. Canadian Community Health Survey Annual Component – 2013 Questionnaire, www23.statcan.gc.ca/imdb-bmdi/instrument/3226_Q1_V10-eng.pdf (2013, accessed 2 December 2014).

14.

Health Canada. Household food security survey module, www.hc-sc.gc.ca/fn-an/surveill/nutrition/commun/insecurit/hfssm-mesam-eng.php (2012, accessed 10 November 2015).

15.

Bickel G, Nord M, Price C, et al. Guide to Measuring Household Food Security, Revised 2000, www.fns.usda.gov/sites/default/files/FSGuide_0.pdf (2000, accessed 2 December 2014).

16.

Association of Public Health Epidemiologists of Ontario. 2A Food Insecurity, www.apheo.ca/index.php?pid=80 (2011, accessed 3 October 2011).

17.

Seligman

Laraia

Kushel

. Food insecurity is associated with chronic disease among low-income NHANES participants. J Nutr 2010; 140: 304–310.

18.

Statistics Canada. Table 5-1. Quality level guidelines. www.statcan.gc.ca/pub/13f0026m/2007001/table/tab5p1-eng.htm (2009, accessed 4 July 2015).

19.

StataCorp. STATA/SE 12. Stata Statistical Software: Release 12. College Station, TX: StataCorp, 2011.

20.

Homer

Lemeshow

Sturdivant

. Applied logistic regression, 3rd edn. Hoboken, NJ: Wiley, 2013.

21.

Nord

Andrews

Carlson

. Household food security in the United States, 2008. Economic Research Service Publication ERR-83, Alexandria, VA: US Department of Agriculture, Economic Research Service, 2009.

22.

Lantéri-Minet

Radat

Chautard

. Anxiety and depression associated with migraine: influence on migraine subjects' disability and quality of life, and acute migraine management. Pain 2005; 118: 319–326.

23.

Whitaker

Phillips

Orzol

. Food insecurity and the risks of depression and anxiety in mothers and behavior problems in their preschool-aged children. Pediatrics 2006; 118: e859–e868.

24.

Kelman

. The triggers or precipitants of the acute migraine attack. Cephalalgia 2007; 27: 394–402.

25.

Saracco

Calabrese

Cavallini

. Relationship between primary headache and nutrition: a questionnaire about dietary habits of patients with headache. Neurol Sci 2014; 35(Suppl 1): 159–161.

26.

Lipton

Bigal

Kolodner

. The family impact of migraine: population-based studies in the USA and UK. Cephalalgia 2003; 23: 429–440.

27.

D'Amico

Grazzi

Usai

. Disability in chronic daily headache: state of the art and future directions. Neurol Sci 2011; 32(Suppl 1): S71–S76.

28.

Burton

Landy

Downs

. The impact of migraine and the effect of migraine treatment on workplace productivity in the United States and suggestions for future research. Mayo Clin Proc 2009; 84: 436–445.

29.

Markson

Lipton

. Burden of migraine in the United States: disability and economic costs. Arch Intern Med 1999; 159: 813–818.

30.

Linde

Gustavsson

Stovner

. The cost of headache disorders in Europe: the Eurolight project. Eur J Neurol 2012; 19: 703–711.

31.

Stewart

Roy

Lipton

. Migraine prevalence, socioeconomic status, and social causation. Neurology 2013; 81: 948–955.

32.

Gervil

Ulrich

Olesen

. Screening for migraine in the general population: validation of a simple questionnaire. Cephalalgia 1998; 18: 342–348.

33.

Russell

Rasmussen

Thorvaldsen

. Prevalence and sex-ratio of the subtypes of migraine. Int J Epidemiol 1995; 24: 612–618.

34.

Lenert

Looman

Agoncillo

. Potential validity of conducting research on headache in internet populations. Headache 2002; 42: 200–203.

35.

Martin

. Prevalence of migraine headaches in Canada. CMAJ 2001; 164: 1481–1481a.

36.

Ramage-Morin

Gilmour

. Prevalence of migraine in the Canadian household population. Health Rep 2014; 25: 10–16.

37.

Marques

Reichenheim

de Moraes

. Household food insecurity: a systematic review of the measuring instruments used in epidemiological studies. Public Health Nutr 2015; 18: 877–892.

38.

Bickel G, Nord M, Price C, et al. Guide to measuring household food security, revised 2000, www.fns.usda.gov/fsec/files/fsguide.pdf (2000, accessed 10 March 2011).

39.

Cafiero

Melgar-Quiñonez

Ballard

. Validity and reliability of food security measures. Ann N Y Acad Sci 2014; 1331: 230–248.

Food insecurity and migraine in Canada

Abstract

Aim

Methods

Results

Conclusions

Keywords

Introduction

Methods

Results

Discussion

Qualifier

Public health relevance

Footnotes

Acknowledgments

Declaration of conflicting interests

Funding

References