![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
3 Division of Nutritional Sciences, Cornell University, Ithaca, NY 14853 and 4 Department of Health Promotion, Education, and Behavior, University of South Carolina, Columbia, SC 29208
* To whom correspondence should be addressed. E-mail: kh289{at}cornell.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
20 y in the 1999–2002 NHANES. Married men were more likely to be overweight than men in all other marital-status categories except for those living with partners. Divorced men were more likely than never-married men to live in a household with very low food security. Compared with fully food-secure men, marginally food-secure men were heavier, whereas men with low food security were lighter. Compared with fully food-secure women, marginally food-secure women also showed a tendency to be overweight (P = 0.05), whereas women with low food security were more likely to be obese. When considering the interaction between food insecurity and marital status, food insecurity was related to a greater likelihood of obesity among married women, those living with partners, and widows, when compared with never-married women. Sensitivity analyses suggest this effect was concentrated among marginally food-secure women. These findings indicate that food insecurity relates differently to body weight for men and women, and that partnering plays a strong part in this relationship for women.
|
Introduction |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
One aspect of associations between food insecurity and body weight that has not been well examined is marital status. Growing evidence reveals an association between marital status and body weight (14,15). Cross-sectional investigations suggest that married people are heavier, particularly married men (16,17), whereas longitudinal analyses suggest that entering marriage is associated with weight gain and leaving marriage is related to weight loss (15,18,19). The extent to which marital status influences weight, or weight influences marital status, or both, is unclear (20). Spouses eat together and influence each others' diet (21,22), suggesting that marriage may influence body weight through food intake.
Little is known about the association between marital status and food insecurity. One study reported a substantial variation in food insecurity across marital-status categories for women, from 2% of widows to 23% of separated women (23). Being married profoundly enhances family income and wealth (24–26). Long-term cohabiting partners typically share resources like spouses do, with the presence of a partner reducing economic hardship (27). Conversely, separation and divorce have negative economic consequences (26) that are disproportionately borne by women (28). Marriage and long-term partnerships also provide social support and other noneconomic resources that help individuals withstand periods of economic uncertainty or stress (29,30), but the benefits of marital support may be smaller for women than men (31,32). Dissolution of marriage or a long-term partnership may therefore exacerbate food insecurity and adversely affect the ability to cope with it. Given these associations, we expect the prevalence of food insecurity to be greatest among divorced and separated individuals, particularly women.
Women are typically household food managers (21,33,34), a role that uniquely ties them to food-insecurity dilemmas. Past research suggests that women sometimes deprive themselves to feed others, especially children, and establish food intake patterns that alternate between restriction in times of scarcity and binging when food is plentiful, with a resultant gain in weight (35). Given these role dynamics, we expect the association between food insecurity and body weight to be larger among women than men, and particularly women without spouses or partners to provide economic and social support.
|
Methods |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
Sample. NHANES 1999–2002 contained records for 10,291 adults aged 20 y or older. We excluded respondents if: 1) marital status was unknown (n = 553); 2) measured BMI and self-reported height and weight were both missing (n = 70); 3) women were pregnant or pregnancy status could not be ascertained (n = 746); and 4) household food security status was missing (n = 412). The final analytical sample included 4338 men and 4172 women. A comparison of available data on age, ethnicity, education, and income, for included and excluded men and women, revealed only one significant difference: excluded men were slightly younger than included men (P < 0.05).
Measures.
Body weight is represented here as relative body weight, or BMI. NHANES included measured heights and weights for most (7699) respondents. We substituted self-reported heights and weights for 811 respondents (449 men and 362 women) who had no measured values. BMI was calculated as kg/m2. For 53 respondents (17 men and 36 women) with extreme values, BMI was constrained to 50 kg/m2. Respondents were classified as overweight if their BMI was 25 kg/m2, and obese if their BMI was 30 kg/m2 (36).
Respondents self-reported their current marital status in 6 categories: never married, living with a partner, married, separated, divorced, or widowed. Age was reported in years, with individuals 85 y or older coded as 85 y. Age-squared was included in multivariate models due to the nonlinear association between age and body weight. NCHS provided 5 race or ethnicity categories: Mexican American, Other Hispanic, White non-Hispanic, Black non-Hispanic and Other Race. Education was reported in 3 categories: less than high school, high school graduate or general equivalency diploma (GED), and more than high school. Family income was analyzed as a percentage of the federal poverty level for family size (% FPL). Family size was not publicly available from NCHS and was not used in these analyses.
The % FPL was missing for 1075 respondents (11%) and education was missing for 26 respondents; these were replaced with values obtained through an expectation-maximization (EM) imputation procedure (37). Imputed variables were simultaneously obtained using age, an indicator variable for whether a respondent's income was <$20,000 or $20,000, and home ownership. Data were imputed separately for men and women. These variables explained 
60% of the variation in family income for respondents with no missing data.
Food security was measured at the household level using an 18-question Food Security Survey Module (FSSM) that was condensed to a 4-category food-security measure considered valid for identifying food-insecure households (38–41). Food security was recorded as: 1) fully food secure, 2) marginally food secure (answered 1 or 2 questions about food insecurity affirmatively, suggesting the household is at-risk for food insecurity), 3) low food security (reported a concern for adequate household food supply, a need to adjust food management strategies, or indicated a decreased diet quality), or 4) very low food security (food intake reduced to the extent that 1 household members repeatedly experienced physical sensations of hunger). Households with higher incomes ($50,000 or 400% FPL) were not administered this survey module and were coded by NCHS as fully food secure. For analyses involving interaction effects, small cell sizes required combining respondents with marginal, low, and very low food security into 1 food-insecure category.
Analyses. All analyses examined men and women separately. Unadjusted differences in food security and body weight between marital categories were tested using Chi-square analysis and bivariate general linear regression models. Multiple logistic regression was used to assess demographic-adjusted differences in household food security across marital categories. Demographic adjustment included age, age-squared, race or ethnicity, education, and % FPL. Because neither being overweight nor obesity was a rare condition, multiple general linear regression models were used to predict BMI, being overweight, and obesity (42). Parameter estimates from multiple regressions can be interpreted in BMI units (kg/m2) or percentage point differences in predicted prevalence of overweight and obesity. Interactions between marital status and our indicator of any food insecurity were also tested. Differences with P < 0.05 were considered significant.
Data were weighted for unequal sample selection probabilities using 4-y sample weights computed by NCHS (43) standardized to this analytic sample. Variances were corrected for complex sampling using masked strata and pseudo primary sampling units provided by NHANES (43) and the SPSS complex samples module (version 14).
Three types of sensitivity analyses were performed on all multivariate models. First, considering the association between food insecurity and food-stamp receipt, we controlled for household receipt of food stamps in the past 12 mo in all multivariate models of food insecurity, with no change to the general pattern of findings. Analysis of the role of food stamps themselves on body weight is beyond the scope of this paper. Second, data about family size or composition were not publicly available in NHANES, but we did construct a variable reflecting childbearing. Women were considered biological mothers if they reported a live birth, and not biological mothers if they were not yet menstruating, had never been pregnant, or never had a live birth. Biological motherhood was missing for 717 female respondents (18%) and was not imputed due to a lack of good predictors. For women with data, all analyses were repeated including biological motherhood as an independent variable, with no change to the results. Third, repeating the interaction analyses, while considering marginally food-secure households as food secure, tested the sensitivity of results from the analysis of "any household food insecurity." This sensitivity analysis produced changes to one set of findings, which are noted in our results.
|
Results |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
|
Controlling for demographic characteristics, divorced men remained significantly more likely than never-married men to report very low food security (OR = 2.8, P < 0.05). Married men remained more likely to be overweight than never-married men (+9.2 percentage points, P < 0.01), and significantly more likely to be overweight than every other marital category except men living with partners (data not shown). Demographically adjusted differences did not occur in food security or body weight across marital categories for women (data not shown).
Compared with fully food-secure men (Table 2), marginally food-secure men had a higher mean BMI (+1.2 kg/m2, P < 0.05), and a tendency to be overweight (+6.4 percentage points, P = 0.06) and obese (+6.5 percentage points, P = 0.08). In contrast, compared with fully food-secure men, low food security was associated with lower BMI (–0.9 kg/m2, P < 0.05) and a lesser likelihood of being overweight (–8.3 percentage points, P < 0.05). Compared with fully food-secure women, marginally food-secure women tended to have higher BMI (+0.8 kg/m2, P = 0.10) and to be overweight (+6.5 percentage points, P = 0.05), whereas women with low food security were significantly more likely to be obese (+10.8 percentage points, P < 0.01).
|
|
|
Discussion |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
Consistent with prior research showing a link between mild food insecurity and being overweight among women (6–10), we found marginally food-secure women tended to be overweight and women with low food security were more likely to be obese. Adjusting for marital status did not explain away observed differences in women's body weight by food security status.
There are several possible explanations for these observed associations. First, the major direction of causation may be that body weight influences reported food insecurity. Overweight and obese women may more often report food insecurity due to their eating habits and the importance of food in their lives. One study that directly addressed this hypothesis found that the association between household food stores and food insecurity did not differ between obese and nonobese women, suggesting that they perceived their food insecurity similarly (11).
Second, these associations may reflect an underlying dose-response relationship. Marginally food-secure women were more likely to be overweight, and women with low food security were more likely to be obese. Although initially appearing counterintuitive (we might expect food insecurity to lead to food restriction and lower body weights), increasingly severe food insecurity may cause declining diet quality and stress about food management rather than reduced energy intake (44) until a very low level of food security is reached.
Men in our sample showed a different pattern of associations between food insecurity and body weight. Like women, marginally food-secure men had heavier body weights than fully food-secure men, whereas men with low food security had lower body weights and a lower likelihood of overweight. Among men, marginal food security may be associated with deteriorating diet quality or irregular eating and therefore higher body weight, whereas low food insecurity may be associated with intake reduction and lower body weight.
The associations observed between food insecurity and body weight for men and women suggest that each gender uniquely experiences food insecurity. As the household food situation intensifies from marginal to low food security, men have lower weights and women higher weights. Although coping strategies vary widely, men more often isolate themselves from others and take individual action (30), which may lead to reducing food intake in response to deteriorating food security. Women more often engage their social network in coping with stress and gear their efforts toward the needs of others (30). As their food security gets worse, women may restrict their food intake in times of scarcity and binge when food is plentiful, with a resultant gain in weight (35). This suggests that an economic or food safety net may help men and women maintain orderly eating, with differential effects on their body weights.
Associations between food insecurity and body weight were not uniform across marital categories. Among married women, food insecurity was associated with a greater likelihood of obesity. This was the largest marital category and therefore statistical significance was most easily attained. Women living with a partner and widows (much smaller categories) had similar associations between food insecurity and obesity. Although most prior studies did not report on marital status, many limited their samples to women aged 20–39 y with children living at home (6,7), women aged 19–55 y who did not live alone (10), or women with children aged 3 to 6 y (11), thereby over-representing married women relative to other marital categories. In light of the marital-specific effects observed here, the strength of associations between food insecurity and overweight found by previous authors may have been amplified by their sample-frames.
When our definition of "any food insecurity" was changed to exclude the marginally food-secure, significant results were not maintained. This suggests that the interaction between food insecurity and partnerships, in relation to body weight, may be concentrated among the marginally food secure. Small cell sizes precluded examining the interaction between marital status and food security reflected in >2 categories.
We hypothesized stronger associations between food insecurity and body weight among women without partners, but found the opposite. Women may deprive themselves to feed adult partners, like they do for children (35). Women also may not reap the same benefits of social support in marriage as do men (31,32). Some men's coping strategies may not integrate well with partners, which place women at increased risk for depression and feelings of loneliness and powerlessness (30). Depression is associated with obesity among women (45), although the direction of that association is unclear (46,47). Targeting nutrition assistance and education toward food-insecure women and their partners could address an important need.
Several limitations of this study deserve note. It used cross-sectional data, so the temporal order of food insecurity, marriage, and/or weight change cannot be discerned. Long-term measures of poverty, food insecurity, or weight gain or loss were not available. Duration measures might have strengthened our ability to extrapolate from these cross-sectional findings. Food insecurity questions were not asked of respondents with relatively higher household incomes and therefore the prevalence of the food-insecure may be underestimated. We had no good measure of family size or the presence of children in the household. Our measure of biological motherhood reflects only childbearing and we had no data about children for men. Some analyses report subgroup findings or interaction effects based upon small samples. For example, only 24 women were living with a partner, food insecure, and obese, limiting the strength of the conclusions drawn from interaction effects.
A prior review of the literature showed that body weight was known to be inversely related to socioeconomic status, with the most consistent relation among women (48). The prevalence of obesity has recently grown in the U.S., however, and most markedly among higher socioeconomic groups (49), thereby narrowing the gap in body weight between rich and poor. Our analyses use 1999–2002 data that show no differences in obesity across socioeconomic measures (49), and we find associations between food insecurity and body weight that are consistent with prior research. We think this reaffirms that the observed links between food insecurity and body weight reflect behavioral responses to food insecurity rather than uncontrolled differences in socioeconomic status. Programs and policies to address the growing prevalence of overweight and obesity would benefit from conjointly considering the problem of food insecurity. These data suggest that marginally food-secure women and men are heavier than their fully food-secure counterparts. This risk escalates for women who have low food security, and was concentrated among partnered women. More research is needed to better understand the temporal order and the processes involved in associations between gender, marital and nonmarital partnerships, food insecurity, and body weight in order to effectively address both obesity and food insecurity.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Author disclosures: K. L. Hanson, J. Sobal, and E. A. Frongillo, no conflicts of interest.
Manuscript received 20 December 2006. Initial review completed 27 January 2007. Revision accepted 10 April 2007.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTIntroductionMethodsResultsDiscussionLITERATURE CITED |
|---|
1. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA. 2006;295:1549–55.
2. American Dietetic Association. Position of the American Dietetic Association: Domestic food and nutrition security. J Am Diet Assoc. 2002;102:1840–7.[Medline]
3. Office of Disease Prevention and Health Promotion, U.S. Department of Health and Human Services. Healthy people 2010. [cited 11 July 2006.] Available from: www.healthypeople.gov.
4. Nord M, Andrews M, Carlson S. Household food insecurity in the United States, 2004. Food Assistance & Nutrition Research Program, Economic Research Report no. 11. U.S. Department of Agriculture, Economic Research Service; 2005.
5. Bickel G, Nord M, Price C, Hamilton W, Cook J. Guide to measuring household food security, revised 2000. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service, Office of Analysis, Nutrition and Evaluation; 2000.
6. Frongillo EA, Olson CM, Rauschenbach BS, Kendall A. Nutritional consequences of food insecurity in a rural New York State County. Institute for Research on Poverty Discussion, Paper no.1120–97. Madison, WI: University of Wisconsin; 1997.
7. Olson CM. Nutrition and hunger outcomes associated with food insecurity and hunger. J Nutr. 1999;129:521S–4S.[Medline]
8. Townsend MS, Peerson J, Love B, Achterberg C, Murphy SP. Food insecurity is positively related to overweight in women. J Nutr. 2001;131:1738–45.
9. Adams EJ, Grummer-Strawn L, Chavez G. Food insecurity is associated with increased risk of obesity in California women. J Nutr. 2003;133:1070–4.
10. Basiotis PP, Link M. Food insufficiency and prevalence of overweight among adult women. Family Economics and Nutrition Review. 2003;15:55–7.
11. Kaiser LL, Townsend MS, Malgar-Quinonez HR, Fujii ML, Crawford PB. Choice of instrument influences relations between food insecurity and obesity in Latino women. Am J Clin Nutr. 2004;80:1372–8.
12. VanEenwyk J, Sabel J. Self-reported concern about food security associated with obesity—Washington, 1995–1999. MMWR Morb Mortal Wkly Rep. 2003;52:840–2.[Medline]
13. Food Research and Action Center. Proceedings of the roundtable on understanding the paradox of hunger and obesity, November 22, 2004. Washington, DC: Food Research and Action Center; 2004.
14. Sobal J. Marriage, obesity and dieting. Marriage Fam Rev. 1984;7(1/2):115–39.
15. Sobal J, Rauschenbach B, Frongillo EA. Marital status changes and body weight changes: a U.S. longitudinal analysis. Soc Sci Med. 2003;56:1543–55.[Medline]
16. Sobal J, Rauschenbach BS. Gender, marital status, and body weight in older U.S. adults. Gender Issues. 2003;21:75–94.
17. Sobal J, Rauschenbach B, Frongillo EA. Marital status, fatness, and obesity. Soc Sci Med. 1992;35:915–23.[Medline]
18. Kahn HS, Williamson DF. The contributions of income, education, and changing marital status to weight change among U.S. men. Int J Obes. 1990;14:1057–68.[Medline]
19. Kahn HS, Williamson DF, Stevens JA. Race and weight changes in U.S. Women: the roles of socioeconomic status and marital status. Am J Public Health. 1991;81:319–23.
20. Jeffery RW, Rick AM. Cross-sectional and longitudinal associations between body mass index and marriage-related factors. Obes Res. 2002;10:809–15.[Medline]
21. Bove CF, Sobal J, Rauschenbach BS. Food choices among newly married couples: convergence, conflict, individualism, and projects. Appetite. 2003;40:25–41.[Medline]
22. Sobal J, Nelson MK. Commensal eating patterns: a community study. Appetite. 2003;41:181–90.[Medline]
23. Jones SJ, Frongillo EA. The modifying effects of food stamp program participation on the relationship between food insecurity and weight change in women. J Nutr. 2006;136:1091–4.
24. Waite LJ. Does marriage matter? Demography. 1995;32:483–507.[Medline]
25. Light A. Gender differences in the marriage and cohabitation income premium. Demography. 2004;41:263–84.[Medline]
26. Zagorsky JL. Marriage and divorce's impact on wealth. J Sociol (Melb). 2005;41:406–24.
27. Bauman K. Shifting family definitions: the effect of cohabitation and other nonfamily household relationships on measures of poverty. Institute for Research on Poverty Discussion Paper no. 1123–97. University of Wisconsin. Madison, WI; 1997.
28. Holden KC, Smock PJ. The economic costs of marital dissolution: why do women bear a disproportionate cost? Annu Rev Sociol. 1991;17:51–78.[Medline]
29. Cohen S, Syme L. Social support and health. Orlando, FL: Academic Press; 1985.
30. Hobfall SE, Cameron RP, Chapman HA, Gallagher RW. Social support and social coping in couples in handbook of social support and the family. In: Pierce GR, Sarason BR, Sarason IG, editors. New York: Plenum Publishing; 1996.
31. Cutrona CE. Social support in couples: marriage as a resource in times of stress. Thousand Oaks, CA: Sage Publications; 1996.
32. Neff LA, Karney BR. Gender differences in social support: a question of skill or responsiveness? J Pers Soc Psychol. 2005;88:79–90.[Medline]
33. DeVault ML. Feeding the family: the social organization of caring work. Chicago, IL: University of Chicago Press; 1991.
34. Kirk MC, Gillespie AH. Factors affecting food choice of working mothers with young families. J Nutr Educ. 1990;22:161–8.
35. Olson CM. Food insecurity in women: a recipe for unhealthy trade-offs. Topics in Clinical Nutrition. 2005;20:321–8.
36. World Health Organization (WHO). Obesity: preventing and managing the global epidemic. Geneva: WHO;1998.
37. Allison PD. Missing data. Thousand Oaks, CA; Sage Publications; 2002.
38. Kendall A, Olson CM, Frongillo EA. Validation of the Radimer/Cornell measures of hunger and food insecurity. J Nutr. 1995;125:2793–801.
39. Kendall A, Olson CM, Frongillo EA. Relationship of hunger and food security to food availability and consumption. J Am Diet Assoc. 1996;96:1019–24.[Medline]
40. Frongillo EA, Rauschenbach BS, Olson CM, Kendall A, Colmenares AG. Questionnaire-based measures are valid for the identification of rural households with hunger and food insecurity. J Nutr. 1997;127:699–705.
41. Frongillo EA. Validation of measures of food insecurity and hunger. J Nutr. 1999;129:506S–9S.[Medline]
42. Cox DR, Snell EJ. Analysis of binary data. 2nd ed. New York: Chapman Hall; 1989.
43. NCHS. NHANES analytic guidelines, June 2004 version. [cited 1 May 2006.] Available from: http://www.cdc.gov/nchs/data/nhanes/nhanes_general_guidelines_june_04.pdf.
44. Frongillo EA. Understanding obesity and program participation in the context of poverty and food insecurity. J Nutr. 2003;133:2117–8.
45. Heo M, Pietrobelli A, Fontaine KR, Sirey JA, Faith MS. Depressive mood and obesity in US adults: comparison and moderation by sex, age, and race. Int J Obes (Lond). 2006;30:513–9.[Medline]
46. Stunkard AJ, Faith MS, Allison KC. Depression and obesity. Biol Psychiatry. 2003;54:330–7.[Medline]
47. Roberts RE, Deleger S, Strawbridge WJ, Kaplan GA. Prospective association between obesity and depression: evidence from the Alameda county study. Int J Obes Relat Metab Disord. 2003;27:514–21.[Medline]
48. Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of the literature. Psychol Bull. 1989;105:260–75.[Medline]
49. Zhang Q, Wang Y. Trends in the association between obesity and socioeconomic status in U.S. adults: 1971 to 2000. Obes Res. 2004;12:1622–32.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Isanaka, M. Mora-Plazas, S. Lopez-Arana, A. Baylin, and E. Villamor Food Insecurity Is Highly Prevalent and Predicts Underweight but Not Overweight in Adults and School Children from Bogota, Colombia J. Nutr., December 1, 2007; 137(12): 2747 - 2755. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
