Food Security Stability and Change Among Low-Income Urban Women

Many welfare-reliant and working poor families experience difficulties making ends meet and thus must develop strategies to cope with competing demands for limited resources. Needs, resources, and the strategies low-income families use to negotiate competing priorities change as family circumstances and contextual conditions evolve in response to changing policy environments. Such changes often have effects, for better or worse, on outcomes that are of concern to parents and policy makers alike. In our research, we draw upon two complementary data sources to examine stability and change in food security in the context of the post-1996 welfare reforms. Food security is a fundamental indicator of family well-being and material hardship that is likely to be affected by welfare reforms and low-income employment because of the association of food security with economic resources, Food Stamps Program participation, and the availability of time to utilize scarce resources to their maximum potential. Three recent studies provide indications of the relationships between welfare, low-income employment, food insecurity stability and change, and physical and mental health. However, differences in sample characteristics, measures, and analytic approaches, make it difficult to develop clear expectations and hypotheses beyond the expectation that food security is related to economic resources and may be related to physical and mental health statuses. Nevertheless, the available evidence does suggest that there is much to be learned by pursuing additional analyses to further our understanding of individual-level state-specific stability and change in food security among low-income women. Two complementary data sources are used in this study: longitudinal survey data collected from random samples of initially welfare-reliant women living in highly disadvantaged neighborhoods in four large urban counties (Cleveland, Los Angeles, Miami-Dade, and Philadelphia), and longitudinal qualitative interview data from initially welfare-reliant women living in neighborhoods of concentrated poverty in Cleveland, Ohio. These data were collected under the auspices of MDRC’s Project on Devolution and Urban Change. In the survey sample (N=3210), 25.3 percent of those who were food secure in 1998 were classified as food insecure in 2001, while 43.7 percent of those classified as food insecure in 1998 were food secure in 2001. Multivariate logistic regression analyses indicated that the number of children under age 18 years in the household, income, and physical and mental health statuses were the most consistent predictors of transitions into and out of food security over time. Among women who were food secure in 1998, the odds of transitioning to food insecurity were increased among those who had more minor children in the household in 1998, those who had an increased number of minor children in the household over time, those with lower income in 1998, those with decreasing income over time, those with higher CES-D depression scale scores in 1998, those with increasing depression over time, those with worse self-reported health, and those with worsening self-reported health status over time. Among women who were food insecure in 1998, the odds of transitioning to food security in 2001 were lower among those who had a larger number of minor children in their households over time, those with decreasing income over time, those with higher CES-D scores in 1998, those with increasing depression scale scores over time, those with worse self-reported health status in 1998, and those with worsening health status over time. In the qualitative sample (N=36), we found a strong correlation between depression and food (in)security: 72.7 percent of those who scored low on the depression scale were food secure in 2000 and 2001 compared to only 21.4 percent of those who scored high on the depression scale. Stated otherwise, nearly 80 percent of the women who scored high on depression in 1999 experienced food insecurity in either 2000 or 2001, or at both later points in time. Inductive analyses of the qualitative interview provide some clues that help us understand the link between depression and physical health problems and food (in)security. These data indicate that these low-income women generally used multiple strategies to prevent or reduce food hardships for their families. The qualitative data also indicate that food secure women tended to have better employment and income outcomes than the food insecure women, and they also tended to be less socially isolated. Given the strong correlation between depression and food (in)security in this sample, and other evidence that physical and mental health problems are barriers to employment, one implication of these results is that the women in the qualitative sample who were more depressed may have been less able to obtain and maintain good jobs and less able to maintain the social connections that would allow them to rely on family and friends for food resources. Women who were depressed or had health problems may also have been less able to engage in the various and multiple strategies necessary to achieve food security with extremely constrained resources. Taken together, these mixed methods results suggest that depression and/or health problems may interfere with the time- and labor-intensive strategies many women must employ to maintain adequate food for their families. These results also suggest that access to economic and social resources contribute to food (in)security among these very low income women. To the extent that depression and other health problems, which are often thought of as barriers to employment, compromise women’s ability to pursue public or community-based food resources or maintain ties with family that would allow them to access assistance, depression or other health problems might also be conceptualized as barriers to food security.