Wendy S. Wolfe

Understanding Food Insecurity in the Elderly: A Conceptual Framework

Abstract In order to determine the potential effects of changes in welfare, health, and nutrition programs on food insecurity in the elderly, it is important to have a comprehensive understanding of the factors influencing food insecurity in this population group. Using a naturalistic paradigm and methods, a conceptual framework of factors that influence food insecurity status was developed based on in-depth interviews with 41 low-income rural white and urban black elderly individuals in 35 households in upstate New York.The conceptual framework depicts how several sets of factors, particularly health, food program participation, and life experiences, relate to food insecurity in the elderly. The use of several food assistance programs by low-income elderly, their views on such programs, and a unique “grocery bag” program for homebound elderly who can cook for themselves also are discussed.

Life-Course Events and Experiences: Association with Fruit and Vegetable Consumption in 3 Ethnic Groups

OBJECTIVE To examine how life-course experiences and events are associated with current fruit and vegetable consumption in 3 ethnic groups. DESIGN A theoretic model developed from previous qualitative research guided the development of a telephone survey. Data were collected on fruit and vegetable consumption, sociodemographic characteristics, ethnic identity, and life-course events and experiences, including food upbringing, social roles, food skills, dietary changes for health, and practice of food traditions. SUBJECTS/SETTING Low- to moderate-income adults living in a northeastern US city were selected randomly from 3 ethnic groups: black (n = 201), Hispanic (n = 191), and white (n = 200). STATISTICAL ANALYSES Bivariate and multiple linear regression analysis of associations between life-course variables and fruit and vegetable consumption. RESULTS Black, Hispanic, and white respondents differed significantly in life-course experiences, family roles, socio-demographic characteristics, and place of birth. Explanatory models for fruit and vegetable consumption differed among ethnic groups and between fruits and vegetables. Among black respondents, a college education was positively associated with fruit consumption; education and family roles contributed most to differences in fruit (R2 = .16) and vegetable (R2 = .09) consumption. Among Hispanic respondents, life-course experiences such as liking fruits and vegetables in youth, making dietary changes for health, and food skills were positively associated with fruit (R2 = .25) and vegetable (R2 = .35) consumption. Among white respondents, socio-demographic characteristics, such as being married with a young child or single with no child and having a garden as an adult, were positively associated with fruit (R2 = .20) and vegetable (R2 = .22) consumption. APPLICATIONS/CONCLUSIONS An understanding of the determinants of food choice in different subcultural groups can be used to design effective nutrition interventions to increase fruit and vegetable consumption. Experiences such as eating fresh-picked fruits and vegetables while growing up or vegetable gardening as an adult may enhance fruit and vegetable consumption among members of some ethnic groups.

Parity-associated weight gain and its modification by sociodemographic and behavioral factors : a prospective analysis in US women

OBJECTIVE: To examine how the relationship between parity increase and weight gain is modified by sociodemographic and behavioral factors. DESIGN: Prospective longitudinal data from the first National Health and Nutrition Examination Survey (NHANES I, 1971–75) and its follow-up of those aged 25 y and older, the NHANES I Epidemiologic Follow-up Survey (NHEFS, 1982–84). SUBJECTS: The analytical sample was nationally representative of the United States and included 2952 white or African-American non-pregnant women aged 25–45 y at baseline, who were re-measured approximately 10 y later. MEASUREMENTS: Statistical interactions in multiple linear and logistic regression models were examined to identify how eight sociodemographic and three behavioral factors modified the effect of parity increase on body weight change and risk of substantial weight gain. RESULTS: Factors that increased parity-associated weight gain included being African-American, living in a rural area, not working outside the home, having fewer children, lower income and lower education, and being unmarried. Among white women, being younger and having higher body weight at baseline increased parity-associated weight gain, while among African-American women, being older and having lower body weight increased parity-associated weight gain. African-American smokers gained less weight with an increase in parity, while the interactions between smoking and physical activity with parity-associated weight gain in whites were complex. CONCLUSION: The effects of sociodemographic and behavioral factors on parity-associated weight gain varied by race and parity change, with the most consistent findings being that unmarried and unemployed white women had greater parity-associated weight gain, while both white and African-American women who smoked, had higher education, or higher parity had lower parity-associated weight gain. This information may contribute to better targeting and more effective interventions to prevent postpartum weight retention.

Building Household Food-Security Measurement Tools from the Ground Up

For impact evaluation and for planning and targeting decisions, local organizations in developing countries need tools for assessing household food security that go beyond measuring food availability to include access to food and perceptions of food insecurity. This paper explores the potential for developing direct measures of household food security that include such components and that are based on an in-depth understanding of the experience of food insecurity at the household level. This process was used successfully to develop the US Food Security Measure. The US approach and examples of efforts in developing countries are reviewed, along with relevant conceptual and measurement issues. The potential portability and challenges to use of the US approach in developing countries are discussed. The elements needed to apply this approach are outlined, along with operations research needed for developing such experientially based measures.

Food pattern, diet quality, and related characteristics of schoolchildren in New York State

OBJECTIVES To examine the food patterns and diet quality of elementary schoolchildren in New York State (outside of New York City) and to determine sociodemographic characteristics correlated with diet quality. DESIGN A nonquantitative 24-hour recall administered to students and a brief questionnaire completed by parents. SAMPLE 1,797 second and fifth graders (51% of those asked) in 51 randomly selected schools (46% of those asked) in New York State outside of New York City. STATISTICAL ANALYSES Multiple regression analyses, chi 2, and t tests. RESULTS On the day they were surveyed, 40% of students did not eat vegetables, except for potatoes or tomato sauce; 20% did not eat fruit; 36% ate at least four different types of snack foods, and 16% of fifth graders did not eat breakfast. Children who ate a school lunch ate significantly more dairy foods and fruits and vegetables, and fewer snack-foods than those who brought lunch from home. Fifth graders ate significantly more snack foods and were more likely to skip breakfast than second graders; boys had lower food-group pattern scores than girls; children of lower socioeconomic status had less diverse diets but ate less snack foods than children of higher socioeconomic status; children with single parents were more likely to skip breakfast and to eat fewer vegetables than those with two parents; and children with mothers employed outside the home had less diverse diets than those with mothers at home. APPLICATIONS/CONCLUSIONS The findings indicate that nutrition interventions are clearly needed for this age group and that targeted messages should be based on sociodemographic characteristics.

Hunger and food insecurity in the elderly: its nature and measurement.

To better understand the nature of food insecurity in the elderly and to improve its measurement, in-depth interviews were conducted with 41 urban Black and rural White elderly in 35 households, followed by telephone administration of commonly used measures of food insecurity in 24 of these elderly. Elderly food insecurity appears to follow a progression of severity, beginning with compromised diet quality, followed by food anxiety, socially unacceptable meals, use of emergency food strategies, and finally actual hunger. The five quantitative measures tested were compared to each elderly persons food insecurity status based on the in-depth interview. All measures had reasonable specificity, and good sensitivity for those experiencing severe food insecurity. However, the Cornell-Radimer, Community Childhood Hunger Identification Project (CCHIP) and Nutrition Screening Initiative (NSI) measures appeared more sensitive than the USDA food sufficiency or Urban Institute measures in correctly identifying those in the lesser stages of food insecurity.

Overweight school children in New York City: prevalence estimates and characteristics

OBJECTIVES: To examine the prevalence of overweight and associated characteristics in a representative sample of children from schools in New York City (NYC).DESIGN: A cross-sectional survey with an in-person interview (including measurements of height and weight) and a household questionnaire.SUBJECTS: Second and fifth grade students from schools in NYC participated. A total of 1 396, including 692 second and 704 fifth grade, children were included in the analysis.MEASUREMENTS: Prevalence of overweight based on the body mass index (BMI) and the 85th or 95th percentile cutoff points from Cycles II and III of the National Health Examination Survey reference recently developed by the Centers for Disease Control and Prevention (CDC). The characteristics of overweight examined were: age, gender, socio-economic status, Hispanic origin, family structure, number of children in the household, meal preparation by children, food diversity and participation in the National School Lunch Program.RESULTS: Overweight based on the 85th percentile of BMI was 37.5% among second grade and 31.7% among fifth grade children. Based on the 95th percentile, overweight was 19.9% and 17.8%, respectively. Overweight was associated with Hispanic origin, fewer children in the household, family structure and meal preparation by children.CONCLUSIONS: Overweight is prevalent among elementary school children in NYC. Studies in diverse geographically defined areas with diverse populations, are necessary to design effective prevention efforts. Primary prevention efforts in NYC should target Hispanics and take into consideration individual behaviour and family characteristics.

Evaluating brief measures of fruit and vegetable consumption frequency and variety : Cognition, interpretation, and other measurement issues

To evaluate whether items from 3 brief measures of fruit and vegetable consumption were understood and interpreted as intended, cognitive testing was conducted in a purposive sample of 31 white, African-American and Hispanic persons. The measurement instruments tested were the fruit and vegetable module from the Behavioral Risk Factor Surveillance System (to measure frequency), and 1 fruit and 1 vegetable variety measurement instrument developed by the investigators. The cognitive testing interviews were analyzed qualitatively to identify interpretation difficulties and other measurement issues. The testing identified a number of measurement issues, including issues related to time frame, wording, interpretation, grouping of items, and serving size. Recommendations based on the findings were incorporated into revised versions of each instrument, which were further tested in a small sample. As revised and presented in this article, these instruments for assessing fruit and vegetable frequency and variety appear to be understood and interpreted as intended across different racial and ethnic groups, and may be useful in situations requiring brief dietary assessment, although further testing is needed.

Food Consumption Patterns of Elementary Schoolchildren in New York City

OBJECTIVE To examine food consumption patterns and related characteristics of schoolchildren. DESIGN A representative survey involving a household questionnaire and child interview to conduct a nonquantitative 24-hour dietary recall. Indexes were developed to score the diets based on the Food Guide Pyramid and 5 A Day for Better Health recommendations. SUBJECTS/SETTING Six hundred ninety-three 2nd-grade students and 704 5th-grade students from public and private schools in New York City. STATISTICAL ANALYSES PERFORMED Weighted means and gender-adjusted analysis of covariance models were used to account for a complex survey design in calculating variance estimates. RESULTS Compared with a maximum of 50, the mean Food Guide Pyramid index score was 29.2 for 2nd-grade students and 30.4 for 5th-grade students. The mean 5 A Day consumption score was 3.4 and 3.6 servings of fruits and vegetables daily for 2nd and 5th graders, respectively. Measures of food consumption were significantly related to socioeconomic status, race/ethnicity, skipping meals, and participating in school lunch for both grades, and with meal preparation for 5th graders. APPLICATIONS/CONCLUSIONS The Food Guide Pyramid and 5 A Day recommendations provide guidelines for food selection beginning in childhood. Knowledge and skills training are needed to improve food consumption patterns as children consume foods away from home and as they take on greater responsibility for meal preparation and food selection.

Impact of Participation in Home-Delivered Meals on Nutrient Intake, Dietary Patterns, and Food Insecurity of Older Persons in New York State

The aim of this study was to determine if (1) participation in Home-Delivered Meals (HDM) results in improved dietary patterns and nutrient intake, lower food insecurity, and reduced loss of weight; (2) subgroups of older persons are more likely to benefit; and (3) nutritional indicators of impact other than nutrient intake may be useful. The design used was quasi-experimental, with longitudinal assessment of individuals on HDM at baseline (before receipt of services), 6, and 12 months, and comparison to non-randomized group receiving other services. Outcomes included measured weight and height, 24-hour dietary recall, and food insecurity. Paired t test, multiple linear regression, and selection models using multiple logistic regression were performed. All older persons in three New York State counties referred for aging services over a 5-month period were asked to participate (n = 456), and 212 agreed (171 on HDM). At 6 months, the sample size was 101 (34 discharged, 42 hospital/died/moved, 26 chose not to continue), and at 12 months it was 68 (similar reasons). After receiving meals for 6 and 12 months, participants showed greater improvement in most dietary intake variables than either a non-HDM comparison group or HDM participants who ate no HDM meal on the day of assessment. Compared to initial values, participants improved significantly in some variables for dietary patterns, nutrient intake, and nutrient density, and were less likely to be food insecure. Furthermore, HDM was more likely to impact those living alone and those with poorer initial status. This study provides strong evidence that HDM has a positive impact on the nutritional well-being of older persons. Food insecurity and dietary patterns are useful nutritional indicators of impact.