Page 49 - IJPS-7-1

Freiria et al.

ability to understand the procedures of the study. Exclusion criteria were: (1) Use of any dietary supplements based on
vitamins and/or minerals, (2) being followed up on by a home care program, and (3) undergoing chemotherapy.
Data were collected in the primary healthcare units where older adults were enrolled. Participants with incomplete
data on social support and food insecurity were excluded from the analysis (n = 13), thus a total of 598 older adults were
included in the study. Signed consent forms were obtained from all participants.
2.2. Measures

2.2.1. Food insecurity
Food insecurity was measured using the abridged version of the Brazilian Food Insecurity Scale, a questionnaire validated
for the Brazilian population (Santos, Lindemann, Motta, et al., 2014; Sperandio, Morais, and Priore, 2018). The scale is
composed of five questions (yes/no): “Were you worried that the food in your home would be finished before you could
buy, receive or produce more food?”, “Did the food finish before you had money to buy more?”, “Did you run out of money
and could not have a healthy and varied diet?”, “Did you or an adult in your home reduce the amount of food in your
meals, or skip meals, because there was not enough money to buy food?”, “Did you eat less than you thought you should
because there was not enough money to buy food?”, and identified individuals who experienced food insecurity during
the previous 3 months. A sixth question that was used in the previous studies on older adults was: “Have you ever been
hungry but did not eat because you were not able to leave the home for food?” was added to assess the possible mobility
difficulties that could influence food access by older adults (Carder, Luhr, and Kohon, 2016; Chung, Gallo, Giunta, et al.,
2012). Participants who answered “yes” to at least one question were evaluated as food insecure.

2.2.2. Social support
Social support was assessed using the Brazilian adapted and validated version (Griep, Chor, Faerstein, et al., 2005) of
the Medical Outcomes Study Social Support Survey (Sherbourne and Stewart, 1991). The scale contains 19 questions
and evaluates five dimensions of social support: material (four questions), affective (three questions), emotional (four
questions), positive social interaction (four questions), and information (four questions). Each question on the scale was
answered by indicating the frequency of the support received by the older adult, and each choice had a different score:
never (1 point), rare (2 points), sometimes (3 points), almost all the time (4 points), and always (5 points). On each
dimension evaluated, the participant obtained a score between 15 points (affective dimension) and 20 points (material,
emotional, positive social interaction, and information dimensions). The sum of the points obtained in each dimension
was multiplied by 100 and divided by the highest possible score in each dimension, with a higher score indicating a
greater perception of social support perceived by older adults.

2.2.3. Sociodemographic covariates
The literature suggests that social, economic, and demographic factors are associated with food insecurity. For example,
different studies have shown that higher income is an important factor influencing access to adequate food, along with
education. In addition, Causasian older adults have food insecurity rates that are substantially lower than that of other ethnicities
(Marín-León, Segal-Corrêa, Panigassi, et al., 2005; Fernandes, Rodrigues, Nunes, et al., 2018; Grammatikopoulou, Gkiouras,
Theodoridis, et al., 2019; Ziliak and Gundersen, 2022). The following variables were included as covariates that could act as
potential confounders in the relationship with social support because they were previously associated with food insecurity:
Sex (male or female); family monthly income, classified into two groups: ≤latimes of the minimum wage, >2 times of the
minimum wage. The minimum wage was R$954.00 (US$261.10) in 2018 and R$998.00 (US$257.20) in 2019 (we used mean
imputation for individuals with missing data (n = 52) in this variable); schooling (0 – 8 years, 9 years or more, not informed);
marital status (married, widowed, and single/divorced); ethnicity (Caucasians, African Americans – African Americans and
Mixed Race, and Others – South native American and Asian); whether they were the head of the household (yes and no);
working for pay (yes, no, and not informed); and age group (60 – 69-years-old, 70 – 79-years-old, and 80 years and older).
2.3. Analytical strategies

All analyses were conducted using Stata version 12.0. In the descriptive analysis, frequencies and percentages were used
®
for categorical variables, and means with standard deviations were estimated for the continuous variables. To examine the
differences in the prevalence of food insecurity, we used Chi-square test for categorical variables and Mann–Whitney U

test for social support dimensions as continuous variables (because they did not adhere to a normal distribution according
to the Shapiro-Wilk test).

International Journal of Population Studies | 2021, Volume 7, Issue 1 43