Data Source and Study Population
This study was based on data from the Korea National Health and Nutrition Survey (KNHANES), a cross-sectional, nationally representative survey conducted by the Korea Disease Control and Prevention Agency (http://knhanes.kdca.go.kr). This study was approved by the Institutional Review Board of the Korea Centers for Disease Control and Prevention (2008-04EXP-01-C, 2009-01CON-03-2C, 2010-02CON-21-C, 2011-02CON-06-C, 2012-01EXP-01-2C, 2013-07CON-03-4C, 2013-12EXP-03-5C, 2018-01-03-P-A). KNHANES is a legal survey conducted based on the Koreans National Health Promotion Act and is designed to assess the health-related behaviors, chronic diseases prevalence status, and food nutritional state of Korean people. As KNHANES corresponds to research projects conducted by nation for public welfare according to Article 2-1 of the Bioethics and Safety Act and Article 2-2-1 of Enforcement Regulations of the same Act, it was conducted in 2015 – 2017 with IRB review exemption. The survey target was 11,250 adults aged 19 – 34 y who participated in the health survey and the dietary intake survey of KNHANES IV-2 to KNHANES VII-3. The exclusion criteria were as follows: persons who had no food security data (n=30), persons who had no health-related quality of life (EQ5D) data (n=359), pregnant women (n=337), persons with a history of cancer (n=43), persons who had no information on household income (n=76), and persons whose daily energy intake was less than 250 kcal or 5000 kcal or higher (n=181). As a result, a total of 10,224 people were included in the analysis.
General Characteristics
Age, sex, body mass index (BMI), marital status, women’s birth experience, residential area, monthly household income, household composition type, occupation, education level, smoking status, alcohol consumption, and subjective health state of participants were analyzed. For monthly household income, monthly income was analyzed using gross household income variable of KNHANES and annual income was analyzed after converting it into monthly income by dividing by 12.
Household Food Security and Household Income
Household food security was investigated by a question on household dietary life status. The survey participants selected one from four response options to the question “Which of the following best describes your household dietary life status for the past one year? The participants who responded “Our family could have many different types of foods as much as we wanted” were classified as the food security group and the participants who responded “Our family could have sufficient amount of food but could not eat various kinds” or “Our family went short of food from time to time due to financial difficulty” or “Our family often went short of food due to financial difficulty” were classified as the food insecurity group. As for household income, using household income quartiles, participants in low and low-middle categories were classified as the low income group and participants in middle-high and high categories were classified as the high income group. The sensitivity and specificity of a single food insufficiency questionnaire with the food security status by 18-item questionnaire were reported as 56.8% and 92.3%, respectively [34].
Euro Quality of Life five Dimensions (EQ5D)
EQ5D was developed by Euro Quality of Life Group and is an index to assess health-related quality of life with five dimensions [35]. It consists of mobility, self-care, usual activity, pain/discomfort, and anxiety/depression and three response options were used (not at all, some problems, many problems). EQ5D was investigated with the approval of EuroQol Group (www.euroqol.org), and validity and reliability were assessed in the Korean population-based survey [35]. In this study, the participants who responded their daily life was hindered or they felt uncomfortable with daily life in mobility, self-care, usual activity, and pain/discomfort areas or the participants who responded they had an anxiety or depression in anxiety/depression area were classified as the any problem group and the subjects who responded they had no discomfort at all in those areas as the no problem group.
Food intake assessment
The food intakes of the participants were estimated by the 24-h dietary recall method. In the study, the food items were categorized into 18 food groups including cereal and cereal products, potatoes and starch products, sugar and sugar products, beans and bean products, nuts and seeds products, vegetables, mushrooms, fruits, meat and meat products, eggs and egg products, fish and shellfish, seaweeds, milk and dairy products, oil and fat, beverages, seasoning, processed foods, others.
Furthermore, we estimated most commonly consumed foods of participants. Among the foods taken by participants, in case the same food ingredients were used even though cooking and processing methods were different, the case was classified as one kind of food. Twenty major most commonly consumed foods were selected by food lists with high intake amount.
Nutrient intake assessment
Daily intake of carbohydrates, protein, and fat in participants and energy composition were calculated. Intake of vitamins and minerals (calcium, phosphorus, iron, sodium, potassium, thiamin, riboflavin, niacin, vitamin C, and vitamin A) was assessed. Calcium and vitamin A are especially nutrients that are prone to lack among Koreans [36, 37]. Therefore, we evaluated nutrient intake status including these nutrients. In addition, to assess the quality of dietary intake, we investigated the intake status compared to 2020 Dietary Reference Intakes for Koreans [38]. The percentages of participants who took less than Estimated Average Requirement (EAR) by nutrients were calculated.
Statistical Analysis
SAS 9.4 software (SAS Institute, Cary, NC, USA) was used for statistical process of all data. As for KNHANES, SURVEY procedure was used with stratified, multistage sampling design, and significance was set to α < 0.05 for the test. In this study, participants were classified into four study groups based on household food security and in-come: Food security & higher-income, food insecurity & higher-income, food security & low-income, and food insecurity & low-income groups. And general characteristics, eating habits, food and nutrient intake, and most commonly consumed foods were compared between four groups and a relationship with health-related quality of life was analyzed. General characteristics and eating habit related matters of young adults according to the food security and household income groups were compared and intake of foods and nutrients was estimated.
For categorical variables, the ratio (weighted %) considering frequency and weight was calculated by conducting chi-square test through SURVEY FREQ procedure and for continuous variables, weighted mean and standard error were calculated using SURVEY MEANS procedure and significance by groups was tested by conducting analysis of co-variance (ANCOVA) using SURVEY REG procedure. Post hoc analysis was performed using Tukey’s test and age and sex were adjusted. SURVEY LOGISTIC analysis was performed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for health-related quality of life in young adults according to food security and household income. We conducted analyses adjusted for age, sex, BMI, marital status, residential area, occupation, education level, smoking status, and alcohol consumption.