Introduction

The World Health Organization (WHO) has emphasized that a healthy diet assists in preventing many chronic non-communicable diseases such as heart disease, diabetes, and cancer1. The WHO recommends that people consume free sugars less than 10% of their total energy intake to prevent diet-related chronic diseases, and those with a total fat energy of at least 20% are consistent with good health1. The COVID-19 pandemic (hereinafter “the pandemic”) has significantly altered individuals’ lifestyles worldwide2,3,4, as it has restricted their ability to dine out at restaurants and encouraged them to prepare their meals in their households2,4,5,6.

Concerns about maintaining a healthy diet have increased during the pandemic. For example, those experiencing financial difficulty owing to the pandemic may have no choice in what food to consume, and thus food labeling is irrelevant2,4,5. Food labeling for processed food is mandatory currently in South Korea. The labeling includes calories, carbohydrates, sugars, proteins, fats, saturated fats, trans fats, cholesterol, and sodium, and is expressed on the information display surface along with other identified terms, which is mostly the back of the pack7. This law for food labeling was enacted in 1996 and has not changed since7. While using food labeling on food packages is crucial in maintaining a healthy diet, there is limited research related to the change in food labeling awareness as a result of the pandemic8,9,10. Given this background, the aim of the study was to investigate the long-term trends and prevalence of food labeling awareness, comprehension, and usage before and during the pandemic as well as to determine pandemic-related vulnerability factors of food labeling awareness. The results of the study may assist in determining individual and personalized policies of food labeling to encourage a healthy diet and, ultimately, improve public health.

Methods

Study population and data sources

This nationwide, large-scale, long-term study is based on data provided by Korea Community Health Survey (KCHS) from 2014 to 202211. The KCHS was conducted to produce community health statistics for the establishment and evaluation of a community health care plan12,13,14,15,16. The participants were recruited based on the method whereby the probability of selection is proportional to the number of households classified by the type of residence. The survey collected a wide range of information, including age, sex, height, weight, region, food labeling awareness, food labeling comprehension, and food labeling usage12. Trained interviewers visited the selected households to conduct interviews with household representatives and individuals aged 19 years or older. The data used in this study were approved by the Korea Disease Control and Prevention Agency (KDCA) and Kyung Hee University (KHUH 2022–06-042). This study was conducted in accordance with the principles of the Declaration of Helsinki and all participants signed a written informed consent form17.

While 2,061,838 interviewees were conducted in total, participants who did not provide all the necessary information on the covariates used in this study were excluded (excluded n = 304,991). Therefore, 1,756,847 participants were included in the study (men: 804,808 [45.89%]; women: 952,039 [54.19%]).

Endpoints

Food labeling awareness was determined by the participants’ answers to the question on whether they were aware of food labeling on processed food packaging12,18. Participants that answered affirmatively were placed in the “aware group.” Participants in the aware group were further questioned to determine their comprehension of food labeling. Those who replied positively were categorized into the “comprehension group.” The participants in the comprehension group were questioned to determine their food labeling usage. Those who responded positively were categorized into the “use group” (Table S1)12,18. Participants who answered 'no' to any one question was considered to have answered 'no' to all subsequent questions.

Covariate definitions

This study included 9 covariates for consideration; age (19 to 39, 40 to 49, 50 to 59, 60 to 69, and ≥ 70 years), sex, body mass index (BMI), residential areas (urban and rural)19, household income (lowest quartile, second quartile, third quartile, and highest quartile), educational level (elementary school or less, middle school, high school, and college or more), smoking status (non-smoker, ex-smoker, and current smoker), alcohol consumption (below a day, once to four days, and five days or more per month), and subjective health level (high, normal, and low). BMI was subdivided into four categories, namely; underweight (< 18.5 kg/m2), normal weight (18.5–23 kg/m2), overweight (23–25 kg/m2), and obese (> 25 kg/m2), with respect to the Asia–Pacific BMI20,21,22,23,24.

Statistical analyses

The study used KCHS data from 2014 to 2022 to analyze long-term trends in the prevalence of food labeling awareness, understanding, and use. β-coefficients with 95% confidence intervals (CIs) were calculated using linear regression models for each endpoint before and during the pandemic (2014–2019; 2019–2022), and the differences in β-coefficients were estimated to assess the variations in trends between the before and during pandemic periods4,25. All β-coefficients and any differences between them were multiplied by 100 to simplify comprehension. To minimize the impact of potential confounders, all models utilized the following adjusting variables: age group, sex, BMI group, residential area, household income, educational level, smoking status, alcohol consumption, and subjective health level (Supplementary material). Additionally, to address the potential relationship between one or more regressors and the error variance, the Eicker-White heteroskedasticity-consistent estimator was used to estimate standard error values (Tables S2–S4).

$${\text{Endpoint}}={\beta }_{0}+{\beta }_{1}\times year+\sum_{i=2}^{10}{\beta }_{i}\times adjusted\,variable+\epsilon$$

A multivariate logistic regression model was used to analyze the coverage from 2019 to 2022 (2020 versus 2019, 2021 versus 2020, and 2022 versus 2021; Tables S5–7) and to express the ratio of Odds Ratios (OR) with a 95% CIs for identifying factors associated with vulnerability to food labeling awareness during the pandemic26. To ensure the representativeness of the study, a survey weighting analysis was conducted using KCHS's unique weighting system, household weighting, for all analyses. This assesses the household extraction rate by incorporating the sample extraction process in sample design, the rate of households suitable for the survey, and the distribution of households based on housing types. The statistical significance was defined as a two-sided p-value less than 0.05. Statistical analyses were performed by utilizing the SAS software (version 9.4; SAS Inc., Cary, NC, USA).

Results

In total, 1,756,847 adults were included in the KCHS between 2014 and 2022. A large proportion of participants were female (54.19%; n = 952,039; Table 1). Of these groups, 12.2% (n = 214,912) were surveyed in 2014, 12.3% (n = 215,297) in 2015, 12.3% (n = 215,667) in 2016, 12.4% (n = 217,659) in 2017, 10.0% (n = 174,726) in 2018, 10.2% (n = 179,182) in 2019, 10.2% (n = 178,295) in 2020, 10.1% (n = 177,461) in 2021, and 10.5% (n = 183,648) in 2022.

Table 1 Baseline characteristics of the participants in the KCHS, 2014–2022 (n = 1,756,847).
Full size table

Tables 2, 3, and 4 present the changes in trends regarding the proportion of individuals who reported awareness of the presence of food labeling on the products they buy, those who read and comprehend the labeling on the food, and those who make use of the food labels when buying products, respectively. All trends exhibited a constant rate of growth in awareness, comprehension, and utilization of food labeling (Fig. 1).

Table 2 Prevalence of food labeling awareness in the KCHS, 2014–2022 (n = 1,756,847).
Full size table
Table 3 Prevalence of food labeling comprehension in the KCHS, 2014–2022 (n = 1,756,847).
Full size table
Table 4 Prevalence of food labeling usage in the KCHS, 2014–2022 (n = 1,756,847).
Full size table
Figure 1
figure 1

Prevalence of food labeling awareness, comprehension, and usage in the KCHS, 2014–2022 (n = 1,756,847).

Full size image

The upward slope in overall food labeling awareness became less pronounced and even exhibited a downward slope momentarily during the pandemic. The slope value was 2.158 (95% CI 2.105–2.210) before the pandemic and 0.399 (95% CI, 0.297 to 0.501) during the pandemic (βdiff − 1.759; 95% CI − 1.874 to − 1.644). (Table 2) Before the pandemic, the rate of food labeling comprehension exhibited a gradual increase, which intensifies during the pandemic. The overall slope value was 0.259 (95% CI 0.215–0.303) before the pandemic and 0.794 (95% CI 0.705–0.882) during the pandemic (βdiff 0.535; 95% CI 0.436–0.634). (Table 3) Similarly, regarding food labeling usage, the overall slope was 0.174 (95% CI 0.133–0.215) before the pandemic and 0.867 (95% CI 0.784–0.949) during the pandemic (βdiff 0.693; 95% CI 0.601–0.785), showing a similar trend change. (Table 4) Adjusted OR of food labeling awareness was 0.900 (95% CI 0.887–0.914) between 2019 and 2020, 1.013 (95% CI 0.998–1.028) between 2020 and 2021, and 1.111 (1.095–1.128) between 2021 and 2022 (Table S5). Conversely adjusted OR of food labeling comprehension was 0.964 (95% CI 0.948–0.980) between 2019 and 2020, 1.082 (95% CI 1.064–1.100) between 2020 and 2021, and 1.050 (95% CI 1.033–1.067) between 2021 and 2022 (Table S6). Similarly adjusted OR of food labeling usage was 0.995 (95% CI 0.977–1.013) between 2019 and 2020, 1.091 (95% CI 1.072–1.110) between 2020 and 2021, and 1.041 (95% CI 1.023–1.059) between 2021 and 2022 (Table S7).

The ratio of OR before to during the pandemic increased in relation to the age group (≥ 65 versus 19–39 years; ratio of OR 1.295; 95% CI 1.270–1.319), being male (ratio of OR 1.179; 95% CI 1.164–1.194), higher BMI (obese versus underweight; ratio of OR 1.237; 95% CI 1.198–1.278), residing in a rural area (ratio of OR 1.062; 95% CI 1.049–1.076); having a lower household income (ratio of OR 1.091; 95% CI, 1.067–1.117), lower educational level (risk of OR 1.163; 95% CI 1.138–1.189), being a smoker (ratio of OR 1.199; 95% CI 1.178–1.220), having heavy alcohol consumption (risk of OR 1.103; 95% CI 1.085–1.123) (Table 5).

Table 5 Pandemic-related factors of food labeling awareness in the KCHS, 2014–2022 (n = 1,756,847).
Full size table

Discussion

Findings of our study

This study analyzed the 9-year trend in the prevalence of food labeling awareness, food labeling comprehension, and food labeling usage based on nationally representative data from adults in South Korea from 2014 to 2022. The overall prevalence of food labeling awareness, food labeling comprehension, and food labeling usage exhibited a continuously increasing trend. Unlike the trend of before the pandemic breakout, there was a deceleration in the trend slope of nutrition label awareness, indicating a slower rate of increase during the pandemic period from 2019 to 2022. On the other hand, the trend slope of food labeling comprehension and usage accelerated, indicating a faster rate of increase during the same period of pandemic. The pandemic-related vulnerability factors of food labeling awareness were older age, male, obesity, residing in rural area, lower household income, lower educational level, smoking, and increased alcohol consumption. This study’s findings suggest personalized nutrition strategies, such as educating the importance of a balanced diet to recognize vulnerable groups with risk factors, and to improve food labeling awareness among Korean adults after the pandemic breakout.

Comparison with previous studies

Few studies have investigated food labeling awareness or the determination of vulnerability factors. A previous Korean study on the prevalence of food labeling reported a constant increase in awareness and utilization of food labeling from 2014 to 201718. However, this study only analyzed a short period (2014–2017) that did not include the pandemic period. In addition, studies conducted in China investigating the current food labeling knowledge, attitude, and practice27 only conducted a survey in a single city with a small sample size (n = 636) and cross-sectional design, and were unable to demonstrate any trend changes. A study conducted in the USA analyzed the food labeling usage of participants with chronic diseases28. However, this study exclusively focused on participants with chronic diseases, limiting its ability to provide insights into food labeling usage among the general population. In addition, a study from Italy discovered that most people do not use food labeling completely consciously and that simplifying the label format would help people with no expertise to read the labeling8. However, the small sample size and short observation period has a difficulty showing the trend of the general population. Results from the present study provide long-term evidence demonstrating how the pandemic affected the awareness, comprehension, and usage of food labeling, which covered 1,756,847 participants and was collected over nine years.

Possible explanations

The constant increase of food labeling awareness, comprehension, and usage despite the unchanging food labeling system may be explained by increasing interest of people towards maintaining a healthy diet. The deceleration of awareness increases and the acceleration of the increase in food labeling comprehension and usage may be attributable to quarantine and social distancing29. A study conducted in Korea revealed that there was a substantial rise in the volume of business for food services (839.7%), food and beverage (203.8%), as well as agricultural goods, meat, and fish products (193.5%) from 2017 to 202130. As a result of the pandemic, a greater number of individuals have resorted to shopping for groceries through the internet (also referred to as online shopping). This movement might have likely hindered the increase in food labeling awareness, as fewer individuals see the product in person and inspect the nutrition label31,32. Additionally, after selecting the product and delivering it, examining the package becomes irrelevant, thereby reducing the increase in awareness. However, individuals who were already aware of food labeling might have become more active in reading and utilizing food labeling, as the pandemic has heightened people’s interest in maintaining a healthy diet33. Furthermore, because of the COVID-19 quarantine, a greater number of individuals have started cooking meals at home as opposed to dining out32,34. This shift in behavior may have led to an increase in comprehension and usage of food labeling.

Policy implication

As previous studies have not analyzed the differences in food labeling awareness before and during the pandemic, they exhibit a somewhat optimistic view of the trend in food labeling awareness18,33. However, this study’s results imply that the pandemic has negatively influenced the overall awareness of food labeling, necessitating further efforts to increase its awareness21. There is no specific advertisement that would assist in increasing the awareness of food labeling. Food labeling awareness significantly affects whether a consumer can maintain a healthy diet when required. Another suggested policy is the front-of-pack labeling (FOPL)35. It indicates that people with lower socioeconomic status (defined according to participants’ education, income, areas of residence or the store’s location) are affected more by FOPL. Policymakers should recognize that the pandemic negatively affected people’s food labeling awareness and that efforts must be made to provide citizens with the right information on the label of food packages36,37,38.

Strength and limitations

This study included a large-scale, nationally representative, and long-term serial survey that covered the pandemic period (2019–2022). However, there were limitations to this study that should be considered. First, the data utilized by the researchers were missing for people who did not answer all the questions related to the covariates used in this study. The missing data may lead to a biased interpretation of the data as the missing data may have similar factors4,39. Second, while this study contained data from the pre-pandemic (2014–2019) to the pandemic (2020–2022) period, it did not contain data from the post pandemic (2023–) period40. Further research of food labeling awareness, comprehension, and usage must be conducted to estimate the future trend. Third, this research examines the trend of food labeling awareness, comprehension, and usage before and after the pandemic, but does not examine the causal relationship between the two factors. Therefore, further research must be conducted to examine the causal relationship and solve the fundamental problem of information inequality in food labeling. Finally, the dataset exclusively comprised individuals in Korea, thereby missing the consideration of racial and cultural variances in the research41. Consequently, additional investigations should be conducted in countries with diverse cultural backgrounds.

Conclusion

This study elucidated that food labeling awareness, comprehension, and usage increased throughout the years of data collection. The COVID-19 pandemic negatively influenced food labeling awareness but positively impacted food labeling comprehension and usage. Through the identification of risk factors for food labeling awareness, it will be necessary to present a political solution to steadily increase food labeling awareness and bridge the gap between the groups.