In recent years, fast foods, take-out foods, and prepackaged foods have become increasingly popular, with a rapid increase in the consumption rate of them (1–3), among which, the consumption rate of prepackaged foods in China has reached 59.8% (4). The poor cooking methods of take-out foods and fast foods caused a large accumulation of unhealthy ingredients in the body, including fat, salt, and sugar (1, 5). Prepackaged foods (including puffed food, beverage, pickled canned food, and leisure food) are generally high in energy, fat, and sodium and low in protein and dietary fiber (6, 7). Whether it is fast foods, take-out foods, or prepackaged foods, its rising consumption rate and accumulation of unhealthy nutrients are the key factors in causing the high incidence of chronic diseases such as obesity and diabetes (8, 9). Diet-related diseases have become more common because of changes in lifestyle and food habits, but researchers have also established that dietary modifications significantly reduce the risk of diseases (10). Individual food choices and eating behaviors are influenced by many interrelated factors which affect the results of nutrition-related public health interventions. To improve the adverse health effects of this situation, recommendations and interventions have been implemented across the globe. Nutrition labeling plays an active role as a dietary strategy as recommended by the WHO (11). In the face of increasing diet-related chronic diseases, many countries have initiated steps to include nutrition labeling on prepackaged food packets and in restaurant menus to standardize the management of nutrition labeling (12).

Nutrition labeling is not only an information tool to interpret the nutrient content and function of food but also a strategy against overweight and obesity (13), which plays a critical role in promoting healthy eating habits. Petimar et al. found that the calorie menu labeling was associated with an immediate decrease of 60 calories per transaction or 4% of total calories purchased (14). A meta-analysis expressed that food labeling decreased consumer intakes of energy by 6.6%, total fat by 10.6%, artificial trans fat by 64.3%, sodium by 8.9%, and other unhealthy dietary options by 13.0% while increasing vegetable consumption by 13.5% (15). The implementation of nutrition labeling and sugar labeling can contribute to the lower risk of cardiovascular diseases and cancer and kidney diseases, thereby reducing the prevalence of chronic diseases and increasing life expectancy (16, 17). In addition, under the Nutrition Labeling and Education Act (NLEA) promulgated by the United States in 1990, nutrition labeling will be required for all retail food products to facilitate consumers to obtain more nutrition information and maintain healthy dietary practices (18). Although the nutrition labeling system has been introduced in China as early as 1994 (19) and has been revised several times, the practice of nutrition labeling has not been actually promoted until the General Rules of National Prepackaged Food Nutrition Labels (GB 28050-2013) were enacted in 2013 (20, 21). Since then, the labeling rate of nutrition labeling in prepackaged foods has been significantly improved (22). Since then, the labeling rate of nutrition labeling in prepackaged foods has been significantly improved. However, the actual utilization of nutrition labeling was in fact lower than that reported (23), possibly because the complex design of nutrition labeling is puzzling, including energy conversion and professional terms description (24, 25). Previous studies have found that the longer a consumer gazed at the nutrition claim, the more likely the product with a nutrition claim was bought (26). It is also reported that consumers who regularly use the nutrition labeling seem to have a higher diet quality (27).

The knowledge, attitude, and practice (KAP) model is a theory to explain an individual healthy behavior (28) and the model figures that there are two key steps to changing behavior: establishing beliefs and changing attitudes. Up to now, the KAP model has been widely applied to health education in the fields of prevention of primary infectious diseases (e.g., schistosomiasis, tuberculosis, malaria, and AIDS) (29–31) and control of chronic diseases (e.g., diabetes and hypertension) (32, 33). However, few studies have explored the relationships among knowledge, attitude, and practice behavior of nutrition labels based on the KAP theory (34–36). The use of nutrition labeling is a dietary self-management behavior and is closely related to their own nutrition knowledge and health beliefs. Therefore, we adopted the KAP model as a framework to explore the relationships between them, and the new findings may contribute to future nutrition education to promote nutrition label use in China.

According to the KAP theory, there is a causal relationship among knowledge, attitude, and practice (37). However, KAP are potential variables that are difficult to measure directly. The traditional statistical methods cannot deal with these potential variables effectively, while the structural equation model (SEM) integrates the traditional statistical analysis methods, such as confirmatory factor analysis, path analysis, and multiple regression analysis, leading to a new multivariate statistical technology. It can not only analyze and deal with measurement errors but also analyze the structural relationship between potential variables (38, 39) and directly display the correlation between the variables through the path diagram. In addition, it can also explore the causal relationship between potential variables and quantitatively evaluate the direct and indirect effects of variables (40), as shown in Figure 1.

Knowledge means the ability of understanding and using nutrition information, through education, learning experience, and identifying the nutrition label. Attitude refers to the feeling or opinions of community residents on nutrition labeling in some situations, including credibility, helpfulness, and necessity. Practice refers to the use or application of nutrition labeling by community residents. Based on the KAP model, it is predicted that nutrition knowledge will positively and indirectly affect practice through attitude change, and nutrition knowledge may also directly affect nutrition labeling practice. We put forward the following five assumptions based on the relevant literature on the knowledge, attitude, and practice structure model published by Zeng Y, Kwak C, Zeying H, and Misra R.

Thus, we attempted to analyze the interactions among community residents' nutrition labeling knowledge, attitude, and practice by using the KAP model to construct a structural equation. Meanwhile, we should also explore residents' cognition and use behaviors of nutrition labeling, as well as the influencing factors so that the resident can have a better understanding of nutrition labels and habits of food choice.

Data analysis was performed in three stages. First, the data were analyzed using EpiData3.1 data entry. Second, SPSS 24.0 (IBM, NY, United States) was used for statistical analysis and reliability and validity tests. If the quantitative data were subjected to the normal distribution, it is described by mean standard deviation (mean ± SD); On the contrary, it is described by median or interquartile value. If the data submitted to normal distribution and homogeneity of variance, a one-way analysis of variance or chi-square test was used. Otherwise, the Wilcoxon rank-sum test is adopted for comparison. Finally, the KAP structural equation model (SEM) of nutrition labeling for community residents was constructed by using AMOS 24.0 (IBM, NY, United States) software, and the model was revised by Modification Indices. The model fitting was evaluated with χ² -value, GFI (goodness-of-fit index), AGFI (adjusted goodness-of-fit index), TLI (Tucker–Lewis index), CFI (comparative fit index), NFI (normed fit index), IFI (incremental fit index), and RMSEA (root mean square error of approximation). The test level was 0.05, and p < 0.05 indicated that the difference was statistically significant.

The results of this research indicated that the overall cognition level of community residents on nutrition knowledge was low, with an awareness rate of 32.7%, which was unsatisfactory and lower than the national average level (47). Residents have a positive attitude toward nutrition labeling. Approximately 76.1% of the residents indicate that it was necessary to mark the nutrition label on the food packaging; 52.5% of the residents believed that nutrition labeling could help healthy eating or shopping choices in the future, and 33.6% of the residents considered that the nutrition labeling was credible. In total, 38.5% of participants indicated that nutrition labeling would affect their shopping behavior. However, only 25.3% of the residents could understand nutrition labeling, indicating that most of the residents had a positive attitude toward nutrition labeling, but they lack a correct understanding of nutrition labeling and doubt their authenticity. The main reason may be that the promotion of nutrition labeling in China is done mainly to increase the reliability and marking rate of labels, rather than educating residents on nutrition knowledge popularization, label content interpretation, and use training.

Previous studies have shown that there are still existing obvious gaps between the identification of nutrition labeling and use behavior in real life. Especially, young people who are active consumers of prepackaged foods, have plenty of chances to contact with nutrition labeling but rarely use them effectively in fact. The practice of nutrition labeling not only depends on whether to establish health belief and implement restaurant menu labeling (48, 49) but also depends on demographic, social, and psychological factors of the population. (50). In this study, we found that residents who were young, female, having high education level, and having high socioeconomic status had higher awareness of nutrition labeling and more positive attitudes, and the frequency of checking nutrition labeling is also higher, which was consistent with previous studies (51, 52). With increasing attention to weight loss, calorie intake restriction, and own health in recent years, nutrition labeling can be an effective tool to directly obtain the nutrition information of packaged food for consumers, which can also help consumers to make a healthy choice. Therefore, nutrition labeling plays an indispensable role in helping residents maintain healthy eating habits (53).

The advantage of this study is its adoption of the mature KAP model to analyze Chinese community residents' cognitive status of nutrition labeling, which was divided into knowledge, attitude, and practice, and then establishing the structural equation. Regarding statistical methodological strength, structural equation modeling is superior to multiple linear regression modeling. The structural equation model can analyze multiple dependent variables at the same time, and its application is helpful to scientifically analyze the relationship between indicators. In this study, SEM is helpful to analyze the direct effects of the nutrition label KAP and to reveal these relationships. However, there are still a few limitations in the present study. First of all, more rigorous survey questions need to be designed. For example, participants were likely to make inaccurate responses, since the nutrition labeling contains a lot of information and residents are likely to confuse the list of ingredients with the nutrition fact table. Furthermore, we only used one topic to assess the residents' previous nutrition education experience, and we also tested the residents' subjective knowledge. In future research, we need more objective scales to measure residents' subjective knowledge of nutrition labeling, rather than using simple self-reporting questions, rather than through the use of simple self-reported questions. Finally, our sample size was small, drawn by the convenience sampling method, and hardly ensured that the findings above could be replicated within behavioral studies. Other mediating factors (e.g., peer or parental impact on their use of nutrition labeling, understanding of diet-related disease information, taste or sensory attributes of the product) might more effectively explain that residents' use of nutrition labeling was not included in the study and need to be explored in future studies.

Behavior changes of community residents were divided into three continuous processes: knowledge acquisition, belief generation, and behavior formation, which are positive relations (path coefficients > 0). In this study, the path analysis demonstrated that the path coefficient between nutrition labeling knowledge and trust was 0.561 (p < 0.05), and the path coefficient between nutrition labeling knowledge and attitude was 0.764 (p < 0.05), with a significant positive correlation between them, indicating that residents could form a more positive attitude toward the nutrition labeling if they were knowledgeable about the nutrition labeling. Evelyn et al. (54), Rimpeekool et al. (55), Jackey et al. (56), and Cannoosamy et al. (57) also reported similar results in their respective investigations.

Previous studies have suggested more nutrition knowledge, and health-motivated residents might be more skeptical about nutrition claims and nutrition function claims, thus limiting residents' practice of nutrition labeling. We also tested this relationship, and the correlation analysis found that there was a significant positive correlation between nutrition knowledge and trust (path coefficient = 0.561, p < 0.05), the trust was negatively correlated with nutrition practice, but it was not significant (path coefficient = −0.059, p > 0.05), which may be the most residents are skeptical about the authenticity of nutrition labeling in this study. Residents' trust score is low, which leads to a negative correlation between trust and the practice of nutrition labeling, which was the same as the previous study.

We found that more nutrition knowledge and positive attitudes could increase the practice of nutrition labeling among residents in this research, which was consistent with the results of previous studies (58–61). It means that based on the model, consumers are likely to establish positive beliefs, and finally change use behaviors, once they receive nutrition education (55). However, it should be noted that the residents' trust in nutrition labeling was not significant with their frequency of using nutrition labeling. Therefore, in this study, attitude is a psychological reaction (including helpfulness and necessity) to convince ourselves that nutrition labeling is helpful and useful to select healthy foods, which will further change our practice of nutrition labeling (62, 63).

In summary, we confirmed that residents' nutrition knowledge could be directly converted into nutrition labeling reading behavior or indirectly through changing their attitudes. Residents with higher nutrition labeling knowledge scores and more positive attitude towards nutrition labeling seem to be more likely to obtain the information provided on nutrition labeling. It reflected that knowledge of nutrition was the basis for changing the practice of nutrition labeling. Rich nutrition knowledge can promote the use of nutrition labeling, while poor nutrition knowledge will limit their practice. Therefore, we must pay attention to the synchronous development of nutrition labeling KAP.

The KAP model is suitable for analyzing the use behaviors of nutrition labeling by local residents. There was a direct and indirect correlation between nutrition knowledge and the practice of nutrition labeling. The attitude of nutrition labeling was positively affected by knowledge, while the use behavior of nutrition labeling was positively affected by knowledge and attitude.

To improve the lifestyle of residents and correctly use nutrition labeling, the following policy recommendations are offered. First, more public education programs (e.g., printing graphic brochures or posters, learning websites, and special lectures) should be implemented in schools and communities. The purpose of public education programs is to make the public aware of “the availability of nutrition information in nutrition labeling and the importance of that information in maintaining healthy dietary practices” to improve their nutrition literacy. Specifically, the interpretation of nutrition labeling needs to be included in the Dietary Guidelines for Chinese Residents and should be disseminated in the annual National Nutrition Week activity. The theme of National Nutrition Week 2022 is Learn How to Cook, How to Select Ingredients Reasonably, and Check Nutrition Labeling. The guideline, “Learn to read food labeling and choose prepackaged foods reasonably”, highlights the core value of “Check Nutrition Labeling”. Second, the concept and function of NRV% and core nutrients on the nutrition facts table, especially sodium and fat, should be conveyed transparent. It is suggested to mark NRV% explanation on food packaging to ensure consumers understanding. Then, it is necessary to strengthen the nutrition education of residents so that they fully understand the meaning of nutrition claims and nutrient function claims and avoid confusing nutrition function claims and health food function claims.

Finally, it will appeal to the relevant departments to implement effective supervision and inspection to ensure the accuracy of nutrition labeling information, which can in turn enhance consumers' confidence in the nutrition labeling. With the government as the leading role and the participation of the whole society, we should strengthen the publicity and education of labeling knowledge and improve residents' nutritional literacy and cognitive attitude toward labeling knowledge to change their behaviors.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Ningxia Medical University Ethics Committee. The patients/participants provided their written informed consent to participate in this study.

JY conceived and designed this research and revised the manuscript. YL was responsible for data analyses and prepared the manuscript. Both authors have read and approved the final version.