COVID-19 Changes Public Awareness about Food Sustainability and Dietary Patterns: A Google Trends Analysis

The COVID-19 pandemic has not only affected healthcare systems and global economies but also directly impacted food security and purchasing behaviors. The aim of this study is to investigate if COVID-19 has induced changes in public interest regarding Food Sustainability and healthy-sustainable dietary patterns across Europe and in European regions. A Google Trends search was performed using the search terms “Food Sustainability + Sustainable Diet + Sustainable Food” (grouped as “Food Sustainability”) and the topics “Sustainability”, “Healthy Diet”, “Mediterranean Diet”, and “Flexitarianism” for the years 2010 to 2022. Data were obtained for 12 countries in Europe. The trends in interest after the COVID-19 outbreak were forecast based on previous data. After the COVID-19 outbreak, an increase in Food Sustainability interest was observed and was higher than forecast based on the previous data. A significant interest increase in Sustainability was observed; nevertheless, this increase was smaller than the forecast increase. Mixed results were obtained for dietary patterns across European regions, yet, considering the mean interest for Europe, it seems that the COVID-19 pandemic outbreak dampened the interest in dietary patterns such as the Healthy Diet and Flexitarianism and promoted an interest in the Mediterranean Diet. Understanding consumers’ beliefs and behaviors toward food choices is crucial for the transition towards sustainable diets, and definitions of educational and behavioral interventions are essential to this transition.


Introduction
The COVID-19 pandemic significantly changed world economy dynamics and brought new concerns for countries [1,2]. To prevent and limit the transmission of the novel coronavirus, confinements and lockdowns were imposed by several governments. The restrictions to individual liberty directly impacted food security, food waste, and purchasing behaviors [3]. In 2020, the Food and Agriculture Organization of the United Nations (FAO), the International Fund for Agricultural Development (IFAD), the World Bank, and the The accessibility of the Internet and the rise of social media have affected our social lives and our dietary and lifestyle behaviors [26,27]. The Internet provides immediate access to an enormous amount of information; its use increased and was the main source of information used during the COVID-19 pandemic [28,29]. The Internet has also become the main source of health-related information including nutrition [30]. A valuable aspect of web-based information transactions is the record of communication itself, which, in aggregate, may reflect population-level behaviors. For example, researchers have used search engine queries and volumes to recognize population behavior-based patterns [31]. The data that are accumulated during Internet search activities are one form of Big Data that may provide valuable insights and information into population behavior and interests. One tool that allows users to interact with Internet search data is Google Trends, a free, publicly accessible online portal of Google Inc. Google Trends analyses a portion of the three billion daily Google searches [32]. Google Trends is the most popular tool to gather information on web-based behaviors, and it can be used to predict or prevent health-related issues [33]. The analysis of Internet search queries offers information on the extent of public attention, thereby reflecting the level of public awareness. Google Trends has been used in many research publications to analyze users' interests across various fields [34].
Currently, limited data are available on how the COVID-19 pandemic affected our dietary and lifestyle-related behaviors and interest in the sustainability of the food system at regional and global levels. Google Trends is a powerful tool to analyze relevant keywords related to these topics and to address this shortcoming. The aim of this study was to investigate the public interest in Food Sustainability and healthy, sustainable dietary patterns across Europe and in European regions using Big Data gathered from Google Trends. More specifically, our objectives were to investigate the changes of public interest in Food Sustainability and healthy sustainable eating patterns between the pre-COVID-19 and post-COVID-19 eras, and to explore the changes in public interest induced by the COVID-19 outbreak by comparing the observed and predicted interest post the COVID-19 outbreak.

Google Trends
Google Trends (https://trends.google.com/trends/ accessed on 11 July 2022) is an online and free tool provided by Google that measures search interest in a particular topic or a set of search terms. It is anonymized, categorized (determining the topic for a search query), and aggregated (grouped into topics). Google Trends data can be accessed in non-real time and go as far back as 2004 and up to 72 h before you search. The data are indexed and normalized, which means the numbers are scaled on a range of 0 to 100, where each data point on the graph represents the proportion of a search topic, as a search term, of total searches in a given country or worldwide during the time period selected. A value of Relative Search Volume (RSV) of 100 is the maximum search interest for the period and location selected. Searches with low volume, duplicate searches, and special characters are excluded from search.
Google trends may qualify analyzed phrases as "search term" or "topic". Search terms are literally typed words, whereas topics may be proposed by Google Trends when the tool recognizes phrases related to popular queries. Topics enable easy comparison of the given term between countries. For example, the search term "London" will be analyzed by Google Trends exactly; thus, RSV will be the highest in English-speaking countries, whereas the topic "London" will include all queries associated with the query in all the available languages, for example, "Capital of the United Kingdom" and "Londres", which is "London" in Portuguese and Spanish [30].
It is important to note that none of the searches in the Google database for this study can be associated with a particular individual. The database does not retain information about the identity, IP address, or specific physical location of any user.
Google accounts for >80% of global search engine use [35], and Google Trends is one of the few open sources of search query data. Google Trends is considered a valid and robust indicator to track interest, attention, and public opinion over time [36,37] and has increasingly been used to quantify trends in public interest in several fields such as health care [32,33], tourism [38], and economics [39,40]. The strengths and advantages of Google trends are evident; nevertheless, this tool also presents limitations. Due to time and geographical normalization, we can only compare relative popularity, meaning that different regions that show the same RSV for a term will not always have the same total search volumes. Another limitation to the use of Google Trends is the fact that not everyone uses the Internet, and Internet use may be higher in younger people; therefore, Google Trends data may not be fully representative and may present a selection bias [41,42].

Data Collection
Google Trends searches and data extraction were performed on 11 July 2022, with a single data extraction for each country and covering the period of 1 January 2010 to 30 June 2022. The search period covered before and after the outbreak of the COVID-19 pandemic to reflect changes in relative interests. We used two different approaches: (1) one consisting of the Sustainability, Healthy Diet, "Flexitarianism" (FlexDiet) and "Mediterranean Diet" (MedDiet) topics, and (2) another being a combination of search terms related with "Food Sustainability" ("Food Sustainability" + "sustainable diet" + "sustainable food"). The search term combinations are listed in Table 1 in the official language of each country. The RSVs for the countries of northern Europe (Denmark, Norway, and Sweden), western Europe (France, Germany, Ireland, the Netherlands, and the United Kingdom) and southern Europe (Italy, Portugal, Spain, and Turkey) were retrieved. No RSVs could be retrieved for other European countries for "Food Sustainability". The filters "Country", in "All Categories" and for "Google Web Search" were selected in our searches.

Statistical Analysis
The RSVs for each country were used to obtain the monthly mean RSVs for the European region in which the countries were included and the mean RSVs for Europe.
All statistical analyses in the present study were conducted using IBM ® SPSS ® Statistics for Windows, Version 28.0.1.0 (IBM Corp., Armonk, NY, USA). Forecast models of trend search terms and topics were created using the "Expert Modeler" function. The Autoregressive Integrated Moving Average (ARIMA) and exponential smoothing models were considered to obtain the best-fitting model. Forecast period was set for 30 months, from 1 January 2020 to 30 June 2022, based on the data obtained for the period from 1 January 2010 to 31 December 2019. The best-fitting model for each of the selected dependent variables was selected. Forecasting in the time series for Food Sustainability query, in European regions and in Europe, was performed using Winter's additive exponential smoothing models. Forecasting in the time series for sustainability topic in southern Europe, northern Europe, and Europe was performed using Winter's multiplicative exponential smoothing models, and for western Europe was performed using Winter's additive exponential smoothing model. Regarding the Healthy Diet topic, forecasting in the time series for southern Europe and Europe was performed using Winter's additive exponential smoothing models, in the time series for western Europe was performed using Winter's multiplicative exponential smoothing models, and in the time series for northern Europe was performed using simple seasonal exponential smoothing models. Forecasting in the time series for FlexDiet in southern Europe, western Europe, and Europe was performed using Winter's additive exponential smoothing models, and in the time series for southern Europe was performed using ARIMA (0, 1, 1) (0, 0, 1) model. Forecasting in all the time series for MedDiet was performed using simple seasonal exponential smoothing models.
Differences in RSVs observed 30 months pre-and 30 months post-COVID-19 outbreak were tested (July 2017 to December 2019 vs. January 2020 to June 2022), using the Mann-Whitney U test. The forecast data were then compared to the actual search interest for the same period (January 2020 to June 2022), using the Wilcoxon signed-rank test, to determine the effect of COVID-19 on the selected search term interest. Descriptive statistics are presented as median (Mdn) and interquartile range (IQR) for each variable. Results were considered statistically significant if p-value < 0.05.

Results
Google Trends data are presented for European regions by term or topic. The RSV evolution for the Food Sustainability query, and the Sustainability, Healthy Diet, FlexDiet, and MedDiet topics, is shown in Figures 1-5, respectively.

Trends of RSVs for Healthy Diet
The evolution of the RSVs for the Healthy Diet topic is shown in Figure 3

Trends of RSVs for MedDiet
The evolution of the RSVs for the MedDiet topic is shown in Figure 5

Discussion
In this study, we explored, for the first time, the impact of the COVID-19 pandemic on the public's interest in Food Sustainability, in Sustainability, and in dietary patterns, namely the Healthy Diet, FlexDiet, and MedDiet, using Big Data. Specifically, we used the RSVs provided by Google Trends to obtain temporal series of public interest in the mentioned topics before the COVID-19 pandemic and to forecast public interest after it. By comparing the predicted and observed values, an image of how this global health event affected public interest in Food Sustainability, in Sustainability, and in dietary patterns, was obtained. Using this strategy, an increase in Food Sustainability public interest after the COVID-19 outbreak was observed, and this increase was higher than what was forecast based on the previous data. A significant increase in interest in Sustainability in general was also observed; nevertheless, this increase was smaller than that forecast by our models. Taken together, these results may indicate that the COVID-19 pandemic outbreak sparked an interest in Food Sustainability while reducing the increase in interest in general Sustainability. Furthermore, mixed results regarding dietary patterns were obtained across European regions; yet, considering the mean interest for Europe, it seems that the COVID-19 pandemic outbreak dampened the interest in dietary patterns such as the Healthy Diet and FlexDiet, and promoted interest in the MedDiet, at least in the western European region.

Changes in Food Sustainability Interest after COVID-19
The COVID-19 pandemic has changed the functioning of food systems all over the world and, despite its stability, Europe has been no exception. The pandemic outbreak caused problems in the operations of all the food system's players, including the actors providing the means of food production, the food producers, the logistics and processing industry and, inevitably, this turbulence went all the way to consumers. The COVID-19 pandemic affected the food system directly and indirectly; for instance, COVID-19 disrupted food supply transportation (e.g., access to container transport) and changed the demands for related services (such as retail food pickup and delivery services) [43]. Incidents of food flow disruption due to the pandemic were largely highlighted by the media and varied from vegetables rotting in fields and milk being dumped, to food processing facilities running short of workers due to the disease, and panic buying in stores. Strains on multiple points of the food supply chain could affect food availability and prices in the retail sector, ultimately impacting people at the receiving end [44,45]. Our results indicate that public interest in Food Sustainability increased after the COVID-19 outbreak to a higher degree than would be expected considering the tendencies of previous years. A possible explanation may reside in the perfect storm that assaulted the food system during the pandemic and the media coverage of those problems that may have led consumers to seek information on food sustainability and security.
Few studies have explored the adoption of sustainable food behaviors and attitudes during the COVID-19 pandemic. Muresan et al. [46] explored consumers' attitudes towards sustainable food consumption during the COVID-19 pandemic in Romania and found a positive attitude towards sustainable food behavior. Despite the positive attitude towards sustainable food behavior, as observed in previous works [47,48], a gap between attitudes and actual behaviors was observed. The authors also noted that respondents supported local producers, were aware of the importance of domestic agriculture, and were concerned about their health and waste reduction when planning a food menu; however, they also noted that even if the respondents' perception of the sustainable food choices was positive, their purchasing behavior was very different when it came to buying local products or supporting third-world farmers. Possible explanations were put forward, such as the higher prices of local foods compared to prices in supermarket chains or simply the fact that Romanian consumers are not ready to change their consumption patterns, as part of a complex and long process (51). For the most part, human behavior is the force of habits determined by cultural, social, and economic factors. Habits are forged in the process of repeating the same responses in particular contexts. Dietary choices are made virtually every day, driven by habitual response, and are extremely stable [49]. However, when the components that make up our environment or circumstances change, our food choice habits may change as well [50]. Therefore, it is possible that the pandemic period, which patently constitutes a disruption to habitual circumstances, could provoke dietary habit changes [49].

COVID-19 Impact on Sustainability Interest
Muresan et al. [46] also reported that respondents remained neutral towards the environmental issue, without any interest in participating in protection actions, with their actual behavior being limited to recycling food packaging. Along the same lines, Rousseau and Deschacht [51], by analyzing online search behavior in twenty European countries regarding how public awareness of nature and the environment has evolved during the COVID-19 crisis, found that public awareness of environmental issues such as climate change was unaffected by the COVID-19 crisis based on online searches. Our results are along the same lines; despite significantly higher levels of interest in Sustainability after COVID-19, the increase was smaller than would be expected if the pre-COVID-19 conditions were maintained. A possible explanation may reside in the perception that the general increasing sense of urgency observed in the last few years for many environmental problems may have been counteracted by the reduction in media attention and the shortterm positive trends reported for several pollution levels [51].

COVID-19 Impact on Dietary Pattern Interest
Surprisingly, public interest in the Healthy Diet and FlexDiet decreased after the COVID-19 pandemic outbreak. These results are unexpected, since health is one of the main concerns that consumers have in mind when making food choices, especially during COVID-19 times [52]. No studies were found regarding the adherence to healthy or flexitarian dietary patterns, yet one study reported an increase in meat consumption after the pandemic outbreak [53]. Furthermore, reports from gray literature indicate a decline in the number of flexitarians in the United Kingdom [54]. We also observed an increase in the public interest in the MedDiet that was greater than would be expected. In agreement with our results, Della Valle and collaborators [55] found evidence suggesting that adherence to the MedDiet during lockdown might have increased in some settings, while the determinants of such a trend are to be further explored.
No data are available regarding the media coverage of dietary patterns before and/or after the COVID-19 outbreak. However, it is possible that the large media attention directed to COVID-19 may have diverted public attention away from dietary-related issues. It is also likely that media coverage on dietary issues varies across European regions and that, along with many other factors, may explain the inconsistent results obtained across European regions.

Strengths and Limitations
Despite the wide use of Google trends to assess public interest in several fields, no research was found regarding Food Sustainability and/or sustainable diets. The assessment of the sustainability of diets is a topic that has received a sizable amount of research attention in recent years. Most of the research conducted in that field has been performed since 2006 [56]. Consumers are relevant players, shaping food systems [22,57,58], and while the academic interest in this topic is clear, public/consumer interest seems to be overlooked [59].
The novelty of this work is a strength, yet some limitations must be acknowledged. Google Trends is considered a valid and robust indicator to track interest, attention, and public opinion over time [36,37]. Nevertheless, Google Trends is not transparent regarding what content is included in each topic. It is possible that the content included in a topic, despite being related, may not be the most obvious. For instance, it is reasonable to assume that searches related to the history or the economy of the MedDiet are included in the MedDiet topic. Despite being related, those searches are not relevant to the MedDiet as a dietary pattern. Furthermore, while the topics of MedDiet and Flexitarianism cover well-defined and recognizable diets, the topic of Healthy Diet may include a broad range of diets, including MedDiet and Flexitarianism. A strict interpretation of the topics used in this study as dietary patterns should be made with caution and considering they are a proxy measure.

Conclusions
This work explored, for the first time, changes in consumer interest in Food Sustainability, Sustainability, and healthy dietary patterns in Europe induced by COVID-19. Across the counties analyzed, public interest in Food Sustainability increased as a result of the COVID-19 outbreak, and this increase was not the result of a general increase in public interest in Sustainability. Also, public interest in healthy dietary patterns was reduced or only regionally increased.
Understanding consumers' behavior and interest toward food choices is crucial for the transition to sustainable diets, and to the definition of the educational and behavioral interventions that are essential to this transition [60]. Engagement, interest, and information seeking are also essential for the construction of beliefs and behaviors. More research is needed in this field; nevertheless, these findings may be important for policy makers to design and select strategies that may promote a transition to a sustainable food system supported by healthy dietary patterns. Funding: This work was supported by the AgriFood XXI project (NORTE-01-0145-FEDER-000041) co-financed by the European Regional Development Fund through NORTE 2020 and by the project UIDB/CVT/00772/2020 funded by the Portuguese Foundation for Science and Technology (FCT). The CQ-VR is supported by FCT UIDB/00616/2020 and UIDP/00616/2020. The CITAB is supported by FCT/UIDB/04033/2020. The participation of CP-N and CS was supported by the projects UIDB/CVT/00772/2020 and LA/P/0059/2020 funded by the Portuguese Foundation for Science and Technology (FCT).

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.