Food Choice Motives among Workers during COVID-19 Pandemic in Jakarta Motif Pilihan Pangan di Kalangan Pekerja Selama Pandemi COVID-19 di Jakarta

ABSTRAK Latar Belakang: Pandemi COVID-19 terjadi secara luas dan mempengaruhi banyak negara di dunia. Situasi ini dapat mempengaruhi motif pemilihan makanan akibat kondisi stres, durasi karantina yang panjang, keterbatasan kontak personal, kekhawatiran terhadap infeksi, dan perubahan finansial. Motif pemilihan makanan masih menjadi salah satu hal penting yang dapat mendasari intervensi pola makan secara efisien dalam upaya mempertahankan kesehatan, terutama untuk meningkatkan produktivitas pekerja. Tujuan: Untuk mengetahui motif pemilihan makanan pada pekerja kantoran di Jakarta selama pandemi COVID-19. Metode: Penelitian ini dilakukan dengan desain cross-sectional pada 290 pekerja berusia 25-54 tahun di Jakarta. Subjek direkrut menggunakan metode consecutive sampling. Motif pemilihan makanan dilaporkan secara online dan mandiri oleh subjek melalui kuesioner Food Choice Value yang tervalidasi selama Januari-Februari 2021. Analisa hubungan motif pemilihan makanan dan karakteristik sosiodemografik dilakukan menggunakan uji Spearman, Mann-Whitney, dan Kruskal Wallis untuk mendapatkan gambaran utuh. Hasil: Keamanan makanan menjadi motif terpenting dalam situasi pandemi dan berhubungan dengan kekhawatiran terhadap COVID-19 (p<0,001, r: 0,232). Pekerja dengan usia lebih tua cenderung lebih memperhatikan aspek organik (p<0,001, r: 0,254), kontrol berat badan/kesehatan (p<0,05, r: 0,124), dan tradisi (p<0,01, r: 0,161) dalam pemilihan makanan, sedangkan akses Food Choice Motives among Workers during COVID-19 Pandemic in Jakarta


INTRODUCTION
The ongoing COVID-19 global pandemic is now affecting more than 20 million people and causing eight hundred thousand deaths in 216 countries, including Indonesia. 1 Indonesian Task Force for COVID -19 countermeasures reported that current cases in Indonesia are reaching more than 150,000 life with 6,594 deaths, which positioned it as the highest mortality rate country in Asia. 2 Jakarta, as the Indonesian epicenter of COVID-19 cases, now remains one of the high-risk regions in Indonesia, with a positivity rate of 7.4%. 2,3 In response to the emerging situation, the governor of Jakarta has proposed the implementation of large-scale social restriction (PSBB; Pembatasan Sosial Berskala Besar) to control the disease. 4 One of the policies included in the large-scale social restriction (PSBB) regulation is to reduce crowd-engaging activities that eventually prevent further spread of COVID-19. 5 Office is one of the places that need to adapt to the regulation. Owners are encouraged to implement a Work from Home (WFH) scheme for their employees, with only 50% of essential workers being allowed to work from the office with shift arrangements. Data by the Jakarta Department of Manpower shows that 4,074 offices have been implementing WFH regulation to more than 1 million workers by June 2020. 6,7 Consumers in a crisis might have different motivations affecting food choice behavior due to stressful events, prolonged duration of quarantine, lack of in-person contact, fears of infection, and financial loss. 8 Increased unstructured time due to social isolation during WFH might also contribute to altering the psychological condition, eating behavior, and physical activity that lead to weight gain. [9][10][11] International Labor Organization (ILO) reported that people who work remotely on a regular, mobile, or occasional basis tend to experience higher stress compared to those who work in the office. 11 Coping with stress will provoke the desire to eat food-related to mood improvement, familiarity, and convenience which drive into unhealthy food choices that eventually contribute to the development of chronic noncommunicable disease in the future. 12,13 Moreover, consumption of palatable energy-dense food which is high in sugar, fat, and calories as comfort food tends to be higher in people who experienced stress. 14 Food choice motives remain important as the basic information to form efficient and adaptable intervention in modifying dietary habits to maintain health and well-being, especially among workers who make up nearly half the global population to improve their productivity. However, little is known about the food choice motives during the COVID-19 pandemic in Indonesia, especially among workers in Jakarta. Therefore, to fill the knowledge gap, this research means to assess the food choice motives among workers during the COVID-19 pandemic in Jakarta.

METHODS
This was an observational study with a crosssectional design. The study was conducted as an online survey using the LimeSurvey platform to obtain quantitative data on food choice motives as well as sociodemographic characteristics in the workers population aged 25-54 years old in Jakarta during the COVID-19 pandemic. 290 respondents resulted from the sample size calculation for the survey were enlisted using a consecutive sampling method. The link of the survey was distributed from January-February 2021 through several channels, such as professional networks, associations, social media, as well as personal contacts to recruit subjects until reaching the minimum sample size. The professional and association network was reached through the WhatsApp group. Social media used were Instagram and Twitter, including personal accounts, two Instagram accounts, and five Twitter accounts of public influencers related to work and human resources to reach more eligible subjects. The survey link was re-announced daily to increase participation. Subjects who already received the survey link were encouraged to forward it to their contacts whenever possible. Subjects were selected based on several inclusion criteria which were have been working in Jakarta for at least one year before the study, can access the online questionnaire, and being willing to fill the informed consent. Those who were in the middle of a special diet, pregnant, breastfeeding, and having irregular work-hour as well as working in shift systems were excluded from the study. The analysis was only done on the completed questionnaire.
To fulfill the study objectives amidst restrictive situations, we used an online questionnaire consisting of two parts. The first part of the questionnaire covered sociodemographic information. In this part, subjects were asked to reveal their age, gender, education, income, marital status, change in income, change in working hours, frequency of work from home, and concern toward COVID-19. Income was categorized based on DKI Jakarta's minimum regional wage (MRW) of IDR 4,200,000. Respondents whose income was higher than the MRW were labeled as above MRW, meanwhile, those with income equal to or under the MRW were included in under MRW category. Change in working hours identified whether the respondents have worked less or more than regular 8 work hours during the pandemic. For the frequency of work from home, the respondents were categorized as mainly, regularly, and occasionally working from home. In terms of concern towards COVID-19, respondents were asked questions related to their worries about themselves and their families being infected by the virus with a maximum score of 3. Food choice motives were assessed in the second part of the questionnaire using a validated food choice value (FCV) questionnaire that might be relevant to the pandemic situation. FCV questionnaire's validation had been done in the US to several races, including White, African American, Latino, Asian, Biracial, and Native American. This questionnaire is an updated version of the Food Choice Questionnaire (FCQ) with additional motives and more neutral items. The number of questions in each motive was varied, thus the score was summed and obtained the average. The highest score among motives is considered the most important motives for food choices and dietary habits. 15 The questionnaire was going through translation and back-translation procedures to ensure questionnaire quality before being used. Both processes were done by different institutions, as well as authorized and sworn translators. As the questionnaire was reviewed by the experts to ensure content validity, the pretesting was done on 30 respondents who have similar characteristics to the subjects. Pearson correlation coefficient was used to measure items validity. The Pearson correlation coefficient for COVID-19 concern was ranged between 0.4-0.7 which exceed the reference value of 0.3, while the FCV had a value of 0.4-0.9, indicating good validity of the items. Moreover, Cronbach Alpha was also measured to know the questionnaire's reliability. The value of each questionnaire's part was 0.769 and 0.889 for COVID-19 concern and FCV, respectively.
Data quality was taken into consideration from the study planning until data analysis. To ensure data quality, questionnaire arrangement and completeness were double-checked to avoid the missing question. Additionally, to prevent duplications, control of the ID, IP, and email was done. Captcha was also included in every questionnaire to avoid filling by robots. As the response is submitted, the author would check the subject's eligibility to be processed into the analysis step. The checking was done through filter questions, occupation column, and age column which derived from the subtraction of submission date by the birth date. Subjects who didn't complete the questionnaire didn't have a regular working hour (e. g entrepreneur), as well as under 25 and above 54 years old were excluded from the study.
After the data was collected, it was then cleaned, edited, coded, and entries to the Statistical Package for Social Science (SPSS) version 20 before being analyzed. Univariate analysis was performed to describe the subject's distribution based on sociodemographic characteristics as well as food choice motives during the COVID-19 pandemic in Jakarta. To know the relationship between independent and dependent variables, bivariate analysis, including Spearman, Mann-Whitney, and Kruskal Wallis test, was performed based on the data category and normality test. Ethical clearance for this study, with the issued number of KET-1455/UN2.F1/ETIK/PPM.00.02/2020, was published by the Ethical Committee of the Faculty of Medicine, Universitas Indonesia -Dr. Cipto Mangunkusumo Hospital before survey conducted. Table 1 shows that the study was dominated by a female (71.4%). Almost all of the respondents were graduated from college/university, which classified as high educational level (94.1%). The rest of the respondents were having junior or high school degrees and were classified as middle educational level (5.9%). Over half of the respondents were married (61.4%). Most respondents were working as private employees (63.4%), while others were civil servants, contract workers, and army/police, respectively. From the study, it was known that the respondents whose income was above the MRW were higher (89.7%) than those who had income under the MRW (10.3%). Based on research done by APJII (Asosiasi Penyelenggara Jasa Internet Indonesia or Indonesia Internet Service Provider Association) in 2017, data shows that the majority of Indonesian internet penetration was in middle to higher education level, including postgraduate degree, also middle and high social-economic level. 16 Thus, the characteristics of the respondents were dominated by those groups.

Data in
Despite the pandemic situation, the income was tended to be constant (66.9%), although around onethird of the respondents experienced income reduction (26.9%). One-third of the respondents had longer working hours (28.3%), while a quarter experienced working in a shorter time during the pandemic (25.9%). The survey by BPS (Badan Pusat Statistik) showed that the full-time worker percentage declined from 71.04% to 63.85% during the pandemic due to the government regulation and company crisis, means that workers who work under a normal working hour were 36.15%, where 25.9% of the total having no side-job. It was being said that reduced working hours might be related to income decrement in the COVID-19 pandemic situation. 17 On the contrary, McKinsey data stated that working hours were increased by 40% or three more hours compared to usual due to the pandemic situation in America. Work from home (WFH) was indicated often caused this condition where workers had to do the job all the time when being in the scheme. 18 Most of the respondents did the WFH scheme due to COVID-19 (88.3%), where 42.8% of them were regularly having WFH scheme for ≥ 1 day/week, but not every day, and 33.1% were worked mainly from home. The previous report shows that more than one million workers in around 4,074 offices have already been working from home by June 2020 in Jakarta. 6,7 Unfortunately, the reliable data on the number of offices implemented full or partial work from the home scheme was not available yet. In terms of concern toward COVID-19, the respondents were having a median score of 3, which means that most of them worried about getting COVID-19 infection for themselves, including from outdoor activities such as grocery shopping or work, as well as for their family. This condition was also observed by Hamouche who stated that people were feeling unsafe and tend to give more concern on their health and the Based on the food choice motives score shown in Figure 1, the most frequent motives chose by the respondents was safety with the median score of 4.3. The median score remained equally high across all sociodemographic aspects, thus there was no significant difference observed (Table 2). It could be explained by the positive significant relationship of the COVID-19 concern with safety motives in food consumption. Uncertainty of the viral outspread has intensified the fear of being infected by the virus that made consumers increasingly put the concern on food safety by knowing where the food they buy come from as well as ensuring it was not containing harmful material. 20 Sensory appeal was the second highest concern of the respondents towards food consumption. It may refer to food sensory properties such as flavor, texture, odor, and appearance. These properties might fulfill the needs of pleasure, enjoyable experience, as well as comfort that perhaps be helpful in the uncertain situation. Sensory appeal has proven to be one of the most significant determinants of fruit and vegetables as well as convenience food consumption in many studies 21 despite the fact that no significant associations and score differences were found within the sociodemographic characteristics in this study, which means this motive remains important in all sociodemographic characteristics.  Access and comfort were having the same median score among all motives. Concern on access involved price, distance, and easiness to find the food. Access became important in current pandemic situation regarding to the viral spread, economic impairment, and the implementation of large-scale social restriction (PSBB; Pembatasan Sosial Berskala Besar) by the government that resulting in movement limitation. 4 Based on the analysis, access concern was having significant negative correlation with age. Younger worker was more likely to have higher concern on access compare to their older counterparts. In terms of price as one of the components of access, young people reported to earn less than older people on average, thus it contributed to the financial stability and concern on the spending habit. It strengthened by the findings that in Denmark, spending reduction was more common among young people than the elderly. 22,23 Besides, young workers were tended to have longer working hour that may limit their time to purchase food. 24 In line with the result, the higher score of access concern was also reflected in the longer working hour group in this study regardless of the insignificancy. Additionally, although the relationship was not significant, higher score of comforts was found in the longer working hour group as well. It might relate to the negative emotions emerged due to work changing during the pandemic so that there was increment in the search of comfort from food consumption to improve mood. 20 Other sociodemographic characteristics were observed to have either no significant association nor score differences of comfort motives.
Organic concern appeared as the last fourth food choice motives in this study. It included the attention to natural ingredients, vitamins and minerals contents, and the impact on the environment. This motive was found to be positively correlated with the age, marital status, and job type of the respondents. Older respondents as well as those who were working as army/police tend to have higher concern on organic motives. A study from China stated that people aged 40 and above attached more importance on organic food than other younger groups. Moreover, along with the higher organic concern on the married respondents, the research mentioned that couple, especially with children in the family became the main consumer of organic food for the sake of their health. 25 The next motive appointed by the respondents was convenience. This motive considered to be related with the ease of food preparation, including the time and skill needed. Greater consumption of food related to the convenience in the COVID-19 pandemic era were observed in the study in Qatar. 20 In relation with the sociodemographic determinants, significant relationship and higher score of convenience motive was found in female. Traditionally, food preparation on the family had been the responsibility of woman. As more women involved in the workforce, little time is available to do the cooking activities. Thus, this might influence the way women prefer to prepare food more conveniently. 26 Based on the data, convenience was then followed by weight control and health motive. This motive emphasizes the concern on the role of food to help controlling or maintaining the body weight, as well as the amount of the calories contained. It was observed to have positive significant relationship with army/police job type and age, where older respondents tend to have more concern on this motive. People with older age are known to be more health conscious and choosing food related to their well-being. 27 The least motive emerged in this study was tradition. It reflected the concern on traditional and cultural aspects of the food, as well as its familiarity. The motive on tradition was having significant positive relationship with age, having army/police job type, and income. Older respondents had higher concern on the tradition motive. In line with the result, research by Xie stated that older generation have more persistent view on food that contribute to their dietary habit. 25 Furthermore, higher score on tradition motive was also found in the respondents whose income under or equal to the minimum regional wage. Those in the lower socioeconomic group were more likely to choose food based on the familiarity. People with familiarity motive tend to have unwillingness to try new foods that lead to lack of dietary variety and quality. 28 Since this study was conducted in online basis, the result can not be generalized to the whole population as it can only be accessible to those with internet access. Thus, it might be beneficial for future study to assess people in different settings to get bigger picture. In addition to the limitation, strength of this study lied in the quality assurance of the data. Several steps were done to make sure that the data is reliable. Moreover, since it was the first study that timely assess the food choice motives among workers during COVID-19 pandemic in Jakarta, it may added some new findings to the literature.

CONCLUSION
The most important food choice motive reported by the respondents was safety concern. It then followed by sensory appeal, access, comfort, organic, convenience, weight/health, and tradition as the least concern. COVID-19 concern was observed to be related with the safety motive. Sensory appeal was tend to be perceived by all sociodemographic characteristics group. Meanwhile, age was significantly associated with access, organic, weight/health, and tradition motives. Single respondents were having more concern on the access, while married ones were paying attention to the organic motive. Female found that convenience motive for food was important. Lastly, people whose income under or equal to the minimum regional wage was having higher score on tradition motive. Based on the findings, it might be beneficial for food and beverage business owner as they pay more concern into the safety insurance of their product in order to get more consumer. Additionally, development and promotion of organic healthy food that is tasteful, accessible, and convenience should be added into consideration.