key: cord-0993922-uka0spkp authors: Bauza, V.; Sclar, G. D.; Bisoyi, A.; Owens, A.; Ghugey, A.; Clasen, T. title: Experience of the COVID-19 pandemic in rural Odisha, India: knowledge, preventative actions, and impacts on daily life date: 2020-11-24 journal: nan DOI: 10.1101/2020.11.20.20235630 sha: 93713604e5c18fc031e8f383b3888ba7604e9fff doc_id: 993922 cord_uid: uka0spkp We conducted 131 semi-structured phone interviews with householders in rural Odisha, India to explore participants' COVID-19 related knowledge, perceptions, and preventative actions, as well as how the pandemic was impacting their daily life, economic and food security, and the village-level response. Interviews were conducted with 73 heads of household, 37 primary caregivers, and 21 members of village water and sanitation committees from 43 rural villages in Ganjam and Gajapati districts in Odisha state. The study took place between May-July 2020 throughout various lockdown restrictions and at a time when many migrant workers were returning to their villages. Most respondents could name at least one correct symptom of COVID-19 (75%), but there was lower knowledge about causes of the disease and high-risk groups, and overall COVID-19 knowledge was lowest among caregivers. Respondents reported high compliance with important preventative measures, including staying home as much as possible (94%), social distancing (91%), washing hands frequently (96%), and wearing a facial mask (95%). Additionally, many respondents reported job loss (31%), financial challenges (93%), challenges related to staying home whether as a preventative measure or due to lockdowns (57%), changes in types and/or amount of food consumed (61%), and adverse emotional effects as a result of the pandemic and lockdown. We also provide detailed summaries of qualitative responses to allow for deeper insights into the lived experience of villagers during this pandemic. Although the research revealed high compliance with preventative measures, the pandemic and associated lockdowns also led to many challenges and hardships faced in daily life particularly around job loss, economic security, food security, and emotional wellbeing. The results underscore the vulnerability of marginalized populations to the pandemic and the need for measures that increase resilience to large-scale shocks. The World Health Organization (WHO) declared COVID-19, the disease caused by the novel 46 coronavirus virus SARS-CoV-2, as a global pandemic on March 11, 2020 (1). The pandemic 47 quickly resulted in lockdowns around the world. India implemented its own nationwide 48 lockdown on March 24, 2020, limiting the movement of 1.3 billion people (2). The national 49 lockdown in India temporarily shut down portions of the economy and substantially altered daily 50 life, generating fear about economic and food security among the many living in poverty and 51 resulting in a mass exodus of millions of migrants workers from cities to their rural villages. 52 Although initially there were limited services for migrant workers, the government began 53 organizing transportation for returning migrants in May 2020 and quarantine centers were set up 54 in many villages or other centralized locations to temporarily house migrants upon returning 55 from other regions. Additionally, while the strict nationwide lockdown eased on May 1, 2020, it 56 was replaced by a series of lockdowns of varying levels of restriction imposed throughout India 57 by national, state, and district governments over the next several months (3). Government and 58 health authorities in India also undertook massive awareness campaigns to educate the public on 59 COVID-19, emphasizing the need for frequent handwashing, social distancing, staying home as 60 much as possible, avoiding touching one's face, and good respiratory hygiene, including wearing 61 a face mask (4,5). 62 The objectives of this research were to understand the COVID-19 related knowledge, 64 perceptions, and preventative actions of rural householders and communities, and how the 65 pandemic was impacting their daily life, economic security, and food security. We conducted 66 semi-structured phone interviews to capture participants experiences. The interviews took place 67 throughout various lockdown restrictions in rural Odisha, India at a time when many migrant 68 workers were returning to their villages. Our aim is to contribute to an understanding of 69 participants' knowledge, actions, and lived experience during this time in order to determine if 70 the government response, in the form of lockdowns, restrictions, and awareness campaigns, were 71 effective at changing behavior and whether they had unintended negative consequences. These 72 findings can be used to help develop and target interventions that minimize adverse effects and 73 improve overall resilience to external shocks, such as a pandemic, increasingly faced by 74 marginalized populations. 75 76 Study site and sampling frame 78 We conducted semi-structured phone interviews with households in Ganjam and Gajapati 79 districts of rural Odisha, India from May-July 2020. In order to capture a range of perspectives 80 and experiences, the phone interviews were conducted with three types of respondents from each 81 district: head of households (HOH), primary caregivers of young children (<5 years old), and 82 members of Village Water and Sanitation Committees (VWSC). We specifically targeted HOH 83 to understand household-level impacts and decision-making, caregivers to understand childcare 84 practices, and VWSC members to understand village-level response and collective action. In 85 some cases the HOH interview was completed by another household member requested by the 86 HOH (such as the HOH's wife or son), and in a few cases the caregiver interview was completed 87 by her husband when the caregiver was not fluent in Odia. Details of specific village and 88 respondent selection are provided in the Supplemental Information (SI). Additionally, each of the 89 two districts has distinct geographic and demographic characteristics to capture a wider range of 90 experiences. Ganjam is a coastal district with relatively large villages that are typically closer to 91 cities or markets and have predominately Hindu populations. Comparatively, Gajapati is a hilly 92 and mountainous district with more remote, smaller villages that have predominantly Christian 93 populations belonging to Scheduled Tribes. 94 95 All included villages had previously participated in a village-level WASH intervention which 96 included the construction of a community piped water system with household connections, as 97 well as village-level mobilization for the construction of latrines with attached bathing rooms at 98 each household (6) . The villages are also a subset of villages that are currently enrolled in a 99 randomized-controlled trial evaluating a child feces management (CFM) intervention 100 (ISRCTN15831099), however intervention delivery had not yet started at the time of this study. 101 or block level, which changed over time to ease or tighten restrictions based on local case counts. 110 A timeline of these activities in relation to the data collection period is provided in Figure 1 . A semi-structured interview tool was developed for each respondent type with a mix of 123 structured and open-ended questions. All respondents were asked questions about their COVID-124 19 related knowledge, risk perception, and practices, including any preventative actions taken. 125 All respondents were also asked about how the pandemic or lockdowns had impacted their daily 126 life and social interactions. The HOH and VWSC interview tools also included questions about 127 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint the impact on finances, food security, and cooking fuel, as well as additional knowledge and 128 actions questions, including challenges related to following recommended preventative actions. 129 The VWSC tool included a specific section on village dynamics and response to the pandemic. 130 Lastly, the caregiver interview tool included a specific section on the impact of the pandemic on 131 childcare practices. The interview tools are provided in the SI. Several of the quantitative 132 questions were modified from the Population Council (7,8) and Y-RISE (9) surveys. 133 134 A team of three research assistants, all originally from Ganjam district and fluent in the local 136 language Odia, were trained on how to conduct the phone interviews. The team underwent a 137 multi-day remote training in May 2020, followed by a pilot to revise and finalize the interview 138 tool. Target respondents were randomly ordered using a computer-generated sequence, and 139 research assistants were provided call lists to contact respondents in random order. Details of the 140 specific calling procedure are provided in the SI. questions, as well as "please explain" follow-ups to some of the structured questions. For the 147 modified thematic analysis, a research team member read through all responses for a given 148 question and wrote memo notes with reflections, noting common and interesting answers. Then 149 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint the researcher read through all responses again and categorized responses into initial themes. A 150 third read through was done, as needed, to finalize the themes. 151 152 During quality checks after data collection was complete, it was found that one of the research 153 assistants had sometimes skipped select interview questions or fabricated data for some of their 154 interviews. To ensure the accuracy of the data collected, a study supervisor listened to every 155 audio recording of this research assistant's interviews and corrected the data accordingly prior to 156 analysis. As a result of these skipped questions in some interviews, our sample sizes vary slightly 157 among different questions. A total of 131 participants were interviewed: 73 HOH, 21 VWSC members, and 37 caregivers. 169 The respondents represented 43 villages (26 in Ganjam and 17 in Gajapati) and two-thirds of all 170 respondents were from Ganjam district. Approximately 8% of respondents (5 HOH, 1 VWSC, 5 171 caregivers) ended the interview early. Overall, the majority of respondents were male, had at 172 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint least a primary education, were self-employed, and their household had access to piped water 173 (Table 1) HOH and VWSC respondents were also asked questions about their household members to 182 assess whether any had characteristics that could put them at higher risk for developing severe 183 illness from COVID-19. Among these respondents, 48% (N=45) had an elderly household 184 member over 60 years old, 17% (N=16) had a household member with a pre-existing condition, 185 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. mentioned was the elderly (45%, N=58), followed by people with weak immune systems (12%, 202 N=15), children or babies (14%, N=18) and 'everyone' (6%, N=8). However, almost a third of 203 respondents (30%, N=39) stated they did not know of any high-risk groups, with caregivers 204 being the most likely to report they did not know any (51%, N=19). 205 206 Causes. While many respondents said they did not know what causes someone to become sick 207 with COVID-19 (44%, N=58), a similar percentage (44%, N=58) correctly explained that the 208 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint virus was transmitted from person-to-person contact with a sick person, and/or described 209 behaviors that put a person at risk for infection. Among the few respondents that gave an 210 inaccurate explanation of what causes COVID-19 (12%, N=15), most described perceived causes 211 of illness more broadly, such as eating cold food, not drinking hot water, or coming from/being 212 in a cold area. One respondent mentioned avoiding foods that make the body more susceptible to 213 colds, specifying bananas, pineapple, and lemon -potentially referencing principles from 214 ayurvedic medicine. (Table S2) . No specific sources were associated with a greater likelihood of a respondent 225 reporting a correct symptom of COVID-19. However, getting their main source of information 226 from social media or the internet was associated with respondents being more likely to know a 227 high risk group (χ 2 =4.47, p=0.04), while getting their main source of information from the news 228 was associated with being more likely to know a correct cause of COVID-19 (χ 2 =9.38, p=0.002). 229 Across the most common main information sources, the majority of respondents reported 230 trusting all of the information received from their main source (73% for news, 70% for 231 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint Anganwadi/ASHA worker, 77% for social media/internet), with fewer reporting that they trusted 232 some of the information (25% for news, 30% for Anganwadi/ASHA worker, 23% for social 233 media/internet), and only one person who got their main information from the television news 234 and social media/internet reporting that they did not trust any of the information from these 235 sources. 236 237 In regard to self-perceived risk of their chances of contracting COVID-19, the majority of 239 respondents (59%, N=74) felt they had no risk of getting COVID-19, about one-quarter (27%; 240 N=35) said they did not know their risk, and only 13% (N=17) reported feeling either a low, 241 medium, or high level of risk (Table S3) . Those who felt no risk explained this was due to them 242 taking precautions, the lack of local cases, or their good health status. 243 244 When asked how concerned they would be if they or a family member contracted COVID-19, 245 the majority of respondents (77%, N=68) stated they would be 'very concerned', while less 246 answered 'somewhat concerned' (13%, N=11), 'not concerned' (5%, N=4), or 'don't know' (6%, 247 N=5) (Table S4) . For the respondents who said they would be 'very concerned', the most 248 common explanations were concern around transmitting the virus to other members of the 249 family, concerns around lethality or severity of the virus, and concern simply because a member 250 of their family contracted the virus. There was no difference in self-perceived risk of their 251 chances of contracting COVID-19 by main source of information, but those who reported getting 252 their main source of information from social media or the internet were less likely to report being 253 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint very concerned if they or a family member contracted COVID-19 (χ 2 =5.73, p=0.02). 254 Additionally, there was no difference in risk perception or level of concern by district. 255 256 Preventative measures. Respondents were read a list of specific actions and asked if they or their 258 family members had practiced that action in the past 7 days to protect themselves from COVID-259 19 ( Table 2 ). Almost all respondents reported practicing key preventative actions: 96% washed 260 hands more often, 95% wore a mask, 94% stayed at home as much as possible, and 91% kept 261 distance from others. There was no difference in proportion of respondents that reported 262 practicing all four of these key preventative measures by district or main source of information. 263 Respondents also commonly reported avoiding public transit and traveling (81%) and sanitizer to clean our hands. When we come home from outside, we must not take the 274 clothes inside. We should keep it outside and change our clothes." 275 -Male respondent, 35 years old, Gajapati district (June 2020) 276 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Actions if Exhibiting Symptoms. HOH and VWSC members were asked what actions they would 288 take if they were to begin experiencing symptoms of COVID-19. The majority of respondents 289 (68%, N=60) said they would go to a health clinic or hospital. Some respondents also described 290 calling a coronavirus helpline (22%, N=16), informing a government worker or elected leader, 291 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint such as an ASHA worker or Village Sarpanch (17%, N=12) , or going to get tested for COVID-292 19 (11%, N=10). 293 294 Challenges related to staying home and social distancing 295 Many respondents experienced challenges with staying home as much as possible (57%, N=50) 296 and some experienced challenges with social distancing (28%, N=25; Table 3 ). Many challenges 297 with staying home were related to the lockdown or curfews, including challenges meeting basic 298 needs, problems accessing healthcare, negative impacts on emotional well-being, and difficulty 299 traveling due to police enforcement of lockdown restrictions on travel. Challenges reported 300 related to social distancing included that not everyone was following the guidelines and it was 301 not always possible to maintain distance (see Table 3 for descriptive themes). 302 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint Table 3 . Reported challenges related to following COVID-19 precautions of staying home as much as possible and social distancing. 303 Quantitative Results Qualitative Themes and Descriptions What are some of the challenges you and your family face in being able to stay at home at all times except when you are performing essential activities such as working or buying food? 57% (N=50) described facing challenges, often due to the lockdown or curfews  Challenge meeting basic needs: The most common challenge respondents faced was being able to meet their basic needs during lockdown or when staying home. Many explained it is a challenge to stay home when one needs to work to have money, repay loans, or feed their family. Others expressed problems getting essential items due to lockdown, curfews, or closed shops, such as food or seeds and fertilizer for their fields. One respondent explained that he needed money but was unable to withdraw it from the bank during the lockdown. Multiple respondents had difficulties selling their fruits/vegetables during the lockdown. Yet another respondent had difficulty getting support for his special needs child because he could not go to the administrative block office or get required paperwork.  Problems accessing healthcare: Multiple respondents explained they would not access the health clinic or hospital if they needed to because they are afraid of catching the virus. One person said that someone had been seriously ill a couple days before our interview, and they had a problem taking the person to the hospital. Another respondent said that he could not get planned treatment for a health condition due to COVID-19.  Negative emotional response: Many respondents expressed emotional challenges with staying home, saying it makes them feel annoyed, irritated, bored, gloomy, or lazy. Others felt it was difficult to stay home because humans are social, and they wanted to meet their friends. A couple of respondents also explained that one needs to move their body to stay healthy and it does not feel good to stay home for a long time.  Difficulty travelling due to police: A few respondents explained that police were not allowing people to travel or go out, even for essential goods. These respondents said a person might get beaten up by the police. One respondent described how the police took pictures of his vehicle when he went to buy medicine and he was worried about getting a fine. What are some of the challenges you and your family face in being able to maintain 2 meters of distance between other non-family members when you are outside of your house? 28% (N=25) reported challenges with social distancing  Not everyone is following guidance: Many respondents explained that not all people are listening to this guidance, that some people will come closer to you even if you try to stay away or ask them to maintain distance.  Not always possible to maintain distance: Many respondents explained that it is simply not always possible to maintain distance from others. Some respondents said it was difficult to maintain distance from their friends. Other respondents explained that their job did not allow them to maintain distance (e.g., auto rickshaw driver, farmers who go to the fields in a group, a man who works with another person). Some respondents said that it is not possible to maintain distance in their village at all. Impacts on economic and food security 305 The pandemic had major impacts on households' economic and food security (Table 4 ). Many 306 respondents (31%, N=28) reported they or a family member had lost a job as a result of the 307 pandemic, mainly the loss of work as a daily laborer. Almost all households (93%, N=84) had to 308 take action in the past week to cover their basic needs, with the majority relying on government 309 assistance (74%, N=67) or using savings (69%, N=62). The majority of respondents also 310 reported a change in the type of foods (59%, N=52) and/or a reduction in the quantity of food 311 that their family consumes daily (25%, N=22). Many respondents reported that their family was 312 consuming less meat or vegetables since the pandemic began, often due to reduced income, 313 rising food prices, or lack of availability in markets. There was a range in experiences across 314 respondents in how the pandemic affected their food security. Some respondents reported 315 relatively minor changes in food consumption: 316 317 "I used to go out and eat street food, which is not possible now. And there are some items 318 that we used to have in home, which are not available anymore. Mostly, the quantity has 319 also reduced." 320 -Male respondent, 30 years old, Ganjam district (June 2020) 321 322 However, others described more extreme reductions in food variety. For example, one 323 respondent described how they are now only surviving on rice and water and two respondents 324 similarly described how they used to eat a variety of foods before the pandemic, but since their 325 income has dropped, they are now only surviving on rice, salt, and green chilies: 326 327 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint "There is some change in the eating practices. We received rice [ration from the 328 government] which is fine, but the practice of eating has changed because there is no 329 curry. Now, we have to eat green chili and salt. The government is helping but if we get 330 some work, we can live a little more peacefully. We are eating green chili and salt in 331 place of having a curry." 332 -Male respondent, 30 years old, Gajapati district (June 2020) 333 Multiple respondents also explained that they had stopped eating meat due to a belief that 335 COVID-19 could be spread through meat consumption: With regard to cooking fuel access, only three respondents reported difficulties in getting wood 343 or gas as a result of the pandemic (3%, N=3). Among the 3 respondents who had difficulties, 344 reasons included lack of finances, inability to travel during the pandemic, and not being given 345 gas assistance (free gas or gas subsidy). 346 347 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint Respondents were asked how their daily life had changed as a result of the pandemic and 352 lockdowns, and many expressed that daily life had become more difficult (Table S5 ). One of the 353 biggest changes was the experience of job loss or having less work due to the lockdown curfews 354 and travel restrictions. Consequently, many also talked about the financial strain they or their 355 household were facing. Another major change described was that everyone was staying at home 356 more and 'not going anywhere.' Some respondents elaborated that this meant they or their family 357 members no longer go to work or school, shop at the market less, do not visit family or friends, 358 do not go to temple, or no longer take part in their normal leisure activities. Several 359 HOH/VWSCs also reported less or restricted travel, such as not visiting other villages or nearby 360 towns, and less 'roaming around.' Respondents also described how they were interacting with 361 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint others less now and maintaining distance or taking precautions while interacting, which was 362 confirmed at a village-wide level by most VWSC members. "Earlier[…]we did not have any tension or fear, because any male member from the 382 family could go out and work as laborers and earn money to ensure that the family is 383 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint managed nicely and safely. Now, there is a challenge because nobody is able to work as 384 a laborer and there is no income." 385 -Female respondent, 20 years old, Gajapati district (July 2020) 386 In contrast to HOH and VWSC members who expressed that their daily life felt very different 388 now, about a third of the caregivers explained that there was not much of a change in their own 389 personal lives because they mostly stayed at home even before the lockdowns. While some 390 HOH/VWSCs said they now have more free time, many caregivers noted that their housework 391 and amount of free time remained the same: 392 393 "We stay at home. We always stay at home. How will life change then? That's it. Work is 394 also like how it always used to be. I never used to go anywhere." 395 -Female respondent, 30 years old, Ganjam district (July 2020) 396 In addition, most caregivers explained that there was no change in their small child's day-to-day 398 life either, although many caregivers said they were limiting or no longer allowing their child to 399 play outside the home and schools were now closed. 400 401 In line with responses from individual respondents, the majority of VWSC members (81%, 403 N=17) reported that most villagers took precautions to prevent coronavirus. Many (62%; N=13) 404 reported holding a village-wide meeting as part of village-level actions, primarily to discuss 405 COVID-19 prevention measures. Several VWSC members (33%, N=7) also reported taking 406 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint precautionary measures against outsiders from the village, including physically barricading the 407 village from outsiders or requiring those returning to stay in quarantine treatment centers before 408 entering the village. Only one VWSC member said their village had taken no action to protect 409 itself from coronavirus. pay for the bus fare after his family took out a loan, because he was cheated out of all of his 439 money by someone that promised to return him home and did not follow through. Although he 440 was told to go to a police station in Surat to register for a train ride home, he did not do so out of 441 fear. The other migrant worker explained that he started to walk home from Andhra Pradesh on 442 his own and was later able to pay a truck driver to take him home. The migrant workers also 443 described having a lack of money and work opportunities now that they have returned home. 444 Additionally, one migrant worker described how depressing it was for him to face discrimination 445 from his fellow villagers, who were afraid of talking to him now because he returned from a 446 quarantine center. 447 448 The COVID-19 pandemic and related lockdowns had a substantial impact on villager's daily life, 450 social interactions, and economic and food security in rural Odisha, India. In particular, we found 451 many individuals reported job loss, challenges related to lockdowns, changes and reductions in 452 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint food consumed, and negative impacts on emotional well-being. In many instances, it was not about COVID-19 was notably lower among caregivers who may be more likely to care for sick 469 or elderly household members. This suggests information campaigns targeted at these less-470 knowledgeable demographics could be beneficial, particularly door-to-door information 471 campaigns to reach women and young caregivers who experience gendered restrictions on 472 freedom of movement and interactions outside of the home. Those getting their main source of 473 information from social media/internet and the news were also more likely to know a high-risk 474 group and the disease cause, respectively, suggesting that greater education on these aspects 475 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Many respondents reported struggles related to COVID-19, particularly related to economic and 488 food security and mental health. Almost all respondents reported relying on some form of 489 support to cover basic needs and almost a third reported job loss due to COVID-19. Although 490 loss of jobs for daily wage workers was common in both areas, reported job loss in our study 491 area was lower than a study in Bihar, India, where job loss effected almost two thirds of 492 households (12). This may be because many respondents in our study worked in agriculture and 493 were largely able to continue farming during the pandemic despite some struggles in selling their 494 fruits and vegetables. Although many respondents reported receiving food rations from the 495 government, reductions in dietary diversity and food quantity were commonly reported, which 496 could lead to negative impacts on nutritional status and health (13,14) . Additionally, migrant 497 workers have been identified as a particularly vulnerable group in India during the pandemic 498 (15). The experience of the few returned migrant workers we interviewed illustrated this 499 vulnerability, including their extreme difficult returning home after job loss and facing 500 discrimination. Mental health is also a substantial concern related to COVID-19, particularly as a 501 result of job loss, financial strain, lockdowns, and fear of the illness (16-18). Many respondents 502 reported declines in mental health as a result of the pandemic and feeling increased anxiety, fear, 503 boredom, and irritation. Future pandemic response by governments and NGOs should consider 504 these factors to prevent greater impacts on mental health. 505 This study had some important limitations. As it relied on phone interviews, we could only reach 507 households with a mobile phone and network connection, which could potentially exclude the 508 poorest and most remote households. However, we were able to talk to some respondents who 509 lived in villages without a mobile network by connecting with them when they were outside the 510 village in an area with network. Responses were also self-reported which may result in reporting 511 bias, as respondents may over report practicing desirable hygienic behaviors, like handwashing 512 (19-21). To try to reduce this bias and capture detailed experiences, the interview included 513 several open-ended questions and asked follow-up explanations to closed-ended questions. 514 Further findings related to handwashing and other WASH-related practices will also be reported 515 in detail in a forthcoming paper. Additionally, although quantitative surveys have been used 516 extensively to capture knowledge and impacts of the COVID-19 pandemic for large population 517 samples, our use of open-ended, qualitative questions allowed us to explore the lived experiences 518 of participants during this public health crisis more deeply than what can be captured by 519 structured questions alone. 520 521 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint Overall, the study results help identify gaps in local capacity building for information providers, 522 content of the awareness messaging, and channel of communication for message delivery to 523 ensure accessibility to all. In rural Odisha, ASHA workers, Anganwadi workers, and auxiliary 524 nurse midwifes (ANMs) typically serve as information providers on health matters. However, 525 they are trained in-person periodically as per the demand of situations, which was not possible in 526 this pandemic, and may have contributed to less reliance on these typical change agents as shown 527 by our finding that less than half as many respondents reported getting their main source of 528 information on the virus from Anganwadi or ASHA workers compared to the news. 529 Additionally, the common message delivery mechanism of person-to-person communication was 530 less feasible as people were more isolated during the pandemic, which may have resulted in a 531 change in information channel to other sources for COVID-19 knowledge. There was also a gap 532 in messaging content, which suggests that messaging needs to be designed keeping in 533 perspective the broader aspects of health impacts such as physical health, mental health, and 534 nutrition needs. Additionally, message delivery channels need to be made accessible to all, 535 including women with limited access to the external world. As such, new delivery channels need 536 to be explored for creating awareness other than the village meetings or in-person consultation 537 and digital literacy may be helpful in this context. 538 539 Additionally, although the lockdowns and information campaigns appeared to be effective at 540 changing villagers' behavior to increase preventative measures taken, the lockdowns also had 541 many adverse effects that should be mitigated during future responses. The evidence base created 542 through understanding lived experiences during this pandemic can also help improve planning 543 and preparedness at village and Gram panchayat (GP) administrative levels. The struggles that 544 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint respondents reported related to difficulties meeting basic needs, accessing healthcare, 545 experiencing negative emotions from staying home more, and police in some areas not allowing 546 people to travel even for essential goods, demonstrate that administrative efficiency in dealing 547 with the deterrent actions against the spread of virus could be improved. In particular, there is a 548 need for greater government response to limit harm to livelihood and mental, social, and 549 nutritional health during any lockdown periods required to reduce disease spread. Although there 550 were already some government initiatives in place targeting economic and food security, our 551 results show that there were still needs remaining for many rural villagers despite these programs 552 and further initiatives may be needed. Overall, we identified high compliance with COVID-19 553 preventive actions despite shortcomings with messaging, many challenges and hardships faced in 554 daily life due to the pandemic and lockdowns, all of which are insights that could be used to 555 improve messaging and preparedness for future crisis response. Formal Analysis: VB, GS, AO 567 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.20.20235630 doi: medRxiv preprint Acquisition: VB, GS, TC 568 Investigation: AB 569 Methodology: VB Writing -Original Draft Preparation: VB Timeline: WHO's COVID-19 response [Internet]. World Health Organization Coronavirus India timeline: Tracking crucial moments of Covid-19 581 pandemic in the country. The Indian Express Government of India. Novel Coronavirus Government of Odisha. CORONAVIRUS Movement and Action Network for Transformation in Rural Areas 589 (MANTRA) COVID-19-related 592 knowledge, attitudes, and practices in urban slums in Low perceived risk: a challenge to adoption of preventive 600 behaviors for COVID-19? Population Council Awareness, risk perception, and stress during the 602 COVID-19 pandemic in communities of Tamil Nadu, India. International journal of 603 environmental research and public health Population Council Institute and UNICEF. COVID-19 research brief #8: Effects of 605 COVID-19 on livelihood, employment, and financial resources Dietary diversity is associated with child nutritional status: evidence 607 from 11 demographic and health surveys Low dietary 610 diversity is a predictor of child stunting in rural Bangladesh. European journal of clinical 611 nutrition COVID 19 pandemic: Mental health challenges of internal migrant workers 613 of India Mental health and the Covid-19 pandemic Rajkumar RP. COVID-19 and mental health: A review of the existing literature. Asian 617 journal of psychiatry Pandemic fear" and COVID-19: mental 619 health burden and strategies 621 Measuring hygiene practices: a comparison of questionnaires with direct observations in 622 rural Zaire. Tropical medicine & international health Observed hand 624 cleanliness and other measures of handwashing behavior in rural Bangladesh. BMC public 625 health Twenty-four-hour recall, knowledge-627 attitude-practice questionnaires, and direct observations of sanitary practices: a comparative 628 study