key: cord-0301817-ibv4jlfm
authors: Gerken, K. N.; Maluni, J.; Mutuku, F.; Ndenga, B.; Ichura, C.; Mwashee, L.; Mwaniki, M.; Shaita, K.; Orwa, S.; Seetah, K.; LaBeaud, A. D.
title: "We just considered it to be over there" : A qualitative study exploring urban perspectives for disease introduction
date: 2022-05-07
journal: nan
DOI: 10.1101/2022.05.05.22274700
sha: 42dbbd58e76336e4260b18dd0bca3327c6479720
doc_id: 301817
cord_uid: ibv4jlfm

Rift Valley fever virus (RVFV) is a zoonotic arbovirus that is also transmitted to humans from fluids and tissues of infected livestock. Urban areas in Kenya have the hosts, dense vector distributions, and source livestock (often from high-risk locations to meet the demand for animal protein ) , yet there has never been a documented urban outbreak of RVFV. To understand the likely risk of RVFV introduction in urban communities and guide future initiatives, we conducted Focus Group Discussions with slaughterhouse workers, slaughterhouse animal product traders, and dairy livestock owners in Kisumu City and Ukunda Town in Kenya. For added perspective and data triangulation, in-depth interviews were conducted one-on-one with meat inspector veterinarians from selected slaughterhouses. Themes on benefits of livestock in the urban setting were highlighted, including business opportunities, social status, and availability of fresh milk in the household. Urban slaughterhouses have formalized meat trading, which , in turn , has resulted in regulation for meat inspection and reduction in local livestock theft. High-risk groups have moderate knowledge about zoonotic diseases and consensus was towards lower personal risk in the urban setting compared to rural areas. Risk assessment was focused on hand hygiene rather than the slaughtering process. There was high reliance on veterinarians to confirm animal health and meat safety, yet veterinarians reported lack access to RVFV diagnostics. We also highlighted regulatory vulnerabilities relevant to RVFV transmission including corruption in meat certification outside of the slaughterhouse system, and blood collected during slaughter being used for food and medicine. These factors, when compounded by urban vector abundance and dense human and animal populations could create ideal conditions for RVFV to emerge in endemic regions and establish an urban transmission cycle. Here, we present a qualitative study that provides context for urban RVFV introduction risks and insight for adapting current prevention and control measures.

This study's main objective was to explore potential pathways of urban RVFV introduction and 115 human exposure from the perspective of those most involved in livestock. Our five specific study 116 objectives are presented in Table 1 . Question guides were created to ensure full coverage of the study 117 objectives and homogeneity between the study sites. The guides were piloted with smaller groups (3-5 118 participants) from adjacent towns to gather baseline information and inform probing techniques for the 119 moderator. 120 Explore pathways of RVFV exposure in an urban setting from the perspective of the livestock associated community 2

Understand how high-risk groups in urban areas perceive personal risk of RVFV 3

Understand the perceived difference in zoonotic disease risk compared to rural areas 4

Explore current preventive measures in urban centers for mitigating zoonotic disease introduction and spread 5

Understand current knowledge and previous experience with RVFV to predict how urban areas might respond to a RVFV outbreak 121 122 Study Areas

The two urban study sites in this study were Kisumu city located on the shores of Lake Victoria 124 in Western Kenya and Ukunda town which is an emerging urban center located on the South coast of 125 Kenya. At both urban sites, nearly all animals are slaughtered in designated slaughterhouses. 139 Participants and participant recruitment 140 We worked with key informants and community health volunteers to identify a representative 141 sample of the urban livestock community. We aimed to recruit seven to ten participants for each FGDs 142 involving slaughterhouse workers, slaughterhouse soup traders (described below), and livestock owners.

143 Participants were identified two weeks prior to being invited to participate in the FGDs and attempts 144 were made to capture an equal number of men and women of all age groups and religions.

145 Slaughterhouse workers 146 Four groups of slaughterhouse workers were recruited from the two main slaughterhouses at 147 each site, and we invited participants that carried out all slaughterhouse activities including the 148 slaughterman, skinners, and meat filleters. The slaughterhouse manager was only included if they 149 participated in the daily activities on the slaughterhouse floor. To be included in this group 150 slaughterhouse employees must have worked at the slaughterhouse for a minimum of six months.

151 Slaughterhouse Soup traders 152 We identified individuals who purchased various products daily from the slaughterhouse, 153 including the heads, hooves, offal, and blood and invited two groups total, one from each site, to 154 participate in this study. These individuals handle animal products shortly after slaughter and boil the 155 heads and hooves to soup that is sold by the roadside. The most common practice among this group is 156 making soup from these products and therefore, we collectively refer this group as the "slaughterhouse . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. ; 157 soup traders" (SSTs) throughout this manuscript. They also collect blood to either make local blood 158 sausage (mtura) or sell blood privately to farmers as livestock feed. To be included, SSTs must have 159 purchased either blood or animal products at least three days per week over the past six months.

160 Livestock owners 161 Four groups of livestock owners, two at each site, were recruited for this study. Ownership in 162 this study was defined as possessing at least one cow, sheep, or goat and we specifically aimed to 163 identify participants involved in local dairy production and business. Although livestock owners with 164 dairy animals were targeted, participants who owned multiple livestock for meat production were also 165 invited.

166 Data collection

The slaughterhouse FGDs took place in a private room or designated area at the respective 168 slaughterhouse. Livestock owners were invited to meet at a central public location that was identified by 169 key informants and deemed an acceptable travel distance from the participants' homes. At enrolment, 170 basic demographics were collected for each study participant. FGDs were carried out in Kiswahili except 171 for the SSTs group in Kisumu, which was conducted in the local language, Dhuluo, preferred by all 172 participants. All groups and interviews were audio recorded with a Sony IC audio recorder (ICD-PX470) 173 and transcribed verbatim in English within one week by the moderators and their assistants.

174 Randomized quality control spot checks were carried out for each transcript by a second transcriber and 175 any discrepancies were discussed.

The first author of this study was present at all the FGDs serving as the note taker and personally 177 led the IDIs. A moderator and assistant moderator were trained from each site and the moderator from 178 the first site (coast) was physically present for the piloting and training at the second site (West) to 179 ensure consistency in data collection. Team training focused on understanding how the study objectives 180 related to RVFV epidemiology and highlighted the importance of minimizing responder bias.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint 181 Data Analysis 182 Transcripts were read from which a codebook was developed for thematic analysis. The 183 transcripts were uploaded into NVivo (Version 1.5, 4577) and transcripts were coded by two 184 independent coders according to the codebook and organization of codes was discussed in real-time.

185 The coded transcripts were verified by a mutual third coder to ensure consistency.

All FGD data were analyzed together for a collective understanding of urban perspectives and 187 where appropriate, differences between the groups and two study sites were noted. A thematic analysis 188 was carried out to identify prominent themes in the qualitative data. IDIs provided data triangulation 189 from a different perspective and IDI analysis focused on themes identified in the FGDs. Text has been 190 summarized in the results section from detailed ethnographic accounts. 200 during consenting and were allowed to ask any questions to the group or privately. Before beginning, 201 participants were informed that the study was focused on Rift Valley Fever and their responses could be 202 used to inform policy and disease control in their area. To decrease the risk of responder bias, questions 203 specific to the disease dynamics were deferred to the end of the FGD. The assistant moderator spent a 204 minimum of ten minutes explaining the study procedures and informing participants their rights as a . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 7, 2022. ; 205 research study participant. Understanding of informed consent was assessed through a checklist at the 206 end of the consent form and each participant signed a consent form and was given a copy. All 207 participants were above Kenya's adult consenting age of 18 years and were compensated for their lost 208 time and transport costs to attend the meeting point.

Ninety-one participants from ten different FGDs (n=87) and four IDIs (n=4) were enrolled in this 211 qualitative study and participated until the end of their session. A summary of FGD participant 212 demographics is represented in Table 2 . Data saturation was achieved in the final FGDs from the 213 slaughterhouse groups and livestock owners. This was confirmed by codebook developers and no 214 additional themes were added after the final groups. FGD length varied from 46 to 94 minutes. At the 215 coast site (Ukunda) , it was challenging to identify female livestock owners despite best efforts from the 216 key informants and snowball sampling techniques. Given the gender ratio for the recruited livestock 217 owners of 13M:4F in Ukunda, it was decided to organize one group of only men and another group with 218 50% men and 50% women. All slaughterhouse workers at both sites were men; however, women were 219 involved at the slaughterhouse as SSTs. All SSTs in Kisumu were women and the years of experience in 220 this group was comparable to other slaughterhouse staff. The IDIs were with two men and two women 221 and to protect the identity of the meat inspectors, age and location has not been reported. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint 232 

The SSTs demonstrated that all animal parts can be used for income generation including the 247 heads, hooves, and blood. Both SSTs groups noted that their products provided customers with a less . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022.

248 expensive protein source. Blood that was less fresh could be sold to farmers as feed for pigs (West SSTs) 249 and poultry (all SSTs). All slaughterhouse sales were conducted as private business and the Ukunda 250 slaughterhouse groups highlighted the businessperson who purchased the live animal at market would 251 lose their money if a carcass was condemned. Fresh milk and blood were animal products reported to be linked to human health. Local fresh 267 milk was preferred, particularly for children and babies, because of concerns for chemicals in 268 commercially packed milk. Trust in fresh foreign milk that arrived from surrounding regions was lower 269 than local milk, as they believed the foreign milk had chemicals, adulterants, and water as this was the 270 only way to compete with commercial milk prices.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint 271 Among dairy owning livestock owners, it was understood that boiling milk was important for 272 health, but local fresh milk was still sold un-boiled, and it was to be decided by the consumer to boil the 273 milk or not. At least two participants agreed with this and added that some people drink milk direct 274 from the cow and they do not have any health consequences, so therefore, "the dangers are higher than 275 the truth." However, most participants in the livestock owners' group recognized milk could cause 276 disease and a commonly known consequence of raw milk consumption was diarrhea. Fresh milk was not 277 just for food but also used as medicine, and tuberculosis (TB), burns, HIV, and arthritis were specifically 278 mentioned to be treated with fresh milk.

Fresh, raw blood was also consumed as a cure for health problems in Kisumu. The 280 slaughterhouse workers shared that blood was collected in plastic jugs and sold to the SSTs who 281 confirmed to purchase these products. The meat inspector veterinarian from this slaughterhouse also 282 affirmed this exchange and recognized that because the blood was pooled from multiple animals it was 283 difficult to remove any if a contributing animal to the pool was condemned. Direct consumption of raw 284 blood from the slaughterhouse was only noted in Kisumu. We did not confirm if the participant's remark 285 regarding doctors suggesting this practice was in reference to conventional doctors or traditional 286 doctors. 287 288 "You know blood from an animal is normally warm, then they drink it, the following day they come as 289 well and might go even for a week and then he regains blood quickly that is what the doctor says. He is 290 just someone who has been directed by a doctor that go and drink raw and direct milk from a cow."

292 293 "People believe that that blood boosts their immunity. When they go to the hospital, they are told, 294 especially pregnant mothers, to take blood from the slaughter." -Vet 3 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint 295 296 Theme 3: Nutritional demands and livestock movement 297 Non-local livestock entered urban center for three main reasons: additions to urban herds, 298 arrival at the slaughterhouse to meet the high urban demand for ASFs, or as situational grazers that 299 would arrive and graze for a given time before returning to their origin. Major drivers of livestock 300 movement into urban centers were related to urban nutritional demands of humans and animals. We 301 confirmed animals and animal products must be sourced from distant locations to meet the high 302 demand from the urban population for ASFs. Meanwhile, urban livestock were reported to have limited 303 grazing areas and/or affordable zero-grazing feed options which meant they either had to be grazed 304 along the peri-urban periphery of the city or to scavenge locally on market vegetable waste. Opportunity 305 to mix with local urban animals, either directly or within vector flight range, was noted for 306 slaughterhouse arrivals and situational grazers. Animals that arrive at the slaughterhouse are not 307 slaughtered immediately on arrival and instead holding times were dependent on the orders received by 308 the butcheries. Both Ukunda and Kisumu slaughterhouse workers noted that animals in the holding area 309 never "ran out" as they were always replaced with new arrivals. In Ukunda, holding time was up to one 310 month and the animals primarily walked to the slaughterhouse overnight, even if their origin was more 311 than 200 kilometers away. Animals in holding grazed around the slaughterhouse grounds while they 312 recovered from their transport either on foot or in a lorry. Milk was also a commodity that entered the 313 urban centers in large quantities to meet human nutritional demands. In Kisumu milk was noted to 314 specifically arrive from Eldoret and Nandi. In Ukunda, milk came from nearby Shimba Hills, Lunga Lunga, 315 and Tanzania. Milk was reported to quickly sell out as the demand for fresh milk from consumers is very 316 high.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. Urban livestock roaming was a common practice primarily due to feed scarcity. Approximately 334 half of all livestock owners at both sites reported to release their animals daily near to marketplaces for 335 scavenging on vegetable scraps as this was deemed necessary for animals to meet their daily nutritional 336 needs. Two participants noted that despite this, their animals struggled to maintain adequate nutrition. 337 There were additional challenges associated with roaming including being detained by police, poisoning, 338 neighbors abusing animals with knives, dog attacks, snake bites, and consumption of plastic bags. 339 Livestock theft was not noted to be a major concern of this practice.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. However, animals were sometimes treated for illness before slaughter and all four IDI 370 respondents reported that every urban slaughtered animal receives an antemortem examination the 371 day before, and they were to report to the slaughterhouse for emergencies to determine safety for 372 slaughter if an animal arrival sick or injured. It was also mentioned in Ukunda that if the treatment was 373 with antibiotics, a withdraw period of two weeks was observed. 374 375 "We have a doctor here who also inspects the animals. He does inspection before we slaughter. So, if it 376 is not okay, he will say that it is supposed to be treated first or it should be returned, it's not okay. So that 377 also is their work." -Kisumu SH4 R6 378 Theme 5: Information-focused determinants of risk differential 379 FGD participants were generally aware that diseases could be transmitted from animals to 380 humans and stated the name of several zoonotic pathogens including anthrax (KSMLO4), bilharzia 381 (KSMLO3), and tetanus (KSM SSTs). The Ukunda SSTs knew they were at risk for zoonotic disease but 382 said that "there is no business which has no risks," and that even the fire for cooking the products was a 383 health risk for them. The Kisumu SSTs recognized that the slaughterman could become infected with an 384 animal disease by inhaling the "hot blood" from the animal and concluded that while they do have some 385 contact with the blood and animal products, the slaughterhouse workers were touching the animals 386 more, so they were at higher risk. When risk was compared among the slaughterhouse workers groups, . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint 387 both sites identified the slaughterman to be at the highest risk but noted that indeed everyone touches 388 the blood.

The perceived occupational risk of slaughterhouse workers was heavily focused on hygiene 390 which nearly all participants determined to be superior to rural areas because of running water 391 availability. However, one participant in a Kisumu livestock owners group felt that urban populations 392 were at higher risk because of congestion and air pollution. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. 

Regulations in place to protect public health were reported to be vulnerable to violations, 437 particularly around meat certification and slaughtering of sick animals. The slaughterhouse workers 438 group firmly affirmed that it was illegal for any dead animals to enter the slaughterhouse premises.

439 However, at least two participants in the livestock owners group anecdotally described that in recent 440 years, these requirements have been circumvented when meat from sick or dead animals is mis-441 marketed and mixed in with an officially approved animal. They also stated that these individuals who 442 slaughter sick animals outside of the official system and may have options for receiving fake or corrupt 443 stamps of approval on their meat. One of the main drivers for corruption in meat certification was 444 explained to be the huge financial loss to the owner of the animal. This was noted by the livestock 445 owners' group in Kisumu rather than by any group at the slaughterhouses.

In the case of an official veterinary meat inspector condemning part or all a carcass, this process 447 has significant oversight by these professionals. In all four IDIs, participants highlighted the importance 448 of strict follow up during the condemnation process and all shared at least one personal story of a 449 community members becoming sick or dying after eating meat that was purchased through a regulatory 450 loophole and therefore, they are strict with follow-up. Notably, one of the meat inspectors from Kisumu 451 revealed that blood is one potential product that can bypass the condemnation process because it is 452 usually pooled early with many other animals and distributed before the inspection occurs.

Vulnerabilities were also present in animal movement restrictions. Data from an Ukunda IDI 454 explained the process for requesting a movement permit and that during high-risk times issuing 455 movement permits was suspended. However, they recognized animals do indeed still move at night 456 undetected or use "short cuts." In Kisumu, an IDI participant added that while there was an official . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. ; 481 slaughterhouse in Kisumu also had some restrictions for animals entering from far away not being able 482 to graze normally with the other animals. At least five slaughterhouse workers recalled animal 483 movement restrictions "nearly closing the meat sector" which greatly impacted their livelihoods. Groups 484 recalled foot and mouth disease outbreaks to be (the most common cause for movement restrictions. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint 504 A slaughterhouse worker in Kisumu noted that he was never told how to identify an animal that 505 was infected with RVFV over the course of the outbreak, and instead heard about it on the radio. Only 506 one slaughterhouse worker in Ukunda thought he had previously seen signs of RVFV in animals and 507 reported that in 2010 there were numerous deaths in cattle with blood-tinged mucus draining from the 508 nose, though this was never confirmed to be RVFV. For those that had not experienced, RVFV 509 specifically commonly confused it with other fevers and diseases, especially when the other disease has 510 "fever" in the name. East Coast Fever and Yellow Fever were specifically mentioned. Slaughterhouse When asked about the potential to be vaccinated against Rift Valley Fever, participants 522 requested more information about the disease and ingredients in the vaccine before they would accept 523 an injection. There was consensus agreement that vaccines worked to prevent diseases and people 524 would be willing to accept a vaccine if they understood their risk. In the slaughterhouse, they were keen 525 to be vaccinated, with some requesting vaccination the same day, because they feared high 526 consequence diseases such as Rift Valley Fever. In Ukunda, livestock owner willingness to be vaccinated 527 was linked to understanding that the livestock community was at higher risk.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) While there was indeed a high acceptance rate, barriers for accepting vaccination were also 537 noted. The main barriers were a lack of a perceived immediate threat, fear of long-term side effects, and 538 confusion with livestock vaccinations. The livestock owners in Ukunda believed that there should be sick 539 animals in their area before they were to accept a vaccination themselves. More hesitancy was present 540 in Ukunda compared to Kisumu as livestock owners were concerned that they would personally receive 541 a cattle injection. A livestock owner in Ukunda said it would be impossible to get people to accept an 542 injection before the disease arrives in their area. In Ukunda, one woman expressed concern the 543 vaccination could cause infertility and all groups wanted to know the potential negative consequences 544 of receiving a vaccine. 545 546 "Without seeing the disease even if you say you are going to pay me to be vaccinated, still I 547 won't accept unless the animals here are sick and can transmit to humans" -Ukunda LO1 R1 548 549 "Cows, maybe their immunity is higher than that of human beings, so the vaccine that would be 550 administered to cow is it the same as the one that would be administered to me? Are we going to run in 551 the same line as that? The same injection for cows is the same as my injection?" -Kisumu LO4 R4 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 7, 2022.

This qualitative study has highlighted factors relative to RVFV transmission and key 555 vulnerabilities in urban livestock systems that could support successful introduction of RVFV via infected 556 livestock. This risk is intertwined with personal sentiments, challenges faced by the urban livestock 557 community, and private business pressure at the slaughterhouse. Inclusion of the human-centric 558 findings presented here in urban risk management strategies could be expected to allow for earlier and 559 targeted approaches to urban RVFV case identification.

RVFV has been shown to adapt to many different climatic conditions, mammalian hosts, and 561 vectors (22). With the possibility of an imported animal seeding an infection and local herds maintaining 562 or amplifying RVFV, urban disease ecology would differ from rural areas and current measures and 563 surveillance approaches do not account for this. This qualitative study foreshadows how high-risk 564 populations in urban centers could be expected to identify and respond to an introduction. First, for 565 recognizing the first urban livestock cases, slaughterhouses are a key entry point for RVFV as animals 566 enter in high volume from a wide geographical range. Passive syndromic surveillance of livestock alone 567 could be expected to fail in this system as the herd context is required for the most recognizable signs of 568 RVFV: abortion storms and death of young animals. Most animals entering the urban center are adult 569 cattle, which often have inapparent infections (28) and could inadvertently infect slaughterhouse 570 workers, local livestock, or urban mosquitos. The risk to local urban livestock exposure from these 571 imported animals is highlighted by theme three (Nutritional demands and livestock movement) as 572 animals have ample opportunity for mixing (either directly or within vector flight range) with non-local 573 animals. If a local animal were to be exposed, the congested living conditions of humans and livestock in 574 the urban setting would deepen the spillover threshold. Notably, lack of feed availability is the root 575 cause of urban livestock roaming, and in the case of an urban outbreak, supporting nutritional demands . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 7, 2022. ; 576 of these animals could limit unauthorized movement. Moreover, if RVFV is already present in urban 577 vectors, livestock movement bans may be too delayed to have significant effects on the outbreak 578 magnitude and the subsequent threat to livelihoods may not be worthwhile. Risk for urban RVFV viral 579 maintenance is suggested by increased urban abundance of Aedes. spp mosquito vectors known to be 580 important in viral amplification phases of a natural outbreak (16), we therefore hypothesize that vector 581 exposures may play a greater role in an urban human exposures. The potential for an undetected 582 introduction coupled with RVFV's complex epidemiology requires that prevention measures account for 583 all aspects of transmission including human behaviors that affect risk, host interactions with vectors, and 584 opportunities for local animals to be exposed.

Theme two (Connecting human health and animal product consumption) of this study 586 demonstrates retention of cultural practices in urban Kenya important in RVFV transmission: 587 consumption of raw blood and fresh milk. Pooled blood purchased directly from slaughterhouses poses 588 a risk to everyone that handles and prepares the product as RVFV is well documented to aerosolize from 589 infected blood, and has been linked to more severe infections (29,30). This exchange of blood from the 590 slaughterhouse also shifts risk into the community where risk factors may be less recognizable, and 591 exacerbates challenges differentiating human RVF disease clinically at health centers from other febrile 592 diseases, including malaria and meningitis (31). Our findings suggest that public health messaging from 593 previous RVFV outbreaks at our study sites to have been primarily focused on consumption of meat and 594 milk, rather than vector exposure, and this is how participants evaluated their risk. This is in contrast to 595 a study conducted in pastoralist communities of Northeastern Kenya where the majority of participants 596 believed RVFV infection to be from mosquito bites rather than from consumption of milk, meat, and 597 blood (32). The authors of that qualitative study reported that while their participants had heard of 598 consumption related risks, their experiences didn't match their health outcomes and, to them, 599 consumption of meat or milk was not risky for contracting RVFV. In the urban setting, meat and milk . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 7, 2022. ; https://doi.org/10.1101/2022.05.05.22274700 doi: medRxiv preprint

Data on viral presence and stability in blood after slaughter, milk and dairy products, 605 and wastewater from urban slaughterhouses could quantify these public health risks and allow for more 606 specific and directed policy that could be used to build trust and/or be leveraged for surveillance

Furthermore, integrated, and practical approaches to managing these culturally connected risks are 608 required as both livestock owners and SSTs expressed that the ability to purchase and sell their products 609 is important for livelihoods and women's empowerment

Our study demonstrated regulatory vulnerabilities in theme six

We highlighted both knowledge-focused and experience-focused 617 factors that affect perceived risk differentials between groups. With these data, we understand current 618 slaughterhouse regulations to be focused on highly prevalent bacterial and parasitic diseases and food 619 safety. This was demonstrated by the participants mentioning hygiene and visual inspection of organs as 620 key measures to determine food safety and reporting that animals are inspected and therefore safe. to determine the ailment of the animal, which neglects their own personal protection from 623 zoonoses. With few individual exceptions, group consensus determined risk to be lower in urban areas 624 compared to rural areas because of improved urban hygiene. Still, fear filled the gaps in participant 625 knowledge, they requested to learn more about their risk of RVFV. Awareness building and education 626 could contribute to better alignment of the knowledge-focused determinants of risk with known RVFV 627 transmission pathways. Messaging would be best focused on the key activities associated with urban 628 slaughterhouse risk and risks in the greater community including handling and consumption of fresh 629 milk and blood

Awareness building for diseases with alternative transmission mechanisms, such as direct aerosolization

Formalizing meat trade in Kenya by requiring a stamp of inspection from qualified veterinarians 634 has undoubtedly improved key zoonotic diseases and contributed to the improvement of Kenya's meat 635 safety

All meat inspector veterinarians in this study shared personal stories about the 637 importance of oversight when condemning a carcass, yet livestock owners reported that corruption and 638 falsification in meat certification does occur, and community-based education of vulnerable groups may 639 alter personal decisions to partake in this risky behavior. For urban introduction of RVFV, slaughterhouse 640 workers and their business partners may be the first line of defense in outbreak control. Full reliance on 641 veterinarians to make key decisions for slaughtering and consumption undermines the ability for 642 slaughterhouse workers to be leveraged in reporting sick animals. In our IDIs, veterinarians understood 643 this public health responsibility, but added that there was insufficient diagnostic support for RVFV, and 644 they fear missing the diagnosis. Underequipping and underpreparing urban meat inspector veterinarians 645 for introduction of RVFV

local One Health initiatives and better administrative cooperation between human 647 health providers and veterinarians are required. If RVFV were to enter an urban center of an endemic 648 region and establish an urban transmission cycle, the human health, and economic effects from loss of 649 livestock would be devastating and concern for re-emergence after would persist every subsequent 650 flooding event. Urban livestock markets would be significantly disrupted and pressure to bypass 651 regulations to maintain livelihoods would be high

Conclusion: Implications for policy to prevent and control urban RVFV 654 This study has informed understanding of potential pathways of introduction in the context of 655 current practices and urban high-risk groups' perceived risk. Tracing the pathway of blood, organs, and 656 fresh milk could inform risks in the greater community and allow for preferential testing of RVFV at for RVF include education of high-risk groups, community-based awareness 659 building, preventive and reactive livestock vaccination, scenario planning, and surveillance. Passive 660 surveillance is less reliable in urban areas, and we recommend active surveillance, particularly during 661 high-risk times

An early and rapid response to RVFV introduction will be required to give urban centers the best 664 opportunity at avoiding infection of urban mosquitos and initiation an urban transmission cycle

Integrating the themes presented here into development of future 668 preventive measures could be expected to be more efficacious and avoid marginalization of vulnerable 669 communities that rely on livestock to support their livelihoods when there is an outbreak. Additionally, a 670 digital record keeping system of urban animal imports could allow veterinarians to carry out local 671 surveillance on suspect animals that enter from high-risk markets known to source animals within high-672 risk zones. High-risk groups, such as affiliates of the slaughterhouse and livestock owners, of urban areas 673 would likely be the first to identify an urban introduction of RVFV and empowering them to recognize 674 and report suspect cases would assist urban slaughterhouse veterinarians in Kenya that often have 675 overwhelming fast-paced workloads

This study in its entirety was approved by the Institutional Review Board (IRB) at Stanford 682 University (IRB-57869) and the Technical University of Mombasa IRB (TUM ERC EXT/004/2019 683 (R)). The funders had no role in study design, data collection and analysis

This study was funded by NIH Fogarty Global Health Equity Scholars Program NIH, 689 D43TW010540 (Gerken) and NIH

Author Contributions 692

Makena 696 Mwaniki, Karren Shaita, Stella Orwa 697 Formal analysis: Keli Nicole Gerken, Justinah Maluni 698 Funding acquisition

Caroline Ichura 707 Visualization: Keli Nicole Gerken 708 Writing -original draft: Keli Nicole Gerken

Review article: Sociocultural and economic 725 dimensions of Rift Valley Fever

High risk for human exposure to 727 rift valley fever virus in communities living along livestock movement routes: A cross-sectional 728 survey in Kenya

Severe Human Rift Valley Fever in Sangailu, Garissa County

Responding to the threat of urban yellow fever outbreaks

Persistence of Yellow fever virus outside the Amazon Basin

Izabela Maurício de Rezende 1☯ , Lívia Sacchetto 1☯ , E ´ rica Munhoz 739 de Mello

Understanding Poverty

Stocktaking of the housing sector in Sub-Saharan Africa: Challenges and 745 opportunities

The role of livestock 747 movements in the spread of rift valley fever virus in animals and humans in Mayotte

TAJ of TM and H. ASTMH 2021 Annual Meeting

English and Mandarin on the novel coronavirus COVID-752 19 . The COVID-19 resource centre is hosted on Elsevier Connect , the company ' s public news 753 and information

Urban livestock keeping in 755 the city of Nairobi: Diversity of production systems, supply chains, and their disease management 756 and risks

Africa's demographic trends

Prospects 2014 revision of the World Urbanization. 2014 revision of the World Urbanization 761 Prospects

Cities and pandemics: Urban areas are ground zero for the 763 transmission of emerging human infectious diseases

Characterization 765 and productivity profiles of Aedes aegypti (L.) breeding habitats across rural and urban 766 landscapes in western and coastal Kenya. Parasites and Vectors

Risk factors for infectious diseases in urban environments of sub-768

Saharan Africa: A systematic review and critical appraisal of evidence

Reemergence of yellow fever virus in southeastern 771

What sparked the spread?

Global perspectives on arbovirus outbreaks: a 2020 snapshot

Managing the COVID-19 776 pandemic in poor urban neighborhoods: The case of Accra and Johannesburg

A framework for assessing 779 the effects of shock events on livestock and environment in sub-Saharan Africa: The COVID-19 780 pandemic in Northern Kenya

Reconstruction of Rift Valley fever transmission dynamics in Madagascar: Estimation of force of 784 infection from seroprevalence surveys using Bayesian modelling

Counting Our People for Sustainable development and devolution of services" VOLUME IV 788 Distribution of population by County and Sub County

Epidemiological investigation 792 of a rift valley fever outbreak in humans and livestock in Kenya

Rift Valley 797 fever in Kenya: History of epizootics and identification of vulnerable districts

Recognizing Rift Valley Fever. FAO Animal. Rome

Fever Virus Causes Fatal Encephalitis in African Green Monkeys and Common Marmosets

Differences in aerosolization of Rift 804 Valley fever virus resulting from choice of inhalation exposure chamber: Implications for animal 805 challenge studies

The 807 challenging management of Rift Valley Fever in humans: Literature review of the clinical disease 808 and algorithm proposal

Lay perceptions of risk factors for Rift Valley fever in a pastoral 811 community in northeastern Kenya

Guidelines for preventing mosquito breeding associated with wastewater 814 treatment and disposal in the Northern Territory

Working 816 conditions and public health risks in slaughterhouses in western Kenya

Source reduction with a 819 purpose: Mosquito ecology and community perspectives offer insights for improving household 820 mosquito management in coastal Kenya