Plague, primarily a disease of rodents, is most frequently transmitted by fleas and causes potentially fatal infections in humans. In Uganda, plague is endemic to the West Nile region.
Here, we describe an animal-based surveillance and early response program (herein referred to as rat fall surveillance or RFS) that engages members of the community, volunteer village health teams (VHTs), subcounty environmental health officers, and local leaders. Through the program, rat falls are reported, carcasses are collected and tested, and Y. pestis-positive results trigger community education and target implementation of vector control (IRS) to prevent human plague cases.
Our specific objectives are to evaluate community participation and timelines of response under the RFS program, and to report frequency of human plague cases in participating and surrounding villages. This program was implemented and evaluated in the Arua and Zombo districts of the West Nile region, Uganda, from July 1, 2013 to June 30, 2016.
Primary prevention for plague includes control of rodent hosts or flea vectors, but targeting these efforts is difficult given the sporadic nature of plague epizootics in the region and limited resource availability.
Here, we present a community-based strategy to detect and report rodent deaths (rat fall), an early sign of epizootics. Laboratory testing of rodent carcasses is used to trigger primary and secondary prevention measures: indoor residual spraying (IRS) and community-based plague education, respectively.
Read also: "Baby" in Nigerian Culture: A Deeper Look
A total of 83 villages, representing a local estimate of approximately 37,000 persons, were selected for participation among 563 local villages with a history of plague and included many of those reporting the greatest number of confirmed or suspect human plague cases between 1999 and 2011 (Centers for Disease Control and Prevention, unpublished data) (Figure 1). To expand the geographic extent of the surveillance network, in instances where neighboring villages reported confirmed cases and high case counts, a village may have been excluded so that a geographically distant village could be included, even if the more distant village had lower case counts.
Seventy-five villages were invited to participate in the program between July and September of 2013, and an additional six villages, which were described in a concurrent study6 were added to the surveillance network in September 2014.
Plague is a life-threatening, flea-borne, rodent-associated zoonosis caused by Yersinia pestis. The plague bacterium has a nearly global distribution; however, in recent decades most plague cases have been reported from East Africa and Madagascar. In Uganda, plague is endemic in the highlands of the far northwest, which are known as the West Nile region. Here, Y. pestis is maintained in enzootic cycles among sylvatic and peridomestic rodents and their fleas, with Arvicanthis niloticus and Crocidura sp. likely playing important roles in plague epizootics.
During plague epizootics, infections are assumed to spill over into Rattus rattus, which is commonly encountered within households in the West Nile region, is highly susceptible to plague infection, and harbors efficient Y. pestis vectors (Xenopsylla cheopis and Xenopsylla brasiliensis).
A schematic overview of the RFS and response program is shown in Figure 2.
Read also: Enduring History of African Baby Carriers
As part of the surveillance program, VHTs investigated reports of dead rodents from village residents, collected any carcasses found using basic universal precautions, and notified Uganda Virus Research Institute (UVRI) staff by cell phone call to a preprogrammed toll-free phone number of the need for sample transfer and testing. Particularly because rodenticide use and other methods of rodent control are not uncommon in the villages, for the purposes of the program; a “rat fall” was defined as the discovery of one or more small mammal carcasses in the absence of rodenticide use or obvious injury.
At the time of carcass collection, VHTs recorded date and location information and searched an area at least 100 m in all directions to locate any additional carcasses, inquiring with neighboring households to determine if any other carcasses had been found.
Processing, testing, and storage of potentially infectious carcasses were conducted in an access-controlled laboratory under BSL-2 conditions. When possible, carcasses were identified to species using a published key. Carcasses were then necropsied and touch-preparation slides of the liver and spleen tissues were tested for the presence of Y. pestis F1 antigen using a direct fluorescent antibody (DFA) assay described elsewhere. Carcasses were considered presumptive positive for Y. pestis and actionable if the presence of the Y. pestis F1 antigen was detected in one or both tissues.
For the purpose of brevity, we later refer to carcasses testing positive or equivocal by DFA as Y.
When a carcass tested positive for Y. pestis, UVRI staff notified the District Health Officer, Uganda Ministry of Health to recommend timely IRS treatment of all the huts in the reporting village to reduce the numbers of hut-associated, potentially infectious fleas. In addition to IRS, a number of community sensitization efforts were initiated in response to Y. pestis-positive carcass results. Immediately after notification of the test result, the VHT reported the finding directly to the household where the carcass was found.
Read also: Becoming a Father: Chad Duell
UVRI staff, together with village leadership, hosted meetings at places of worship, schools, markets, and other community spaces to raise awareness of plague at the village level and to share messages of primary and secondary plague prevention. When Y. pestis was not detected in a carcass, the laboratory staff directly notified the submitting VHT who then gave the result to the individual(s) who reported the carcass.
To improve the surveillance and response program and address specific programmatic issues, UVRI and Centers for Disease Control and Prevention (CDC) leadership elicited feedback from VHTs, district- and sub-county-level representatives, environmental health officers, field and laboratory staff, and others annually at refresher training workshops and through organized periodic meetings with key stakeholders.
Village-level reporting data, notification, and test result dates were used to evaluate the timeliness of the RFS program for both Y. pestis-positive and Y. pestis-negative carcasses. Statistical comparisons of various response times were made for all carcasses between years 1, 2, and 3 using analysis of variance among them, assuming unequal variance, and if any statistically significant difference was found; a Tukey’s multiple comparison of means test was used to identify where the difference lay.
Difference of medians were compared using the Wilcoxon-Mann-Whitney exact test. Proportions were compared using a mid-p exact test for two proportions and a Fisher’s exact test for three proportions. For all statistical comparisons, significance was declared at the alpha = 0.05 level.
Before the initiation of the program, all protocols were reviewed and approved by the UVRI Research Ethics Committee, Uganda National Council for Science and Technology, and the Uganda President’s Office.
During the first 3 years of the program, individuals from 142 villages reported 580 small mammal deaths; 24 of these tested presumptive positive for Yersinia pestis by fluorescence microscopy. In response, for each of the 17 affected communities, village-wide IRS was conducted to control rodent-associated fleas within homes, and community sensitization was conducted to raise awareness of plague signs and prevention strategies. No additional presumptive Y. pestis-positive carcasses were detected in these villages within the 2-month expected duration of residual activity for the insecticide used in IRS. Despite comparatively high historic case counts, no human plague cases were reported from villages participating in the surveillance program; five cases were reported from elsewhere in the districts.
Of the 83 participating VHTs, 68 (80%) reported and submitted at least one small mammal carcass through the program during the 3 years of evaluation. In addition, small mammal deaths were reported through spontaneous channels of communication by community members from another 74 villages throughout the region that were not initially selected for participation in the program (Figure 1). During brief follow-up interviews, those who submitted carcasses for testing from these “out of network” villages reported hearing about RFS through community sensitization efforts or through work or personal contacts.
Overall, more carcasses were submitted from participating villages than “out of network” villages (432 versus 148). Between July 1, 2013 and June 30, 2016, a total of 580 small mammal carcasses were submitted through the RFS program. Of the 523 calls received over the 3-year surveillance period, most (N = 484; 92.5%) were made to report the observation of a single carcass, whereas the remainder (N = 39; 7.5%) were made to report multiple carcasses on a single day.
Small mammal deaths were reported to the surveillance program in every month of the year (Figure 3), with a median of 15 carcasses submitted per month (range: 2-40, mean: 16.1) and a statistically significant decreasing trend over time (P < 0.0001). The most commonly reported species was R. rattus (N = 432), which represented 74.6% of the carcasses submitted, followed by A. niloticus (N = 90; 15.5%), and Mastomys sp. (N = 16; 2.8%); the remaining nine other species combined (N = 27) represented 4.7% of total submissions. A small number of carcasses were not identifiable because of poor condition of the carcass (N = 14; 2.4%).
Species belonging to the genera Mastomys and Crocidura are difficult to distinguish morphologically; therefore, these identifications were made at the genus level. Based on earlier molecular identifications from the same region, however, small mammals identified as Mastomys spp. were likely either Mastomys natalensis or Mastomys erythroleucus and Crocidura spp.
Of 580 carcasses submitted, 24 (4.1%) tested positive by DFA for Y. pestis, 555 carcasses (95.7%) tested negative for Y. pestis, and 1 (< 1%) was too desiccated to test. All carcasses that tested Y. pestis-positive were reported between the months of September and April (Figure 3). Most carcasses testing Y. pestis-positive were R. rattus (19, 79.2%), whereas A. niloticus and Mastomys sp. comprised 16.7% and 4.2% of the total, respectively.
There was not a statistically significant difference in infection prevalence among R. rattus (4.4% positive of 433 submitted), A. niloticus (4.4% of 90), and Mastomys sp. (6.3% of 16) (P = 0.72). Although less commonly reported, small mammal carcasses submitted in groups (> 1 per day from the same village), were significantly more likely to test positive for Y. pestis (10.3% of 39 carcass groups) than carcasses submitted singly (2.5% of 484 carcasses) (Difference: 7.8%, 95% confidence interval [CI]: 1.3-21.2%, P = 0.03). There was no significant difference in the percent of carcasses testing positive for Y.
During the 3 years of program evaluation, 17 village-wide IRS applications were enacted in response to the submission of at least one Y. pestis-positive carcass, including one village that was treated twice, once in November 2013 and again in November 2014 (Figure 4). Ten sprays were conducted in participating villages, whereas seven were conducted in “out of network” villages. The number of family huts per village ranged from 112 to 544, with a median village size of 230 huts. Hut-level IRS coverage in these villages ranged from 52% to 100%, with a median of 93.5%.
In each of the 17 affected villages, community-wide meetings were held to communicate messages about plague prevention, the signs and symptoms of plague infection, and the importance of early treatment. The average number of days between critical communication and action steps are shown in Table 1. For all small mammal carcasses submitted during the 3 years of program evaluation, the average time elapsed from notification to response (including the report from the community, carcass collection and transfer to the laboratory, DFA testing, and notification of the test result to the VHT) was fewer than 2 days.
When this same metric was evaluated separately for each year of the program, average time to response for all carcasses showed a statistically significant decrease from year 1 to year 3 (Table 1). The timeliness of specific action steps, including retrieval and submission of carcasses to the laboratory and the communication of test results, likewise improved significantly over time (Table 1). Overall time from community report of a carcass to VHT being notified of the test result was similar for both Y. pestis-positive and negative carcasses (data not shown). Community reports carcass.
The number of days between the detection of the first Y. pestis-positive carcass from a village and completion of the two main response efforts, IRS and community sensitization, are shown in Figure 4. For the 17 village-wide interventions, IRS was completed within a median of 10 days (range: 4-113) and mean of 29.9 days (95% CI: 11.5-48.3) after the initial report of a carcass by community members, and village-level community sensitization efforts were conducted within a median of 4 days (range: 1-9) and mean of 4.3 days (95% CI: 3.3-5.3).
Intermediate communication steps after the detection of Y. pestis-positive carcasses were evaluated to identify delays. Test results were communicated from the laboratory through the UVRI epidemiology coordinator, and the recommendation to use IRS was made to district staff within a median of 1 day (range: 0-5) and mean of 1.1 days (95% CI: 0.3-1.9). District-level coordination to implement IRS initially added an additional 26.9 days on average to the total response time (95% CI: 8.7-45.2, range: 3-108, median: 7).
Black Rat (Rattus rattus)
Arvicanthis niloticus
Crocidura sp.
Mastomys sp.
Under Paris: The Rat-and-Mouse Infestation Tourists Never See
Popular articles:
tags: #Africa
