Evaluation of the impact of chemical control on the ecology of Rattus norvegicus of an urban community in Salvador, Brazil.

BACKGROUND: The presence of synanthropic rodents, such as Rattus norvegicus, in urban environments generates high costs of prophylaxis and control, in large part due to the environmental transmission of the pathogenic spirochete Leptospira interrogans, which causes leptospirosis. In Salvador, Brazil, The Center for Control of Zoonosis (CCZ) is responsible for planning and implementing Rodent Control Programs (RCP) which are based on chemical rodenticide. However, these strategies have not been standardized for use in developing countries. AIM: This study aimed to identify the effect of a chemical control campaign on the demographic variables of urban R. norvegicus, analyzing relative abundance, sex structure, body mass, and age of the population, as well as the characterization of spatial distribution among households, rodent capture campaigns and interventions.
METHODS: This study was carried out during 2015 in three valleys of an urban poor community in Salvador. Individuals of R. norvegicus were systematically captured before (Pre-intervention) and three months (1st post-intervention) and six months (2nd post-intervention) after a chemical control intervention conducted by the CCZ in two valleys of the study area while the third valley was not included in the intervention campaign and was used as a non-intervention reference. We used analysis of variance to determine if intervention affected demographic variables and chi-square to compare proportions of infested households (Rodent infestation index-PII).
RESULTS: During the chemical intervention, 939 households were visited. In the pre-intervention campaign, an effort of 310 trap nights resulted in 43 rodents captured, and in the 1st and 2nd, post-intervention campaigns resulted in 47 rodents captured over 312 trap nights and 36 rodents captured over 324 traps-nights, respectively. The rodent infestation index (PII) points did not show a reduction between the period before the intervention and the two periods after the chemical intervention (70%, 72%, and 65%, respectively). Regarding relative abundances, there was no difference between valleys and period before and two periods after chemical intervention (trap success valley 1: 0,18; 0,19; 0,18 / Valley 3 0,15; 0,17; 0,13/ P>0,05). Other demographic results showed that there was no difference in demographic characteristics of the rodent population before and after the intervention, as well as there being no influence of the application of rodenticide on the areas of concentration of capture of rodents between the campaigns.
CONCLUSION: Our study indicates that the chemical control was not effective in controlling the population of R. norvegicus and provides evidence of the need for re-evaluation of rodent control practices in urban poor community settings.

Introduction

Unplanned urbanization and the increase in human population have favored the establishment of poor urban settlements or slums, which are characterized by highly populated residential areas and a lack of essential public services, such as garbage collection, sewage systems, water supply, electricity, and public lighting [1]. According to UN-Habitat, currently, 1.6 billion people live under such conditions, with urbanization, leading to environmental degradation and growing inequality [2], with drastic impacts on hydrologic cycles, green area suppression, pollution, and climate change both local and global [3]. These changes often cause loss of native biodiversity and increased abundance of introduced and generalist species, such as synanthropic rodents [4].

Synanthropic rats are among the most impactful pests to human activity, causing economic impacts in the hundreds of billions of dollars worldwide [5], and can transmit zoonotic pathogens responsible for significant human morbidity and mortality around the world [6]. Among these pathogens, Leptospira interrogans is a bacterium that causes leptospirosis, an emerging infectious disease in urban centers in developing countries [7]. Each year more than one million cases occur worldwide causing approximately 60,000 deaths [7]. In Brazil, more than 10,000 severe cases of leptospirosis are reported during epidemic events in periods of high precipitation in urban poor communities [8]. Mortality for severe forms of the disease, such as Weil’s syndrome is >10% [9], exceeding 50% for pulmonary hemorrhage syndrome [10]. In urban poor communities, the transmission of this disease is more intense due to the proximity between humans and rodent populations. In addition, the characteristics of urban poor communities (e.g., open sewers, accumulated trash, dirt floors) create a habitat ideal for rats, which leads to high infestation rates and frequent contact with residents [11].

One of the main strategies of the Brazilian Ministry of Health, and many other developing countries, to prevent or reduce human exposure to leptospires is to control rodent reservoirs through chemical control, the most widely used method to eliminate rats on a large scale [12,13]. Studies on rats and possible eradication methods are largely concentrated in the USA and Europe, in northern temperate climates [14]. In these regions, control with rodenticides has proved effective in reducing the density of these animals by 50–90% [15]. Additionally, studies have shown that the demographic structure of rodent populations is altered by the removal of dominant individuals and the potential resulting immigration influx [6].

Few studies have evaluated the impact of chemical interventions on Rattus norvegicus populations and they have reported a high reduction in infestation rates from 69% to 100%. One exception was a study performed in London [16], where poisoning interventions reduced rat population sizes by 10% and those population sizes returned to prebaiting population estimates within six months. In a study in Brazil, in Sao Paulo, reductions in infestation rates of 64% and a general reinfestation rate of 80% six months after the intervention were reported [17]. However, most of these studies provide no data on the effects of rodent control on population characteristics such as the number of embryos, percent of pregnant females etc, and the few studies that report demographic data on rat populations are limited to temperate regions [14]. Considering that patterns of pathogen infection are shaped by population features of reservoir hosts [18-20], it is critical to understand the effect of chemical control on the abundance and demographic characteristics of R. norvegicus, the main reservoirs host of L. interrogans in Brazil [21]. This information can help us to establish effective control interventions, in order to reduce the risk of pathogen transmission to humans [22].

The objective of this study, therefore, was to evaluate the effects of a rodent chemical control campaign carried out by the Center for Control of Zoonosis (CCZ) in the city of Salvador on the relative abundance and demographic markers of Rattus norvegicus in an urban community with a high risk of leptospirosis transmission.

Material and methods

Ethics statement

The ethics committee for the use of animals from the Oswaldo Cruz Foundation, Salvador, Brazil approved the protocols used in this study (protocol number 003/2012) These protocols were also approved by the Yale University’s Institutional Animal Care and Use Committee (IACUC), New Haven, Connecticut (protocol number 2012–11498).

Study area

The study was conducted in the neighborhood of Pau-da-Lima, in the city of Salvador (BA, Brazil), which is 0.17 km2 in extent (Fig 1A) with 3,717 residents [23]. The site is a slum area described in detail previously [23] and is characterized by three geographic valleys (referred to as valleys 1, 2, and 3) where there is no regular refuse collection and sanitary structure, with open sewers. The mean annual incidence of leptospirosis is 35.4 per 1000 inhabitants [24] and, with a high abundance of rodents (Rattus norvegicus is a more dominant species) [23] and a prevalence of L. interrogans in rats [25].

Fig 1

A. Study area in Brazil. B. Schedule of events in Pau da Lima, Salvador, Brazil. Capture, environmental assessment, chemical intervention stratified by valley during the study period.

In 2015, the CCZ carried out rodent control activities in 11 areas of the city of Salvador with a high risk of leptospirosis, which was defined by the density of cases registered in the areas by the Municipal Secretariat of Health, as well as previous data on rodent infestation collected by the CCZ. These activities included the application of rodenticide, educational campaigns, and environmental management. In the urban poor community of Pau da Lima, control activities focused mostly on chemical control which was performed in two of the three valleys. To evaluate this intervention, we performed rodent trappings before and after the intervention in valleys 1, 2, and 3 and monitored the CCZ chemical intervention performed in valleys 1 and 3. Valley 2 was used as a control area with no chemical intervention. (see Fig 2 for detailed steps).

Fig 2

A. Area of study and distribution of households visited with or without need for rodenticide application in Pau da Lima, Salvador, Brazil. B. Number of rodenticide applications among households in need in Pau da Lima, Salvador, Brazil.

Rodent control intervention carried out by the Zoonoses Control Center

The CCZ carried out its intervention in the period from July 14 to September 15, 2015. As per its standard practice, the CCZ’s intervention was performed in three steps: 1) an initial evaluation of the houses for rodent infestation (pre-intervention); 2) three rounds of rodenticide application (chemical intervention), and 3) evaluation of post-intervention infestation. This intervention follows standard methodologies described previously [26] and is used extensively by other CCZs throughout Brazil [12].

The pre-intervention assessment of rodent infestation was performed from July 14 to August 4, 2015. The CCZ team visited all abandoned houses, occupied housing, and public spaces in the study area. The team visited each household and classified them as ‘inspected’ when owners accepted the intervention, “abandoned”, “refused” when owners rejected the intervention, or “closed” when owners were absent. In the inspected houses, a standardized questionnaire for the evaluation of rodent infestation adapted from CDC protocols [27] was applied as previously tested [28]. During the evaluation, 12 variables were assessed, including the type of domicile, availability of food, water, access/entry for rodents to houses, and signs of rodent infestation (feces, burrows, trails, and grease marks on external walls).

Application of rodenticide (round 1). Simultaneously with the previous activity (evaluation of pre-intervention infestation), CCZ Endemy Control Agents (ECAs) applied rodenticides to households that showed signs of rodent infestation. The following commercial formulations were used: Coumatek® contact powder (Cumatetralil 0.75%) and Klerat paraffin block (Brodifacoum 0.005%). The contact powder was applied to the tracks and entrance of burrows. The baits were tied with pieces of wire so they would not be dragged by rodents or other animals. Rodenticides were only applied when the risk to children or domestic animals was excluded. The amount of rodenticide applied in each household depended on the area of the property, environmental conditions, and level of rodent infestation.

Application of rodenticide in rounds 2 and 3. Subsequent visits were performed at 10 (round 2) and 20 (round 3) days after round 1. The status of the baits (totally consumed, partially consumed, or not consumed) was recorded and new baits were placed according to the need.

Post-intervention rodent infestation evaluation. Fifteen days after round 3, the last visit was conducted to each household’s, and the same questionnaire used for the initial evaluation was applied to evaluate the level of infestation after the intervention.

All these activities were carried out between 08:30 and 12:00 in the morning, in valleys 1 and 3. In valley 2 no visit and/or chemical control by the CCZ was carried out because the institution prioritized other risky areas for the prevention of leptospirosis. Therefore, in this study, this valley was used as a reference. Four people from the research team followed the CCZ teams during all stages–recording household classification, infestation, and rodenticide data–to ensure the quality of the information. The researchers had no influence on the nature or timing of the intervention.

Rodent trapping and processing

We performed rodent trapping campaigns 10 months before CCZ’s chemical intervention and three and six months after to evaluate demographic changes in the rodent populations (Fig 2).

The data collected during a previous study [23] conducted between October and December 2014 was used as pre-intervention trapping campaign. The previous study included trapping in 108 randomly selected sites in the three valleys of the study area. For the present study, we initially considered the points of the previous study in which trapping success was different from zero in order to guarantee the success of the analyses of abundance. Of these, 40 sampling points were randomly selected in the three valleys. Subsequently, we returned to the same 40 points to carry out new rodent trappings three- and six months post-intervention (November-December 2015 and March-April 2016, respectively).

During trapping campaigns, protocols already described were used [23,29]. The research team applied another questionnaire [28] on the 40 points to identify signs of rodent infestation (e.g.: feces, rodent runs, grease marks). In parallel, two Tomahawk traps were installed in each household within a 10-meter radius buffer. Trapped rodents were transported to the laboratory where they were euthanized. For each rat, demographic characteristics were recorded (sex and age structure, pregnancy, number of embryos, body mass/body condition) [19,23]. Rodent’s body condition was estimated using a "scaled mass index" (SMI) [30]. All data were collected following previously validated methods using REDCap software [31].

Data analyses

To calculate the infestation rate of valleys 1 and 3 we used the infestation index (IIP), a reference method utlised by the Brazilian Ministry of Health to estimate the level of vector infestation in urban centers [13] that uses number of houses with signs of rodent infestation/number of the households evaluated. To verify if there was variation between IIP before and after the intervention, a binomial probability test was used. To estimate the relative abundance of rats we uthe sed trap success rate. This index was calculated by dividing the total number of trapped rats by the total trap effort [32].

Differences in the means of age (days), SMI, and a number of embryos (response variables) of the trapped individuals between campaigns and valleys (independent variables) were first assessed through analysis of variance (ANOVA), followed by Tukey Honestly Significant Difference. Permutation ANOVA and Fisher’s Least Significant Difference were the alternative tests when ANOVA test assumptions were not met. In addition, to determine if the sexual ratio varied from 1:1 between campaigns, a binomial probability test was used. We compared the proportion of pregnant females between the campaigns using a chi-square test of homogeneity.

To analyze the effect of the number of rodenticide baits on trapping success, a beta-inflated regression was used, which has a wide range of distribution shapes (Attachment 1) [33]. We also tested whether trap success, as a response variable, was associated with the proportion of household visits receiving rodenticide applications using a generalized linear model (glm) with a binomial distribution.

We assessed whether the need for rodenticide application (as a binomial response variable) was associated with the household distance to the main avenue (here used as a proxy for quality of infrastructure), accounting for the effect of the valley by applying a glm using household distance to the main avenue (continuous) and valley (factor) as explanatory variables.

Model simplifications were performed considering the Akaike Information Criterion (AIC). In all statistical analyses a significance level of p <0.05 was considered and was performed in R [34] using packages lmPerm, agricolae, and zoib [35-37].

Spatial descriptive analysis

A database with georeferenced aerial photographs provided by the Company for Urban Development of the State of Bahia (CONDER) was constructed in QGIS version 2.18.20. Photographs of the study site, with a scale of 1:2,000 and spatial resolution of 16 cm, were taken in 2006. The study team identified households within the study site and marked their positions onto hard copy 1:1,500 scale maps. A survey was conducted during the period of April 2015 to August 2016 to geocodify the location of the households and traps.

Kernel Density Estimation analysis (KDE) was performed to evaluate smoothed spatial distributions of sites of the three rodent capture campaigns, considering as a weight factor in the analysis the success rate of capture [38]. We use a distance matrix analysis to measure the shortest distance of households to an asphalted road >5 meters in width. To determine the smoothed population-adjusted risk distribution we calculated the ratio of the KDE for households that need rodent control to all households evaluated (‘inspected’). The same analysis was performed for closed households relative to all households, as well as for households with the application of rodenticide relative to households that need rodent control.

Results

Chemical intervention

During the chemical intervention, all households in valleys 1 and 3 (N = 939) were visited, 283 in valley 1 and 656 in valley 3. Table 1 presents data on the coverage of the chemical intervention stratified by the valley. However, only two-thirds of the households (634, 67.5%) were inspected in order to evaluate rodent infestation. Closed households (32.5%) were the main reason why households were not inspected. Only 0.2% refused to participate.

Table 1

The proportion of inspections households per intervention site (valley 1 and 3).

Valley	Total households (n)	With inspection	House in need of treatment(total inspected)	Number of treatments
				0	1	2	3
1	283	177 (62,5)	130 (73,4)	67 (37,8)	50 (28,2)	44 (24,8)	16 (9)
3	656	457 (69,7)	341(74,6)	248 (54,2)	112 (24,5)	70 (15,3)	27 (5,9)
Total	939	634 (67,5)	471 (74,3)	315 (49,7)	162 (25,5)	114(18)	43 (6,8)

Among the inspected households, 60% (380/634) showed signs of rodent infestation. Seventy-eight percent (297/380) of infested houses were treated with rodenticide. The remaining 22% (83/380) were not treated due to the presence of children or animals that could be at risk. Forty percent (n = 254) showed no signs of rodent infestation or environmental deficiencies and therefore did not qualify for intervention. However, 16% of these (n = 41) received rodenticide applications at the request of the resident.

Among households that received rodenticide applications, 67% received only paraffin blocks and 6% received only contact powder, while 27% received both types of rodenticides. The average number of paraffin blocks per household was 3.2, although there was a high variation in the intensity of application (range from 1 to 21). Only 12.3% of the households visited, which required application of rodenticide, received the three applications (Table 1) that are recommended for rodent control guidelines in urban settings.

There was high spatial heterogeneity in the distribution of households with infestation or environmental conditions that required the application of rodenticide. We also observed high heterogeneity in the application of rodenticides (Fig 2B).

During the second and third applications of rodenticide, it was not possible to estimate the bait consumption adequately, since the majority of them (73%, 500/685 in the first application) were not found in the following applications. Of the baits found during the second application (185), 96 (51.9%) were totally consumed, 37 (20%) partially consumed and 52 (28.1%) were not consumed.

Regarding rodent infestation as evaluated by CCZ, the pre-intervention survey revealed 39% of the households with signs of rodent infestation. In the post-intervention evaluation, this proportion dropped significantly to 21.1% (p <0.05), indicating a 54% reduction. The availability of resources such as water and food did not show a significant reduction in the environmental assessments carried out by CCZ. As examples, the proportion of households with available water decreased from 44.1% to 40%, while the proportion of households with available food increased from 37.2% to 39.7%.

Rodent trapping

During the three trapping campaigns performed in valleys 1, 2, and 3 (one pre- and two post-intervention), 126 individuals of Rattus norvegicus and two of Rattus rattus were trapped in the 40 household points sampled in the three valleys (see the summary of the sampling in Table 2). During the pre-intervention campaign, trapping effort was 310 trap nights which resulted in 43 R. norvegicus caught at 23 points. During the campaign performed three months after the intervention 47 R. norvegicus and 2 R. rattus were trapped at 21 points out of an effort of 312 trap nights. Finally, in the campaign that was carried out six months after the intervention, 36 R. norvegicus were trapped at 24 points, with a trapping effort of 324 trap nights. In this case, only R. norvegicus data was analyzed.

Table 2

Summary of population characteristics of R. norvegicus before and after chemical intervention in Pau da Lima, Salvador, Brazil.

	Pre-intervention		Post-intervention
	Oct-Dec 2014		Nov–Dec 2015		Apr-May 2016
Valleys	1 and 3	2	1 and 3	2	1 and 3	2
No. of rats	31	12	32	15	25	11
Males	20 (64.5)	7 (58.3)	16 (50)	7 (46.7)	10 (40)	4 (36.4)
Females	11 (35.4)	5 (41.7)	16 (50)	8 (53.3)	15 (60)	7 (63.6)
Mass mean (SMI)	264.9	248.9	212.2	211.9	237.8	221.5
Males	272.5	242.5	195.2	212.1	206.6	221.9
Females	251.2	257.8	229.2	211.6	258.6	221.3
Age mean (days)	82.77	108.9	91.67	81.2	84.25	99
Males	87.82	97.61	98.55	89.2	90.89	131
Females	69.03	124.8	84.8	74.1	79.84	80.6
No. sexually active males	17 (85)	7 (100)	14 (87.5)	7 (100)	8 (53.3)	4 (100)
No. pregnant rats *	3 (37.5)	2 (100)	9 (75)	4 (80)	4 (50)	1 (25)
Lactating rats*	6 (75)	0 (0)	8 (66.7)	2 (25)	5 (62.5)	3 (75)
Pregnant lactating	2 (33.3)	0 (0)	5 (62.5)	1 (50)	1 (20)	0 (0)
Trap Success	0.194	0.148	0.196	0.167	0.164	0.127

*Considering only sexually active females. The values in brackets are represented in percentages (%).

The rodent infestation index (PII) as evaluated at the 40 points did not show a reduction between the period before the intervention and the two periods after the chemical intervention (70%, 72%, and 65%, respectively). No significant differences were identified in mean trapping success between valleys or between pre-and post-intervention campaigns (Table 2).

As no pre and post-intervention differences were identified between valleys, and the chemical intervention was not homogeneous in the study area (Fig 3), trapping success comparisons were performed by point to evaluate if the number of baits had an influence on the trapping success at each point. The beta regression pointed to there being no relation between the number of baits applied and the capture success at each point. However, a positive relationship between capture success and three applications of contact powder was detected, but the number of houses that received the three applications of this rodenticide was very low (12.3%) and also had a greater number of signs of rodent activity (see the supplementary material).

Fig 3

Valley 1 and Valley 3 respectively: a and h) kernel Density Estimation (KDE) of the 1st rodent capture campaign, b and i) Kernel Ratio (KR) of closed households by the total of households, c and j) KR of households that need rodenticide control by evaluated households, d and l) KR of rodenticide application in households by households that need for rodent control, e and m) KDE of the 2nd rodent capture campaign, f and n) KDE of the 3rd rodent capture campaign, g and o) Polygons of hot areas of households that need for rodent control identified by agents of the Zoonoses Control Center of Salvador, Bahia, Brazil.

In valley 1 and valley 3 the upper part of the area was not shown in some figures due to the absence of rodent capture campaign at this site.

The demographic characteristics of the pre-and post-intervention rodent populations are described in Table 2. Among males, 88.8% (56/63) were sexually active, in females at least one characteristic of sexual activity was observed in 61.9% of animals (39/63). Of the sexually active females, 58.9% (23/39) were pregnant and 61.5% (24/39) were lactating (37.5% of the lactating females were also pregnant). Among the pregnant females, the mean number of embryos was 11 (IQR 9–12).

No significant differences were found in the proportion between males and females, between the valleys (which received the chemical intervention or not) and between the pre and post-intervention campaigns (p> 0.05). There was no significant difference detected by the ANOVA with permutations in mean SMI between campaigns or valleys (Iter = 869, P = 0.485). In addition, no significant differences were detected in the number of embryos per pregnant female (ANOVA with permutation, Iter = 1.113, P = 0.354), the mean age (Iter = 1.091, P = 0.365) or average age of pregnant females (valleys 1 and 3: Iter = 368, P = 0.527, 2: iter = 295, P = 0.152) between campaigns and valleys with and without chemical intervention. Finally, no significant result was found for the variation in the proportion of pregnant females between campaigns and valleys with or without intervention (p> 0.05).

Spatial distribution rodenticide application analysis

Rodenticide intervention is designed to reduce abundance, with the success of rodent capture as a proxy. However, it was observed in the valleys 1, 2, and 3 that there was no influence of rodenticide application on the areas with a concentration of rodent captures. In sites 1 and 3, the areas of rodenticide application (Fig 3C and 3J) overlapped some of the concentration areas of the first campaign, which preceded the intervention. After the intervention, there was no observed reduction of areas with the concentration of capture of rodents. In site 2, a new area of capture concentration appeared in the two campaigns after the intervention, even with a record of application of rodenticide in this location (Fig 3C). Only in site three was it observed that after the intervention there was a reduction of rodent capture concentration. The closed households reached 37.8% in valley 1 (107/283) and 30.0% in valley 3 (197/656) and formed spatial aggregation in the area (Fig 3B and 3I), which may have influenced the intervention process.

Discussion

Few previous studies have reported data on the efficiency of control programs in affecting urban populations of rodents [17,38,39]. Additionally, none of them systematically and longitudinally assessed the quality of the intervention in terms of coverage, intensity, and the number of visits performed in each household in the studied area. In the present study, no changes were identified in the demographic characteristics of R. norvegicus populations before and after a chemical intervention carried out by the CCZ in Pau da Lima. Also, most of the spatial analyses did not present a negative influence of areas with the application of rodenticide on the areas with a successful capture. Likewise, we did not observe differences in the proportion of infested households during the application of rodenticide in a sample of the households evaluated. These findings contrast with the results obtained by the CCZ in the same intervention, which indicated a reduction in the proportion of households with signs of rodent infestation of 54% among all households evaluated during the intervention. The agents of the Zoonoses Control Center identified households that required intervention against rodents. This identification generated clusters and part of these clusters overlap areas with a concentration of rodents (Fig 3). The main challenges encountered by the CCZ during the implementation of the intervention were moderate household access (67.5%) and the low completion rate (12.3%) of the protocol, which includes three applications of rodenticide among the households in need of intervention. These results are key to identifying the barriers that are behind the low effectiveness in the control of synanthropic rodents and the prevention of zoonotic diseases, such as leptospirosis, in Brazil and other developing countries.

The relative abundance of rodents did not show significant differences between pre and post-intervention periods. We also did not find the correlation between the number of baits applied with the relative abundance (trap success) estimated at each evaluated point. These results are consistent with a previous study carried out in São Paulo [39], which used a similar methodology where there were no changes in the relative abundance of R. rattus after an intervention using rodenticides with the same active principles and formulations as those used in our study. In other intervention studies that combined measurements of sanitation, environmental education, and chemical control, a reduction in rodent activity/abundance was observed in poor communities in the cities of Buenos Aires (Argentina) and São Paulo (Brazil), and within ninety days the populations had not yet returned to pre-intervention levels [17,40]. In developed temperate countries where these methodologies were conducted, studies show a 50–90% decrease in the relative abundance of rodents, as reported in Baltimore [15]. This may be related to the availability of food throughout the year and high breeding rates in tropical environments where climate-related ecological limitans are relaxed [23].

In the evaluations carried out by the research group, no differences were observed in the proportion of households with signs of infestation. This contrasts with the CCZ results during the intervention, which reported a 54% drop in the proportion of infested households (from 39% to 21%). The divergence between the research group and the CCZ assessment can be attributed to differences at the time of evaluation, and the number of households evaluated.

In addition, we did not observe changes in the demographic characteristics of the population of rodents–sex ratio, age, SMI, and the number of embryos–after the chemical intervention. As far as we know, no previous study has described these population characteristics in detail in urban areas after chemical control to compare with our results. However, in Baltimore [41], when analyzing the composition of stationary, growing, and decreasing populations of R. norvegicus, it was observed that after a removal campaign, the rats of growing populations gained weight faster than those in stationary populations, even if younger. Thus, a population consisting mainly of young individuals in the post-intervention period would have been expected if control had reduced the abundance of the rodent population in Pau da Lima, which was not observed. On the other hand, no difference in sex ratio after removal was found, always remaining close to 1: 1 [41], and these results were similar to those observed in our study.

It is expected in small mammals that fecundity increases after a decrease in population density, given that fecundity and recruitment are regulated by density [42]. This phenomenon was observed in urban populations of R. norvegicus in Baltimore as an increase in pregnant females was recorded [43]. We did not observe differences in the proportion of pregnant females or the number of embryos, again suggesting that the intervention did not have an impact on population decrease. Alternatively, it would have been expected that after control, the average age of the pregnant females would decrease if the pregnancy rate remained constant [44], which was also not observed in our study.

The CCZ’s chemical control campaign faced challenges common to other community-based strategies for controlling vectors or reservoirs [45]. The proportion of inspected households was moderate (67.6%). One of the main limitations of the program was the number of inaccessible houses (32.4%). The proportion of houses where intervention was not conducted is similar to that reported in rodent control programs elsewhere in Brazil, in São Paulo and Recife (Masi personal communication) and dengue control programs worldwide [46-48]. The visiting hours of the control agents were restricted to the morning shift, excluding the possibility that homes, where people were not available during this period, would receive visits from the agents and, consequently, the applications of rodenticides. The present study records 32.3% (304/939) of closed residences, which were not evaluated and formed areas of concentration (Fig 3b & 3i), which may influence intervention actions. Changes in visiting times/schedules may be an alternative to increase program efficiency [46]. However, the logistics and cost of these changes should be considered.

Another fact observed in this study was the heterogeneous distribution of rodenticide application in the study area (Fig 2B). Some areas received more coverage while others had little or no treatment. Additionally, only 12% of the households that had signs or environmental characteristics associated with rodent infestation had the complete intervention scheme (3 applications of rodenticide). Unfortunately, previous studies do not report data on the completeness of reported interventions [17,39].

There are several factors that might be playing into the lack of success of the interventions as rodent population control: baits might be poorly deployed, could be eliminated from the environment (either by being picked up by an animal or human, or displaced by environmental factors) or the baits chosen are being ignored due to the presence of more palatable foodstuffs [49], or due to neophobic/bait resistance behaviors [50,51]. It is also possible that local rodent populations are developing resistance against rodenticides [52]. It is possible that baits might lose palatability with time and exposure to the elements [53].

This study has important limitations. The measurements used to evaluate the effectiveness of the program by the research team and CCZ are necessarily comparable, however, since it is known that there is no relevant correlation between relative abundance estimated by capture methods and infestation levels evaluated by signs [54]. In addition, the pre-intervention evaluation carried out by the research group was conducted seven months before the chemical intervention campaign. Changes in environment or rodent population could have occurred between the pre-intervention evaluation and intervention. However, there was a prior expectation that the rodent population is temporally stable in the study area, in the absence of interventions [23].

The evidence presented above suggests that the current intervention design based mainly on rodenticide application was not efficient in reducing the abundance of rodents in this community. It is known that the methodology proposed by the Ministry of Health to control rodents includes, in addition to the application of rodenticides, measures of sanitation and environmental management as well as educational actions [12,13]. However, due to the operational restrictions faced by most of the rodent control programs in Brazil and other countries, this control is done using mainly chemical rodenticides, which has not been an effective long-term solution [5]. The use of rodenticides often is used as a short-term measure against the lack of structure to fight against neglected diseases, such as leptospirosis [21]. In addition, inefficient use of rodenticides may generate a condition for rapid reproduction by surviving rodents and the perpetuation of genetic qualities associated with resistance [55]. In addition, educational actions, without environmental improvements, also do not seem to have an effect on residents’ practices. In this study, even with the recommendations of endemic agents during the intervention, the availability of water and food for rodents remained the same before and after educational actions.

The control of rodents in urban areas of Brazil faces a series of restrictions, mainly operational, such as a lack of human and financial resources that do not allow the maintenance of routine actions. These programs are negatively affected by outbreaks of other more visible diseases such as dengue and Zika. Other challenges are the political instability that leads to rotation in leadership positions in the control bodies, the difficulty of access to some places with high levels of violence and drug trafficking, the lack of access and cleaning of vacant lots and houses, and the lack of a continuous training plan for CCZ. These difficulties can be overcome in part by the availability of technical support related to control methods and local ecological factors, which is accessible to endemic agents working in the field, and by carrying out integrated activities with the implementation of sanitation infrastructure.

The importance of a programmatic approach to urban rat population management that incorporates long-term planning, programming, data management, and mapping capabilities is clearly needed to address the task of rodent control [56-58]. The CCZ of Salvador, for example, does not have a team using Geographic Information System (GIS) tools, important in the planning and management of actions for this type of entity. For these purposes, municipal management of zoonoses control must be closely associated with pest management science, and efforts must be made at the political and administrative levels to ensure that zoonoses control programs are associated with universities and research institutions and can be better designed, implemented, and sustained. Some of the few examples in the literature of successful control programs were coordinated in Budapest, Hungary [59,60], and Kuwait [61], and all emphasized the importance of investments in sanitation and environmental management in conjunction with the rodent control measures mentioned above.

Evaluation of the effectiveness of control programs on the population of rodent reservoirs of zoonoses is essential for the prevention of human diseases affecting the most vulnerable communities in developing countries. These evaluations allow the identification of barriers in rodent control programs as in the coverage pattern and intensity of the interventions, allowing direct actions to overcome these limitations. Interventions to improve environmental deficiencies, for example, closing open sewers, and implementing garbage collection, should also be integrated into chemical control, since in these areas only chemical control seems insufficient to reduce the abundance of rodents and consequently, the transmission of rodent-borne zoonosis.

(DOCX)

Click here for additional data file.

8 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Bi-Song Yue, Ph.D

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We suggest you thoroughly copyedit your manuscript for language usage, spelling, and grammar. If you do not know anyone who can help you do this, you may wish to consider employing a professional scientific editing service.

Whilst you may use any professional scientific editing service of your choice, PLOS has partnered with both American Journal Experts (AJE) and Editage to provide discounted services to PLOS authors. Both organizations have experience helping authors meet PLOS guidelines and can provide language editing, translation, manuscript formatting, and figure formatting to ensure your manuscript meets our submission guidelines. To take advantage of our partnership with AJE, visit the AJE website (http://learn.aje.com/plos/) for a 15% discount off AJE services. To take advantage of our partnership with Editage, visit the Editage website (www.editage.com) and enter referral code PLOSEDIT for a 15% discount off Editage services.  If the PLOS editorial team finds any language issues in text that either AJE or Editage has edited, the service provider will re-edit the text for free.

Upon resubmission, please provide the following:

The name of the colleague or the details of the professional service that edited your manuscript

A copy of your manuscript showing your changes by either highlighting them or using track changes (uploaded as a *supporting information* file)

A clean copy of the edited manuscript (uploaded as the new *manuscript* file)

3. Thank you for stating the following in your Competing Interests section:

"no competing interests"

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors evaluated the success of a chemical intervention of the CCZ (Center for Control of Zoonosis, Salvador de Bahia, Brazil) to control rat infestation in a neighbourhood of 3,700 people. The intervention was justified according to potential transmission of Leptospira interrogans from rats to humans, which causes severe disease (10,000 cases in Brazil, according to authors). Giving the Brazil human population (more than 200 million) this means a prevalence of 0.05% of severe disease, and only two persons in the neighbourhood would be prone to severe disease according to that small rate. Readers do not know anything about actual prevalence of Leptospira in rats in the country, but this reduced rate of infection maybe is “a needle in a haystack”. Maybe no one in the neighbourhood studied suffers from severe leptospirosis.

The authors showed that the chemical intervention had no significant effects on rats’ populations, which can describe a possible case of resistance to rodenticides, but also some uncontrolled biases in the study. One of the main concerns of this study is the way of handling baits, since the setting scheme seems to be very exposed to other non-target animals, including humans and their pets. Baits should be placed inside boxes specially designed for rats, without the possibility of other species having access. This does not seem to be the case, and it was worrying the loss of 73% of baits without knowing their fate.

I think that there is important missing information to understand the biological cycles of both the rats and the bacteria, and their relationships with the humans. One of the main shortcomings of the study is that interventions to reduce rat infestation are not well justified. I mean, there is no information about the prevalence of Leptospirosis in the population of rats studied, which can be the trigger to perform an intervention by the Administration (CCZ). If you are fighting against a zoonosis, but you don’t know anything about the actual rates of prevalence in reservoirs, this has not very much sense. If the prevalence was low, maybe an intervention is not justified since it will be completely inoperative. If the population of reservoirs is high, but the prevalence of disease is low, the risk of infection to humans maybe low. I think that an assessment of the number of rats carrying Leptospira is necessary to know if an intervention can be justified, even solely based on the economic costs of the intervention. If you are collecting animals, you can perform tests for Leptospirosis in the rats. At least, citation of other studies in Brazil is necessary.

I would like to know why the intervention was in such a small study area (0.17 km2). There were notified many cases of Leptospirosis?. Is 3.5% of cases a Low-Medium-High prevalence of this disease? There were high previous rates of rat infestation?

Abstract: correct “…in three valleys (two )…”

Line 124: What are the 11 areas related to the three valleys finally studied? What means high risk of disease?

Line 137: use a different quotation of authors…”This intervention follows standard methodologies described by Davis, Casta (25)…”, use the form “This intervention follows standard methodologies (25)…”, change throughout the text. This form of citation is repeated sometimes and is wrong, because the number format avoids using the authors’ names, and you are using the names plus the numbers, which represents a double citation.

Line 142: Describe “abandoned”

Line 144: If questionnaires are used for the intervention…, how did you rate each questionnaire to classify them into intervention/non-intervention?

Line 154: You use two different baits? Powder and Klerat paraffin? How were the baits placed to be inaccessible to other animals? This can be a shortcoming since they were not used when potential risk to children and pets. I think that poison always needs to be placed inaccessible to humans and pets, i.e., within especial boxes only accessed by rats. If intervention is not performed because of risk, the success to reduce infestation maybe limited.

Line 175: Describe better the sampling protocol... Why 10 months before intervention? What is the normal breeding cycle in rats?

The protocol is very confusing, since it is formed by two different (independent) interventions. One by the CCZ which is based on detection of signs of rats, and the other performed by the authors based on trapping.

The trapping design is insufficiently described. Did you trap in the peak/low of the breeding cycle? Only two traps per site, but for how many days? Did you analyse probability of capture to deal with false negatives? At least in my country, rats are very intelligent and difficult to trap without some days of pre-baiting, with traps set during some days open with access to bait.

Line 191: what means a 10m-radius buffer? If only two traps were set, saying that they were placed 10m apart is enough, isn’t it?

Line 265: What are the reasons to use one, two, both kind of rodenticides in every household? Can this be a source of biases?

Line 274: What means that you observed high heterogeneity in the application of rodenticides? They were not applied properly?

Line 277: This is strange, you did not find the first bait in 73% of sites. This means that baits were placed without much care, and surely were found by non-target species. This can represent some kind of non-ethical praxis. Baits need to be placed in areas inaccessible to other animals and need to be found in next campaign and retired at the end of the intervention for human safety.

Line 309: “A positive relationship between capture success and three applications of contact powder was detected.” This means that when more poison used, more captures success? This has nonsense.

Figure 1: Can you include the surrounding of the valleys to have information of the geography of the whole area? May be in shaded colours?

Line 718: Table 2: Not completely seen

Reviewer #2: General Comments

The current paper is an important contribution to our understanding of the challenges associated with managing rodent populations amongst the urban poor in developing countries. The key finding is that rodenticide applications did not significantly reduce rodent populations, nor affect the demographic “machinery” of the target pest population. These are extremely important findings given the high risk of rodent borne infections for humans in urban poor villages.

At the end of the “Introduction” it would be helpful to state a hypothesis based on what you expected to find. Then this hypothesis/hypotheses should be the lead to the discussion.

The text needs to be tightened considerably; I have highlighted some examples of where text can be deleted. I am surprised by the number of grammatical errors given three of the co-authors are professors at Yale University and University of Liverpool. A revised submission needs to be carefully reviewed for grammatical accuracy by one of these co-authors.

The paper can be improved in its readership reach if there is discussion of research on rodent management in poor villages in Africa (see Taylor et al. (reference given in detailed comments)) and more recent European studies (see Walther et al 2021 https://doi.org/10.1016/j.scitotenv.2021.147520 and references therein). There also has been lot of research publications on ecologically-based rodent management since 1999. Given the ecological and demographic focus of this paper, then it may be useful to place the findings and future directions for research in this context.

Baseline rodent trapping: I am surprised that the baseline trapping was conducted in October 2014 and the first round of rodenticides was applied in July 2015. Why is there such a long gap between baseline data and application of the treatment? This is a major limitation to the study. The delay is not mentioned until line 442 during the discussions of shortcomings of the study. Even then there is no reason given for such a long delay. I am not convinced by the glib argument that “However, there was a prior expectation that the rodent population is temporarily stable in the study area, in the absence of interventions.” Moreover, it is not description of what rodent control measures, if any, were applied by residents prior to the baseline trapping.

Although the trapping protocols are covered in a previous paper, as a minimum there needs to be a description of the type and number of traps set each night, and their location. You cannot expect readers to be familiar with the previous paper published 6 years ago.

Results:

In the methods there is mention that the ‘Rodent body condition was estimated

using a "scaled mass index" (SMI)’. I did not see mention of SMI in the results.

Detailed feedback:

Line 94: ‘to less of 10%’ should be ‘to less than 10%’; ‘those population sizes’ change to ‘those populations’

Line 96: ‘a high a general’ delete the second ‘a’

Line 99: ‘and others,’ and other what? Demographic parameters?

Line 100: ‘conscripted’? please find a more suitable word.

Line 103: first use of R. norvegicus in the text therefore should be Rattus norvegicus

Line 155: entrance should be plural.

Line 160: What does ‘topic 2’ mean?

Line 193: age structur (spelling)

Line 249: evaluated (spelling)

Line 270: There is no Table 2. This information appears in Table 1!

Line 272: ‘It was possible to observe’, replace with ‘There was’

Line 274: Delete ‘as observed in valleys 1 and 3’.

Line 278: ‘Among the baits placed in the first application that were found’ change to ‘Of the baits found…’

Line 281: ‘Regarding to rodent infestation as evaluated by CCZ, the…’ change to The CCZ….

Line 298: ‘data was analized’ should be ‘data were analyzed’

Line 299: Delete ‘in the 40 points’

Line 303: There is no Table 3. This should be Table 2.

Line 335; ‘was not observed reduction of areas’ change to ‘was no observed reduction of area’

Line 336: Delete ‘in the following campaigns’ this is redundant.

Be clearer with use of ‘sites’ here and elsewhere (refer to Figure 3 if need be). For example, ‘In the site section 2,’ should simply be ‘In site 2,’

Line 342: ‘in the intervetion process’ should be ‘the intervention process’ (delete in and correct the spelling of intervention)

Line 343: ‘identifyed households that requiring’ change to ‘identified households that required’

Line 344: spelling of identification

Line 375: “than those” change to “as those”

Line 377: abundance spelling

Line 380: conducted spelling

Line 380: ‘these methodologies were contucted’ change to ‘similar methodologies…’

Line 381-383: Might there also be differences in the efficacy of the rodenticides because of increased genetic resistance. The Baltimore study was done in 1948! This point should be raised here rather that brief mention later (line 435). Another factor is possible behavioral resistance to rodenticide baits (e.g. Brunton CFA, Macdonald DW, Buckle AP (1993) Behavioural resistance toward poison baits in brown rats, Rattus norvegicus. Appl Anim Behav Sci 38:159–174.)

At some stage in the Discussion I expected a comparison of more relevant studies such as that by Taylor PJ, Arntzen L, Hayter M, Iles M, Frean J, Belmain SR. Understanding and managing sanitary risks due to rodent zoonoses in an African city: beyond the Boston Model. Integrative Zoology. 2008; 3: 38-50. https://doi.org/10.1111/j.1749-4877.2008.00072.x

Figures and Tables

- As indicated above there is no Table 2 and Table 3 is obviously Table 2.

- Table 1 requires more explanation in the title.

- The legend for Figure 3 appeared after Table 1, which was slightly confusing initially

Table 2 (not 3). I assume that the numbers in brackets are percentages.

‘Pos intervention’ change to ‘Post Intervention’

The last column of the Table is not complete in my version. For the April

sampling there are no data shown for Village 2.

Figure 3. Need to explain the introduction of ‘sites’ (1 to 4), which appear in the text.

References

There are references that are incomplete. Often there are no page numbers. Please be consistent with the use of issue numbers – some references have them, some do not. Usually, it is best not to include issue number. Please also be consistent with species names in italics.

Also, book chapters are variable in their style, and sometimes the names of the authors are incorrect. For example, the correct citation of Cavia et al 2012 is:

Cavia R, Cueto GR, Suárez OV (2012) Techniques to estimate abundance and monitoring rodent pests in urban environments. In: Integrated Pest Management and Pest Control—Current and Future Tactics, Larramendy ML, Soloneski S (editors). Rijeka: InTech, pp 147–172.

Given that PLOS ONE does minimal editing of accepted manuscripts, it is essential to pay particular attention to the required detail of references.

Reviewer #3: This manuscript aims to assess the effectiveness of rodenticide application campaigns in infested households in poor urban areas of San Salvador, Bahia.

The information come from two sources: the CCZ and the research team, which only performed infestation evaluation and captures for data about rats populations, as abundance and population structure, but did not designed the control application.

In general the manuscript is clear, the methods adequate and the results and discussion are consistent. Results are novel and relevant to public health concerning the transmission of leptospirosis and other zoonoses by rats.

I have some comments and doubts:

1- It is not clear when was conducted the pre infestation evaluation by CCZ, and why the capture evaluation was conducted 10 months before the intervention. Along 10 months rat populations may have changed their abundance, and in consequence the comparison according to the effect of chemical application is not valid. The authors, however, took this fact in account and discuss it.

2- Lines 216-220 are not clear.

3- It is not clearly stated how was defined “the need of rodenticide application”

4- line 344. It must be identification instead of indetification.

5- Is available information about the prevalence of rodenticide resistance in rats populations of the area?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

30 May 2022

We thank the reviewers for their thoughtful and constructive comments and suggestions. We are submitting our responses along with a revised manuscript, "Evaluation of the impact of chemical control on the ecology of Rattus norvegicus of an urban community in Salvador, Brazil", for consideration as a research article to be published in PLOS ONE.

Again, we appreciate your consideration in reviewing our manuscript and please contact me if there are any questions on the preparation of the revision.

Sincerely,

Caio Graco Zeppelini,

Corresponding Author

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

Reviewer #1: The authors evaluated the success of a chemical intervention of the CCZ (Center for Control of Zoonosis, Salvador de Bahia, Brazil) to control rat infestation in a neighbourhood of 3,700 people. The intervention was justified according to potential transmission of Leptospira interrogans from rats to humans, which causes severe disease (10,000 cases in Brazil, according to authors). Giving the Brazil human population (more than 200 million) this means a prevalence of 0.05% of severe disease, and only two persons in the neighbourhood would be prone to severe disease according to that small rate. Readers do not know anything about actual prevalence of Leptospira in rats in the country, but this reduced rate of infection maybe is “a needle in a haystack”. Maybe no one in the neighbourhood studied suffers from severe leptospirosis.

The authors showed that the chemical intervention had no significant effects on rats’ populations, which can describe a possible case of resistance to rodenticides, but also some uncontrolled biases in the study. One of the main concerns of this study is the way of handling baits, since the setting scheme seems to be very exposed to other non-target animals, including humans and their pets. Baits should be placed inside boxes specially designed for rats, without the possibility of other species having access. This does not seem to be the case, and it was worrying the loss of 73% of baits without knowing their fate.

I think that there is important missing information to understand the biological cycles of both the rats and the bacteria, and their relationships with the humans. One of the main shortcomings of the study is that interventions to reduce rat infestation are not well justified. I mean, there is no information about the prevalence of Leptospirosis in the population of rats studied, which can be the trigger to perform an intervention by the Administration (CCZ). If you are fighting against a zoonosis, but you don’t know anything about the actual rates of prevalence in reservoirs, this has not very much sense. If the prevalence was low, maybe an intervention is not justified since it will be completely inoperative. If the population of reservoirs is high, but the prevalence of disease is low, the risk of infection to humans maybe low. I think that an assessment of the number of rats carrying Leptospira is necessary to know if an intervention can be justified, even solely based on the economic costs of the intervention. If you are collecting animals, you can perform tests for Leptospirosis in the rats. At least, citation of other studies in Brazil is necessary.

I would like to know why the intervention was in such a small study area (0.17 km2).

There were notified many cases of Leptospirosis?. Is 3.5% of cases a Low-Medium-High prevalence of this disease? There were high previous rates of rat infestation?

Line 124: What are the 11 areas related to the three valleys finally studied? What means high risk of disease?

Response: the 11 areas represent priority points detected in the entire territory of Salvador, of which the community here studied is one of them. High risk areas were detected based on the historical records of incidence, as well as the occurrence of the main risk exposure phenomena (such as rat complaints, floods). The text has been amended to better convey the information.

Line 142: Describe “abandoned”

Response: abandoned in the literal dictionary sense, the houses are no longer inhabited or sought after as such.

Line 154: You use two different baits? Powder and Klerat paraffin? How were the baits placed to be inaccessible to other animals? This can be a shortcoming since they were not used when potential risk to children and pets. I think that poison always needs to be placed inaccessible to humans and pets, i.e., within especial boxes only accessed by rats. If intervention is not performed because of risk, the success to reduce infestation maybe limited.

Response: this was part of the internal protocol of the CCZ and was, at the time of the study, beyond the realm of our possibilities to propose any alterations. This is an aspect that has been part of the discussion between our group and the CCZ in order to design better interventions

Line 144: If questionnaires are used for the intervention…, how did you rate each questionnaire to classify them into intervention/non-intervention?

Response: areas of intervention were classified according to the performance of the CCZ, the chemical intervention conducted by the CCZ was carried out in two areas (valleys 1 and 3). Valley 2 was classified as control (non-intervention), as this area would not have CCZ action. Baseline infestation rates were carried out in all areas, but in the intervention areas this survey was carried out by the CCZ and in the control area it was carried out by our research team using questionnaires standardized for evaluating Building Infestation Index. The same has variables raised by the CCZ.

Line 175: Describe better the sampling protocol... Why 10 months before intervention? What is the normal breeding cycle in rats?

The protocol is very confusing, since it is formed by two different (independent) interventions. One by the CCZ which is based on detection of signs of rats, and the other performed by the authors based on trapping.

The trapping design is insufficiently described. Did you trap in the peak/low of the breeding cycle? Only two traps per site, but for how many days? Did you analyse probability of capture to deal with false negatives? At least in my country, rats are very intelligent and difficult to trap without some days of pre-baiting, with traps set during some days open with access to bait.

Response: We restructured the text according to the comments for a better understanding of the text.

Line 191: what means a 10m-radius buffer? If only two traps were set, saying that they were placed 10m apart is enough, isn’t it?

Response: the traps are places somewhere within a 10m buffer, with no indication of the distance between traps, as is in the text: “Tomahawk traps were installed in each household within a 10-meter radius buffer”,

Line 265: What are the reasons to use one, two, both kind of rodenticides in every household? Can this be a source of biases?

Response: These are decisions made internally by the CCZ team, beyond of our team’s scope of authority. The purpose of applying both contact powder and provide poison blocks is to try and maximize the chances of exposing the rats to a lethal dose of rodenticide (powder to try and reach those who might not reach or might display neophobia to the blocks, and the blocks for those residing outside of the area applied. While we recognize that, at a finer level of analysis, this could introduce a lot of noise to the assessment of the effects of rodenticides in particular, this limitation has been taken into account on our analytical design.

Line 274: What means that you observed high heterogeneity in the application of rodenticides? They were not applied properly?

Response: The high heterogeneity of rodenticide applications cited means that some areas received greater coverage while others had little or no rodenticide treatment. Only 12% of the households that had environmental signs or characteristics associated with rodents had the complete intervention schedule (which includes 3 applications of rodenticide). These differences are due to the difficulty of access by CCZ agents to some households, refusal of residents due to the presence of domestic animals in the house, or absence of a resident at home at the moment of the visit. In addition, sites within the entire study area remained inaccessible due to drug trafficking-related violence.

Line 277: This is strange, you did not find the first bait in 73% of sites. This means that baits were placed without much care, and surely were found by non-target species. This can represent some kind of non-ethical praxis. Baits need to be placed in areas inaccessible to other animals and need to be found in next campaign and retired at the end of the intervention for human safety

Response: we recognize this as a problem. However, observe the previous answers regarding the CCZ protocols. While there are provisions in their protocol to not apply rodenticide blocks in areas where there are children, pets or animal rearing (such as people raising hens) to minimize the risk of contact with non-targets, we do consider that a part of the baits might be removed as bycatch (likely by possums), or more likely being washed out by torrential rain.

Line 309: “A positive relationship between capture success and three applications of contact powder was detected.” This means that when more poison used, more captures success? This has nonsense.

Response: this has been observed in other studies we have conducted. Several possibilities for the phenomenon have been considered, and our current stance is that the relationship observed is not causal, but the areas being baited are the areas with higher rat presence/activity.

Figure 1: Can you include the surrounding of the valleys to have information of the geography of the whole area? May be in shaded colours?

Response: We agree with your suggestion, however, when including the surroundings of the shaded valleys to have information on the geography of the entire area, we did not obtain a quality view of the area, we decided to keep the figure with the areas of the valleys of studies highlighted in another figure.

Abstract: correct “…in three valleys (two )…”

Line 137: use a different quotation of authors…”This intervention follows standard methodologies described by Davis, Casta (25)…”, use the form “This intervention follows standard methodologies (25)…”, change throughout the text. This form of citation is repeated sometimes and is wrong, because the number format avoids using the authors’ names, and you are using the names plus the numbers, which represents a double citation.

Line 718: Table 2: Not completely seen

Response: corrections were made to the text as pointed.

Reviewer #2: General Comments

The current paper is an important contribution to our understanding of the challenges associated with managing rodent populations amongst the urban poor in developing countries. The key finding is that rodenticide applications did not significantly reduce rodent populations, nor affect the demographic “machinery” of the target pest population. These are extremely important findings given the high risk of rodent borne infections for humans in urban poor villages.

At the end of the “Introduction” it would be helpful to state a hypothesis based on what you expected to find. Then this hypothesis/hypotheses should be the lead to the discussion.

The text needs to be tightened considerably; I have highlighted some examples of where text can be deleted. I am surprised by the number of grammatical errors given three of the co-authors are professors at Yale University and University of Liverpool. A revised submission needs to be carefully reviewed for grammatical accuracy by one of these co-authors.

The paper can be improved in its readership reach if there is discussion of research on rodent management in poor villages in Africa (see Taylor et al. (reference given in detailed comments)) and more recent European studies (see Walther et al 2021 https://doi.org/10.1016/j.scitotenv.2021.147520 and references therein). There also has been lot of research publications on ecologically-based rodent management since 1999. Given the ecological and demographic focus of this paper, then it may be useful to place the findings and future directions for research in this context.

Baseline rodent trapping: I am surprised that the baseline trapping was conducted in October 2014 and the first round of rodenticides was applied in July 2015. Why is there such a long gap between baseline data and application of the treatment? This is a major limitation to the study. The delay is not mentioned until line 442 during the discussions of shortcomings of the study. Even then there is no reason given for such a long delay. I am not convinced by the glib argument that “However, there was a prior expectation that the rodent population is temporarily stable in the study area, in the absence of interventions.” Moreover, it is not description of what rodent control measures, if any, were applied by residents prior to the baseline trapping.

Although the trapping protocols are covered in a previous paper, as a minimum there needs to be a description of the type and number of traps set each night, and their location. You cannot expect readers to be familiar with the previous paper published 6 years ago.

Response: We restructured the text according to the comments for a better understanding of the text.

Results:

In the methods there is mention that the ‘Rodent body condition was estimated

using a "scaled mass index" (SMI)’. I did not see mention of SMI in the results.

Response: the SMI is used for the population characterization, see table 2.

Line 381-383: Might there also be differences in the efficacy of the rodenticides because of increased genetic resistance. The Baltimore study was done in 1948! This point should be raised here rather that brief mention later (line 435). Another factor is possible behavioral resistance to rodenticide baits (e.g. Brunton CFA, Macdonald DW, Buckle AP (1993) Behavioural resistance toward poison baits in brown rats, Rattus norvegicus. Appl Anim Behav Sci 38:159–174.)

Response: we have added a passage on behavioral resistance on the baits.

At some stage in the Discussion I expected a comparison of more relevant studies such as that by Taylor PJ, Arntzen L, Hayter M, Iles M, Frean J, Belmain SR. Understanding and managing sanitary risks due to rodent zoonoses in an African city: beyond the Boston Model. Integrative Zoology. 2008; 3: 38-50. https://doi.org/10.1111/j.1749-4877.2008.00072.x

Line 99: ‘and others,’ and other what? Demographic parameters?

Response: We restructured the text according to the comments for a better understanding of the text.

Detailed feedback:

Line 94: ‘to less of 10%’ should be ‘to less than 10%’; ‘those population sizes’ change to ‘those populations’

Line 96: ‘a high a general’ delete the second ‘a’

Line 100: ‘conscripted’? please find a more suitable word.

Line 103: first use of R. norvegicus in the text therefore should be Rattus norvegicus

Line 155: entrance should be plural.

Line 160: What does ‘topic 2’ mean?

Line 193: age structur (spelling)

Line 249: evaluated (spelling)

Line 270: There is no Table 2. This information appears in Table 1!

Line 272: ‘It was possible to observe’, replace with ‘There was’

Line 274: Delete ‘as observed in valleys 1 and 3’.

Line 278: ‘Among the baits placed in the first application that were found’ change to ‘Of the baits found…’

Line 281: ‘Regarding to rodent infestation as evaluated by CCZ, the…’ change to The CCZ….

Line 298: ‘data was analized’ should be ‘data were analyzed’

Line 299: Delete ‘in the 40 points’

Line 303: There is no Table 3. This should be Table 2.

Line 335; ‘was not observed reduction of areas’ change to ‘was no observed reduction of area’

Line 336: Delete ‘in the following campaigns’ this is redundant.

Be clearer with use of ‘sites’ here and elsewhere (refer to Figure 3 if need be). For example, ‘In the site section 2,’ should simply be ‘In site 2,’

Line 342: ‘in the intervetion process’ should be ‘the intervention process’ (delete in and correct the spelling of intervention)

Line 343: ‘identifyed households that requiring’ change to ‘identified households that required’

Line 344: spelling of identification

Line 375: “than those” change to “as those”

Line 377: abundance spelling

Line 380: conducted spelling

Line 380: ‘these methodologies were contucted’ change to ‘similar methodologies…’

Figures and Tables

- As indicated above there is no Table 2 and Table 3 is obviously Table 2.

- Table 1 requires more explanation in the title.

- The legend for Figure 3 appeared after Table 1, which was slightly confusing initially

Table 2 (not 3). I assume that the numbers in brackets are percentages.

‘Pos intervention’ change to ‘Post Intervention’

The last column of the Table is not complete in my version. For the April

sampling there are no data shown for Village 2.

Response: We restructured the text according to the comments for a better understanding of the text. We have modified the sentences following the reviewer’s suggestion.

Figure 3. Need to explain the introduction of ‘sites’ (1 to 4), which appear in the text.

Response:We have modified following reviewer’s suggestion.

References

There are references that are incomplete. Often there are no page numbers. Please be consistent with the use of issue numbers – some references have them, some do not. Usually, it is best not to include issue number. Please also be consistent with species names in italics.

Also, book chapters are variable in their style, and sometimes the names of the authors are incorrect. For example, the correct citation of Cavia et al 2012 is:

Cavia R, Cueto GR, Suárez OV (2012) Techniques to estimate abundance and monitoring rodent pests in urban environments. In: Integrated Pest Management and Pest Control—Current and Future Tactics, Larramendy ML, Soloneski S (editors). Rijeka: InTech, pp 147–172.

Given that PLOS ONE does minimal editing of accepted manuscripts, it is essential to pay particular attention to the required detail of references.

Response: the references have their information retrieved by DOI through EndNote, and contain the standard information deposited in the DOI registry, and presented under the guidelines of the Vancouver style. Information for items without DOI were extracted in its entirety from the original files. The references were re-checked on their data, and updated.

Reviewer #3: This manuscript aims to assess the effectiveness of rodenticide application campaigns in infested households in poor urban areas of San Salvador, Bahia.

The information come from two sources: the CCZ and the research team, which only performed infestation evaluation and captures for data about rats populations, as abundance and population structure, but did not designed the control application.

In general the manuscript is clear, the methods adequate and the results and discussion are consistent. Results are novel and relevant to public health concerning the transmission of leptospirosis and other zoonoses by rats.

I have some comments and doubts:

1- It is not clear when was conducted the pre infestation evaluation by CCZ, and why the capture evaluation was conducted 10 months before the intervention. Along 10 months rat populations may have changed their abundance, and in consequence the comparison according to the effect of chemical application is not valid. The authors, however, took this fact in account and discuss it.

2- Lines 216-220 are not clear.

3- It is not clearly stated how was defined “the need of rodenticide application”

4- line 344. It must be identification instead of indetification.

5- Is available information about the prevalence of rodenticide resistance in rats populations of the area?

Response: No information on the subject is available.

Submitted filename: carta_revisao.docx

Click here for additional data file.

14 Jun 2022

Evaluation of the impact of chemical control on the ecology of Rattus norvegicus of an urban community in Salvador, Brazil

PONE-D-21-36428R1

Dear Dr. Zeppelini,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Bi-Song Yue, Ph.D

Academic Editor

PLOS ONE

27 Jun 2022

PONE-D-21-36428R1

Evaluation of the impact of chemical control on the ecology of Rattus norvegicus of an urban community in Salvador, Brazil

Dear Dr. Zeppelini:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Bi-Song Yue

Academic Editor

PLOS ONE