Assessing trends in rat populations in urban and non-urban environments in the Netherlands

Abstract

Rats in urban areas pose health risks as they can transmit various zoonotic pathogens. Monitoring rat populations in urban areas is therefore a key determinant in risk assessments for taking adequate control and preventive measures. However, large-scale and long-term monitoring of rat populations is labor-intensive and time consuming. The aim of this study was to develop a low-cost and low-time- consuming method to gain insight into the trends of rat populations in urban and non-urban environments in the Netherlands, and to identify potential drivers of these trends. From 2014 to 2018, local municipalities or their pest control organizations voluntarily submitted quarterly overviews of rat nuisance reports in urban areas. For non-urban areas, a nationwide record of reported bycatch species from the muskrat control was used to assess a potential trend. To identify potential drivers of observed trends, employees of nine municipalities were interviewed. Rat nuisance reports from 25 municipalities were analyzed. An increasing trend in rat nuisance reports was observed in 12, a decreasing trend in 3 and no trend in 10 municipalities. In non-urban areas, no trend in the bycatch of rats was detected. The increase in rat nuisance reports was associated with a large municipality resident size. No consistent drivers could be identified, but potential drivers were discussed in the interviews. Although it was not possible to quantify their influence on the rat population trends seen, they provide direction for future studies on drivers of rat populations.

INTRODUCTION

Brown rats (Rattus norvegicus) and black rats (Rattus rattus) are among the most widespread animals of the world (Feng and Himsworth 2014). Rats have been associated with the spread of diseases for centuries, for example with the black death. Nowadays, rats are more prominent carriers of zoonotic pathogens such as Leptospira spp., Streptobacillus monoliformis, Seoul orthohantavirus and Rickettsia typhi (Meerburg et al. 2009; Himsworth et al. 2013b; Firth et al. 2014). Urban rats live in close proximity to humans, which facilitates the transmission of zoonotic pathogens. The public health risk depends on the exposure to zoonotic pathogens. To assess this exposure and thus the public health risk, both knowledge about the presence of zoonotic pathogens is needed, as well as about the population densities and dynamics. The rat-borne zoonotic pathogens are increasingly studied (Himsworth et al. 2013a; Firth et al. 2014), but information about rat populations in urban areas is scarce (Parsons et al. 2017; Byers et al. 2019).

Brown rats concentrate in urban environments, where the population is heavily influenced by human factors, e.g. through food availability and pest management, and by non-human factors such as climate, e.g. through different survival and reproduction rates in relatively warm or cold winters (Andrews et al. 1972; Vadell et al. 2014). Assessment of the size of rat populations is costly on a large scale, but reports of rat nuisance may serve as a proxy for the rat population in an area, as was recently studied in Chicago, USA (Murray et al. 2018). In the Netherlands, no central registry of rat infestations exists and municipalities are responsible for rat control in public areas. Information about rat populations on a larger time scale is currently largely based on anecdotal information from professional pest control technicians from municipalities or pest control organizations (PCOs). They often lack the overview on a larger geographical scale, which confuses the overall picture of the dynamics of rat populations. Consequently, no nationwide overview is available of trends in (urban) rat populations, nor is it clear why certain cities or areas experience more nuisance than others.

To gain more insight into the densities and to follow trends of rat populations in Dutch municipalities, the National Institute for Public Health and the Environment (RIVM) started a monitoring of rat nuisance reports in a limited number of municipalities, as a proxy for rat populations. Of these municipalities, nine were selected for interviews to assess potential drivers for the trends that were seen. Furthermore, rat populations in non-urban areas were monitored by the use of bycatch data from muskrat control to assess whether trends would be similar for non-urban areas. If so, this could offer information about the underlying causes of population trends. The data were combined to investigate trends in the rat populations in the Netherlands.

METHODS

Urban rat monitoring

From 2014 to 2018, a voluntary urban rat monitoring program was coordinated by the RIVM. A number of the participating municipalities were suggested by regional health officers throughout the Netherlands, complemented with selected municipalities to achieve an equal spread of small and large municipalities across the Netherlands. Three municipalities referred us to their contracted PCO. One of these PCOs not only reported information from the initial municipality but also included other municipalities (see Fig. 1).

All of the municipalities and PCOs were represented by a professional in charge of pest control. The organization of pest control varies between municipalities and thus these professionals varied from a city ecologist, the head of the municipal pest control department, pest control technician (municipal or commercial), to municipal field employees. The professional was requested to send in an overview of nuisance reports to RIVM every 3 months. The nuisance reports usually originated from citizens or commercial businesses. A nuisance report was logged by the municipalities and PCOs into the database only when a professional pest control technician confirmed the report on-site to ensure identification of the correct rat species and to ensure an accurate estimated size of the reported population. A standardized reporting sheet was used, which contained the following variables: date, location (based on the four-digit postal code), rat species (brown/black/unknown), estimate of the size of the reported population (<10 rats, 10–25 rats and >25 rats) and the location type (private ground/business/public space). These variables were determined by the professional pest control technician.

Interviews

After analysis of the rat nuisance reports from 2014 to 2017, nine municipalities were selected based on trends of the size of the rat population: three increasing, three decreasing and three stable trends. Preference was given to municipalities with high numbers of reports to reduce the effects of chance (e.g. high numbers due to media attention or building activities).

A descriptive study was performed with an emphasis on understanding the policies of the selected municipalities with regards to rat nuisance. Interviews were held with the representatives of the municipalities that also sent in the quarterly nuisance reports. In the interviews, the main questions concerned changes that had taken place in the past 4 years in the municipality regarding habitat, rat control policy and rat nuisance reporting. Semi-structured interviews were conducted, using a standard list of topics and questions and a list of subquestions/-topics that could be used for detailed questioning. Topics included the nuisance and reports, written policies, control, prevention, citizens, municipal activities and personal explanations of the trend seen in their municipality. An overview of the questions that were used to structure the interviews can be found in Supplement 1.

Additionally, from each municipality, an area was visited that had relatively many rat nuisance reports. This was done to confirm the information retrieved during the interview (e.g. confirm whether litter is a problem in the area) and to assess potential similarities between areas with the same rat population trends. The area was scanned using a checklist, including features such as type of buildings, waste disposal policy, demography etc. The various topics were compared between municipalities to gain insight into potential explanations of why different trends in rat populations were seen. Finally, it was assessed what associations could be found between changes in policies of municipalities and the trends in the reported rat nuisance.

Non-urban rat monitoring

In the Netherlands, the invasive muskrat (Ondatra zibethicus) and nutria (Myocastor coypus) populations are controlled nationwide to protect the dikes. The control activities result in bycatch of a variety of species, including brown rats. Bycatch of the brown rat is regarded as beneficial, due to its status as a pest animal. Registration of all bycatch is mandatory and is done in 5 × 5 km squared areas. An overview of the brown rat bycatch data from 2009 to 2018 was purchased. The bycatch data were corrected for the hours that a person had worked within an area.

Analysis of urban rat monitoring

We counted all nuisance reports for each municipality in each calendar quarter in the 5-year period (2014–2018) and subsequently inserted the municipality name, its resident population size, calendar year plus calendar quarter and the nuisance count to an excel database. The 5-year period was divided into 20 different time points. Our null hypothesis was that the nuisance remains constant over time, with an alternative hypothesis that the nuisance changes with time. We fitted the negative binomial distribution (R Core Team 2015) to the nuisance count using the set of 20 consecutive time points as a predicting variable and tested using the likelihood ratio test the significance of the time variable. We chose the alternative hypothesis when the P-value was <0.05 and consequently the null hypothesis was rejected. The analysis considered changes over the years but neglects the seasonality in rat reproduction and activity.

Next, we searched for an association between the rat nuisance trend and the size of the municipality resident population. This is solely an associative study because our dataset contains no information to identify a causation. We applied a generalized linear model analysis to the resident population size using the three categories of rat nuisance trends (increase, decrease, or stable) as a regressor variable. We assumed that the resident population size follows the negative binomial distribution. The significance of the rat nuisance trends over the stable model is determined by applying the likelihood ratio test. We rejected the stable model when the P-value was <0.05.

Analysis of non-urban rat monitoring

We counted in each 5 km × 5 km area all bycatch of brown rats and the working hours in the area. Subsequently, we inserted into a database the area identifier, the year of the monitoring, the bycatch count and the total working hours. Our null hypothesis was that bycatch of brown rats remains constant, and then alternative hypothesis was that bycatch of brown rats changes in frequency with the calendar year. We fitted the negative binomial distribution (R Core Team 2015) to the bycatch using the total hour and the year of monitoring as predicting variables. Bycatch of brown rats is expected to increase with the total hours. Hence, we performed the likelihood ratio test between the full model and a sub-model in which the total hours are the only predicting variable.

RESULTS

Urban rat monitoring

Fifty-two municipalities/regions were invited to join the voluntary rat monitoring program, with an equal distribution of large and small municipalities across the Netherlands. However, not all agreed to participate, or only participated for 1 or 2 years. Other municipalities missed quarterly reports, resulting in incomplete overviews. We excluded municipalities with 10 or more missing quarterly reports. Nine municipalities/PCOs had a complete set of nuisance reports for the study period and 16 had an incomplete set that could still be analyzed.

The analysis was performed on a total of 25 municipalities. Of these, 12 municipalities showed a significant rise in reports of rat nuisance over the 5-year study period and 3 showed a significant decrease. Of the remaining 10 municipalities, neither a significant increase nor decrease was detected. The results are shown in Fig. 1 (selected municipalities), Supplement 2 (total overview) and Supplement 3 (statistical results). An increase in reports of rat nuisance was significantly associated with a large municipality (P-value = 0.01: Supplement 4). The decreasing rat nuisance lacks an association with a small municipality.

Interviews and observations

The trends of rat populations in nine municipalities (Table 1) were further investigated by semi-structured interviews with professionals of these municipalities, to assess whether drivers could be detected for increasing or decreasing trends. No drivers were found that were mentioned by all interviewees, but the interviews provided direction for future studies on possible drivers, for example: built in the ‘50’s–‘60’s or old stoneware sewers, but also the relatively large green spaces and many bodies of water in new neighborhoods. The three municipalities with increasing rat nuisance reports all emphasize the green environment and green ecological connections in their cities, but this is not exclusively for these three municipalities. In two of the three municipalities with increasing numbers, waste bags next to underground containers were mentioned specifically in the interviews and were also observed on site to be an issue; however, this was not exclusively for this category.

The professionals reported very different rat control policies between municipalities, ranging from non-existent to embedding of rat control within larger programs on wildlife in the municipality and a dedicated rodent control manager. Especially in smaller communities, rat control is part of a larger work task. This is also reflected in the (lack of) effort of public education, e.g. information leaflets, advertisements in local newspapers, letters, information evenings and education at schools in several municipalities. The information regarding rat control that the municipalities display on their websites ranges from detailed to minimal. Two of the municipalities with an increasing trend and one municipality with a decreasing trend do not cover the costs of rat control, in contrast to the other municipalities. One of the municipalities with a decreasing trend stated that they stopped offering free rat control on private property within the study period, likely leading to fewer rat nuisance reports. Little collaboration is present between municipalities, except for a successful collaboration of municipalities in the south of the Netherlands regarding the control of black rats.

The three municipalities exhibiting an increasing trend all have the possibility to report by use of a mobile application, and the professionals have the opinion this increases the number of nuisance reports. The other six municipalities make use of online forms and telephone reports. It was also stated that media-attention (for both The Hague and Eindhoven) and increased citizen education had a stimulating effect on the number of reports. In Amsterdam, a public campaign in February 2018 resulted in an increased number of nuisance reports.

Ethnicity or religion of people is sometimes argued to be of influence on rat populations, due to various views on food wastage (Al-Naggar et al. 2019; Chang et al. 2019). About half of the municipalities mentioned having a mixed population consisting of different ethnic groups, but no pattern in this could be determined. Human factors, for example, the lack of cleanliness, feeding of animals and keeping pet animals were mentioned as contributing factors to the rat nuisance.

In the Netherlands, the outside use of rodenticides has been restricted since 2017. Pest control technicians should first implement preventive measures, before using rodenticides. Of the three interviewed municipalities with increasing trends, one had changed its control policy by reducing the use of rat rodenticides. Another had already limited the use of rodenticides before 2017, while one used rodenticides very frequently even after the restrictions set in 2017. When compared with the three municipalities that showed a decreasing trend, these three indicated that they had used rodenticides often or always, and now changed to preventive measures and use of rodenticides only in extreme cases when other control measures failed.

Non-urban rat monitoring

The monitoring area is a collection of 5 × 5 km unique square areas (n = 1630). We assembled 15 592 records of muskrat and nutria control events in 2009–2018. In that period, muskrat and nutria control registered a total 4 533 327 working hours and 46 976 bycatch of brown rats (Fig. 2 and Table 2). A significant positive association of the total working hour to the bycatch of brown rats was identified (P-value <0.001) in line with the expectation. A positive association of year with the bycatch of brown rats was not statistically significant (P-value = 0.29). Hence, we concluded that the bycatch of brown rats by muskrat and nutria control remained constant from 2009 to 2018.

DISCUSSION

The aim of this study was to gain more insight into the trends in the time of rat populations in the Netherlands in both urban and non-urban environments. Moreover, potential drivers of the trends were studied. In the 25 municipalities studied, 48% (12 out of 25) showed a significantly increasing trend and 12% (3 out of 25) a significantly decreasing trend. Of the remaining municipalities, almost half showed an increase, half a decrease. However, it can be discussed how well the nuisance reports, even when confirmed by professionals, reflect the true rat populations. In Chicago, this was assessed and it was found that complaints about rats were most associated with rat trap success, indicating that the use of public reporting can serve as a useful tool to identify areas of greater rat activity (Murray et al. 2018). In the interviews, it was discussed that whether a citizen will report a rat also depends heavily on the ease of reporting, with a mobile application facilitating this, and whether or not actions will be taken after reporting. Therefore, it needs to be taken into account that the trends in the nuisance reports may not reflect the actual population trends but are merely an indication. However, currently, it is our best method to monitor multiple cities longitudinally.

The study originally included 52 of the 380 Dutch municipalities (CBS 2018), but only 25 had enough quarterly reports to be analyzed. Another challenge of the monitoring was to collect the data in a uniform manner. This was accounted for by designing a reporting sheet, with clear instructions. Even so, it was noted during the study that the reporting sheet was used in different ways by the various municipality representatives, which was confirmed in the interviews. This complicated the comparison between municipalities. For example, municipalities that follow-up all rat nuisance reports, will have a high numbers of reports, in contrast to municipalities with limited actions on rat nuisance reports. Furthermore, some municipalities took the lead in control of rat populations in public and/or private space, whereas other municipalities generally referred to private control agencies, except when large infestation in public space were present. Therefore, comparisons between the municipalities should be avoided and extrapolation of the results to a national level is difficult. Trends of urban rat populations should be evaluated per municipality, with the assumption that municipalities did not change their reporting method during the monitoring period.

Site visits were made to confirm statements from the interviews and to compare the characteristic of areas with an increasing trend versus a decreasing trend of rat nuisance reports. In future studies, comparing the characteristics of areas with low and high numbers of nuisance reports would also be informative.

In 12 of the 25 municipalities, the number of rat nuisance reports increased from 2014 to 2018. The increase was associated with a large municipality resident size (Supplement 4, Table 2a and b). No other specific drivers stood out from the interviews explaining the various trends seen across the nine municipalities. Sewer type has previously been shown to be of influence for rat populations (Heiberg et al. 2012), as well as natural soil (Traweger and Slotta-Bachmayr 2005). Overall, it appears to be a sum of various drivers, of which availability of food, water and shelters are primary. Translated to the urban environment, the biodiversity and greenness of the city seem to play an important role, as well as food availability due to waste and earthen soil for burrowing.

In 2017, Integrated Pest Management was implemented in the Netherlands. This new work practice focuses on the prevention of rats, by restricting access to food, shelter, etc. and restricts the outside use of rodenticides. Some interviewees experienced this as an important factor. However, from the interviews, it appears that not necessarily the restricted use itself is a problem, but merely the change to new work practice, as there are also municipalities that have stopped using rodenticides many years ago, that do not show increasing numbers of rat nuisance reports in their municipalities. In that respect, the perception of the rat nuisance may also be misleading. For example, the pest control technician of The Hague thought the rat nuisance remained the same, but the number of complaints increased. On the other hand, Amersfoort had a non-significant increase in reports, but the interviewee experienced an increase of nuisance. Naturally, this may also reflect changes in the reporting effort of citizens, complicating interpretation of the data.

The increase in rat nuisance reporting observed in 12 out of 25 municipalities, was not identified in rural areas using the bycatch data. Apparently, other (human) drivers occur in urban areas compared to non-urban areas.

In our study, 12 out of 25 municipalities exhibited an increase in rat nuisance reports. Increasing urban rat populations may not only decrease the comfort of residents in an area but are also relevant for public health, as rats may carry various zoonotic pathogens with them. To assess the public health risk of urban rats, a closer look at pathogens and the complex pathogen–host–ecology relationship is needed. This is illustrated by the complex associations of leptospirosis with both population size (Himsworth et al. 2013a) and lethal control methods (Lee et al. 2018). The current study describes a low-cost and low-time-consuming method to collect data about trends in rat populations in municipalities based upon which informed decisions can be made when confronted with rat infestations.

Supplementary data

Supplementary data are available at JUECOL online.

ACKNOWLEDGMENTS

The authors would like to thank all participating municipalities and pest control organizations for their essential contribution to the study. We also thank Sanne van den End and Annika van Roon for their contributions to the monitoring program. This study was financed by the Ministry of Health, Welfare and Sport.

DATA AVAILABILITY

Data are available, but to a restricted detail due to the necessary permission of the involved municipality/municipalities. Readers interested in underlying data are encouraged to contact the author.

Conflict of interest statement. None declared.

REFERENCES