To Continue to Collect Ecological and Epidemiological Data from the Woodchester Park Badger

1. Background

This report describes work carried out under Defra funded project SE3032 from April 1st, 2007 to March 31st, 2008. This includes all data collection, statistical analyses and updates of some long-term demographic and epidemiological trends.

2. Objectives

To continue to collect ecological and epidemiological data from the Woodchester Park badger population, consistent with that obtained in previous years. Specifically this involved,

1. Obtaining data on the spatial configuration of badger social groups in the study area (by bait-marking).

1. Collecting data on the size, structure and infection status of the population (by capture-mark-recapture and clinical sampling).

1. The collation of collected data onto the existing Woodchester Park epidemiological and spatial databases.

1. Following consultations with Defra the following exploratory demographic and epidemiological analysis packages were also agreed to be delivered during the duration of the project

(a). Demographic Analyses

• Formal comparative assessment of approaches to population size estimation, and identification of an optimum standardised method.
• Investigation of density dependent and independent correlates of reproduction and recruitment.
• Quantification of the annual number of unmarked adult captures, and their demographic correlates.

(b) Epidemiological Analyses

• Calculation of confidence limits for prevalence and incidence estimates.
• Assessment of the evidence for cyclicity in disease trends.
• Preliminary analysis of pre-disposing factors for infection (a retrospective analysis relating infection status to condition and reproductive status).
• Duration of infection/disease progression (building on what was included in the 2007 report).
• Assessment of the evidence for transient seropositivity in cubs (a retrospective analysis).
• Description of a localised depopulation and immigration event, and its epidemiological consequences.
• Preliminary description of spatio-temporal patterns of M. bovis spoligotype occurrence in badgers and cattle.
• Consideration of potential approaches to estimating disease-induced mortality.

We also undertook to provide copies of work carried out with CSL Seedcorn funding, on refining estimates of age-dependent fecundity. The report relating to this work will be completed in June 2008, and a copy will be forwarded to Defra.

3. Extent of Delivery of Objectives

All objectives have been fully met.

4. Methods

All methods described below are detailed in Standard Operating Procedures (SOPs; see Appendix 7.1 for a complete list) which are reviewed regularly and in which all staff are trained. Below we describe the basic methodology (as in previous reports) employed to collect and collate data. Methods related to specific analysis packages (Objectives 4a & b) are described separately.

4.1. The Study Population

The Woodchester Park study area is located on the Cotswold limestone escarpment in Gloucestershire, South-west England. This includes an intensively studied core area of approximately 7 km2 where the resident badgers have been monitored in a consistent manner since 1981. This area is dominated by mixed woodland, grassland and arable (see Delahay et al., 2006a for details). Central to the study area is a steep-sided wooded valley, surrounded by agricultural grasslands used largely for grazing beef and dairy cattle. Badger main setts were relatively regularly spaced throughout the study area, following a band of exposed Cotswold sandstone on the valley sides.

4.2. Bait-Marking (Objective 1)

Bait marking was carried out each year during the trapping ‘close season’ (see below) in order that social group territories could be delineated. This involved feeding a mixture of peanuts and syrup laced with small coloured plastic pellets at active setts for up to 12 days. For the following 6-8 weeks the study area was searched for badger latrines to identify the undigested plastic in faeces (Delahay et al., 2000a). As a uniquely coloured/shaped plastic marker was fed at each sett, the marked faeces could be related to particular badger social groups and their territorial boundaries determined. The distribution of plastic pellets at latrines and field records of boundary runs were used to digitise social group territories using a geographical information system (ArcView 3.2; ESRI, 1996).

4.3. Trapping (Objective 2)

A routine capture-mark-recapture (CMR) programme provided the basis for the collection of demographic and epidemiological data on the resident badger population. The methods and level of trapping effort have been conistent since 1981 (see Rogers et al., 1997). Trapping was suspended each year during a ‘close season’ in the spring (February to April inclusive) when female badgers may have young dependent cubs that cannot be left underground for protracted periods (see also Woodroffe et al., 2005).

For the purposes of trapping, the study area was divided into three zones of approximately equal size. Each zone (and therefore each badger social group) was trapped four times during each year. A sett activity survey was conducted in each zone before each trapping event, so that the location of active setts and the number of traps to deploy, could be assessed. Although levels of activity and knowledge of the number of animals previously caught were used to guide the number of traps deployed per sett, the general approach was to put out more traps than were likely to be needed (i.e. saturation trapping).

Box traps of steel mesh construction, with a spring-loaded door mechanism were dug into the substrate in the immediate vicinity of each active sett. Once traps had been deployed they were secured in the open position and baited with peanuts for four to eight days to habituate the badgers to their presence (Cheeseman & Mallinson, 1979). On the last day of this pre-baiting period they were set for two consecutive nights and checked on each following morning. From December until the ‘close season’ began (February 1st), the traps were also checked at night so that small cubs and lactating females could be released (in the case of adult animals an attempt was made to read and record their tattoo prior to release). After capture, each badger was transferred to a holding cage for transportation to the sampling facilities. In order to reduce opportunities for the transmission of infection between badgers from different social groups, they were segregated during transport in vehicles and arranged into their respective social groups on racks in the sampling facility. In addition, holding cages were not exchanged between social groups during trapping operations and were thoroughly cleaned between operations.

Animals were trapped and examined for two consecutive days. Those caught during the first night of trapping were held overnight following examination, and usually released the following morning. This prevented them from being re-captured on the second trapping night. However, lactating females and small cubs were marked (with an aerosol stock marker) and released on the afternoon of the day of capture so as to minimise the amount of time away from the sett. If recaptured on the following day these animals could be identified and released immediately. Animals caught and examined on the second day were released later that day. Badgers were released at the point of capture, following a period of recovery and subject to approval by the named Animal Care and Welfare Officer (NACWO), Named Veterinary Surgeon (NVS) or another experienced person with delegated authority.

4.4. Sampling (Objective 2)

Prior to examination each captured badger was anaesthetised by intramuscular injection of medetomidine hydrochloride (Domitor), butorphanol tartrate (Torbugesic) and ketamine hydrochloride (Vetalar), with the option of reversing the effects of medetomidine hydrochloride with atipamezole (Antisedan) (de Leeuw et al., 2004).

On the occasion of its first capture each badger was aged (cub vs adult) according to body size, and the degree of development and wear of the teeth. Each new capture received a unique tattoo on the belly (Cheeseman & Harris, 1982) as a means of permanent identification. The weight, body length (from the tip of the snout to the end of the terminal vertebra), body condition (assessed by palpation of the musculature and scored on a scale from ‘poor’ to ‘very good’) and reproductive status (in females based on the condition of the teats with respect to lactation, and in males on the level of descent of the testes) of each animal, was recorded at each capture event.

A number of samples were taken to assess TB infection status. Clinical samples (for microbiological culture) provided evidence for bacterial shedding and blood samples were taken from the jugular for serology. Prior to 2006 blood was only subjected to detection of antibodies to the single M. bovis antigen MPB83 using an ELISA test (Goodger et al., 1994). In July 2006 additional blood samples began to be routinely taken for the detection of a cellular response using a gamma-interferon test (Dalley et al., 2004) and to detect antibodies to multiple M. bovis antigens (MPB83, MPB70, CFP10) using the Stat-Pak ELISA (Kampfer et al., 2003). Preliminary results of these tests appear in Appendix 7.2, and a systematic comparative examination of all test results is the subject of work agreed for project SE3032 during the next three years.

Clinical samples included sputum obtained by oesophageal aspiration until May 2006, but subsequently also by tracheal aspiration (on the basis of veterinary advice which suggested that the latter may be more sensitive). The collection of samples by both methods will allow direct comparisons of rates of M. bovis recovery to be made (see Appendix 7.3 for preliminary results). Urine was obtained by palpation of the bladder, faeces by administration of an enema and swabs of pus and exudate were taken from bites and open abscesses. Where appropriate, a small sample of fluid was drained from abscesses by needle and syringe. All clinical samples were cultured on media selective for M. bovis growth (see Appendix 8.5 in the 2007 Project Report for SE3032) and isolates were spoligotyped (Aranaz et al., 1996).

The two principal diagnostic approaches that have been historically used to detect TB infection in badgers in this study (i.e. microbiological culture of clinical samples and the ELISA test) have their limitations (see Clifton-Hadley et al., 1995; Pritchard et al., 1986), and the excretion of bacilli in clinical samples may occur intermittently in individuals. Consequently, every effort was made to take a full suite of samples from each captured badger in order to maximise the chances of detecting infection. In addition, repeated captures of an individual within a year reduce the likelihood of missing infected cases and provide data on the progression of disease and its consequences in individual animals. However, it is recognised that this sampling regime is likely to underestimate the true level of infection in the population, although the magnitude of this effect is unknown. Nevertheless, the data generated in this study provides an index of the true level of infection that is consistent in time and space, and can therefore identify differences between demographic groups and temporal and spatial trends.

4.5. Post Mortem Examinations

All badgers found dead within the study area were submitted for post-mortem examination whereupon selected tissues (see Appendix 8.6 of the 2007 Project Report for SE3032) were cultured and M. bovis isolates were spoligotyped.

4.6. Data Collation (Objective 3)

Data from individual capture events and the results of the culture of clinical samples, spoligotyping, blood tests and post mortem examinations were entered onto an ACCESS database. Quality checks were routinely carried out on a sub-sample of observations in the database by reference to the original sample sheets.

4.7. Data Analyses (Objective 4)

The methods described here were common to most analyses. Those approaches related to specific analysis packages (Objectives 4a & b) are described separately in section 5.

4.7.1. Allocation of Disease Status and Social Group

The disease status of badgers captured in the study area is currently based on their assignment at each capture event to one of three disease states, based on the results of the culture of clinical samples and the ELISA antibody test. This assignment is based on a cautious and pragmatic interpretation of the results of these imperfect diagnostic tests, and in the full knowledge that they are likely to reflect minimum levels of infection in the population as a whole. The three categories are ‘negative’, ‘exposed’ and ‘excretor’, with an additional sub-sample of the excretor category known as ‘super-excretors’, used in some analyses. The ‘negative’ label is assigned to any individual which has not previously returned a single positive culture or ELISA result. An animal is considered an ‘exposed’ case following returning at least one ELISA positive result, and an ‘excretor’ case if M. bovis has been isolated from any clinical sample. The ‘super-excretor’ category consists of individuals which have yielded two M. bovis positive culture results either at a single or consecutive captures. Where a post mortem examination was carried out, any badger from which a positive culture result was obtained from tissues was classified as an ‘excretor’ for that year only if it had not tested positive before. Detailed discussion of the assumptions underlying these assignments can be found in section 8 (sub-section 4.7) of the 2007 Project Report to Defra for SE3032.

An individual became an incident case the first time it was recorded in a given disease state. These disease states were assumed to be progressive, such that an animal could not revert to an ‘earlier’ state regardless of subsequent test results. Hence an animal was only an incident case once with respect to any particular disease state. An individual was considered a prevalent case for any given disease state if at some time in its previous capture history it had been assigned to that disease state, and up until the time at which it progressed to another disease state.

As each badger was often trapped many times during its lifetime, the dataset consisted of repeated observations of individuals in each year. Prior to analysis, each captured badger was allocated to one social group and disease status category (see above) in each calendar year, using established protocols. Criteria for allocation of badgers to a social group were, in order of priority: (1) allocate to the group where most frequently caught in that year; (2) refer to allocation(s) in adjacent years and allocate to the group where most frequently caught; (3) consider the number of captures in current and both adjacent years and allocate to the group where most frequently trapped; (4) consider the badger’s last capture before the year in question and the first capture after that year. In both cases calculate the time lag to/from the first/last date of the year in question and allocate to the group for which this time period was the shortest (assuming this is one of the most frequent groups as defined in step 3); (5) allocate to the first relevant group caught in if otherwise indeterminable. Amongst animals trapped and allocated to a core social group in a given year approximately 91% were assigned on the basis of the first criterion (1).

4.7.2. Data Analysis Packages (Objectives 4a & b)

The analyses described in the present report can be broadly classified as either demographic or epidemiological in character. As each specific analysis package represents a discrete piece of work we describe their objectives, methods (where they differ from the general methods described above), results and conclusions separately below.

Collateral Research

The routine capture of badgers in the study area and the wealth of pre-existing information on the population, provided opportunities for collateral research. This has involved the use of the well described population in other fieldwork, the use of existing samples, the collection of supplementary samples during the routine work and the analysis of data. Appendix 7.4 lists all publications arising from project SE3032 during the period covered by this report, including those resulting from projects that benefited from the above arrangements.

5. Results & Discussion

5.1 General trends

During the period 1982 to 2006 inclusive, the number of badger social groups identified in the core of the study area each year by bait-marking remained relatively constant (mean = 24.1, range = 20 - 27). During the period of study the total annual badger population size (i.e. cubs and adults) estimated using the minimum number alive (MNA) method varied from 98 in 1982 to 311 in 1999 (n = 25 years, Figure 1). Variations in population size were largely driven by fluctuating group size (annual mean group size (MNA) = 9.26  0.38, range = 4.9 - 12.4). MNA estimates indicated that the overall trend in population size for the intensively studied core area followed an increase from 1982 onwards, rising to a peak in 1999, and decreased thereafter (Figure 1.). During the study a total of 10270 individual capture events took place in the core of the study area, involving 1878 different individuals. Each badger (conditional on it being trapped in a given year) was captured on average twice (1.99) each year (range of annual means 1.59 ± 2.23).

5.2. Data Analysis Packages (Objectives 4a & b)

5.2.1. (a). Demographic Analyses

5.2.1.1. Formal comparative assessment of approaches to population size estimation, and identification of an optimum standardised method.