7  Exploration of herd-level bTB breakdowns (BDs)

\(~\) \(~\)

7.1 Breakdown (BD) definition

Here, we defined the start date of a BD in a herd as the date when an animal was disclosed as a bTB case, as determined by the SICTT, IFN-γ or by the detection of visible lesions at routine slaughter which are then laboratory confirmed with M. bovis infection (also known as ‘factory lesion cases’). In the situation when a BD started due to the detection of a visible lesion, we chose the start date as the date when the sample was sent to the lab and not the date in which it was identified as culture positive which could be a number of weeks later. Depending on the circumstances, the herds trading_status on AHCS will generally change from free to withdrawn/suspended. We defined the BD end date as the date of the next SICTT (whole herd) test where the trading_status was restored to free. Once a bTB case is detected, a herd will usually remain in a BD until it has had two consecutive, clear whole-herd SICTT tests (usually 60 days apart). There are certain exceptions to the above criteria e.g., singletons.

There were rare instances when a herd was in a BD based on trading status but it did not have a bTB case (these could be for awhole host of reasons including fraud investigations (so a herd may be in suspended mode although no cases), error on system etc.). Of all BDs after 2007 (n = 87761), there were only 612 of these zero case BDs (0.7%). The initiating/index SICTT of these BDs is given below:

\(~\) \(~\)

 bd_first_skin_test_type   n     percent
                       1  55 0.089869281
                     10A  16 0.026143791
                       2   2 0.003267974
                       3   1 0.001633987
                       4  21 0.034313725
                      5A  20 0.032679739
                      5B  14 0.022875817
                      5C   8 0.013071895
                      5E   5 0.008169935
                      5F  69 0.112745098
                      5G   4 0.006535948
                      5H   4 0.006535948
                       6  15 0.024509804
                      7B  17 0.027777778
                       8  12 0.019607843
                      9A 349 0.570261438

\(~\) \(~\)

For the purposes of this work, we have excluded these cases along with extremely short BD (<60 days in length):

# DROP short and zero cases ----
bd_df <- bd_df %>%
  filter(all_cases != 0) %>% 
  filter(bd_duration_days >= 60)

\(~\) \(~\)

\(~\) \(~\)

7.2 Number of BDs over time

7.2.1 Per year

Figure 7.1: Number of BDs per year.

\(~\) \(~\)

\(~\) \(~\)

7.2.2 Per month

Figure 7.2: Number of BDs per month.

\(~\) \(~\)

\(~\) \(~\)

7.2.3 Per month and by herd-type

Figure 7.3: Number of BDs per month and herd-type.

\(~\) \(~\)

\(~\) \(~\)

7.3 BD by herd-type

Figure 7.4: Proportion of BDs ending per year by herd-type.

\(~\) \(~\)

7.4 BD size

Here, we explore the number of bTB cases disclosed within herd BDs (i.e. BD size). This illustrates that BD size (e.g., median) is increasing particulaly within dairy herds. The variance in the number of cases within BDs is also increasing within dairy. Mixed herd-types are also increasing in size (these are herds that have a component of their herd as dairy). The fact that dairy herds have more bTB cases during a BD is not that surprising as dairy are also the herds with the largest number of animals (herd size **add link here) in their herd.

\(~\) \(~\)

7.4.1 AHCS herd-type by BD size

\(~\)

Figure 7.5: Boxplots of the number of bTB cases in BDs that ended in each year by AHCS herd-types.

\(~\)

Figure 7.6: Boxplots of the number of bTB cases in BDs that ended in each year by AHCS herd-types.

\(~\)

Figure 7.7: Boxplots of the number of bTB cases in BDs that ended in each year by AHCS herd-types.

\(~\) \(~\)

7.4.2 Brock et al. herd-type by BD size

\(~\)

Figure 7.8: Boxplots of the number of bTB cases in BDs that ended in each year by Brock et al. herd-types (only including the 4 most common herd-types).

\(~\)

Figure 7.9: Boxplots of the number of bTB cases in BDs that ended in each year by Brock et al. herd-types (only including the 4 most common herd-types).

\(~\)

Figure 7.10: Boxplots of the number of bTB cases in BDs that ended in each year by Brock et al. herd-types (only including the 4 most common herd-types).

7.4.3 Brock et al. dairy sub-class herd-types by BD size

Rather than plotting all sub-classes, here we focus on dairy sub-classes only as we have seen above that dairy tend to have the largest BDs in terms of the number of bTB cases disclosed.

\(~\)

Figure 7.11: Boxplots of the number of bTB cases in BDs that ended in each year by Brock et al. sub-classes by dairy herd-types only.

\(~\)

Figure 7.12: Boxplots of the number of bTB cases in BDs that ended in each year by Brock et al. sub-classes by dairy herd-types only.

\(~\)

Figure 7.13: Boxplots of the number of bTB cases in BDs that ended in each year by Brock et al. sub-classes by dairy herd-types only.

7.4.4 BD size by category groups

7.4.4.1 All herds combined

Figure 7.14: Proportion of BD size by category over time (all herds)

7.4.4.2 AHCS

Figure 7.15: Proportion of BD size by category over time

7.4.4.3 Brock et al.

Figure 7.16: Proportion of BD size by category over time

7.5 BD duration

\(~\)

Figure 7.17: Boxplots of the number of BD durations (in years) by AHCS herd-types (zoomed into 1 year).

7.6 Repeat BD herds

Some herds will have more than one BD over time (i.e., they have a BD, become clear, and then have another BD). This is the proportion of herds with different number of BDs:

 bd_no     n       percent
     1 23940 0.49076484697
     2 12704 0.26042926549
     3  6114 0.12533568398
     4  2939 0.06024886739
     5  1482 0.03038068100
     6   792 0.01623582952
     7   383 0.00785141756
     8   203 0.00416145630
     9   108 0.00221397675
    10    56 0.00114798795
    11    23 0.00047149505
    12    17 0.00034849634
    13    11 0.00022549763
    14     1 0.00002049978
    15     6 0.00012299871
    16     1 0.00002049978
    17     1 0.00002049978

7.6.1 Boxplots of the number of BDs per herd:

7.6.1.1 AHCS:

Figure 7.18: Boxplots of the number of bTB BDs per herd in the dataset by AHCS herd-types.

7.6.1.2 Brock et al.:

Figure 7.19: Boxplots of the number of bTB BDs per herd in the dataset by Brock et al. herd-types (only including the 4 most common herd-types).

7.6.1.3 Brock et al. sub-classes:

Figure 7.20: Boxplots of the number of bTB BDs per herd in the dataset by Brock et al. sub-class herd-types.

7.7 Additional bTB cases after the initial detection

7.7.1 SICTT initiated BDs

Figure 7.21: The percentage of SICTT initiated BDs where additional cases will be disclosed during the BD (top 4 most common herd-types are selected).

7.7.2 Slaughter-house initiated BDs

Figure 7.22: The percentage of slaughter-house initiated BDs where additional cases will be disclosed during the BD (top 4 most common herd-types are selected).

7.8 Count of no. BDs by detection method

7.8.1 SICTT initiated BDs

Figure 7.23: The count of the number of BDs initiated by SICTT, by herd-type.

7.8.2 Slaughter-house initiated BDs

Figure 7.24: The count of the number of BDs initiated by a slaughter-house case, by herd-type.

7.9 Tables

7.9.1 Overall BD statistics

Table 7.1: bTB BD statistics

Variable

level

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

No. of unique herds with at least one case

7479

6363

5764

5238

5218

4776

4449

4108

3949

4189

4209

4384

5202

5079

5034

5696

6899

6642

No. of BDs that ended in year

7005 (8.1)

6449 (7.4)

5540 (6.4)

5023 (5.8)

4727 (5.4)

4547 (5.2)

4410 (5.1)

3918 (4.5)

3642 (4.2)

3730 (4.3)

3813 (4.4)

3916 (4.5)

4392 (5.1)

4610 (5.3)

4561 (5.3)

4728 (5.4)

5413 (6.2)

6388 (7.4)

Median (IQR) no. of cases per BD

2 (1-4)

2 (1-4)

2 (1-4)

1 (1-3)

2 (1-4)

1 (1-3)

1 (1-3)

1 (1-3)

1 (1-3)

1 (1-3)

1 (1-3)

1 (1-3)

2 (1-4)

2 (1-4)

2 (1-4)

2 (1-5)

2 (1-5)

2 (1-6)

Range of no. of cases in BDs that started in the year

1-237

1-208

1-169

1-176

1-226

1-296

1-224

1-295

1-303

1-265

1-306

1-291

1-193

1-171

1-396

1-209

1-245

1-374

BDs by no. of bTB cases

1

3350 (47.8)

3024 (46.9)

2761 (49.8)

2575 (51.3)

2326 (49.2)

2311 (50.8)

2306 (52.3)

2099 (53.6)

1959 (53.8)

1970 (52.8)

1965 (51.5)

1986 (50.7)

2097 (47.7)

2105 (45.7)

2205 (48.3)

2176 (46)

2259 (41.7)

2372 (37.1)

2

1063 (15.2)

959 (14.9)

852 (15.4)

770 (15.3)

770 (16.3)

756 (16.6)

677 (15.4)

562 (14.3)

584 (16)

581 (15.6)

622 (16.3)

636 (16.2)

670 (15.3)

697 (15.1)

700 (15.3)

696 (14.7)

867 (16)

971 (15.2)

3

608 (8.7)

552 (8.6)

445 (8)

440 (8.8)

387 (8.2)

350 (7.7)

362 (8.2)

305 (7.8)

272 (7.5)

281 (7.5)

287 (7.5)

326 (8.3)

348 (7.9)

398 (8.6)

363 (8)

417 (8.8)

441 (8.1)

607 (9.5)

4

390 (5.6)

346 (5.4)

308 (5.6)

281 (5.6)

250 (5.3)

221 (4.9)

214 (4.9)

187 (4.8)

155 (4.3)

140 (3.8)

149 (3.9)

196 (5)

234 (5.3)

292 (6.3)

216 (4.7)

247 (5.2)

351 (6.5)

386 (6)

5

255 (3.6)

257 (4)

208 (3.8)

173 (3.4)

176 (3.7)

154 (3.4)

163 (3.7)

122 (3.1)

100 (2.7)

73 (2)

88 (2.3)

111 (2.8)

154 (3.5)

165 (3.6)

148 (3.2)

175 (3.7)

202 (3.7)

278 (4.4)

6

227 (3.2)

208 (3.2)

157 (2.8)

135 (2.7)

138 (2.9)

133 (2.9)

101 (2.3)

97 (2.5)

78 (2.1)

75 (2)

67 (1.8)

79 (2)

112 (2.6)

143 (3.1)

135 (3)

129 (2.7)

159 (2.9)

211 (3.3)

7

154 (2.2)

138 (2.1)

100 (1.8)

81 (1.6)

105 (2.2)

92 (2)

77 (1.7)

69 (1.8)

56 (1.5)

58 (1.6)

62 (1.6)

81 (2.1)

96 (2.2)

96 (2.1)

81 (1.8)

117 (2.5)

111 (2.1)

167 (2.6)

8

108 (1.5)

127 (2)

85 (1.5)

66 (1.3)

80 (1.7)

69 (1.5)

62 (1.4)

52 (1.3)

42 (1.2)

67 (1.8)

47 (1.2)

66 (1.7)

76 (1.7)

78 (1.7)

79 (1.7)

80 (1.7)

102 (1.9)

149 (2.3)

9

95 (1.4)

102 (1.6)

63 (1.1)

68 (1.4)

57 (1.2)

61 (1.3)

49 (1.1)

50 (1.3)

34 (0.9)

40 (1.1)

32 (0.8)

51 (1.3)

61 (1.4)

54 (1.2)

60 (1.3)

69 (1.5)

81 (1.5)

95 (1.5)

10-14

312 (4.5)

300 (4.7)

217 (3.9)

186 (3.7)

182 (3.9)

164 (3.6)

160 (3.6)

134 (3.4)

126 (3.5)

149 (4)

172 (4.5)

129 (3.3)

221 (5)

216 (4.7)

216 (4.7)

217 (4.6)

282 (5.2)

358 (5.6)

15-19

150 (2.1)

140 (2.2)

120 (2.2)

54 (1.1)

83 (1.8)

71 (1.6)

91 (2.1)

76 (1.9)

66 (1.8)

72 (1.9)

92 (2.4)

89 (2.3)

106 (2.4)

120 (2.6)

101 (2.2)

116 (2.5)

166 (3.1)

234 (3.7)

20-24

96 (1.4)

92 (1.4)

64 (1.2)

58 (1.2)

53 (1.1)

36 (0.8)

36 (0.8)

54 (1.4)

54 (1.5)

63 (1.7)

64 (1.7)

46 (1.2)

64 (1.5)

82 (1.8)

76 (1.7)

85 (1.8)

109 (2)

129 (2)

25-29

64 (0.9)

46 (0.7)

36 (0.6)

29 (0.6)

31 (0.7)

29 (0.6)

20 (0.5)

32 (0.8)

24 (0.7)

34 (0.9)

51 (1.3)

35 (0.9)

49 (1.1)

43 (0.9)

59 (1.3)

57 (1.2)

55 (1)

106 (1.7)

30-34

36 (0.5)

31 (0.5)

31 (0.6)

26 (0.5)

23 (0.5)

19 (0.4)

15 (0.3)

7 (0.2)

18 (0.5)

34 (0.9)

29 (0.8)

25 (0.6)

21 (0.5)

32 (0.7)

31 (0.7)

39 (0.8)

61 (1.1)

74 (1.2)

35-39

24 (0.3)

24 (0.4)

13 (0.2)

17 (0.3)

12 (0.3)

24 (0.5)

21 (0.5)

5 (0.1)

17 (0.5)

22 (0.6)

21 (0.6)

17 (0.4)

22 (0.5)

16 (0.3)

24 (0.5)

32 (0.7)

35 (0.6)

54 (0.8)

40+

73 (1)

103 (1.6)

80 (1.4)

64 (1.3)

54 (1.1)

57 (1.3)

56 (1.3)

67 (1.7)

57 (1.6)

71 (1.9)

65 (1.7)

43 (1.1)

61 (1.4)

73 (1.6)

67 (1.5)

76 (1.6)

132 (2.4)

197 (3.1)

BD initiated by

SICTT

4881 (69.7)

4552 (70.6)

3701 (66.8)

3556 (70.8)

3446 (72.9)

3270 (71.9)

3188 (72.3)

2881 (73.5)

2725 (74.8)

2758 (73.9)

2821 (74)

2936 (75)

3308 (75.3)

3318 (72)

3008 (66)

3227 (68.3)

3842 (71)

4677 (73.2)

Slaughter (non permit animal)

2124 (30.3)

1897 (29.4)

1839 (33.2)

1467 (29.2)

1281 (27.1)

1277 (28.1)

1222 (27.7)

1037 (26.5)

917 (25.2)

972 (26.1)

992 (26)

980 (25)

1073 (24.4)

1111 (24.1)

1187 (26)

1218 (25.8)

1293 (23.9)

1397 (21.9)

IFN-γ

-

-

-

-

-

-

-

-

-

-

-

-

11 (0.3)

181 (3.9)

366 (8)

283 (6)

278 (5.1)

314 (4.9)

BD: herd-level bTB breakdown; SICTT: Single Intradermal Comparative Tuberculin Test; IFN-γ: interferon-gamma assay; AHCS: Animal Health Computer System
IFN-γ cases prior to May 2019 were recorded as SICTT (i.e., negative SICTT reinterpreted as a positive SICTT), after this date there was a facility to record them separately on AHCS
BD figures relate to BDs that ended in a particular year (i.e., BDs can span over more than one year so when data was summarised, values relate to the year the BD began in)

7.9.2 BD duration

Table 7.2

Year

N

mean

sd

median

min

max

2008

7,005

0.5474375

0.3288191

0.4

0.2

3.7

2009

6,449

0.5546906

0.3803079

0.4

0.2

4.7

2010

5,540

0.5426354

0.3927235

0.4

0.2

5.8

2011

5,023

0.5531356

0.4363118

0.4

0.2

6.9

2012

4,727

0.5328538

0.4219806

0.4

0.2

6.8

2013

4,547

0.5361777

0.4642283

0.4

0.2

8.0

2014

4,410

0.5539229

0.5242621

0.4

0.2

8.5

2015

3,918

0.5450230

0.5881386

0.4

0.2

10.3

2016

3,642

0.5256727

0.5270829

0.4

0.2

10.0

2017

3,730

0.5308311

0.5454840

0.4

0.2

12.1

2018

3,813

0.5381327

0.5451194

0.4

0.2

11.2

2019

3,916

0.5222421

0.4011200

0.4

0.2

10.3

2020

4,392

0.5377505

0.5369815

0.4

0.2

12.4

2021

4,610

0.5300651

0.5094364

0.4

0.2

13.5

2022

4,561

0.5494190

0.5518869

0.4

0.2

15.6

2023

4,728

0.5221658

0.3726920

0.4

0.2

8.8

2024

5,413

0.5323481

0.4232406

0.4

0.2

17.9

2025

6,388

0.5388541

0.3323242

0.4

0.2

6.6

Year represents year the BD ended in (it could overlap multiple years)

7.10 Notes

  • Explore CFUs, perhaps I need to drop them or have them as a seperate group