dfzljdn9uc3pi.cloudfront.net · Web viewSupplementary figure 6A – Sensitivity analysis with...
Transcript of dfzljdn9uc3pi.cloudfront.net · Web viewSupplementary figure 6A – Sensitivity analysis with...
Supplementary Material for
Evaluation of outbreak response immunization in the control of
pertussis using agent-based modeling
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Table of content
1. Supplementary table 1A. Distribution of individuals in the population by vaccine attitude:
model input at the initialization
2. Supplementary table 1B. Probabilities of vaccination by attitude: model input at the
initialization
3. Supplementary figure 1A. Auto-ORI algorithm overview.
4. Supplementary figure 1B. Auto-ORI algorithm – rating sub-outbreak levels
5. Supplementary figure 1C. Auto-ORI algorithm – evaluating/determining outbreak levels
6. Supplementary figure 2. 15-years cumulative incidence data for two reference
populations: (A) Alberta Central Zone; (B) Alberta South Zone
7. Supplementary figure 3. Calculating waning immunity algorithm
8. Supplementary figure 4A. Illustration of vaccine-induced waning immunity output by
years of simulation, 4-16 age group
9. Supplementary figure 4B. Illustration of natural infection-derived waning immunity
output by years of simulation, 4-16 age group
10. Supplementary figure 5. Scatterplot illustrating differences in case counts in a single
paired simulation run, all ages
11. Supplementary table 2A – Sensitivity analysis with vaccine coverage for dose 7 equals
27%. Number of pertussis cases averted and numbers needed to vaccinate by time periods
after the outbreak-response immunization campaign and by age groups: modeling-
generated results
12. Supplementary figure 6A – Sensitivity analysis with vaccine coverage for dose 7 equals
27%. Boxplot of pertussis cases averted over time, up to 10 years after outbreak response
immunization campaign: All ages (A), infants under 1 year of age (B) and adolescents
10-14 years of age (C)
13. Supplementary table 2B – Sensitivity analysis with τ value increased to 3 for individuals
born before 1997. Number of pertussis cases averted and numbers needed to vaccinate by
time periods after the outbreak-response immunization campaign and by age groups:
modeling-generated results
14. Supplementary figure 6B – Sensitivity analysis with τ value increased to 3 for individuals
born before 1997. Boxplot of pertussis cases averted over time, up to 10 years after
2
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
outbreak response immunization campaign: All ages (A), infants under 1 year of age (B)
and adolescents 10-14 years of age (C)
15. Supplementary table 2C – Sensitivity analysis with α value increased to 10. Number of
pertussis cases averted and numbers needed to vaccinate by time periods after the
outbreak-response immunization campaign and by age groups: modeling-generated
results
16. Supplementary figure 6C – Sensitivity analysis with α value increased to 10. Boxplot of
pertussis cases averted over time, up to 10 years after outbreak response immunization
campaign: All ages (A), infants under 1 year of age (B) and adolescents 10-14 years of
age (C)
17. Supplementary table 2D – Sensitivity analysis: Restricting ORI eligibility to those who
did not receive vaccine within last 6 months. Number of pertussis cases averted and
numbers needed to vaccinate by time periods after the outbreak-response immunization
campaign and by age groups: modeling-generated results
18. Supplementary figure 6D Sensitivity analysis: Restricting ORI eligibility to those who
did not receive vaccine within last 6 months. Boxplot of pertussis cases averted over
time, up to 10 years after outbreak response immunization campaign: All ages (A),
infants under 1 year of age (B) and adolescents 10-14 years of age (C)
19. Supplementary table 2E. Sensitivity analysis: Enhancing boosting effect. Number of
pertussis cases averted and numbers needed to vaccinate by time periods after the
outbreak-response immunization campaign and by age groups: modeling-generated
results
20. Supplementary figure 6E. Sensitivity analysis: Enhancing boosting effect. Boxplot of
pertussis cases averted over time, up to 10 years after outbreak response immunization
campaign: All ages (A), infants under 1 year of age (B) and adolescents 10-14 years of
age (C)
21. Supplementary table 2F. Multi-way sensitivity analysis: τ value increased to 3 for
individuals born before 1997 and α value increased to 10. Number of pertussis cases
averted and numbers needed to vaccinate by time periods after the outbreak-response
immunization campaign and by age groups: modeling-generated results
3
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
22. Supplementary figure 6F. Multi-way sensitivity analysis: τ value increased to 3 for
individuals born before 1997 and α value increased to 10. Boxplot of pertussis cases
averted over time, up to 10 years after outbreak response immunization campaign: All
ages (A), infants under 1 year of age (B) and adolescents 10-14 years of age (C)
4
86
87
88
89
90
91
92
93
94
95
96
97
98
99100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
Supplementary table 1A. Distribution of individuals in the population by vaccine attitude:
model input at the initialization
Vaccination attitudeAcceptor Hesitant Rejecter
Percentage in the population 50% 40% 10%
Supplementary table 1B. Probabilities of vaccination by attitude: model input at the
initialization
Dose Acceptor Hesitant Rejecter1 0.95 0.6 02 0.95 0.6 03 0.95 0.6 04 0.9 0.6 05 0.9 0.6 06 0.9 0.6 07 0.1 0.05 0
Table 1A demonstrates the distribution of vaccination attitudes in the simulated population.
Vaccination attitudes for children are assumed to be those of their parents. Table 1B shows final
values of the dose-specific vaccination probabilities based on vaccine attitudes. Values were
varied during the calibration process to derive model-generated dose-specific vaccine coverage
judged to be comparable with reference vaccine coverage values. References values were
derived from Alberta Health, Public Health Ontario and Public Health Agency of Canada.
5
116
117
118
119
120
121
122
123124
125
126
127
128
129
130131132133134135136137
138
139
140
Supplementary figure 1A. Auto-ORI algorithm overview.
We considered ORI to be supplementary immunization delivered to a specified age-group based
on the magnitude of the outbreak. Therefore the decision to implement ORI was based in “real
time” on analysis of current and historical incidence rates for each age-group separately. The
auto-ORI algorithm in our model is based on evaluation/determination of the outbreak level,
while the outbreak level is further decided based on historical and current sub-outbreak levels as
demonstrated in supplementary figure 1A above.
6
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
Supplementary figure 1B. Auto-ORI algorithm – rating sub-outbreak levels.
We rated sub-outbreak levels/states by examining past five years’ monthly incidence rates for the
specific age-groups and used both fixed (but modifiable) thresholds and standard deviations
above 5-year mean as criteria to classify monthly incidence rates into S and S+ states, as shown
in supplementary figures 1B.
7
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
Supplementary figure 1C. Auto-ORI algorithm – evaluating/determining outbreak levels
We classified potential outbreaks into three levels: non-outbreak, outbreak and outbreak
triggering ORI (ORI-able outbreak), based on the magnitude and duration of monthly incidence
rates, which were expressed as sequences (monthly runs) of sub-outbreak levels as shown in
supplementary figure 1C. The outbreak was defined as at least two consecutive S states, while
ORI-able outbreak was defined as three consecutive S states or two consecutive S+ states. S+
state would equal to the S state for the purpose of defining outbreaks but not vice versa (e.g., S,
S+ sequence would be classified as outbreak but not ORI-able outbreak; adding either S or S+
state in the following month would upgrade this outbreak to the ORI-able outbreak).
8
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
Supplementary figure 2. 15-years cumulative incidence¶ data for two reference populations:
(A) Alberta Central Zone; (B) Alberta South Zone
¶ 15-years cumulative incidence was up-scaled to 30-years cumulative incidence for calibration
purposes. For Central zone – up-scaled 30-years cumulative incidence is 504 per 100,000; for
South zone – up-scaled 30-years cumulative incidence is 456 per 100,000. Data derived from
Alberta Heath, Interactive Health Data Application, 2015.
Supplementary figure 2 demonstrates 15-years cumulative incidence of pertussis in two
jurisdictions in Alberta, Canada: Central and South zones. These jurisdictions were selected as
reference based on having comparable population to the population size simulated in the model.
As only 15 years of reliable data were available, 15-years cumulative incidence was up-scaled to
30 years. Two reference population presented here have comparable cumulative incidence, but
dis-similar shape of epidemiological curves (e.g. Central zone had one year with a larger peak
while South zone had several smaller peaks). The variability of epidemiologic curve shapes
(distribution of incidence over time) is represented in our model simulations. The implication of
having such variability is that the simulation resulting in a higher peak is more likely to trigger
outbreak response (as may happen in reality), while smaller-scale outbreaks will be managed
without outbreak response immunization.
9
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
Supplementary figure 3. Calculating vaccine-induced waning immunity algorithm.
Vaccine-induced waning immunity was calculated by identifying qualifying group of individuals
(age group) at the beginning of tracing year and then count individuals who transitioned from V3
state to V2 state during a tracing year while excluding those who received natural boosting (by
virtue of contracting disease). Waning immunity was expressed as a percentage (or fraction of
population) per year.
10
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
Supplementary figure 4A. Illustration of vaccine-induced waning immunity§ output by
years of simulation, 4-16 age group
Supplementary figure 4B. Illustration of natural infection-derived waning immunity§
output by years of simulation, 4-16 age group
§ Based on “No-ORI” runs [no outbreak response immunization] and expressed as annual
percentage transition from the V3 or R3 state to V2 or R2 state
Supplementary figures 4A and 4B demonstrate distribution of vaccine-induced and natural
infection-derived waning immunity over time. For each given year of 30-year study period,
distribution reflects different values of waning immunity in different simulations. There is a
smaller fluctuation of vaccine-induced waning immunity in comparison to natural disease-
derived waning immunity over time. This is likely be explained by more predictable vaccination
patterns for agents in our model, while natural infection immunity is less predictable due to
outbreaks’ occurrence.
11
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
Supplementary figure 5. Scatterplot illustrating differences§ in case counts in a single
paired simulation run, all ages¶
§ Difference is between count of cases in “No-ORI” runs [no outbreak response immunization]
versus ORI [outbreak response immunization triggered] run. Positive number [dots above zero
line on y-axis] represent “cases averted”, while negative number [dots below zero line on y-axis]
represent “cases added” as a result of ORI¶ Case count for all ages
Supplementary figure 5 illustrates a representative example of the single individual paired “No-
ORI” – “ORI” simulation. The “No-ORI” simulation serves as a control group. The timing of
triggered-ORI is shown as a vertical red line. The figure captures the difference in the number of
pertussis cases occurring after the implementation of the ORI. Prior to ORI implementation the
number of cases in both “No-ORI” and “ORI” simulations are identical and both these
simulations are run from the same randomly-selected seed. The y-axis represents the difference
in the number of cases between paired “No-ORI” and “ORI” runs with positive number (i.e. dots
above horizontal zero line) representing cases averted (greater number of cases during “No-ORI”
in comparison to “ORI” run), while negative number (i.e. dots below horizontal zero line)
indicates additional cases occurring as a results of having ORI (“harmful” effect of ORI).
12
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
Supplementary tables 2A through 2F and figures 6A to 6F demonstrate results of sensitivity
analyses performed. Numbers of cases averted reported in these tables are compared to those
reported in the main simulation to check for the robustness of the model and its sensitivity to
changes in various parameters.
Supplementary table 2A. Sensitivity analysis with vaccine coverage for dose 7 equals 27%.
Number of pertussis cases averted and numbers needed to vaccinate by time periods after
the outbreak-response immunization campaign and by age groups: modeling-generated
results
Age groups Post-outbreak-response immunization period, years
Number of cases averted*
Minimum number needed to vaccinate§
Maximum number needed to vaccinate§
All ages 1 119 176 228All ages 3 256 82 106All ages 10 410 51 66Under 1 year 1 5 4067 5254Under 1 year 3 12 1789 2311Under 1 year 10 20 1038 134110-14 years old 1 51 409 52910-14 years old 3 102 205 26510-14 years old 10 156 134 173
*p<0.00001 for all groups of comparisons of counts of cases in outbreak-response immunization
(ORI) versus no-ORI simulations, one-way Mann Whitney U test;§ Number needed to vaccinate (NNV) was calculated directly from the model by dividing a
number of vaccinations delivered during the ORI by a number of cases averted in a respective
group. NNV only applies to a current model and for a given population size.
13
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290291292293294295296297298299300301
Supplementary figure 6A. Sensitivity analysis with vaccine coverage for dose 7 equals 27%.
Boxplot of pertussis cases averted over time, up to 10 years after outbreak response
immunization campaign: All ages (A), infants under 1 year of age (B) and adolescents 10-14
years of age (C)
14
302
303
304
305
306
Supplementary table 2B. Sensitivity analysis with τ value¶ increased to 3 for individuals
born before 1997. Number of pertussis cases averted and numbers needed to vaccinate by
time periods after the outbreak-response immunization campaign and by age groups:
modeling-generated results
Age groups Post-outbreak-response immunization period, years
Number of cases averted*
Minimum number needed to vaccinate§
Maximum number needed to vaccinate§
All ages 1 112 188 245All ages 3 228 92 120All ages 10 409 51 67Under 1 year 1 5 4181 5448Under 1 year 3 10 2005 2612Under 1 year 10 20 1041 135610-14 years old 1 48 434 56510-14 years old 3 92 229 29810-14 years old 10 156 135 175
¶ τ value represents the transition between V states in the model. Increasing its value reduces
vaccine-induced waning immunity and increase protection. This was applied only to individuals
born before 1997, who were recipients of whole-cell pertussis vaccine.
*p<0.00001 for all groups of comparisons of counts of cases in outbreak-response immunization
(ORI) versus no-ORI simulations, one-way Mann Whitney U test;§ Number needed to vaccinate (NNV) was calculated directly from the model by dividing a
number of vaccinations delivered during the ORI by a number of cases averted in a respective
group. NNV only applies to a current model and for a given population size.
15
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321322323324325326327328329330331332333
Supplementary figure 6B. Sensitivity analysis with τ value increased to 3 for individuals
born before 1997. Boxplot of pertussis cases averted over time, up to 10 years after
outbreak response immunization campaign: All ages (A), infants under 1 year of age (B)
and adolescents 10-14 years of age (C)
16
334
335
336
337
338
Supplementary table 2C. Sensitivity analysis with α value¶ increased to 10. Number of
pertussis cases averted and numbers needed to vaccinate by time periods after the
outbreak-response immunization campaign and by age groups: modeling-generated results
Age groups Post-outbreak-response immunization period, years
Number of cases averted*
Minimum number needed to vaccinate§
Maximum number needed to vaccinate§
All ages 1 148 143 183All ages 3 262 81 103All ages 10 422 50 64Under 1 year 1 7 2998 3850Under 1 year 3 13 1661 2132Under 1 year 10 21 1016 130510-14 years old 1 61 348 44710-14 years old 3 104 203 26110-14 years old 10 161 131 168
¶ α value represents the transition between R states in the model. Increasing its value reduces
natural infection-derived waning immunity and increase protection.
*p<0.00001 for all groups of comparisons of counts of cases in outbreak-response immunization
(ORI) versus no-ORI simulations, one-way Mann Whitney U test;§ Number needed to vaccinate (NNV) was calculated directly from the model by dividing a
number of vaccinations delivered during the ORI by a number of cases averted in a respective
group. NNV only applies to a current model and for a given population size.
17
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
Supplementary figure 6C. Sensitivity analysis with α value increased to 10. Boxplot of
pertussis cases averted over time, up to 10 years after outbreak response immunization
campaign: All ages (A), infants under 1 year of age (B) and adolescents 10-14 years of age
(C)
18
361
362
363
364
365
Supplementary table 2D. Sensitivity analysis: Restricting ORI eligibility to those who did
not receive vaccine within last 6 months. Number of pertussis cases averted and numbers
needed to vaccinate by time periods after the outbreak-response immunization campaign
and by age groups: modeling-generated results
Age groups Post-outbreak-response immunization period, years
Number of cases averted*
Minimum number needed to vaccinate§
Maximum number needed to vaccinate§
All ages 1 108 184 235All ages 3 241 82 106All ages 10 429 46 60Under 1 year 1 4 4555 5836Under 1 year 3 10 1910 2448Under 1 year 10 21 949 121610-14 years old 1 47 422 54110-14 years old 3 96 208 26610-14 years old 10 162 123 158
*p<0.00001 for all groups of comparisons of counts of cases in outbreak-response immunization
(ORI) versus no-ORI simulations, one-way Mann Whitney U test;§ Number needed to vaccinate (NNV) was calculated directly from the model by dividing a
number of vaccinations delivered during the ORI by a number of cases averted in a respective
group. NNV only applies to a current model and for a given population size.
19
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
Supplementary figure 6D. Sensitivity analysis: Restricting ORI eligibility to those who did
not receive vaccine within last 6 months. Boxplot of pertussis cases averted over time, up to
10 years after outbreak response immunization campaign: All ages (A), infants under 1
year of age (B) and adolescents 10-14 years of age (C)
20
388
389
390
391
392
Supplementary table 2E. Sensitivity analysis: Enhancing boosting effect¶. Number of
pertussis cases averted and numbers needed to vaccinate by time periods after the
outbreak-response immunization campaign and by age groups: modeling-generated results
Age groups Post-outbreak-response immunization period, years
Number of cases averted*
Minimum number needed to vaccinate§
Maximum number needed to vaccinate§
All ages 1 41 521 667All ages 3 93 227 290All ages 10 148 143 183Under 1 year 1 1** 17238 22050Under 1 year 3 4 5572 7127Under 1 year 10 6 3536 452310-14 years old 1 19 1110 142010-14 years old 3 40 537 68610-14 years old 10 60 351 449
¶ In this scenario, booster immunization for an individual in lower V states would lead to a V4
state (full protection) without transitioning through V2 and V3 states as is in the main
experiment
*p<0.00001 for all groups of comparisons of counts of cases in outbreak-response immunization
(ORI) versus no-ORI simulations, one-way Mann Whitney U test;
** p=0.12§ Number needed to vaccinate (NNV) was calculated directly from the model by dividing a
number of vaccinations delivered during the ORI by a number of cases averted in a respective
group. NNV only applies to a current model and for a given population size.
21
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
Supplementary figure 6E. Sensitivity analysis: Enhancing boosting effect. Boxplot of
pertussis cases averted over time, up to 10 years after outbreak response immunization
campaign: All ages (A), infants under 1 year of age (B) and adolescents 10-14 years of age
(C)
22
413
414
415
416
417
Supplementary table 2F. Multi-way sensitivity analysis: τ value increased to 3 for
individuals born before 1997† and α value increased to 10¶. Number of pertussis cases
averted and numbers needed to vaccinate by time periods after the outbreak-response
immunization campaign and by age groups: modeling-generated results
Age groups Post-outbreak-response immunization period, years
Number of cases averted*
Minimum number needed to vaccinate§
Maximum number needed to vaccinate§
All ages 1 130 162 209All ages 3 229 92 119All ages 10 378 56 72Under 1 year 1 6 3590 4623Under 1 year 3 10 2012 2591Under 1 year 10 18 1164 149810-14 years old 1 54 390 50210-14 years old 3 90 235 30210-14 years old 10 143 148 191
† τ value represents the transition between V states in the model. Increasing its value reduces
vaccine-induced waning immunity and increase protection. This was applied only to individuals
born before 1997, who were recipients of whole-cell pertussis vaccine. ¶ α value represents the transition between R states in the model. Increasing its value reduces
natural infection-derived waning immunity and increase protection.
*p<0.00001 for all groups of comparisons of counts of cases in outbreak-response immunization
(ORI) versus no-ORI simulations, one-way Mann Whitney U test;§ Number needed to vaccinate (NNV) was calculated directly from the model by dividing a
number of vaccinations delivered during the ORI by a number of cases averted in a respective
group. NNV only applies to a current model and for a given population size.
23
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
Supplementary figure 6F. Multi-way sensitivity analysis: τ value increased to 3 for
individuals born before 1997 and α value increased to 10. Boxplot of pertussis cases averted
over time, up to 10 years after outbreak response immunization campaign: All ages (A),
infants under 1 year of age (B) and adolescents 10-14 years of age (C)
24
440
441
442
443
444