key: cord-0334020-1evwakjf
authors: Hardelid, P.; Favarato, G.; Wijlaars, L.; Fenton, L.; McMenamin, J.; Clemens, T.; DIBBEN, C.; Milojevic, A.; Macfarlane, A.; Taylor, J.; Cunningham, S.; Wood, R.
title: Risk of SARS-CoV-2 testing, PCR-confirmed infections and COVID-19-related hospital admissions in children and young people: birth cohort study
date: 2021-12-17
journal: nan
DOI: 10.1101/2021.12.17.21267350
sha: 1fc47eec43d25485a4be537d8640df9bfb635f46
doc_id: 334020
cord_uid: 1evwakjf

Background There have been no population-based studies of SARS-CoV-2 testing, PCR-confirmed infections and COVID-19-related hospital admissions across the full paediatric age range. We examine the epidemiology of SARS-CoV-2 in children and young people (CYP) aged <23 years. Methods We used a birth cohort of all children born in Scotland since 1997, constructed via linkage between vital statistics, hospital records and SARS-CoV-2 surveillance data. We calculated risks of tests and PCR-confirmed infections per 1000 CYP-years between August and December 2020, and COVID-19-related hospital admissions per 100,000 CYP-years between February and December 2020. We used Poisson and Cox proportional hazards regression models to determine risk factors. Results Among the 1226855 CYP in the cohort, there were 378402 tests, 19005 PCR confirmed infections and 346 admissions, corresponding to rates of 770.8/1000 (95% confidence interval 768.4-773.3), 179.4 (176.9-182.0) and 29.4/100,000 (26.3-32.8) CYP-years respectively. Infants had the highest COVID-19-related admission rates. Chronic conditions, particularly multiple types of conditions, was strongly associated with COVID-19-related admissions across all ages. The hazard ratio for >1 chronic condition type was 12.2 (7.9-18.82) compared to children with no chronic conditions. 89% of admitted children had no chronic conditions recorded. Conclusions Infants, and CYP with chronic conditions are at highest risk of admission with COVID-19, however the majority of admitted CYP have no chronic conditions. These results provide evidence to support risk/benefit analyses for paediatric COVID-19 vaccination programmes. Studies examining whether maternal vaccine during pregnancy prevents COVID-19 admissions in infants are urgently needed. Funding UK Research and Innovation-Medical Research Council

4 19-related admission for children with asthma, indicating an increased risk that supports 93 vaccine strategies targeted to this group. 18 Our aim was to provide population based 94 estimates of risk of SARS-CoV-2 testing, PCR confirmed infections and COVID-19 related 95 admissions in children and young people (CYP) based on age, presence of multiple chronic 96 conditions, and socioeconomic status, that could support vaccination and other policy 97 recommendations across the whole paediatric population. 98

Data sources 100

We used a national birth cohort of all children and young people (CYP) born in Scotland from 101 1997 onwards, developed from administrative health datasets linked to public health 102 surveillance data on SARS-CoV-2 test results, originally constructed for the PICNIC study. 19 103 CYP born in 1997 onwards were included. Birth registrations comprised the cohort spine, 104 and CYP are linked over time and between databases using the Community Health Index 105 (CHI) number, a unique personal identifier recorded at all interactions with the Scottish 106 National Health Service (NHS). Table 1 summarises the databases and variables used in 107 this study. 108 109 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; https://doi.org/10.1101/2021.12.17.21267350 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; https://doi.org/10.1101/2021.12.17.21267350 doi: medRxiv preprint

We included CYP born in Scotland from 1 st April 1997 to 31st December 2020. Children born 115 at less than 24 completed weeks' gestation or with a birthweight <500 grammes were 116 excluded to ensure stillbirths were not inadvertently included. 20 CYP whose mothers were 117 not resident in Scotland at the time of delivery, and CYP who migrated out of Scotland 118 before 1 st February 2020 were also excluded. For analyses of SARS-CoV-2 tests and 119 positive test results (from now on referred to PCR-confirmed infections), CYP were followed 120 from birth or the 1 st August 2020 (whichever occurred last), until death, migration from 121 Scotland, or their 23 rd birthday, whichever occurred first. 1 st August 2020 was chosen as the 122 follow-up start date for analyses of tests and PCR-confirmed infections since this is when 123 testing for SARS-CoV-2 became commonly available in the community (rather than solely in 124 hospitals) for children of all ages. 21 For calculation and analyses of rates of COVID-19-125 related admissions, we used the follow-up start date as the 1 st February 2020. This allowed 126 us to include all COVID-19-related hospital admissions since PCR testing for SARS-CoV-2 127 was available for suspected cases in hospital from January 2020. 128

Outcomes 129

Our primary outcomes were rates of a) SARS-CoV-2 PCR tests (positive or negative); b) 130 PCR-confirmed SARS-CoV-2 infections; c) COVID19-related hospital admissions. We also 131 considered as secondary outcomes d) PIMS-TS admissions and e) COVID19-related 132 test to 21 days prior to the start of, or during, the ICU stay. ICU episodes where the 168 difference between the ICU admission date and previous ICU discharge date was ≤ 1 day 169 were assumed to indicate the same ICU stay. 170

We examined four risk key factors for outcomes a-c): age group, sex, family socio-economic 172 position (SEP) and history of chronic conditions. Age group was defined on 1 st February 173 2020 as the following five categories: <1 year (this also includes children born during 2020), 174 1-4 years, 5-11 years, 12-17 years and 18-22 years. We chose these age groups to reflect 175 likely mixing patterns based on age (i.e. prior to formal childcare, nursery/preschool, primary 176 school, secondary school, and higher/further education or work). Sex was recorded on the 177 birth record. Family SEP was defined using parents' (father's, or mother's if the birth was not 178 jointly registered) occupation recorded on birth registration. This was coded using the UK 179 National Statistics Socio-economic Classification (NS-SEC). 23 We collapsed the NS-SEC 180 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. developmental/mental health, blood/cancer, chronic infections, respiratory, 188 metabolic/gastrointestinal/endocrine/genitourinary, musculoskeletal/skin 189 neurological/sensory, and cardiac conditions. We grouped the history of chronic conditions 190 as none, one type of condition, and more than type of chronic condition. 191

We further explored whether gestational age and the number of older siblings affected 192 outcomes b) and c) in children aged <5 years, and BMI in CYP aged 5-17, after adjusting for 193 the four key risk factors. Gestational age was coded into a binary variable: preterm (<37 194 weeks) and term/late term (≥37 weeks). Number of older siblings (indicated by parity 195 recorded on maternity records) was grouped into a three-category variable: no older siblings, 196 one older sibling and two or more older siblings. BMI was derived from the Child Health 197

Surveillance Programme -School dataset, which includes height and weight measurements 198 for children starting their first (reception) year at primary school (at age 5 years), categorised 199 using the British 1990 growth reference standards 25 as underweight (<5 th percentile), healthy 200 weight (5 th to <85 th percentile) and overweight/obese (≥85 th percentile). 201

We calculated rates of outcomes a) and b) per 1,000 CYP years and outcome c) per 203 100,000 CYP-years with 95% confidence intervals, according to each of the risk factors. We 204 estimated the median length of stay with interquartile ranges (IQRs) for COVID-19 related 205 hospital admissions. 206

We examined the association between risk factors and testing rates using Poisson 207 regression models with robust standard errors to account for multiple tests per child. To 208 examine the association between risk factors and PCR-confirmed SARS-CoV-2 infection, 209 and COVID19-related admission risk we used Cox proportional hazards regression models. 210

Where a child had multiple COVID-19-related admissions, only the first was included in the 211

Cox proportional hazards models. For each primary outcome, we first fitted an overall model 212 including all ages. We included age group, sex, SEP and history of chronic conditions a 213 priori as risk factors in the main model. We tested for interaction with age group and each of 214 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; the other main risk factors using the Wald test. Two-sided p-values <0.05 were considered 215 statistically significant. 216

Based on these results, we then fitted models for each primary outcome stratified by age 217 group if a statistically significant interaction with age was identified for any of the other 218 variables or if we identified non-proportional hazards. In further analyses for ages <5 years 219 old we included parity and gestational age as additional risk factors in the models; and for 220 ages 5-17 years we included BMI category. We tested the proportional hazards assumption 221 of the Cox model by inspecting plots of Schoenfeld residuals 26 and survival curves according 222 to each main risk factors. 223

As there was only a small number of events for our secondary outcomes, we did not carry 224 out detailed statistical analyses. We report the number of cases, median length of stay and 225 age (with IQRs) for these outcomes. 226

All analyses were based on complete cases, as only a small number of CYP and missing 227 values for any of the main variables. All statistical analyses were performed using Stata 228 16.0. 229

We repeated the analyses for outcome c) using a more specific definition of a COVID-19-231 related admission restricted to emergency admissions with U07.1 or U07.2 27 as the primary 232 diagnosis. 233

This study included 1,226,855 CYP ( Figure 1 ). Baseline characteristics of the participants 235 are presented in Table 2 . The median age on February 2020 was 11 years (IQR 5-17), and 236 5.4% of the cohort (66,974/1,226,855 CYP) had at least one chronic condition recorded in 237 their hospital or birth record in the previous five years. 238 239 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. , r  e  c  o  r  d  s  b  o  r  n  b  e  t  w  e  e  n  0  1  A  p  r  1  9  9  7  a  n  d  3  1  D  e  c  2  0  2  0   E  x  c  l  u  d  e  :  4 ,

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; 22 years), and those with more than one chronic condition recorded had the highest testing 254 rates (Appendix Table 1 & 2). Children aged <5 years born preterm were also more likely to 255 be tested compared to children born at term (Appendix Table 3 ). 256

Models that mutually adjusted for all four main predictors suggested increasing age (and 257 particularly an age of 18-22 years) and a history of chronic conditions were strongly 258 associated with being tested (Table 3) . As there was a statistically significant interaction term 259 between age group and each of the other risk factors, we conducted age group-stratified 260 analyses (Table 4) . A history of chronic conditions was strongly associated with higher 261 testing rates, with the size of the effect being the largest among infants 262 263 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. and those with no history of chronic conditions had a higher risk of PCR-confirmed infection 295 (Table 3) . As we identified statistically significant interaction terms between age group and 296 each of the other main risk factors, we conducted stratified analyses by age group (Table 5) . 297

These results showed that among infants and preschool children, PCR-confirmed infection 298 rates were lower among higher SEP groups, whereas among CYP aged 12 and above, 299 children from higher SEP groups were more likely to be infected. Infants with chronic 300 conditions had higher infection rates than those without; the opposite was true for CYP aged 301 Table 6 ). Further, in children aged 12-17 years, being 306 overweight/obese increased the risk of a PCR-confirmed infection compared to being of 307 normal BMI (HR: 1.07 (95%CI: 1.00-1.15), Appendix Table 6 ). 308 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. in 318 CYP were identified as COVID-19-related admissions. 25 CYP were admitted more 317 than once. The median time to re-admission was 5 (IQR 2-12) days. The median length of 318 stay was 2 days (IQR 1-4 days). There were 110 admissions between February and July 319 (31.8%) and 236 (68.2%) between August and December (Figure 3) . 320

Of the 346 COVID-19-related admissions 203 (58.7%) had a U07.1/U07.2 recorded as 321 primary diagnosis and 258 (74.6%) were temporally associated with a SARS-CoV-2 positive 322 test with or without a primary or secondary U07.1/U07.2 diagnosis (Figure 4) Of the CYP with chronic conditions who had a COVID-19 related admission, 331 neurological/sensory conditions were the common condition type recorded among children 332 aged<12 years, whereas among those aged 12-22 years the most common conditions were 333 developmental/mental health conditions followed by 334 metabolic/gastrointestinal/endocrine/genitourinary conditions. 335 Across all ages, presence of one or more chronic condition increased the risk of COVID-19-336 related admissions and CYP from lower SEP groups had the highest risk of COVID-19-337 related admission (Table 3) . We identified statistically significant interaction terms between 338 age group and each of the other main risk factors, and therefore conducted age-stratified 339 analyses (Table 6) . Having more than one chronic condition remained the strongest risk 340 factor for COVID-19-related admission across all age groups. There was no statistically 341 significant association between prematurity, number of older siblings, or BMI category and 342 COVID-19 related hospital admission risk (Appendix Table 9 ). 343 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

Thirteen (3.8%) of the 346 COVID-related admissions involved an ICU attendance, 359 accounting for 1.2% of the 1,238 ICU admissions in CYP during the study period. Around 360 half of these admissions were in CYP with a history of one or more types of chronic 361 conditions. The median age of CYP admitted to ICU was 14 years (IQR 9-19 years) and the 362 median length of stay at ICU was 6 days (IQR 2-7 days). Half of COVID-related ICU 363 admissions were in boys. 364

We identified less than five admissions with a diagnosis suggestive of PIMS-TS and 365 temporally associated with a positive PCR test (<28 days prior admission by definition), all in 366 males with an age spanning from 9 to 14 years. The median length of stay at admission was 367 10 days (IQR 6-14). 368

Among the 346 COVID related admissions, 199 (57.5%) were emergency admissions with a 370 primary diagnosis of U07.1/U07.2. Using the more specific definition, the COVID-19 related 371 hospital admission rates were 107.0 (95% CI 80.4-142.4), 13.4 (9.0-19.8), 8.0 (5.6 -11.7), 372 9.3 (6.3-13.6) and 27.7 (21.6-35.5) per 100,000 CYP years in age groups <1, 1-4, 5-11, 12-373 17 and 18-22 years respectively (Appendix Table 10 &11 ). This is 1.8 to 2 times higher than 374 the more inclusive definition for CYP aged ≥ 1 year. Refitting the Cox proportional hazard 375 models by age using the specific definition confirmed that presence of one or more chronic 376 conditions remained the only significant risk factor for hospital admission (Appendix Tables  377 12 & 13). The median length of stay remained 2 days (IQR 1-5). 378

Over one fifth of CYP in Scotland had at least one SARS-CoV-2 PCR test during 2020, and 380 1.5% had a PCR-confirmed infection. Testing rates increased with increasing age. PCR-381 confirmed infection rates were highest in 18-22-year-olds, followed by secondary school 382 children and infants. CYP with chronic conditions were more likely to be tested, but CYP with 383 chronic conditions of secondary school age or older were less likely to have a PCR 384 confirmed infection. Whilst COVID-19 related hospital admissions were uncommon (less 385 than 3 per 10,000 CYP admitted in 2020), infants and those with multiple types of chronic 386 conditions recorded had the highest COVID-19 related-admission rates. Preschool children 387 from lower SEP groups were more likely to have PCR-confirmed infection, and secondary 388 school children and young adults from lower SEP groups less likely. Across all ages, SEP 389 was significantly associated with COVID-related hospital admission. We identified no 390 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; https://doi.org/10.1101/2021.12.17.21267350 doi: medRxiv preprint statistically significant associations between prematurity, nor BMI and COVID-related 391 admission risk; however the number of children admitted was small. 392

The well-established Scottish data linkage infrastructure allowed us to include data for all 393 CYP born in Scotland since 1997, thereby minimising selection bias and loss to follow-up. 394

We relied on linkage between hospital admission and public health surveillance data to 395 define COVID-19-related admissions, allowing us to examine the robustness of our 396 definitions in the linked data, rather than only relying on time difference between SARS-CoV-397 2 positive test and hospital admission alone. Importantly, by using a national birth cohort for 398 this study, we could examine variations in population-based rates of SARS-CoV-2 testing, 399 PCR confirmed infections and COVID-19-related hospital admissions across the full CYP 400 age range, rather than only examining risk factors for ICU admission and death in 401 hospitalised children. Using a population-wide study sample also minimises the risk of 402 collider bias 28 domains (e.g. school absences), further studies will need to examine whether the COVID-19 424 vaccination programme has amended the admission risks reported in this study. However, 425 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; https://doi.org/10.1101/2021.12.17.21267350 doi: medRxiv preprint the results reported here provide a baseline during the first pandemic year against which 426 more recent data can be compared. 427

We based our classification of chronic conditions on coded information in SMR-01 and SBR 428 records; this may mean we have missed some conditions that are primarily managed in 429 primary or community care settings. Our classification of socio-economic position was based 430 on parental occupation derived from birth certificates, however, particularly among older 431 CYP this may not reflect current socio-economic circumstances. 432

Our study confirms that COVID-19-related hospital admissions among children and young 433 adults are uncommon, with admission rates in CYP between 2% and a third of rates 434 observed in adults over 45 years (depending on age group). 33 Most admissions were in 435 children with no prior health condition recorded. We highlight that infants have the highest 436 admission risk among CYP, despite higher testing rates in older children and young adults. A 437 previous systematic review (preprint) has indicated that infants are also at highest risk of 438 requiring PICU admission once in hospital with COVID-19 disease. 14 However, admission 439 rates in infancy related to SARS-CoV-2 (1/1000 child-years) is lower than admission rates 440 associated with confirmed influenza (2/1000 child-years) 34 We demonstrated that a history of chronic conditions, and particularly living with multiple 446 different types of chronic conditions, was the most prominent risk factor for COVID-19 447 related hospital admission rates among CYP, however the vast majority of admitted children 448 had no chronic conditions recorded. Children with chronic conditions also experience higher 449 hospital admission rates for other conditions and injuries. 36 CYP with chronic conditions were 450 more likely to be tested than those without, however in older age groups CYP with chronic 451 conditions were less likely to be positive. This may reflect lower threshold for testing among 452 high-risk groups. 453

Preschool children from lower SEP groups (indicated by parental occupation recorded on the 454 birth certificate) had higher risks of PCR-confirmed infection than children from higher SEP 455 groups. This may be because younger children are spending more time in the home with 456 their parents, and their risk of infection is therefore more strongly associated with their 457 parents' occupation (and ability to work from home). In older CYP, we instead identified 458 higher PCR-confirmed infection rates among higher SEP groups, despite lower testing rates. 459 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint Low SEP at birth was also a risk factor for COVID-19-related admission in CYP in the overall 466 model (incorporating all ages). This confirms previous reports which have indicated higher 467 risk of severe outcomes in hospitalised adult COVID patients from more deprived areas, 38 468 and higher all-age hospital admission rates in areas with higher area deprivation scores. 39 40 469

Although we did not identify SEP as a significant risk factor for admission in age group 470 specific models, COVID-19 related admission rates in children are much lower than in 471 adults. Therefore, systematic differences in admission rates by SEP among specific age 472 groups of CYP are harder to detect, even when using national data. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint We are grateful to Professor Bianca De Stavola (UCL) for her advice on statistical modelling, 499 and Diane Rennie from Public Health Scotland for her help with data access and disclosure 500 checking. This work uses data provided by patients and collected by the NHS as part of their 501 care and support 502

None 504 505 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted December 17, 2021. ; https://doi.org/10.1101/2021.12.17.21267350 doi: medRxiv preprint

Age-dependent effects in the transmission and control 507 of COVID-19 epidemics

European Centre for Disease Prevention and Control. COVID-19 in children and the role 510 of school settings in transmission -first update

512 the-role-of-school-settings-in-transmission-first-update_1.pdf Accessed: 02/03/2021 513 3. Centre for Disease Control and Prevention. Coronavirus Disease 2019 in Children -514 United States

Characteristics of and Important Lessons From the Coronavirus 517 Disease 2019 (COVID-19) Outbreak in China: Summary of a

Cases From the Chinese Center for Disease Control and Prevention

Screening and Severity of Coronavirus Disease 521 2019 (COVID-19) in Children in

Clinical characteristics of children and young 529 people admitted to hospital with covid-19 in United Kingdom: prospective multicentre 530 observational cohort study

COVID-19 in children and 532 adolescents in Europe: a multinational, multicentre cohort study. The Lancet Child & 533 Adolescent Health

Clinical characteristics, treatment and outcomes of 535 paediatric COVID-19: a systematic review and meta-analysis. Archives of Disease in 536

Association between SARS-CoV-2 infection and 538

Kawasaki-like multisystem inflammatory syndrome: a retrospective matched case-539 control study

SARS-CoV-2-related paediatric inflammatory 543 multisystem syndrome, an epidemiological study

Clinical Characteristics of 58 Children With a 547

Pediatric Inflammatory Multisystem Syndrome Temporally Associated With SARS-548

Paediatric multisystem inflammatory syndrome 550 temporally associated with SARS-CoV-2 (PIMS-TS): Prospective, national 551 surveillance, United Kingdom and Ireland, 2020. The Lancet Regional Health -552

Which children and young people are at higher 554 risk of severe disease and death after SARS-CoV-2 infection: a systematic review 555 and individual patient meta-analysis

Risk factors for intensive care admission and death 558 amongst children and young people admitted to hospital with COVID-19 and PIMS-559 TS in England during the first pandemic year

Myocarditis after Covid-19 Vaccination in a Large

Health Care Organization

Acute Myocarditis Following COVID-19 mRNA 565 Vaccination in Adults Aged 18 Years or Older

Risk of COVID-19 hospital admission among children 568 aged 5-17 years with asthma in Scotland: a national incident cohort study. The 569 Lancet Respiratory Medicine

Air Pollution, housing and respiratory tract 571 Infections in Children: NatIonal birth Cohort study (PICNIC): study protocol

The contribution of child, family and 574 health service factors to respiratory syncytial virus (RSV) hospital admissions in the 575 first 3 years of life: birth cohort study in Scotland

Risk of hospital admission with covid-19 among 579 teachers compared with healthcare workers and other adults of working age in 580

population based case-control study

Readmission after discharge from acute mental 583 healthcare among 231 988 people in England: cohort study exploring predictors of 584 readmission including availability of acute day units in local areas

Child deaths due to injury in the four UK countries: 591 a time trends study from 1980 to 2010

British 1990 growth reference centiles for weight, 594 height, body mass index and head circumference fitted by maximum penalized 595 likelihood

Partial residuals for the proportional hazards regression model

Joint Committee of Vaccination and Immunisation. JCVI statement on COVID-19 608 vaccination of children and young people aged 12 to 17 years

vaccination-of-children-and-young-people-aged-12-to-17-years/jcvi-statement-on-616 covid-19-vaccination-of-children-and-young-people-aged-12-to-17-years-4-august-617 2021 618 32. Joint Committee of Vaccination and Immunisation

Weekly COVID-19 Statistical Data in Scotland

Risk factors for admission to hospital 626 with laboratory-confirmed influenza in young children: birth cohort study

The contribution of child, family and 629 health service factors to respiratory syncytial virus (RSV) hospital admissions in the 630 first 3 years of life: birth cohort study in Scotland

Influence of socioeconomic deprivation on 639 interventions and outcomes for patients admitted with COVID-19 to critical care units 640 in Scotland: A national cohort study. The Lancet Regional Health

A Case-Control Study to Estimate the 651 Effectiveness of Maternal Pertussis Vaccination in Protecting Newborn

Infant Hospitalizations and Fatalities Averted by 655 the Maternal Pertussis Vaccination Program in England

Influenza Vaccine Given to Pregnant Women 659 Reduces Hospitalization Due to Influenza in Their Infants

Effectiveness of maternal 662 immunization with trivalent inactivated influenza vaccine in pregnant women and their 663 infants