Skip to main content
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2019 May 22.
Published in final edited form as: Vaccine. 2018 Nov 23;37(1):76–79. doi: 10.1016/j.vaccine.2018.11.038

Similar relative risks of seizures following measles containing vaccination in children born preterm compared to full-term without previous seizures or seizure-related disorders

David L McClure a,*, Steven J Jacobsen b, Nicola P Klein c, Allison L Naleway d, Elyse O Kharbanda e, Jason M Glanz f, Lisa A Jackson g, Eric S Weintraub h, Huong Q McLean a
PMCID: PMC6530777  NIHMSID: NIHMS1019380  PMID: 30478005

Abstract

Background:

Febrile seizures are associated with the first dose of measles-containing vaccines and the risk increases with chronologic age during the second year of life. We used the Vaccine Safety Datalink (VSD) to determine if the relative increase in risk of seizures following receipt of measles-containing vaccine differs by gestational age at birth.

Methods:

Children were eligible if they received their first dose of measles-containing vaccine at age 12 through 23 months from January 2003 through September 2015. Children were excluded if they had a history of seizure or conditions strongly related to seizure prior to 12 months of age. Seizures were identified by diagnostic codes in the inpatient or emergency department settings. Using risk-interval analysis, we estimated the incidence rate ratio (IRR) for seizures in the 7 through 10 days (risk period) vs 15 through 42 days (control period) following receipt of measles-containing vaccines in children born preterm (<37 weeks gestation age) and those born full-term (≤37 weeks).

Results:

There were 532,375 children (45,343 preterm and 487,032 full-term) who received their first dose of measles-containing vaccine at age 12 through 23 months. The IRRs of febrile seizures 7 through 10 days compared with 15 through 42 days after receipt of measles-containing vaccine were 3.9 (95% CI: 2.5–6.0) in preterm children and 3.2 (2.7–3.7) in full-term children; the ratio of IRRs: was 1.2 (0.76–1.9), p = 0.41. IRRs were also similar across gestational age groups, by vaccine type received (measles-mumps-rubella [MMR] or measles-mumps-rubella-varicella [MMRV]) and age at vaccination (12–15 or 16–23 months).

Conclusion:

Vaccination with a measles-containing vaccine in the second year of life is associated with a similar relative risk of a first seizure in children born preterm as in those who were born full-term.

Keywords: Measles containing vaccine, MMR, MMRV, Seizures, Preterm, Premature

1. Background

Combination measles-mumps-rubella (MMR) and measles-mumps-rubella-varicella (MMRV) vaccines are two measles-containing vaccines currently available in the United States [1]. Measles-containing vaccines are associated with an increased risk of febrile seizures in the second week after vaccination [28]. Among children aged 12 through 23 months, the risk of febrile seizures is approximately 2 times higher in those who received the first dose as MMRV vaccine compared to those who received separate MMR and varicella vaccines at the same visit [5,9]. Age at vaccination has also been associated with risk of seizures. Children who received their first dose of measles-containing vaccine at age 16–23 months had a higher risk of seizures than those vaccinated at age 12–15 months, the age recommended by the U.S. Advisory Committee on Immunization Practices (ACIP) [1,6,10]. Although the biological mechanism for febrile seizures is unclear, the age associated risk is likely due to multiple factors, including peak incidence of febrile seizures occurring at age 15–18 months [11] and maturation of the brain’s sensitivity to fever [12]. Additionally, differences in the underlying medical conditions or health-seeking behaviors may exist between those who are vaccinated at the recommended age versus those who receive it later [6]. Parents of preterm infants may delay MMR vaccination for their child [13,14], as the incidence of febrile seizures is higher in children who are born preterm than those born full-term [1517]. However, data concerning the safety of vaccination in children born preterm compared to children born full-term is limited particularly for vaccinations received during the second year of life [18,19].

We aimed to examine if the risk of febrile seizures during the 7 through 10 days versus 15 through 42 days following receipt of measles-containing vaccine is increased in children born preterm compared to children born full-term. We also assessed whether the difference in risk of febrile seizures following receipt of measles-containing vaccine (if any) might vary by vaccine type received (MMR or MMRV) or age at vaccination (12 through 15 or 16 through 23 months).

2. Methods

2.1. Study population

The study population consisted of children who received their first dose of measles-containing vaccine at age 12 through 23 months from January 1, 2003 through September 30, 2015. These children were members at seven integrated health care organizations (sites) located throughout the United States that participate in the Centers for Disease Control and Prevention (CDC) sponsored Vaccine Safety Datalink (VSD) [20,21]. Institutional Review Boards at each VSD site approved this study with a waiver of informed consent.

Children were included if they were born at a study site facility, were continuously enrolled as members of the study site to age 24 months, and had gestational age data. Estimated gestational age at birth was determined from the VSD pregnancy algorithm [22]. In brief, the algorithm captures gestational age from each site’s electronic medical record databases (e.g., procedure dates, ultrasound reports, laboratory tests and dates) and supplemented with birth certificate data from the corresponding state vital records office. When gestational age was not available, it was imputed using published outcome specific averages or definitions. [22]. We classified children born before 37 weeks gestational age as preterm and children born 37 weeks gestational age as full term [23]. We also further classified preterm into those born <35 weeks (early preterm) and 35 through 36 weeks (late preterm) gestational age. Children with epilepsy, cerebral palsy, severe head trauma, intraventricular hemorrhage, intracranial tumor, meningitis, or encephalitis based on documented International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnostic codes in the electronic medical record were excluded (Supplemental Table). Children with an ICD-9-CM coded diagnosis of seizure prior to 12 months of age were also excluded. The exclusion criteria were determined by consensus of coauthors and were conditions known to be associated with increased risk of seizures or recurrence of seizures.

Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.vaccine.2018.11.038.

2.2. Outcome

A seizure was defined as the first emergency department or inpatient hospital encounter with ICD-9-CM diagnostic code of 780.3 (convulsions) during the 42-days following vaccination; we did not distinguish between febrile and afebrile seizures. However, previous VSD studies have found a high confirmation rate (94% for acute seizures, 87% for acute febrile seizures) for this diagnostic code when assigned from emergency department or inpatient hospital encounters [5,6,16,24].

2.3. Analytic approach

We conducted a risk-interval analysis among vaccinated children [25] with each child having 42 days of follow-up following receipt of a measles-containing vaccine. Days 7 through 10 following vaccination were defined as the risk interval and days 15 through 42 following vaccination were defined as the control interval. Days 0 through 6 and 11 through 14 following vaccination were excluded. The first exclusion reduced possible short-term effects with concomitant vaccines [3,2628] and the latter exclusion was to avoid residual exposure effects in the control interval. Children with seizures occurring in these excluded intervals were excluded in the analysis.

Seizure incidence rates and 95% confidence intervals per 1000 person-years were calculated for the risk and control intervals. To examine the possible effect of gestational age differences, we compared the incidence rate ratios (IRRs) of seizures between children born preterm and those born full-term using a Poisson regression model with an effect-modification term for gestational age category by exposure status (risk or control interval). The effect modification estimate was expressed as the ratio of IRRs with its 95% confidence interval and p-value [29]. Separate models were generated by vaccine received (MMR and MMRV) and by child’s age at vaccination (12 through 15 and 16 through 23 months) since seizure risk following vaccination differ by vaccine type and age at vaccination, respectively [57].

We also performed a temporal scan of all cases within 1 through 42 days post vaccination interval [30] to determine if the 7 through 10 day risk interval was appropriate for the gestational age categories, vaccine received, and child’s age at vaccination. Periods in the 1 through 42 days post vaccination among the largest log-likelihood were considered clusters for further analysis.

All analyses were performed in SAS 9.4 (SAS Institute, Cary, NC).

3. Results

There were 556,864 children who received their first dose of measles-containing vaccine at age 12 through 23 months. Of these, 24,489 children were excluded because of documented history of seizure before age 12 months (<1%) or other exclusionary diagnoses (4%). The remaining 532,375 children were included in the analysis; 45,343 were born preterm (8.5%) and 487,032 born full-term (91.5%). There were no meaningful differences between those born preterm and full-term by sex, age at vaccination, or maternal age (Table 1). As expected, children born preterm had lower mean birth weights (30% less) than those born full-term.

Table 1.

Characteristics of children aged 12–23 months at the time of receipt of first dose of measles-containing vaccine by gestational age.

Gestational age category (weeks)
Preterm (<37) Full-term (≤37)
Number of children 45,343 487,032
Vaccine received, n (%)
MMR 37,262 (82.2) 403,238 (82.8)
MMRV 8081 (17.8) 83,794 (17.2)
Sex, n (%)
Female 20,779 (45.8) 239,775 (49.2)
Male 24,564 (54.2) 247,257 (50.8)
Age at vaccination in months, mean (SD) 13.30 (1.98) 13.33 (2.01)
Calendar month of vaccination, n (%)
November through April 22,197 (49.0) 234,430 (48.1)
May through October 23,146 (51.0) 252,602 (51.9)
Birth weight in grams, mean (SD) 2417 (6 4 9) 3449 (4 7 4)
Maternal age in years, mean (SD) 31.9 (5.7) 31.1 (5.3)

Abbreviations: MMR, measles-mumps-rubella; MMRV, measles-mumps-rubella-varicella; SD, standard deviation.

Of the 996 children with a seizure episode during the 42 days following vaccination, 167 seizures occurred 0 through 6 or 11 through 14 days after vaccination and were excluded. The remaining 829 children with seizures were included in the analysis; 87 in children born preterm (32 in children born <35 weeks gestational age and 55 in children born 35 through 36 weeks) and 742 in those born full-term. Seizures occurred in the 7 through 10 days post-vaccination risk period in 263 children and in the 15 through 42 days post-vaccination control period in 566 children. The mean age at seizure events were similar for both children born preterm (13.5 months) and children born full-term (13.3 months).

In the overall primary analysis, the IRR point estimate of children born preterm was 20% larger than children born full-term, although the precision was low [ratio of IRRs: 1.2 (0.76–1.9), p = 0.41], (Table 2). From the secondary analyses, similar IRR results were observed for children born <35 weeks or 35 through 36 weeks gestational age, among those receiving MMR or MMRV, and children vaccinated at age 12 through 15 months. For children vaccinated at age 16 through 23 months, the point estimate IRR was 20% lower for children born preterm vs full-term; precision was also low [ratio of IRRs: 0.82 (0.20–3.4), p = 0.79]. Overall, and for all vaccine and age strata, the risk-interval (exposed) incidence rates were 20–40% larger for children born preterm vs. full-term. The largest incidence rate point estimate was for children born preterm who received MMRV (100 seizures per 1000 person-years). The lowest point estimate (10 seizures per 1000 person-years) was observed for full-term children vaccinated at age 16 through 23 months during the unexposed control interval.

Table 2.

Incidence rate (per 1000 person-years) and rate ratios of seizures following vaccination overall and by gestational age category, child’s age at vaccination, and vaccine received.

7 through 10 days after vaccination (risk interval)
15 through 42 days after vaccination (control interval)
Gestational age category Number of children Number of seizures Incidence ratea (95% CI) Number of seizures Incidence ratea (95% CI) IRR (95% CI)b Preterm vs full-term, ratio of IRR (95% CI), p
Overall
Preterm (<37 weeks) 45,343 31 62 (42–89) 56 16 (12–21) 3.9 (2.5–6.0) 1.2 (0.76–1.9), 0.41
<35 weeks 16,586 10 55 (26–100) 22 17 (11–26) 3.2 (1.5–6.7) 1.0 (0.47–2.1), 0.99
35–36 weeks 28,757 21 67 (41–100) 34 15 (11–22) 4.3 (2.5–7.4) 1.4 (0.77–2.4), 0.29
Full-term (≥37 weeks) 487,032 232 43 (38–49) 510 14 (12–15) 3.2 (2.7–3.7)
Vaccine received
MMR
  Preterm 37,262 22 54 (34–82) 48 17 (12–22) 3.2 (1.9–5.3) 1.2 (0.70–2.0), 0.51
  Full-term 403,238 163 37 (31–43) 425 14 (12–15) 2.7 (2.2–3.2)
MMRV
  Preterm 8081 9 100 (47–190) 8 13 (5.6–25) 7.9 (3.0–20) 1.4 (0.51–3.8), 0.52
  Full-term 83,794 69 76 (59–95) 85 13 (11–16) 5.7 (4.1–7.8)
Child’s age at vaccination
12–15 months
  Preterm 41,391 27 60 (39–87) 51 16 (12–21) 3.7 (2.3–5.9) 1.3 (0.77–2.1), 0.36
  Full-term 442,919 200 41 (36–47) 477 14 (13–15) 2.9 (2.5–3.5)
16–23 months
  Preterm 3952 4 92 (25–237) 5 17 (5.4–39) 5.6 (1.5–21) 0.82 (0.20–3.4), 0.79
  Full-term 44,113 32 66 (45–94) 33 10 (6.7–14) 6.8 (4.2–11)

Abbreviations: MMR, measles-mumps-rubella; MMRV, measles-mumps-rubella-varicella; IRR: incident rate ratio, CI, confidence interval.

a

Per 1000 person-years.

b

IRR (7–10 days versus 15–42 days after vaccination).

Temporal scans showed that the 7 through 10 day post-vaccination period was identified as a high-risk cluster overall and in the gestational age and vaccine type subgroups (Supplemental Figures).

4. Discussion

We examined the incidence rates and incidence rate ratios of seizures following receipt of measles-containing vaccine in the second year of life by gestational age for children without previous seizures or seizure-related disorders. Generally, the exposed incidence rate and IRR point estimates were similar among children born pre-term compared to full-term. The ratio of IRRs in the two groups were near 1, with 95% confidence intervals that ranged from <1 to substantially >1. Infants born preterm are recommended to receive vaccinations at the same chronological age recommended for infants born full-term [31]. Existing data for vaccination of preterm infants are reassuring with no apparent safety concerns, and although immune response to vaccines in the first year of life may be lower for infants born preterm than full-term, the differences are likely to be not clinically relevant [32]. For vaccinations received during the second year of life, differences in immune response between preterm and full-term are less significant [33]. In our study, the overall incidence rate of seizures following vaccination was low and the incidence rate ratios essentially did not differ between children born preterm and those born full-term. The increased risk with delayed receipt of measles-containing vaccine or with receipt of MMRV is consistent with previous VSD studies [57], and is unrelated to gestational age.

This was a large population based study with highly accurate vaccination dates [34] and use of previously validated diagnostic codes for the seizure outcome [5,6,16,24]. However, our study had several potential limitations. First, we did not include children who may have delayed receipt of measles-containing vaccine past age 23 months. Second, there were relatively few seizure cases among children born preterm, particularly among those who were vaccinated after age 15 months and those who received MMRV vaccine. Third, most of the preterm children were born at 35 through 36 weeks gestational age (late preterm) while few children were born at <35 weeks gestational age. Finally, we excluded children with a previous history of seizure or risk factors related to seizure so our results may not apply to this population. Notably, “personal or family history of seizures” is a precaution (but not a contraindication) of measles-containing vaccines [1]. We excluded children with a personal history of seizures before 12 months of age, based on electronic diagnostic codes, in our attempt to independently assess the risk of seizures due to measles-containing vaccines among children born preterm. We were not able to assess family history, as data were not readily available in the electronic health record.

In conclusion, our results support the current ACIP recommendations to administer the first dose of measles-containing vaccine at age 12 through 15 months for all children, including those born preterm [1,13,31]. Delaying vaccination of measles-containing vaccines may increase the risk of seizures following vaccination.

Supplementary Material

suppl figure
supplmental table

Acknowledgments

We gratefully acknowledge the contributions of Edward Belongia, Erica Scotty, Elizabeth Vickers, Melissa Simpson and VSD study site project and data management staff.

Funding statement

This work was supported by the Centers for Disease Control and Prevention [contract number 200-2012-53587].

Disclaimer

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

ALN receives research support from Pfizer, Merck, and MedImmune (now AstraZeneca) for unrelated studies. NPK receives research support from GlaxoSmithKline, Sanofi Pasteur, Protein Science, Pfizer, Merck, and MedImmune (now AstraZeneca) for unrelated studies.

Footnotes

Conflicts of interest

The remaining authors report no conflicts of interest.

References

  • [1].McLean HQ, Fiebelkorn AP, Temte JL, Wallace GS. Prevention of measles, rubella, congenital rubella syndrome, and mumps, 2013: summary recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep 2013;62(RR-04):1–34. [PubMed] [Google Scholar]
  • [2].Barlow WE, Davis RL, Glasser JW, et al. The risk of seizures after receipt of whole-cell pertussis or measles, mumps, and rubella vaccine. N Engl J Med 2001;345(9):656–61. [DOI] [PubMed] [Google Scholar]
  • [3].Farrington P, Pugh S, Colville A, et al. A new method for active surveillance of adverse events from diphtheria/tetanus/pertussis and measles/mumps/rubella vaccines. Lancet 1995;345(8949):567–9. [DOI] [PubMed] [Google Scholar]
  • [4].Griffin MR, Ray WA, Mortimer EA, Fenichel GM, Schaffner W. Risk of seizures after measles-mumps-rubella immunization. Pediatrics 1991;88(5):881–5. [PubMed] [Google Scholar]
  • [5].Klein NP, Fireman B, Yih WK, et al. Measles-mumps-rubella-varicella combination vaccine and the risk of febrile seizures. Pediatrics 2010;126(1): e1–8. [DOI] [PubMed] [Google Scholar]
  • [6].Rowhani-Rahbar A, Fireman B, Lewis E, et al. Effect of age on the risk of Fever and seizures following immunization with measles-containing vaccines in children. JAMA Pediatr 2013;167(12):1111–7. [DOI] [PubMed] [Google Scholar]
  • [7].Klein NP, Lewis E, Fireman B, et al. Safety of measles-containing vaccines in 1-year-old children. Pediatrics 2015;135(2):e321–9. [DOI] [PubMed] [Google Scholar]
  • [8].Klein NP, Lewis E, McDonald J, et al. Risk factors and familial clustering for fever 7–10days after the first dose of measles vaccines. Vaccine 2017;35 (12):1615–21. [DOI] [PubMed] [Google Scholar]
  • [9].Jacobsen SJ, Ackerson BK, Sy LS, et al. Observational safety study of febrile convulsion following first dose MMRV vaccination in a managed care setting. Vaccine 2009;27(34):4656–61. [DOI] [PubMed] [Google Scholar]
  • [10].Hambidge SJ, Newcomer SR, Narwaney KJ, et al. Timely versus delayed early childhood vaccination and seizures. Pediatrics 2014;133(6):e1492–9. [DOI] [PubMed] [Google Scholar]
  • [11].Hauser WA. The prevalence and incidence of convulsive disorders in children. Epilepsia 1994;35(Suppl 2):S1–6. [DOI] [PubMed] [Google Scholar]
  • [12].Waruiru C, Appleton R. Febrile seizures: an update. Arch Dis Child 2004;89 (8):751–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Sisson H, Gardiner E, Watson R. Vaccination timeliness in preterm infants: An integrative review of the literature. J Clin Nurs 2017. [DOI] [PubMed]
  • [14].Davis RL, Rubanowice D, Shinefield HR, et al. Immunization levels among premature and low-birth-weight infants and risk factors for delayed up-to-date immunization status. Centers for Disease Control and Prevention Vaccine Safety Datalink Group. JAMA 1999;282(6):547–53. [DOI] [PubMed] [Google Scholar]
  • [15].Vestergaard M, Basso O, Henriksen TB, Ostergaard JR, Olsen J. Risk factors for febrile convulsions. Epidemiology 2002;13(3):282–7. [DOI] [PubMed] [Google Scholar]
  • [16].Tartof SY, Tseng HF, Liu AL, et al. Exploring the risk factors for vaccine-associated and non-vaccine associated febrile seizures in a large pediatric cohort. Vaccine 2014;32(22):2574–81. [DOI] [PubMed] [Google Scholar]
  • [17].Gavrielov-Yusim N, Hoshen M, Singer SR, Neumann L, Balicer RD. The weight of MMRV-related febrile convulsions among other clinical factors contributing to febrile convulsions in children. Vaccine 2014;32(39):4954–9. [DOI] [PubMed] [Google Scholar]
  • [18].Institute of Medicine. The childhood immunization schedule and safety: stakeholder concerns, scientific evidence, and future studies Washington (DC): National Academies Press; 2013. [PubMed] [Google Scholar]
  • [19].Klein NP. Vaccine safety in special populations. Hum Vaccin 2011;7 (2):269–71. [DOI] [PubMed] [Google Scholar]
  • [20].Baggs J, Gee J, Lewis E, et al. The Vaccine Safety Datalink: a model for monitoring immunization safety. Pediatrics 2011;127(Suppl 1):S45–53. [DOI] [PubMed] [Google Scholar]
  • [21].McNeil MM, Gee J, Weintraub ES, et al. The Vaccine Safety Datalink: successes and challenges monitoring vaccine safety. Vaccine 2014;32(42):5390–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Naleway AL, Gold R, Kurosky S, et al. Identifying pregnancy episodes, outcomes, and mother-infant pairs in the Vaccine Safety Datalink. Vaccine 2013;31(27):2898–903. [DOI] [PubMed] [Google Scholar]
  • [23].Spong CY. Defining “term” pregnancy: recommendations from the Defining “Term” Pregnancy Workgroup. JAMA 2013;309(23):2445–6. [DOI] [PubMed] [Google Scholar]
  • [24].Shui IM, Shi P, Dutta-Linn MM, et al. Predictive value of seizure ICD-9 codes for vaccine safety research. Vaccine 2009;27(39):5307–12. [DOI] [PubMed] [Google Scholar]
  • [25].Baker MA, Lieu TA, Li L, et al. A vaccine study design selection framework for the postlicensure rapid immunization safety monitoring program. Am J Epidemiol 2015;181(8):608–18. [DOI] [PubMed] [Google Scholar]
  • [26].Duffy J, Weintraub E, Hambidge SJ, et al. Febrile seizure risk after vaccination in children 6 to 23 months. Pediatrics 2016;138(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Kawai AT, Li L, Kulldorff M, et al. Absence of associations between influenza vaccines and increased risks of seizures, Guillain-Barre syndrome, encephalitis, or anaphylaxis in the 2012–2013 season. Pharmacoepidemiol Drug Saf 2014. [DOI] [PubMed]
  • [28].Tse A, Tseng HF, Greene SK, Vellozzi C, Lee GM, Group VSDRCAIW. Signal identification and evaluation for risk of febrile seizures in children following trivalent inactivated influenza vaccine in the Vaccine Safety Datalink Project, 2010–2011. Vaccine 2012; 30(11): p. 2024–31. [DOI] [PubMed] [Google Scholar]
  • [29].Knol MJ, VanderWeele TJ. Recommendations for presenting analyses of effect modification and interaction. Int J Epidemiol 2012;41(2):514–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].McClure DL, Xu S, Weintraub E, Glanz JM. An efficient statistical algorithm for a temporal scan statistic applied to vaccine safety analyses. Vaccine 2012;30 (27):3986–91. [DOI] [PubMed] [Google Scholar]
  • [31].Saari TN. American Academy of Pediatrics Committee on Infectious D. Immunization of preterm and low birth weight infants. American Academy of Pediatrics Committee on Infectious Diseases. Pediatrics 2003;112(1 Pt 1):193–8. [DOI] [PubMed] [Google Scholar]
  • [32].Gagneur A, Pinquier D, Quach C. Immunization of preterm infants. Hum Vaccin Immunother 2015;11(11):2556–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Esposito S, Fumagalli M, Principi N. Immunogenicity, safety and tolerability of vaccinations in premature infants. Expert Rev Vaccines 2012;11 (10):1199–209. [DOI] [PubMed] [Google Scholar]
  • [34].Mullooly J, Drew L, DeStefano F, et al. Quality of HMO vaccination databases used to monitor childhood vaccine safety. Vaccine Safety DataLink Team. Am J Epidemiol 1999;149(2):186–94. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

suppl figure
supplmental table

RESOURCES