Publication: Bulletin of the World Health Organization; Type: Research in emergencies Article ID: BLT.16.171009

DISCLAIMER

This paper was submitted to the Bulletin of the World Health Organization and was posted to the Zika open site, according to the protocol for public health emergencies for international concern as described in Christopher Dye et al. (http://dx.doi.org/10.2471/BLT.16.170860 ).

The information herein is available for unrestricted use, distribution and reproduction in any medium, provided that the original work is properly cited as indicated by the Creative Commons Attribution 3.0 Intergovernmental Organizations licence (CC BY IGO 3.0).

RECOMMENDED CITATION

Majumder MS, Cohn E, Fish D & Brownstein JS. Estimating a feasible serial interval range for Zika fever [Submitted]. Bull World Health Organ, E-pub: 9 Feb 2016. doi: http://dx.doi.org/10.2471/BLT.16.171009

Estimating a feasible serial interval range for Zika fever Maimuna S Majumder a, Emily Cohn b, Durland Fish c & John S Brownstein d a Massachusetts Institute of Technology, Cambridge, Massachusetts, 02139, United States of America (USA) bBoston Children’s Hospital, Boston, USA cYale School of Public Health, New Haven, USA dHarvard Medical School, Boston, USA Correspondence to Maimuna S. Majumder (email: [email protected]). (Submitted: 9 February 2016 – Published online: 9 February 2016)

Page 1 of 6 Publication: Bulletin of the World Health Organization; Type: Research in emergencies Article ID: BLT.16.171009 Funding Statement: This work was supported by the National Library of Medicine of the National Institutes of Health (R01LM010812). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The authors declare no conflicts of interest.

Abstract

Objective: In this paper, we determine a feasible serial interval range for Zika fever.

Methods: Existing estimates of intrinsic and extrinsic incubation, as well as viraemic period, are used to develop a simple life cycle model of Zika fever to determine expected time between two consecutive in a chain of . Case count alerts from the HealthMap digital surveillance system are then used to assess the feasibility of our estimated serial interval range.

Findings: Our model suggests that 10 to 23 days may be a feasible serial interval range for Zika fever. Though sparse, preliminary data from the Zapaca department of Guatemala and the Caribbean island of Martinique provide supporting evidence for this range.

Conclusion: Serial interval estimates from Zacapa (serial interval = 15 days) and Martinique (serial = 16 days) are reasonable when compared against estimates for similar arbovirus diseases, such as dengue [15, 17 days]. This suggests that our estimated serial interval range for Zika fever in Central and South America is likely feasible; moreover, it lends credibility to the recent practice of repackaging existing dengue models for predicting the forthcoming burden of Zika fever and potential Zika-related diseases (e.g. Guillain-Barré syndrome, microcephaly).

Page 2 of 6 Publication: Bulletin of the World Health Organization; Type: Research in emergencies Article ID: BLT.16.171009 Objective. Growing concern over the potential relationship between Zika virus in pregnant women and microcephaly in newborn babies has brought global attention to an ongoing outbreak of the disease in Central and South America. 1,2 Though evidence of vertical and sexual transmission exists, the virus is primarily transmitted via Aedes mosquitoes. 3,4,5 It is possible that Zika fever was introduced to the Americas during the 2014 FIFA World Cup and has since been propagated by the density of competent vectors in the region. 5

Prior to 2015, events related to Zika virus were considered to be fairly rare; because of this, epidemiological research has been somewhat limited to date. Markedly, there currently exists no estimate for the serial interval (generation time) of Zika fever in the literature. The serial interval (SI) of a given disease may be defined as the average length of time between two consecutive infections in a chain of transmission. While largely context-driven and variable across both outbreaks and individuals, establishing mechanistically feasible upper and lower bounds is necessary for mathematical modeling and disease mitigation.

Methods. Since the discovery of Zika virus in 1947, estimates for intrinsic incubation in humans (three to 12 days) and extrinsic incubation in mosquitos (four to six days post- meal) – as well viraemic period in infected humans (between days three and five after symptom onset) – have been established. 6,7 Using these parameter estimates, we developed a simple life cycle model for Zika fever to determine the expected time between two successive cases (figure). We then assessed the feasibility of our estimated serial interval range by examining Zika fever case count data over time from the HealthMap digital disease surveillance system. 8

Findings. Assuming that historical parameter estimates for intrinsic incubation, extrinsic incubation, and viraemic period remain accurate, our model suggests that 10 to 23 days may be a feasible serial interval range for Zika fever (figure).

After narrowing our search criteria to settings with minimal likelihood of reintroduction and confirmed autochthonous Zika virus transmission, we found case count data from two suitable locations in the HealthMap system: the Zacapa department of Guatemala and the small Caribbean island of Martinique. An was reported in Zacapa on 24-November-2015; then, 15 days later, 22 subsequent cases were reported on 9-December-2015. More recently, an index case was reported in Martinique on 23- December-2015; this was followed by a report of 11 subsequent cases 16 days later on 8- January-2016.

Conclusion. Properly estimating the serial interval of any given disease is essential for the assessment of real-time dynamics, in turn enabling more effective deployment and evaluation of interventions. Though sparse, preliminary digital surveillance data from the HealthMap system suggest that our estimated serial interval range of 10 to 23 days is likely feasible for Zika fever in the current context of Central and South America.

Page 3 of 6 Publication: Bulletin of the World Health Organization; Type: Research in emergencies Article ID: BLT.16.171009 Notably, these data-driven serial interval estimates from Zacapa (SI = 15 days) and Martinique (SI = 16 days) are reasonable when compared against existing estimates for similar arbovirus diseases, such as dengue [15, 17 days]. 9 This congruence lends credibility not only to our estimated serial interval range, but also to the ongoing repackaging of existing dengue models for predicting the forthcoming burden of Zika fever and potential Zika-related diseases (e.g. Guillain-Barré syndrome, microcephaly). 10

Continued collection and analysis of case count data over time, via both digital and traditional means, will allow us to further refine our understanding of Zika fever and its transmissibility. If data similar to what we acquired for Zacapa and Martinique become available for other locations, it may be possible to determine an empirical probability distribution for the serial interval associated with Zika fever – allowing us to build more accurate projection models in the months and years ahead.

Page 4 of 6 Publication: Bulletin of the World Health Organization; Type: Research in emergencies Article ID: BLT.16.171009

Figure Legend. A simple life cycle model depicting a feasible serial interval range for Zika fever. Two examples are shown to demonstrate the lower (MIN = 10 days) and upper (MAX = 23 days) bounds of the range. *Time between viraemia (infectivity) and bite time is assumed to be negligible.

Page 5 of 6 Publication: Bulletin of the World Health Organization; Type: Research in emergencies Article ID: BLT.16.171009 References

1. European Centre for Disease Prevention and Control. Rapid risk assessment: Zika virus epidemic in the Americas: potential association with microcephaly and Guillain- Barré syndrome – 10 December 2015. Stockholm: ECDC 2015.

2. WHO. WHO Director-General summarizes the outcome of the Emergency Committee regarding clusters of microcephaly and Guillain-Barré syndrome. http://www.who.int/mediacentre/news/statements/2016/emergency-committee-zika- microcephaly/en/. Accessed February 5, 2016.

3. Besnard M, Lastère S, Teissier A, Cao-Lormeau VM, Musso D. Evidence of perinatal transmission of Zika virus, French Polynesia, December 2013 and February 2014. Euro Surveill 2014; 19 (13): pii=20751.

4. Musso D, Roche C, Robin E, Nhan T, Teissier A, Cao-Lormeau VM. Potential sexual transmission of Zika virus. Emerg Infect Dis 2015; 21 (2): 359-61.

5. Marcondes CB, Ximenes MF. Zika virus in Brazil and the danger of infestation by Aedes (Stegomyia) mosquitoes. Rev Soc Bras Med Trop 2015; pii=S0037- 86822015005003102.

6. European Centre for Disease Prevention and Control. Rapid risk assessment: Zika virus outbreak, Brazil and the Pacific region – 25 May 2015. Stockholm: ECDC 2015.

7. Li MI, Wong PSJ, Ng LC, Tan CH. Oral Susceptibility of Singapore Aedes (Stegomyia) aegypti (Linnaeus) to Zika Virus. PLOS Neglect Trop D 2012; 6(8): e1792.

8. HealthMap. Zika Virus. http://www.healthmap.org/zika/. Accessed February 5, 2016.

9. Aldstadt J, Yoon IK, Tannitisupawong D et al. Space-time analysis of hospitalised dengue patients in rural Thailand reveals important temporal intervals in the pattern of dengue virus transmission. Trop Med Int Health 2012; 17 (9): 1076-85.

10. STAT News. What’s behind the prediction of 3 to 4 million cases of Zika virus this year? http://www.statnews.com/2016/01/29/zika-case-estimate-4-million/. Accessed February 5, 2016.

Page 6 of 6