Disease burden and seasonal impact of improving rotavirus vaccine coverage in the United States: A modeling study.

BACKGROUND: Prior to vaccine introduction in 2006, rotavirus was the leading cause of severe diarrhea in children under five years of age in the U.S. Vaccination of infants has led to major reductions in disease burden, a shift in the seasonal peak and the emergence of a biennial pattern of disease. However, rotavirus vaccine coverage has remained relatively low (70-75%) compared to other infant immunizations in the U.S. Part of the reason for this lower coverage is that children whose care is provided by family practitioners (FP) have considerably lower probability of being vaccinated compared to those seen be pediatricians (PE). We used a dynamic transmission model to assess the impact of improving rotavirus vaccine coverage by FP and/or PE on rotavirus gastroenteritis (RVGE) incidence and seasonal patterns.
METHODS: A deterministic age-structured dynamic model with susceptible, infectious, and recovered compartments (SIRS model) was used to simulate rotavirus transmission and vaccination. We estimated the reduction of RVGE cases by 2 doses of rotavirus vaccine with three vaccination scenarios: (Status Quo: 85% coverage by pediatricians and 45% coverage by family practitioners; Improved FP: 85% coverage by pediatricians and family practitioners; Improved FP+PE: 95% coverage by pediatricians and family practitioners). In addition, we tested the sensitivity of the model to the assumption of random mixing patterns between children visiting pediatricians and children visiting family practitioners.
RESULTS: In this model, higher vaccine coverage provided by family practitioners and pediatricians leads to lower incidence of severe RVGE cases (23% averted in Improved FP and 57% averted in Improved FP+PE compared to Status Quo) including indirect effects. One critical impact of higher total vaccine coverage is the effect on rotavirus epidemic patterns in the U.S.; the biennial rotavirus epidemic patterns shifted to reduced annual epidemic patterns. Additionally, assortative mixing patterns in children visiting pediatricians and family practitioners amplify the impact of increasing vaccine coverage.
CONCLUSION: Other high-income countries that introduced vaccine have not experienced biennial patterns, like the U.S. Our results suggest that increasing overall vaccine coverage to 85% among infants would lead to an overall reduction in incidence with annual epidemic patterns.

Introduction

Globally, rotavirus is the leading cause of severe diarrhea, hospitalization, and diarrhea related deaths in infants and children younger than 5 years old [1]. Before the introduction of rotavirus vaccines in 2006, rotavirus caused an estimated 200,000 emergency room visits, 55,000 to 70,000 hospitalizations, and 20 to 60 deaths annually in children younger than 5 years of age, leading to approximately $1 billion in direct and indirect costs to the U.S. [2].

Data from the National Respiratory and Enteric Virus Surveillance System (NREVSS) in the U.S. indicates that rotavirus gastroenteritis (RVGE) incidence has declined between 57% - 89% since the introduction of vaccines in 2006 [3]. Rotavirus hospitalization rates have reduced between 70% - 98%, and all cause diarrhea-associated hospitalization rates declined between 9% - 76% in children under the age of 5 [4]. Rotavirus vaccines have also provided indirect benefits to unvaccinated individuals across the age range [5].

By 2016, full two dose coverage of rotavirus vaccines reached 74.1% in U.S. children 19–35 months of age. However, rotavirus vaccine coverage is still lower than other routine childhood recommended vaccines (DTap≥3 doses: 95%, Poliovirus≥3 doses: 93.7%, MMR ≥1 doses: 91.1%, Heb≥3 doses: 90.5% in 2016) and below the Healthy People 2020 goal of 80% coverage [6, 7]. Therefore, to meet the Healthy People 2020 goal of 80% coverage [8] and decrease further rotavirus disease and economic burden, promotion of rotavirus vaccine coverage needs to be considered.

Prior to rotavirus vaccination, RVGE showed a winter-spring seasonality and geographic patterns that begin in the southwest in December-January, extending across to the U.S. and ending in the northeast during April-May. In the post-vaccine era, the rotavirus season was shorter, delayed and of smaller magnitude compared to seasons in the pre-vaccine era, but a biennial pattern of RVGE incidence emerged [3, 9, 10]. This RVGE reduction was accompanied by biennial patterns with higher seasonal peak in 2009, 2011, and 2013 compared to lower seasonal peak in 2008, 2010, and 2012 [3, 7]. In general, biennial patterns of RVGE could be induced by factors that govern the rate of new susceptibles. These include incomplete vaccine coverage rates, imperfect vaccine efficacy, and birth rates [11-13]. In one study, a RVGE transmission model predicted that biennial patterns of RVGE emerge after the introduction of vaccines when birth rates are low, while an annual pattern of RVGE was predicted when birth rates are high [12]. In the U.S, birth rates have remained fairly stable in the pre and post-vaccine era. Thus, birth rate is likely not a significant driver of the biennial pattern of RVGE in the U.S. Incomplete vaccination coverage that leads to a build-up of susceptible children may be driving this biennial pattern [13]. Unvaccinated children drive the higher rotavirus hospitalizations in the biennial patterns [5]. In contrast, other developed countries with high coverage of rotavirus vaccination (>85%), such as Belgium, Austria, Australia, Finland, and Germany, did not have the biennial epidemic patterns after the introduction of vaccines [14-18].

Health care providers play a critical role in promoting vaccines [19]. Pediatricians and family practitioners are two essential rotavirus vaccine providers in the U.S. Pediatricians provide rotavirus vaccine to all eligible infants at considerably higher levels than family practitioners do (85% and 45%, respectively) [20]. Family practitioners may be more concerned about vaccine safety and the burden of adding additional vaccines to the childhood vaccination schedule [20, 21]. Accordingly, family practitioners could play an essential role in increasing rotavirus vaccine coverage and could impact RVGE incidence in the U.S. [22].

Our research aim is to assess the potential impact of improved rotavirus vaccine coverage in terms of disease incidence and epidemiological patterns of rotavirus in children under 5 in the U.S. Our main hypothesis is that the incidence of RVGE in the U.S. can be further decreased, with a shift from biennial to annual cycles if family practitioners were to provide rotavirus vaccine at the same level of pediatricians. We adapted and analyzed a dynamic transmission model of rotavirus disease and vaccination that included heterogeneity in vaccine coverage and contact patterns among children provided care by pediatricians or family practitioners.

Methods

Model design and model parameters

We used a dynamic deterministic model of rotavirus transmission with susceptible, infectious, and recovered compartments (SIRS model) to represent rotavirus transmission and vaccination. This model is age-structured with six age groups: 0–1 month, 2–3 months, 4–11 months, 1–4 years, 5–24 years, and above 25 years. The model equations are shown in the S1 Appendix. Individuals are born with maternal immunity which wanes at rate e. Susceptible individuals are infected at a rate λ and enter the infectious compartment. Infected individuals either recover from infection and gain long-term immunity at rate γ or become susceptible to subsequent infections. Immunity wanes at rate ω and individuals become susceptible again. We also included seasonal forcing in our model.

The model assumes individuals can have up to four rotavirus infections with decreasing probabilities of infection, disease and severe disease (εi, αi, σi) for each subsequent exposure (Table 1) [23]. In addition, we assumed only symptomatic individuals are infectious and primary infections contribute more to transmission than subsequent infections [24-27]. Only primary and secondary rotavirus infection were assumed to develop severe diarrhea. Additionally, we assumed primary infection, secondary infection, tertiary infection, and quaternary infection have the same duration of infectiousness.

Table 1

Natural history, demographic and estimated parameter values used in epidemiological model.

Parameter	Symbol	Parameter value	Description	Reference
Transmission probability	qi	q1 = 0.9998q2 = 0.4494q3 = 0.0472q4 = 0.0019	Probability of transmission per contact.q = 1…4 represent age group <1 year, 1–4 years; 5–24 years, > 25 years, respectively	Estimated
Seasonal transmission amplitude	A	0.0866	Proportion change in disease incidence	Estimated
Seasonal offset	θ	0.4942		Estimated
Reporting rate	δ	0.0538	Probability that severe RVGE case is reported	Estimated
Vaccine Efficacy	ψ	0.5		Calibrated
Daily rate of loss of immunity	ω	1/21,154	Rate at which immune individuals become re-susceptible infection	Atchison, 2010 [28]
Daily rate of loss of maternal immunity	e	1/90	Maternal immunity against rotavirus infection wane at a constant rate on average 90 days	Heymann, 2015 [29]
Daily rate of loss of infection	γ	1/5	Symptoms last 2–7 days but on average 5 days	Heymann, 2015 [29]
Risk of infection after previous infection	εi	ε1 = 0.62ε2 = 0.37ε3 = 0.37	After first infectionAfter second infectionAfter third infection	Velazquez et al., 1996 [23]
Proportion of symptomatic infection in nth infection	αi	α1 = 0.47α2 = 0.25α3 = 0.32α4 = 0.20	At first infectionAt second infectionAt third infectionAt fourth infection	Velazquez et al., 1996 [23]
Proportion of symptomatic infection associated with severe disease at nth infection	σi	σ1 = 0.28σ2 = 0.19	At first infectionAt second infection	Velazquez et al., 1996 [23]
Relative infectiousness of non-primary infections	r	r = 0.25		Velazquez et al., 1996 [23]
Daily aging rates for age group j	aj	a1 = 1/60a2 = 1/120a3 = 1/365a4 = 1/7,300	j = 1 …4 represent age group 0–3 months, 4–11 months, 1–4 years, and 5–24 years, respectively
Counts of total contacts	ci	c1 = 5.43c2 = 8.56c3 = 15.65c4 = 14.16	Counts for age <1 yearCounts for age 1–4 yearsCounts for age 5–24 yearsCounts for age >25 years	Mossong et al., 2008 [30]
Birth rate (Daily)	μ	1/30,827.7	U.S. 2017 birth rate	CDC Wonder [31]

We assumed the proportion of children visiting pediatricians was 84.4% while the proportion of children visiting family practitioners was 15.6% based on data from the MarketScan Research Database [32]. We separated children under one year of age into pediatrician and family practitioner-attending groups in the model to predict the impact of different vaccine coverage in these two groups. In the model, children get the first dose vaccine at two months of age and second dose at four months of age. U.S. birth rates were informed by the CDC Wonder database [31]. We assumed maternal immunity does not have an effect on vaccine efficacy and assumed that death rates equaled the birth rate so that the total population remains constant. In baseline analysis, we assumed an assortative contact structure between different age groups based on the POLYMOD study [30].

Parameters estimates

We estimated four age specific transmission parameters (q1 for <1 year, q2 for 1–4 years, q3 for 5–24 years, and q4 for > 25 years), seasonality parameters (A,ϴ) and a reporting rate. We used maximum likelihood by fitting the model to data on monthly counts of severe RVGE cases from the MarketScan Research Database. All analyses were conducted using the statistical program R version 1.1.423 [33], and the deSolve package to solve differential equations [34]. We calibrated a parameter for rotavirus vaccine efficacy to allow the model to capture observed biennial epidemic patterns (Table 1). Biennial epidemic patterns in the U.S. may be driven by a slower accumulation of susceptibles following modest vaccine coverage. By calibrating the vaccine efficacy and rate of loss of immunity parameters (i.e., increasing vaccine efficacy and lowering the rate of loss of immunity) we can slow the accumulation of the simulated susceptible population which allows the model to capture biennial epidemic patterns.

Vaccine scenario

We estimated the impact on severe RVGE cases of three vaccination scenarios, all based on 2-dose coverage: 85% coverage of the pediatrician (PE) population and 45% coverage of the family practitioner (FP) population, which is the present vaccine coverage (Status Quo); 85% coverage of the PE population and 85% coverage of the FP population, (Improved FP Coverage); and 95% coverage of the PE population and FP populations, (Improved FP + PE). For the Improved FP and Improved FP + PE scenarios, we assumed a change in vaccine coverage beginning in 2018. As outcomes, we calculated the percent of severe RVGE cases averted by comparing the rate of severe RVGE cases in Improved FP and Improved FP + PE to the average rate of severe RVGE cases in 2000–2006, prior to the introduction of vaccines, and each year from 2007 to 2030 in Status Quo. To quantify the impact of higher vaccine coverage on severe RVGE on the Status Quo population, the percent of severe RVGE cases averted for each vaccination scenario compared to Status Quo was calculated as follows: where i is the vaccine scenario (i = 2, 3). To quantify the impact of higher vaccine coverage in the pre-vaccine era, the percent of severe RVGE cases averted for each rotavirus vaccination scenario compared to pre-vaccine era was calculated as follows: where i is the vaccine scenario (i = 1, 2, 3).

Sensitivity analysis to assumptions about mixing patterns

Initially we assumed random mixing patterns between children visiting PEs and FPs. However, it is possible that there is assortative mixing within these groups such that children who attend FPs are more likely to mix with other such children. Therefore, we tested the sensitivity of the model to this assumption of mixing by setting contact within a group to be higher than contact between groups to depict an assortative mixing pattern. We assumed that 80% of contacts occur within a group and 20% of contacts occur between groups in assortative mixing patterns.

Results

Incidence reduction of severe RVGE

Before rotavirus vaccine introduction, the model estimated that the average incidence of severe RVGE cases was 327 per 10,000 population between 2000 and 2006 (S1 Table in S1 Appendix). To account for biennial patterns of incidence post vaccine introduction, we present four-year averages of severe RVGE cases to compare the incidence of severe RVGE cases in new vaccine scenarios to Status Quo. The four-year average rates of severe RVGE cases in Status Quo, Improved FP, and Improved FP+PE were 75, 57, and 33 per 10,000 population, respectively (Table 2). The four-year average percent of severe RVGE averted for Improved FP and Improved FP+PE compared to Status Quo were 23% and 57%, respectively.

Table 2

Four-year average incidence rates and percent of severe RVGE cases averted in new vaccination strategies assuming random mixing patterns between children visiting pediatricians and family practitioners.

	Vaccine Scenario
Time period	Status Quo aRated	Improved FP bRated (Averted)e	Improved FP + PEcRated (Averted)e
2018–2021	73	57 (23%)	35 (52%)
2022–2025	74	56 (24%)	32 (57%)
2026–2029	74	57 (23%)	33 (56%)

a. 85% vaccination coverage for children visiting PEs and 45% for children visiting FPs (total 78.6% current vaccination coverage).

b. 85% vaccination coverage for children visiting PEs and FPs.

c. 95%vacccination coverage for children visiting PEs and FPs.

d. Rate of severe RVGE per 10,000 population

e. Percent of severe RVGE averted compared to Status Quo in same time period

In Status Quo, biennial patterns were observed with incidence at around 80 and 65 cases per 10,000 in odd and even years, respectively; The model predicted that rotavirus epidemic patterns shift from biennial epidemic patterns to reduced annual epidemic in Improved FP and Improved FP+PE (Fig 1).

Fig 1

Monthly number of severe RVGE cases in children under 5 years of age with Status Quo (grey), Improved FP (red), and Improved FP + PE (blue) vaccine coverage assuming random mixing patterns between children visiting PEs and FPs.

Indirect benefits of improved rotavirus vaccine coverage in population attending family practitioners

In children 0–11 months old and Improved FP, the model estimated the 2018–2029 average percent of severe RVGE cases averted were 23% in the PE population and 37% in the FP population compared to Status Quo (S2 Table in S1 Appendix). The percent of severe RVGE cases averted in the PE population compared to the pre-vaccine era was 85% in Status Quo and 89% in Improved FP. Since there is no improved vaccine coverage in the PE population in Improved FP, this additional 4% of severe cases (about 7000 cases) averted in PE population are indirect benefits of improved vaccine coverage from FP population.

Sensitivity to assumptions about mixing patterns of children attending pediatricians and family practitioners

In the sensitivity analysis (S3 Table in S1 Appendix) we assumed assortative mixing patterns with 80% of contacts occurring within a group and 20% of contacts occurring between groups. This resulted in an increase of 5 per 10,000 population of the four-year average severe RVGE incidence (Table 3) compared to random mixing patterns. However, assuming assortative mixing resulted in around 5% greater reduction in the four-year average of severe RVGE cases averted in Improved FP and Improved FP + PE, respectively, compared to Status Quo than the random mixing patterns (Table 3). The epidemic patterns of RVGE stayed the same as random mixing patterns (Fig 2).

Table 3

Four-year average incidence rates and percent of severe RVGE cases averted in new vaccination strategies with assortative mixing patterns assuming 80% of contacts occur within a group and 20% of contacts occur between groups.

Time period	Status Quo aRated	Improved FP bRated (Averted)e	Improved FP + PEcRated (Averted)e
2018–2021	78	55 (30%)	34 (56%)
2022–2025	79	57 (28%)	31 (60%)
2026–2029	79	57 (28%)	33 (58%)

a. 85% vaccination coverage for children visiting PEs and 45% for children visiting FPs (total 78.6% current vaccination coverage).

b. 85% vaccination coverage for children visiting PEs and FPs.

c. 95%vacccination coverage for children visiting PEs and FPs.

d. Rate of severe RVGE per 10,000 population

e. Percent of severe RVGE averted compared to Status Quo in same time period

Fig 2

Monthly number of severe RVGE cases in children under 5 years of age with Status Quo (grey), Improved FP (red), and Improved FP + PE (blue) vaccine coverage with assortative mixing patterns assuming 80% of contacts occur within a group and 20% of contacts occur between groups.

Discussion

We modeled the epidemiological impact of higher vaccine coverage by U.S. family practitioners and pediatricians and note several key findings. First, as expected, higher coverage leads to lower incidence of severe RVGE cases through both direct and indirect vaccine effects. Second, under both Improved FP and Improved FP + PE vaccine coverage scenarios, we predicted current biennial rotavirus epidemic patterns would shift to more predictable, annual epidemic patterns. Third, while the mixing patterns of populations attending FPs versus PEs is unknown, we found that assuming assortative mixing among children visiting PEs and FPs amplified the impact of increasing vaccine coverage.

A number of results from this model are consistent with what has been predicted with previously published rotavirus transmission models; high rotavirus vaccine coverage (>85%) resulted in reductions of annual severe RVGE incidence by 56% [35], 70% [28, 36] and 84% [37] for children under 5 years of age, compared to pre-vaccination levels in different model studies. One modeling study for Germany [37] predicted no rotavirus biennial epidemic patterns even after high national rotavirus vaccine coverage (90%). However, other models had different predicted effects on rotavirus epidemic patterns after high national rotavirus vaccine coverage (90%) [28, 36, 38]. These models predicted biennial epidemic patterns in medium vaccine coverage (70%) and elimination of rotavirus at 90% coverage, whereas other models predicted potential biennial epidemic patterns with 90% vaccine coverage. Differences in the assumptions and parameters of these models may explain the different predictions for epidemic patterns. For example, in our model we assumed vaccines and natural infection induce partial immunity (i.e. subsequent infection can still occur but at reduced rate) and a long duration of immunity. Another modeling study that predicted high vaccine coverage with biennial patterns assumed one-year duration of complete immunity after previous infection, then, individuals were assumed to be susceptible to infection again with partial immunity [38]. Moreover, our model fit to data from the U.S. whereas other rotavirus models were fit to rotavirus data from England and Wales. Differences in demographic conditions between the U.S. and England and Wales may also partly explain the differences in these results.

The emergence of biennial epidemic patterns of RVGE incidence in the U.S. after the introduction of a national vaccine program may be driven by accumulation of susceptible following modest vaccine coverage. Other developed countries with high coverage of rotavirus vaccination (>85%), such as Belgium, Austria, Australia, Finland, and Germany, have not experienced the biennial epidemic patterns after vaccine introduction [14-18]. Our model predicted that RVGE incidence shifted from a biennial pattern to an annual pattern when vaccine coverage reached 85%. Thus, promoting rotavirus vaccination in children visiting FPs lowers disease incidence rates and results in a shift from biennial to annual epidemic patterns. Higher vaccination coverage leads to a smaller susceptible population compared to the susceptible population in Status Quo. Therefore, the rate of accumulation of susceptible children over two successive birth cohorts may not be sufficient to drive biennial ‘mini-epidemics’. Shah et al. suggested that increased rotavirus vaccine coverage in the U.S. may change rotavirus epidemiological patterns since the biennial epidemic patterns may be driven by lower vaccine coverage [13]. The shift of epidemic patterns from biennial patterns to reduced annual patterns may be beneficial for public health preparedness as the burden of disease would be more consistent year to year.

Our predictions were somewhat sensitive to assumptions about mixing patterns in children visiting PEs and FPs. When we assumed assortative mixing, we found greater impact of increasing vaccine coverage on severe RVGE incidence. Our model had a higher force of infection with assortative mixing patterns than that in random mixing patterns. Effelterre et al. found that the estimated basic reproductive number (R0) of rotavirus is higher when mixing patterns are assumed to be assortative. However, that study showed that reduction in any grade of severe RVGE incidence in children under 5 years of age after vaccination is higher when the assortative mixing is lower, which contradicts our results [39]. Effelterre et al. focused on the reduction of any RVGE but our study focused on the reduction of severe RVGE. This difference influences the vaccine impact on the reduction of RVGE since rotavirus vaccines have better efficacy against severe RVGE [4, 40, 41]. In another study, Choe and Lee indicated that the higher degree of assortative mixing had higher R0 than random mixing [42]. Furthermore, after vaccine introduction, a higher degree of assortative mixing resulted in lower incidence over time. We have no data to inform mixing patterns between PEs and PFs populations. But, under either scenario, promoting vaccine coverage in children visiting FPs can have significant impacts on reducing severe RVGE incidence in children under 5 years of age in the U.S. Future rotavirus vaccine promotion strategies in the U.S. can emphasize covering children visiting FPs.

Our model had several limitations. First, though we took assortative contact patterns for children visiting PEs and FPs into account, there currently are no data that describe the true assortative contact patterns within and between patients who attend physician groups. Furthermore, the assortative contact structure between different age groups used in this model is based on the POLYMOD study, which is a population-based contact survey in Europe. This may not accurately represent the contact structure in the U.S. Second, while our model did capture the intensity in the year to year variability in the present biennial epidemic patterns, the observed RVGE cases in even rotavirus season years were much lower than the value predicted by our model [4, 43]. Third, we fit this model to a large, commercial insurance dataset. While this database covers most states, it may not be representative of the whole U.S. population. For example, children who fall under the coverage of Medicaid, may have lower childhood vaccination coverage and higher incidence of RVGE than the children represented in the MarketScan commercial insurance database [6].

In conclusion, we used a dynamic transmission model to predict the impact of increasing immunization rates among children who attend family practitioners. Under these higher vaccine coverage levels, we predicted that biennial patterns would shift to annual patterns with lower magnitude of RVGE incidence.

S1 Table. Rate of severe rotavirus cases (per 10,000) and percent of severe cases averted for each rotavirus vaccination scenario in 2000 to 2030 assuming random mixing patterns between children visiting pediatricians and family practitioners. S2 Table. Children 0–11 months old percent of severe rotavirus cases averted in post-vaccine era and new vaccine scenario after 2018 in pediatrician and family practitioner populations assuming random mixing patterns between children visiting pediatricians and family practitioners. S3 Table. Rate of severe rotavirus cases (per 10,000) and percent of severe cases averted for each rotavirus vaccination scenario in 2000 to 2030 with assortative mixing patterns assuming 80% of contacts occur within a group and 20% of contacts occur between groups.

(DOCX)

Click here for additional data file.

2 Dec 2019

PONE-D-19-26596

Disease burden and seasonal impact of improving rotavirus vaccine coverage in the United States: a modeling study

PLOS ONE

Dear Mr. Ai,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Jan 16 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Constantinos I. Siettos, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

3. Thank you for stating the following in the Acknowledgments Section of your manuscript:

'This work was supported by NIH/NIAID (R01-AI112970) and the Vaccine Impact Modeling Consortium.'

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

'NO - Include this sentence at the end of your statement: The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.'

Please amend your Financial disclosure statement to declare sources of funding, or state that the authors received no specific funding.

Please provide an amended Funding Statement that declares *all* the funding or sources of support received during this specific study (whether external or internal to your organization) as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.

Please state what role the funders took in the study.  If any authors received a salary from any of your funders, please state which authors and which funder. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

c. Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this paper, Ai et al use a standard mathematical model to investigate the impact of improving coverage with rotavirus vaccine on rotavirus disease burden, and the effect it could have on the biennial patterns that have emerged post vaccine introduction. They test increasing coverage in two patient populations: those attended by pediatricians and those attended by family practitioners. They project significant reductions in disease burden (compared to current levels) when coverage is increased among patients of family practitioners alone, and larger reductions when coverage is increased among patients of both family practitioners and pediatricians. With increased coverage, they note a disappearance of the biennial pattern of rotavirus incidence (it is replaced by a reduced annual pattern), supporting the hypothesis that biennial patterns arise from an inter-year accumulation of susceptible persons in settings with moderate coverage. This is a well written paper, that is concise and understandable even for non-modelers, with clear conclusions that are directly derived from their results, and with an important public health message. The model structure has been built to emulate natural rotavirus disease and transmission. I only have a few minor comments.

Introduction:

Line 60: Typographical error, it should say “$1 billion in direct and indirect costs to the U.S.”

Line 63: Typographical error, it should say “rotavirus incidence has declined between 57%” (erase ‘in the’)

Methods:

Line 110 and Table 1: The text mentions the model is age-structured with 6 age-groups, yet the table shows parameters only for 4 age-groups, would clarify. Perhaps estimated transmission parameters are reported as larger age-groups.

Line 145: Would expand on, explain more, how the calibration of vaccine effectiveness allows the model to capture the biennial patterns.

Discussion:

Line 278: The line “partial immunity when individuals are susceptible” might need to be rephrased, this is not clear.

Line 311: The public health relevance of the additional 4 percent in reductions in severe rotavirus gastroenteritis when assuming assortative transmission by practitioner is unknown, and authors might want to acknowledge this, or translate to absolute numbers.

Line 320: If the intensity in the year to year variability of the biennial patterns were not captured, then the authors might want to acknowledge that other factors might be at play, or hypothesize which ones.

Reviewer #2: The authors focus on an important aspect of mechanisms which may influence the effectiveness of vaccines in general and in the particular setting of RV vaccination. To estimate the reduction of RVGE incidence by efforts to improve vaccination rates by FP by applying a deterministic age-structured dynamic model is a feasible approach. Overall, this is a well-written manuscript with sound results.

However, I have some comments to deal with:

I doubt whether the biennial pattern of RVGE in U.S. may be explained only by incomplete vaccine coverage. Authors should provide more evidence for this hypothesis by literature or own experience. Trend for biennal patterns or oscillations in RVGE associated hospitalizations was seen by others (e.g. Prelog M et al., J Inf Dis 2016). Authors should comment on this. The immunological or virological mechanisms behind biennial patterns or disappearance of biennial patterns should be explained to understand the association with vaccine coverage.

Assumptions in table 1 are mainly based on few articles. Authors should comment why they took this choice and did not perform a systematic review approach previous to the modeling procedure. Authors should state why the articles by Velazquez or Heymann are prior to others if there are any on these parameters.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Jan 2020

Response to Reviewers

Comments from the reviewers:

Reviewer #1: In this paper, Ai et al use a standard mathematical model to investigate the impact of improving coverage with rotavirus vaccine on rotavirus disease burden, and the effect it could have on the biennial patterns that have emerged post vaccine introduction. They test increasing coverage in two patient populations: those attended by pediatricians and those attended by family practitioners. They project significant reductions in disease burden (compared to current levels) when coverage is increased among patients of family practitioners alone, and larger reductions when coverage is increased among patients of both family practitioners and pediatricians. With increased coverage, they note a disappearance of the biennial pattern of rotavirus incidence (it is replaced by a reduced annual pattern), supporting the hypothesis that biennial patterns arise from an inter-year accumulation of susceptible persons in settings with moderate coverage. This is a well written paper, that is concise and understandable even for non-modelers, with clear conclusions that are directly derived from their results, and with an important public health message. The model structure has been built to emulate natural rotavirus disease and transmission. I only have a few minor comments.

Response: Thank you for your comments.

Introduction:

Line 60: Typographical error, it should say “$1 billion in direct and indirect costs to the U.S.”

Response: The typographical error is corrected

Edits to text (line 60): “$1 billion in direct and indirect costs to the U.S.”

Line 63: Typographical error, it should say “rotavirus incidence has declined between 57%” (erase ‘in the’)

Response: The typographical error is corrected

Edits to text (line 63): ” that rotavirus gastroenteritis (RVGE) incidence has declined between 57% - 89%”

Methods:

Line 110 and Table 1: The text mentions the model is age-structured with 6 age-groups, yet the table shows parameters only for 4 age-groups, would clarify. Perhaps estimated transmission parameters are reported as larger age-groups.

Response: Indeed, age groups 0 – 1 months, 2 – 3 months, and 4 – 11 months in the model shared the same group of estimated age specific transmission parameters: age-group <1 year.

Edits to text (line 143 – 144):” We estimated four age specific transmission parameters (q_1 for <1 year, q_2 for 1-4 years, q_3 for 5-24 years, and q_4 for > 25 years)”

Line 145: Would expand on, explain more, how the calibration of vaccine effectiveness allows the model to capture the biennial patterns.

Response: A likely cause of biennial patterns of disease in the U.S. is the accumulation of susceptibles that results from modest vaccine coverage. Vaccine efficacy and the rate of loss of immunity both impact the accumulation of susceptibles over time so calibrating these parameters allows the model to capture the biennial patterns that are observed in the U.S.

We add the explanation of the factors driving biennial patterns to the introduction and methods to explain why we calibrate vaccine effectiveness and rate of loss immunity to capture biennial patterns. Edits to text (lines 83 – 85): “In general, biennial patterns of RVGE could be induced by factors that govern the rate of new susceptibles. These include incomplete vaccine coverage, imperfect vaccine efficacy, and birth rate (Pitzer 2009; Pitzer 2011, Shah 2018).

Edits to text (156 – 164): “Biennial epidemic patterns in the U.S. may be driven by a slower accumulation of susceptibles following modest vaccine coverage. By calibrating the vaccine efficacy and rate of loss of immunity parameters (i.e., increasing vaccine efficacy and lowering the rate of loss of immunity) we can slow the accumulation of the simulated susceptible population which allows the model to capture biennial epidemic patterns.”

Discussion:

Line 278: The line “partial immunity when individuals are susceptible” might need to be rephrased, this is not clear.

Response: In lines 286 – 288, partial immunity (i.e. subsequent infection can still occur but at reduced rate) was explained

Edits to text (lines 288 – 291): Another modeling study that predicted high vaccine coverage with biennial patterns assumed one-year duration of complete immunity after previous infection, then, individuals are susceptible to infection again with partial immunity

Line 311: The public health relevance of the additional 4 percent in reductions in severe rotavirus gastroenteritis when assuming assortative transmission by practitioner is unknown, and authors might want to acknowledge this, or translate to absolute numbers.

Response: Our raw data shows that an additional 7000 RVGE cases are averted when assuming assortative mixing patterns compared to random mixing patterns in Improved FP.

Edits to text (lines 233 -235): Since there is no improved vaccine coverage in the PE population in Improved FP, this additional 4% of severe cases (7000 cases lower) averted in PE population are indirect benefits of improved vaccine coverage from FP population.

Line 320: If the intensity in the year to year variability of the biennial patterns were not captured, then the authors might want to acknowledge that other factors might be at play, or hypothesize which ones.

Response: We rephrased our statement: Our model did capture the intensity in the year to year variability in the present biennial epidemic patterns, however, the actual RVGE cases in even year rotavirus season were much lower than the value predicted by our model (Aliabadi 2015; Getachew 2018).

Edits to text (lines 333 – 336): Second, relatedly, our model did capture the intensity in the year to year variability in the present biennial epidemic patterns, however, the actual RVGE cases in even year rotavirus season were much lower than the value predicted by our model.

Reviewer #2: The authors focus on an important aspect of mechanisms which may influence the effectiveness of vaccines in general and in the particular setting of RV vaccination. To estimate the reduction of RVGE incidence by efforts to improve vaccination rates by FP by applying a deterministic age-structured dynamic model is a feasible approach. Overall, this is a well-written manuscript with sound results.

However, I have some comments to deal with:

I doubt whether the biennial pattern of RVGE in U.S. may be explained only by incomplete vaccine coverage. Authors should provide more evidence for this hypothesis by literature or own experience. Trend for biennal patterns or oscillations in RVGE associated hospitalizations was seen by others (e.g. Prelog M et al., J Inf Dis 2016). Authors should comment on this. The immunological or virological mechanisms behind biennial patterns or disappearance of biennial patterns should be explained to understand the association with vaccine coverage.

Assumptions in table 1 are mainly based on few articles. Authors should comment why they took this choice and did not perform a systematic review approach previous to the modeling procedure. Authors should state why the articles by Velazquez or Heymann are prior to others if there are any on these parameters.

Response: Thank you for your comments and suggestions

Response to biennial patterns:

In general, biennial patterns of RVGE could be caused by(a combination of) vaccine coverage, vaccine efficacy, and birth rate (Pitzer 2009; Pitzer 2011). Since birth rates in the U.S have been low and fairly stable through the pre and post – vaccine era, and previous studies have indicated that incomplete vaccine coverage can drive biennial patterns, we considered vaccine coverage as the main factor in our model driving biennial patterns in the U.S. Note that these patterns do not occur in most other high-income countries with rotavirus immunization. We incorporated a detailed explanation of this rationale in the text (lines 83-89). Additionally, we also mentioned the disappearance of biennial patterns due to increasing of vaccine coverage predicted in the Weidemann 2014 modeling study in the text (lines 288 – 289)

Edits to text (lines 83 – 89):

In general, biennial patterns of disease could be caused by a combination of vaccine coverage rates, vaccine efficacy, and birth rate, such that susceptibles are accumulated at a certain rate (Pitzer 2009; Pitzer 2011, Shah 2018). In one study, a RVGE transmission model predicted that biennial patterns of RVGE after introduction of vaccine when birth rates are lower while an annual pattern of RVGE was predicted when birth rates are high (Pitzer 2011). In the U.S, birth rates have remained fairly stable0 in the pre and post-vaccine era. Thus, birth rate is likely not a significant driver of the biennial pattern of RVGE in the U.S.

Response to parameters selection:

We used data from the Velazquez 1996 study to inform the proportions of infection, symptomatic infection and severe disease and relative infectiousness of non-primary infections because of its rigorous study design and study population. Velazquez 1996 was a birth cohort study conducted in Mexico quantifying risk reduction after infection by collecting stool weekly from 200 infants from birth to 2-year-old. There are three birth cohort studies conducted in Mexican, Indian and African birth cohorts that quantify the reduced risk of subsequent infections. As there are no published birth cohort studies conducted in the U.S., we selected the Velazquez study of the Mexican birth cohort which would have more similar demographic characteristics to the U.S. than the birth cohorts from India and Africa. Moreover, the data from the Velazquez study has been used to parameterize many other rotavirus transmission models (Pitzer 2012; Atchison 2010; Atkins 2012).

The work by Prelog et al (Jid 2018) is an important paper showing the range of indirect effects of rotavirus immunization in Australia, but does not provide parameter input for our model.

The Hymann 2015 study provides data on the natural history of rotavirus such as daily rate of loss of maternal immunity and daily rate of loss infection. These data have been used to inform parameters in several transmission model (Atchison 2010, Atkins 2012).

Submitted filename: Response to Reviewers_ChinEn Ai.docx

Click here for additional data file.

28 Jan 2020

Disease burden and seasonal impact of improving rotavirus vaccine coverage in the United States: a modeling study

PONE-D-19-26596R1

Dear Dr. Ai,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Constantinos I. Siettos, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed and there are no further questions or concerns regarding the manuscript.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

6 Feb 2020

PONE-D-19-26596R1

Disease burden and seasonal impact of improving rotavirus vaccine coverage in the United States: a modeling study

Dear Dr. Ai:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Constantinos I. Siettos

Academic Editor

PLOS ONE