Note: this post is an update to the COVID-19 School Reopening Part I, posted on June 28, 2020.

A few weeks ago, I wrote about my own personal decision as a phd epidemiologist and mother of 3 for why I would feel okay with sending my kids back to school if the schools were able to implement recommended public health measures. Three weeks have now gone by and we have again hit new milestones for COVID-19 in the United States and in my home state of GA. Community spread is increasing in many states, particularly in the Sunbelt region and South. In AZ, FL, GA, TX, LA, SC and AL in the past 2 weeks, hospitalizations are up about 90% and deaths are up about 75%. And we know this is just the tip of the iceberg, as many people are waiting more than a week for tests and test results. The outbreaks in many states suggest we need substantial public health measures in place to reduce the exponential spread of this virus, before we can safely reopen schools.

Many epidemiologists have advocated since the beginning of this pandemic that it is likely we will need multiple cycles of social distancing or shelter in place orders to help manage this pandemic. There was a great webinar from covid19conversations on this topic that provided scientific evidence around this as one approach, and you can read the transcript here. And in many states, we need big policy initiatives to help slow the spread of COVID-19 (e.g., mask ordinances or closing of indoor dining, etc).

The purpose of our Spring shelter in place was to help flatten the curve and to give us time to prepare for the coming waves of the virus by increasing testing capactiy, building our infrastructure to do contact tracing and isolation, and increasing hospital capacity and PPE supply. And yet, here we are, with many states not anywhere close to where we need to be to do this effectively. We have seen news reports that health care capacity is close to its limit in places like Texas and Florida, and I know in Atlanta where I live, many of the ICUs are at saturation or on diversion. For a look at where your state falls with hospital capacity, check out this map by CDC. Though you would think that in all this time we would know how important it is to stock up on PPE, apparently we STILL have a shortage of PPE for medical workers. Side note, I listened to a good podcast episode from Sanjay Gupta’s Fact vs. Fiction about Peter Tsai, the scientist who invented the N95 mask. He had great scientific advice on safely re-using these single use N95 masks, which is exactly what my spouse is doing here in the trunk of our hot car (!). Others - take note!

In addition, as I’ve seen in the last couple of weeks, our testing capacity is still strained in many areas across the US. Increased testing demand has led to substantial delays in testing access and results. For example, there are several people I personally know who have been exposed to a COVID positive contact within the last week. They sought testing but waited a week to get the test and another week to get test results. These testing delays have substantial implications for schools (and daycares, businesses, and really anywhere). The scarcity of testing and delays in test results makes contact tracing almost useless.

With many governors not making any movements to institute mask ordinances (and with my own governor SUING cities and counties in our state that have introduced mask ordinances), with testing capacity still not where it should be, and with contract tracing infrastructure still limited, these “hot spots” are going to have a really difficult time re-opening schools anytime soon. We need real leadership here, or attempts at opening schools will be futile because too many teachers, staff, and potentially children will have high levels of community exposure. High community exposure will lead to closings of classrooms for a 14 day quarantine period, an unstable educational environment, and a riskier environment for teachers and staff (and potentially a smaller amount of children).

Within the last week or so, the issue of schools reopening has become very political. With Trump emphasizing that schools should reopen (while disagreeing with CDC guidance for doing so) and threatening witholding funding for schools that don’t, our conversations and planning have set us back by making this decision political, rather than an evidence-based and collaborative approach to solving a major problem. After Trump and Education Secretary Betsy DeVos weighed in, the American Academy of Pediatrics provided additional clarification of their initial strong stance to reopen schools face to face, stating that schools should not go against local public health and scientific recommendations and reopening schools will require substantial financial investment. You can read their updated comments here. This should not be a debate among people who want schools to reopen vs. people who do not want to make schools reopen, or a debate of left vs. right — we need to use a data-driven approach to answer this question with the best available evidence, and recognize that the recommendations may change as the science and evidence is continually gathered to help inform our decisions from a population level AND from an individual level.

Other than the rising community spread in many states across the U.S., what else has changed in the last couple of weeks? Do we have any new evidence to help guide our decisions? I have had several people contact me since my last post asking about my level of concern for the large number of cases of COVID in childcare facilities in Texas as reported in the media. Among 883 childcare facilities in the state, 894 staff members and 441 children tested positive. This does sound high but it’s also somewhat expected with the amount of active community spread in Texas. Note, they have 12,220 total facilities in the state, so this has affected about 7% of the total number of open facilities. It’s hard to tell from the news media (and no data on contact tracing) but there does not seem to be clear evidence of clustering of cases here. Brown University economist Emily Oster (author of one of my favorite great evidence-based parenting books, Cribsheet) does a fantastic job of breaking this down with some assumptions, and I encourage those who have interest in a data-based explanation of what we do and don’t know about these cases here. As community spread increases, we are likely to see similar media reports like these. I keep seeing various news articles that moms and others are sharing over and over again in a panic. I have seen these too - yes, there are kids that get COVID, yes there will be daycares and teachers that have COVID, and yes we may see examples that are sensationalized in the media. But I would urge anyone reading this to try not to get immediately alarmed, and put this in context with the denominator and the reassuring evidence around the rarity of children getting seriously ill. And, keep looking for more and better data beyond the media reports. Check out the data on COVID-19 in children by state from the American Academy of Pediatrics; this website is frequently updated, and the detailed report is great.

Many have been concerned about the news reports highlighting large percent increases of COVID in children. Nationally, we have seen a 45% increase in new cases in children. Headlines noting the very high “percent increase” in kids can be very scary, but it’s not surprising we would see this with increased testing and increased community spread. And many have been worried about the FL data in children. I have looked into the FL Department of Public Health data in the last few days, after many news stories went around showing 31% COVID test positivity among Florida children < 18 years (16,797 cases among 54,022 tested). This of course is not a population-based sample, it’s only of those TESTED, and we don’t know anything about who is getting tested. There are some concerns being raised about FL’s % positivity rate being wrong, because some labs are ONLY passing along positive results and not negative results, which artificially deflates the denominator. Also, if you look at the proportion of COVID-19 cases that are children in FL, it’s roughly 5.5% (16, 797/291,629), which is similar (slightly lower) than the perecentage reported in U.S. data. And consistent with what we have heard all along, hospitalizations and deaths in FL are rare (hospitalizations range from 0.6% in age 5-9 years to 3.7% in <1 year olds; 2 deaths total reported in the 10-14 age group and 2 in the 15-17 age group). It will be important to continue monitoring hospitalizations and deaths in this age group because these are lagging indicators — so as we see rises in community spread, how do hospitalization rates change? Nationally, there have been a total of 63 deaths in children < 18, and while hospitalization data in states is not reported among all states, hospitalization rates among those that have reported these data show hospitalization is uncommon.

Even with some community spread, if teachers and children in schools are wearing masks, we will substantially reduce spread within a school. Take for instance the example of this well conducted MMWR recently published about two symptomatic hair stylists who exposed 139 clients, but not a single client tested postive for COVID. This week in JAMA, we also saw that universal masking requirements for healthcare workers AND patients substantially reduced the previous exponential transmission of COVID-19 in the hospital (where COVID-19 exposures are rampant). And as the corresponding commentary noted on how this is relevant to the community: “Data from a large health care system may be generalizable to the greater community insofar as the findings represent the contribution of masking when individuals are physically close to one another and social distancing is not possible. Like herd immunity with vaccines, the more individuals wear cloth face coverings in public places where they may be close together, the more the entire community is protected.” In summary - masks work!

I was reassured somewhat by this recent report (though it wasn’t peer-reviewed) from the Public Health Agency of Sweden that came out last week on data I have been anxiously awaiting from Sweden, a country that kept schools (and bars and restaurants) open in the midst of the pandemic. And because of their controversial approach to not lock down the country, they had much more community spread of COVID than other European countries. So while Sweden and the U.S. are very different countries, we have had (up until recent days) similarities in community spread. The figure shows data on COVID-19 cases per million population from Sweden and US, and while our curves have gone in different direction within the last couple of weeks, the biweekly cases per million population were relatively similar at Sweden’s peak. Much of the data from other countries across the world showing limited spread among children is difficult to interpret because we don’t know the direction of causality — was the transmission between children low BECAUSE schools were closed? But in Sweden, their schools (<grade 10) remained open without any major adjustments to class size, recess, or lunch procedures. The Swedish report shows a comparison between Finland and Sweden, which found no difference in the crude incidence of COVID-19 in school-aged children and no evidence of additional risk to children and teachers. It is hard to know if the experience in Sweden would also be the same as the US though, and the report has its limitations. This non peer-reviewed, ecological comparison does not account for differences in individual or population characteristics and the specific behaviors of the population or any mitigation efforts put into place in the schools by country, and this report also cannot tell us whether the same approach would work in a completely different country like the U.S. I will be closely following and awaiting more rigorous analyses and peer-reviewed data from Sweden to look into this more, but I am somewhat reassured with this initial report.

Several good news articles have been published recently summarizing lessons learned in school reopenings from other countries. This new document by Washington State Health Department and the Universty of Washington is also a really helpful summary of some of the literature and comparisons of data by country. The summary states that there is a lack of scientific consensus on the susceptibility of school-age children to COVID, the role of children in community transmission, and the impact of school closures and re-openings on transmission. This is also a great website and resource that summarizes existing data and generally concludes that of the published literature thus far, children are considerably less likely to get COVID-19 when exposed, high quality serosurveillance studies conducted in countries that have done a lot more community testing show that children are less likely to have the disease, and children can get sick but are rarely the index case in spread. Of course, we need more data to guide our decision making. Some of this work is in progress (see this example of a study of >24,000 US child care programs that is in progress). Unfortunately, some don’t have the luxury of waiting for more data to make a decision, because they need to send kids back to daycares and schools ASAP. So we have to make decisions between two bad options (keep kids home until there is a vaccine vs. send them to school without fully knowing the risks) with the best available data we have.

I am still reassured by the many daycares that have remained open to essential workers (who are at higher risk of bringing COVID-19 home to their family members) and the very few cases and virtually no outbreaks that we can see from these schools or daycares. For example, YMCA daycares that remained open in NY during the midst of the peak of the pandemic saw only a few cases among >10,000 children (ages 1-14) at 170 different facilities.

Overall, there is still so much we don’t know about COVID-19. Of course we have no long-term data on the effects of COVID-19 among either adults or kids, and we need to study this more. This is a new virus, and as I described a couple of weeks ago, good science takes time to plan, conduct, and then publish. The evidence and science will continue to help shape our minds about this topic and anyone claiming to have definitive proof about the answer to schools reopening and the implications of COVID-19 among children with one news article or one single study is misguided.

Our large school district announced last week that they will open as full time virtual for the first 9 weeks. I am disappointed personally because I know this will be incredibly hard for both my spouse (an essential healthcare worker) and I (an epidemiologist working on COVID research) to work full time and facilitate virtual learning for a kindergartner and second grader. Side note: I have VERY little faith that this virtual learning will work for a 5 year old who loses his attention span after 10 minutes of a zoom meeting, despite getting to show up in school wearing no pants.

But, I ultimately agree with the decision by the school board to start as virtual because of the large number of cases in our community right now, the many issues around delays in testing and contact tracing in our community, and my concerns that our districts do not have enough time to implement the necessary public health measures to SAFELY reopen schools. Right now the risk is too high for teachers and staff, many of whom are older, higher risk, or may be immunocompromised or living with someone who meets those categories.

This summary of the science from Harvard/Mass General Hospital in a guide for clinicians on advising the public about COVID and school re-openings is the best I have seen so far on the updated literature in this area, AND on the limitations of the data and what we don’t know. This generally supports the rationale I still stand by that the existing evidence shows that 1) COVID-19 occurrs less in children than adults, 2) poor outcomes such as hospitalization are rare in children, and 3) transmission from child —> adult seems to be less common, but with all the caveats I described in my initial post a couple of weeks ago. For example, it could be that the reason we are seeing low transmission from child —> adult is because of less testing among children (because they are more likely asymptomatic and cases are less severe so we don’t seem them in the health care system or data).

So what should you do if you have the option to send your kid back to school this Fall? It’s a personal choice, but my answer would be that IT DEPENDS. For those that are in an area where you currently have low spread of COVID (e.g., declining case counts), access to testing, and good infrastructure for contact tracing, you do not live with anyone who is high risk, sending your kids back to school may be a reasonable option. For others, even the smallest amount of risk is not acceptable and you choose to keep your kids home. THAT IS OKAY. For others still (e.g., essential workers, dual working households, etc), there is little choice in the matter and you may have to send your child to school or risk losing your job. Forthis group, you should be reassured for the health of your children. You should be asking a lot of questions to your daycare or school about the measures they are putting in place to keep not only your children safe, but also the teachers and staff.

For me personally, I’m going to be volunteering and offering as much help and guidance to our public school system on reopening safely as possible when it’s safe to do so. I have already worked on trying to encourage local businesses to donate supplies where they can (e.g., cleaning supplies, air filtration units, etc) to schools, particularly those in lower income areas where their PTA cannot just donate more money to help. My hope is that we will get to a place later this Fall where community spread declines, and where we have a good plan for the safe reopening of schools for both students and the staff/teachers. I also hope that there are other regions in the country and communities with less spread, schools that are smaller and able to better implement public health measures appropriately (as we have seen daycares do) that can still move forward with reopening because of all of the benefits of school I described in my last post.

In addition, we need more FUNDING and support for both K-12 schools and for daycares to implement what approaches we know will mitigate the spread of COVID-19 in schools. As I have heard from educators and school administrators across the country, many schools just simply do not have the resources to open safely. State budgets are being cut everywhere. In Georgia, our legislature just made a substantial cut in K-12 public education -- $950 million in the FY 2021 budget – precisely when we need more funding to manage this crisis and keep our children and teachers safe. We need federal help and the help of community businesses and members to achieve safety in our schools, and we need parents to help volunteer their time and resources to help (e.g., urge businesses to help donate equipment for cleaning schools or improving ventilation with better air filtration systems as a win-win for them to help their workforce return to work faster). There are so many practical issues around implementation of these measures, and we need time, money, and resources to pilot implementation on the local level.

Nearly one-third of our nation’s workforce have children at home without school or daycare. This is disproportionately harming women, communities of color, families with low income and essential workers who cannot work from home. We must strive for the safe reopening of schools. We have all of the tools we need to stop this virus from substantial community spread and we are FAILING in many places. We need bold action to stop the community spread. We are going to have to make really tough choices: do we want to go out to eat at a restaurant and host a small party with a few families? Or, do we want to safely send our kids to school? It’s time to mask up, stay home, and start planning NOW for what a safe reopening of schools could look like when we can get community spread controlled and on a downward trajectory.

In a future post, I would really like to write more about practical considerations from a public health standpoint on moving forward from here. For many of us who are in districts that will not start in person learning this Fall, what is the safest way we can conduct virtual learning at home with a small pod of children, and how do we do this equitably? How can we stay safe, and nurture our children’s social and emotional development at home? What can we do to address the educational gap that will be furthered between socioeconomic groups and communities of color with these school closures? I would love input, comments, and questions on these to help inform this content, or on other topics that are academic + mama that you would like reviewed.