Disclaimer: We summarize the available evidence to the best of our knowledge. Yet, we always recommend to read the original studies. If you find any error, please inform us.
With Covid19-cases on the rise, re-opening schools seems not to be on the table at the moment. Indeed, during the “Lockdown 3.0”, schools throughout India are to remain closed as per the order of the Ministry of Home Affairs, irrespective of the zone classification of the District.
According to media reports, there will be guidelines by the MHRD’s Department of School Education and Literacy. An official is quoted "The guidelines are being formed and will also be shared with states so they can prepare accordingly before reopening schools and colleges. Districts will be tasked with the implementation of the guidelines and certain spaces in the campuses will have to be revamped to ensure social distancing".
While social distancing measures, regular disinfection and well-equipped washrooms might be an option for expensive private schools catering to the rich, Bihar’s government schools are among the most crowded in the country. According to reliable estimates, Bihar needs 75% more classrooms than it has and around double the number of teachers to comply with basic minimum norms. Even those, however, will not allow social distancing.
The question, then, is when and how government schools in Bihar should reopen. There are several questions to consider in this regard, some of which we try to answer in the following paragraphs.
Are school closures curbing transmission?
An article in The Lancet made big waves as it summarized available evidence and referred to an article that suggested that school closures alone might prevent “only 2-4% of deaths”, a number that went into many media reports about the study. However, the article also cautioned that “More research is urgently needed on the effectiveness of school closures and other school social distancing practices to inform policies related to COVID-19,” a sentence that went into far fewer media reports.
Since then, several new studies on the role of children in transmission, but also about their own risk of infection and long-lasting harm have emerged. It must however be cautioned that these data come from a variety of countries that might have starkly different child population characteristics. Many children in Bihar have a weak immune system, high levels of malnutrition and insufficient access to high-quality health care. While the focus in many countries is on “flattening the curve” to stay below ICU capacity limits, there might be good reasons for a different strategy in Bihar.
Another study that received a lot of attention comes from the Netherlands. It suggested that “children have a small role in the spread of COVID-19”. Also, in around 40 GP practices, not a single infection among patients under 20 years who were tested were found. The Netherlands announced on April 21 that primary schools could reopen.
The age-wise distribution of COVID-19 patients in Bihar is, however, in stark contrast to the data from the Netherlands. On April 25, Sanjay Kumar, Principal Secretary in the Health Department, Government of Bihar, tweeted the latest data on age wise distribution of COVID-19 cases in Bihar. There is a high share of 0-20 years olds among those tested positive.
A recent study analysing contact surveys data for Wuhan and Shanghai before and during the outbreak and contact tracing information from Hunan Province suggest, based on transmission models to study the impact of school closures on transmission, that “while proactive school closures cannot interrupt transmission on their own, they can reduce peak incidence by 40-60% and delay the epidemic” (illustrated in Figure 1). While children were found to be less susceptible to getting infected, they had many more contacts as compared to adults (Figure 2). These two counteracting aspects approximately evened out. Yet, the authors caution that there are various limitations to their study, e.g. they estimated the effect of social distancing alone and not in combination with other measures. Further, “it is likely that population wide social distancing, case-based strategies, and decontamination efforts, all contributed to achieve control in Wuhan and Shanghai, and their effect is difficult to separate out in retrospective observational studies.” In addition, the patterns modelled in their study might not be representative for e.g. Bihar on important aspects. Hence, as all studies, it can inform policy but not decide it.
Figure 1: Effect of limiting school contacts on the epidemic spread (Source)
Figure 2: Contact matrices by age (lower left corner is of interest for inter-child contact) (Source)
Are children infecting others?
Another recent study from Germany looked at the viral load by patient age. They “found no significant difference between any pair of age categories including children. In particular, these data indicate that viral loads in the very young do not differ significantly from those of adults. Based on these results, we have to caution against an unlimited re-opening of schools and kindergartens in the present situation. Children may be as infectious as adults.”
(Source)
To interpret this figure, it is important to notice that the fewer dots at lower age are due to a smaller sample size of patients at a lower age. The viral load is on the y-axis. Hence, the distribution of the dots along the y-axis is of interest, not their number.
The available preliminary evidence does not allow to provide straightforward policy recommendations. Interaction patterns in schools, WASH-infrastructure or immune system strengths might differ significantly from country to country or even region to region. Further, several studies might suffer from methodological weaknesses or selection bias. In addition, the effectiveness of school closures also depends on whether children stay indeed at home if schools are closed or whether they e.g. graze animals and visit banks or other public places with their parents instead of going to school. What seems clear is that the claim that children are not affected by Covid-19 is dubious. Children can get infected and they can infect others as the above study clarifies.
The current need for timely policy advise oftentimes makes pre-publications the only source of information. Peer-reviewed, high-quality studies are still missing. Nonetheless, use should be made of the available evidence, but taken with some caution. See also this recent opinion piece in the New York Times on the dangers of preprints, i.e. articles that have not undergone rigorous peer reviews, when it comes to assessing the dangers of Covid-19.
WASH infrastructure in Bihar’s schools
As per U-DISE data, more than 95% of schools have functional toilets for boys and girls (2018-19) and more than 98% of schools have drinking water available. Yet, as the ASER report from 2018 for rural Bihar indicates, usable toilets were available in 75.6% of schools and drinking water was available in 89.7%. While Bihar has made major efforts to improve infrastructure, even having a water pump in school will be insufficient to ensure that children can safely wash their hands multiple times a day.
One solution to this problem might for instance be for children to carry a personal water bottle together with a personal soap that they can use to wash their hands. Alternatively, a single person might operate the handpump while children can wash their hands with a personal soap. Yet, this might lead to crowds forming in front of the pump.
Creative, school-level solutions will have to found to make up for the limited infrastructure available.
Getting the priorities right: learning, dropouts and malnutrition
To answer the question when to reopen schools, several aspects have to be considered to weigh negative and positive aspects, both immediate and in the long-run.
Learning
With schools shut, most government school children in Bihar currently have no access to education. While smartphone apps and online materials are available, the devices are not.
The ASER 2017 report provides some further information. In Muzaffarpur District in Bihar, around 85% of youth (14-18 years) had used a mobile in the last week, while internet was only available to 26.3%. There is a stark difference between males (39.2%) and females (14.9%) indicating inequity along gender lines when it comes to intra-household distributions. A computer was available to 20.4% in the last week, again showing a massive gender imbalance (male 28.9%, female 12.9%) with 72.7% of sampled youth claiming they had never used a computer. Hence, any tech-intensive bridging solutions have highly limited reach, especially among government school students, with numbers possibly being even lower for elementary school students. The reach of ed-tech solutions might further decline if the health pandemic is accompanied by an economic shock, especially in the informal sector, as internet packages might be discontinued due to a lack of disposable income.
Hence, both the TV program started by the government and the mobile app do not reach most government school students in Bihar as evident from latest data on the availability of electricity, smartphones and the internet.
At the same time, a realistic assessment must recognize that learning trajectories are extremely flat for Indian school students. That means that not much learning is happening in a business-as-usual classroom. Any calculation of “human capital loss” must keep this reality in mind.
Looking at data from Andhra Pradesh Randomized Evaluation Study (APRESt), the above paper summarized the findings:
Even for the most mechanical arithmetic operations there is amazingly slow progress from second to fifth grade. For a “level two” competency like “two digit addition without carry” 40 percent of children in grade 2 answer correctly and yet only 70 percent answer correctly by grade 5. This means that of the 60 percent of children who did not already master addition by grade 2 less than half gained the skill in three full years of additional schooling. There are three of every ten children who enrolled school, stayed in school until grade 5, and never learned even the simplest addition.
Hence, the warnings of children falling behind due to the digital divide and school closures have some validity, yet one must also acknowledge that most children have already fallen behind long before any pandemic hit.
Adding a positive note, Bihar has shown that it can achieve extraordinary gains in learning within a very short time. As a study on Mission Gunvatta showed, learning can accelerate quickly if bureaucrats and frontline staff buy in on efforts to improve learning in schools and leave a business-as-usual mode of doing things:
(Source)
While we would not call a pandemic that throws millions of families into abysmal poverty an opportunity, maybe it can serve as an impetus to again come together around a common purpose of improving learning in Bihar’s government schools. At the same time, fiscal spaces shrink and even paying teachers might be a major challenge, potentially running counter such efforts.
Dropouts
Evidence indicates that the longer a child is out of school, the less likely he or she is to return. With family incomes collapsing, out-of-pocket spending might be difficult for families to afford and even “affordable” private schools might turn out to be not affordable at all. Whether these children will enter government schools remains to be seen.
The government must prepare a major exercise to ensure that children return to schools once they reopen. Even a one-day-school-week may allow children to remain connected to schools. This can be accompanied by full-fledged radio programs, accessible Interactive Voice Response Systems, loudspeaker announcements and mass-SMS facilities and visits by teachers to children at risk. It is the duty of the government to ensure that children get their Right to Education fulfilled. The frontline administration must stay alert on the issue and have the resources and staff to keep track of every child, in particular those at risk (e.g. children with disabilities).
With lakhs of migrant workers returning to Bihar, a State-wide enrolment hotline can be established. Migrants returning by train and bus can be registered with their mobile number and home addresses so that they can be called up or visited to instruct and support them to get their children enrolled in government schools. Alternatively, for families arriving at bus stations or train stations, on-the-spot enrolment can be facilitated and their village's schools can be informed to ensure that they are on the enrolment register. This will also ensure that they get access to crucial mid-day meals.
Malnutrition
An important announcement was recently made by the MHRD:
The Minister announced a land mark decision that Approval is being given for providing mid-day meal during summer holidays of schools, on which an additional expenditure of about 1600 crores will be made.
[...]the annual central allocation of cooking cost (for procurement of pulses, vegetable, oil, spices and fuel) under Mid Day Meal Scheme is enhanced to Rs. 8100 crore from Rs 7,300 crore ( an increment of 10.99%).
While the Mid-day Meal Scheme is a Centrally Sponsored Scheme (CSS) and requires States to contribute their shares, this decision opens the door to ensure mid-day meals during the summer holidays.
The Government of Bihar had decided to opt for Direct Benefit Transfer (DBT) instead of in-kind provision of meals. We strongly suggest switching to in-kind provision as banks are often at a distance, intra-household allocations of food might discriminate children at risk (e.g. disabled children, girls) and DBT does not help maintaining a link to schools.
Bottom line
Re-opening schools while new cases per day keep growing is probably too risky. The exact time for re-opening will depend on school characteristics (e.g. the availability of WASH infrastructure, masks, number of children per square meter of classroom space, etc).
Localizing decisions while recognizing State-level responsibility
While preparations should be made now, available evidence implies that children visiting schools with worse infrastructure will be either exposed to higher risks, visit schools less often or their schools re-opening later. State governments should prepare guidelines for re-opening schools and support them with procurement and distribution of masks, soaps, etc. which might be home-delivered together with Mid-Day Meals. Yet, doing so in a centralized fashion will hardly be possible. Districts and Blocks must be given freedoms and be supported. A cross-departmental cooperation is required to coordinate various aspects. A clear communication from the top is required. In a business-as-usual scenario, “elementary education administrators at the block level primarily perceive themselves, or report themselves to be, disempowered cogs in a hierarchical administrative culture that renders them powerless. They refer to their own roles and offices as “post offices,” used simply for doing the bidding of higher authorities and ferrying messages between the top and bottom of the education chain”, a study indicates. The post office must be left behind now in favour of a pro-active, empowered bureaucracy with a shared purpose.
Gearing up supplementary education modes that are low- and non-tech
It might well be that schools cannot run normally for many months to come. Yet, children have a right to a minimum number of instruction days. Hence, the government must ensure that children have access to education, even if they do not own a TV or smartphone. This will require to combine various low- and non-tech means (IVRS, radio, worksheets, public loudspeakers, …) to bridge the digital divide. Lessons in the open air might be another option, depending on the availability of space and weather conditions.
Frameworks to guide decision making
A joint framework for reopening schools by UNESCO, UNICEF, the World Bank and the World Food Programme provides further guidance and orientation for policy makers in these times of uncertainty.