Unintended consequences of measures implemented in the school setting to contain the COVID‐19 pandemic: a scoping review

Abstract

Background

Throughout the COVID‐19 pandemic, schools were a key setting for intervening with public health and social measures (PHSM) to reduce transmission of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). Consequently, there is a need to assess the varied unintended consequences associated with PHSM implemented in the school setting, for students, teachers, and school staff, as well as for families and the wider community. This is an update of a Cochrane scoping review first published in 2022.

Objectives

To comprehensively identify and summarise the published literature on the unintended consequences of public health and social measures implemented in the school setting to reduce the spread of SARS‐CoV‐2. This will serve to identify critical knowledge gaps to inform future primary research and systematic reviews. It may also serve as a resource for future pandemic management.

Search methods

We searched MEDLINE, Embase, CENTRAL, PsycINFO, ERIC, and Web of Science on 5 and 6 January 2023. We also searched two COVID‐19‐specific databases (Cochrane COVID‐19 Study Register and WHO COVID‐19 Global literature on coronavirus disease). Finally, we reviewed the included studies of all relevant systematic reviews and guidelines identified through the searches.

Selection criteria

We included studies that empirically assessed the impact of PHSM implemented in the school setting to reduce the spread of SARS‐CoV‐2. We imposed no restrictions with regard to the types of populations and specific interventions. Outcomes of interest were consequences that were measured or experienced, but not anticipated consequences. This review focused on real‐world evidence: empirical quantitative, qualitative, and mixed‐method studies were eligible for inclusion, but modelling studies were ineligible.

Data collection and analysis

The review was guided by a logic model. In line with the latest Cochrane effectiveness review of school measures to contain COVID‐19 and a conceptual framework of PHSM, this logic model distinguishes between measures to make contacts safer (related to individual protection and the physical environment), measures to reduce contacts (related to social interactions, movement, and services) and surveillance and response measures. Unintended consequences comprise the following categories: health and well‐being, health system and social welfare services, human and fundamental rights, acceptability and adherence, equality and equity, social and institutional, economic and resource, and ecological.

The review team screened all titles and abstracts, then potentially eligible full‐text articles, in duplicate. Across the included studies, we summarised and presented types of measures, consequences, and study designs using the predefined categories of the logic model, while allowing for emerging categories.

Main results

We included 60 studies (57 new to this update) from 25 countries. There were 31 quantitative studies, 17 qualitative studies, and 12 mixed‐method studies. Most targeted either students (26 studies), teachers and school staff (11 studies), or students and school staff (12 studies). Others evaluated measures aimed at parents (2 studies), staff and parents (1 study), students and teachers (3 studies), or the whole school (5 studies). The measures were related to individual protection (26 studies), the physical environment (20 studies), social interactions (25 studies), services (1 study), movement (3 studies), surveillance (9 studies) and response (7 studies). Nine studies evaluated the combined effect of multiple measures. The main consequences assessed were from the categories health and well‐being (29 studies), acceptability and adherence (31 studies), and social and institutional (23 studies). Fewer studies covered consequences from the categories equality and equity (2 studies), economic and resource (7 studies), and ecological (1 study). No studies examined consequences for the health system and social welfare services or for human and fundamental rights.

Authors' conclusions

This scoping review provides an overview of the evidence on the unintended consequences of PHSM implemented in the school setting to reduce the spread of SARS‐CoV‐2. The 60 included studies describe a broad body of evidence and cover a range of measures and unintended consequences, primarily consequences for health and well‐being, acceptability and adherence, social and institutional aspects, and economic aspects. The main gaps identified relate to consequences of school measures for the health system and social welfare services, human and fundamental rights, equality and equity, and the environment.

Further research is needed to fill these gaps, making use of diverse methodological approaches. Future studies should explore unintended consequences – whether beneficial or harmful – in more depth and over longer time periods, in different population groups, and across different contexts. A more robust evidence base could inform and facilitate decisions about whether, how, and when to implement or terminate COVID‐19 risk mitigation measures in school settings, and how to counter negative unintended consequences.

Funding

This publication was partially funded by the German Federal Ministry of Education and Research (BMBF) within the Network of University Medicine (NUM) 1.0, Grant No. 01KX2021 in the context of the project CEOsys, and NUM 2.0, Grant No. 01KX2121 in the context of the projects PREPARED and coverCHILD.

Registration

The protocol is registered on the Open Science Framework (osf.io/bsxh8).

The previous review is published in the Cochrane Library (10.1002/14651858.CD015397).

Plain language summary

What were the unintended consequences of school‐based measures to manage the COVID‐19 pandemic?

Why is this question important?

During the COVID‐19 pandemic, different measures were implemented in schools to help stop the virus spreading. These measures included changes to school rules or actions with the following goals.

• To reduce the spread of the virus when people were together

• To reduce the number of face‐to‐face contacts between people

• To test whether people had COVID‐19

• To isolate those with COVID‐19 to prevent it spreading

School measures like these can have both planned (intended) effects and unplanned (unintended) effects, which we need to monitor. For example, a rule to improve hand hygiene should lead to more handwashing (a planned effect) but could also lead to skin problems (an unplanned effect). These unplanned effects can be negative or positive, and can affect students, teachers, other school staff members, parents, and the wider community.

What are school measures to help contain the COVID‐19 pandemic?

We identified four types of school measures.

• Measures to make contacts safer. These work by reducing risk during face‐to‐face contact. Examples include mask wearing, ventilation, and hand hygiene.

• Measures to reduce the opportunity for contacts. These work by reducing face‐to‐face contact between students and others. One example is lowering the number of students in a classroom.

• Surveillance and response measures. These work by testing whether people have COVID‐19 and taking action to prevent it spreading, for example by isolating people who test positive.

• Multi‐component measures. These combine aspects of the above three types of measures.

What did we want to find out?

We aimed to find and describe evidence on the unplanned effects of school measures implemented to help contain the spread of COVID‐19.

What did we do?

We searched for studies that looked at side effects of school measures. We grouped the side effects into the following categories.

• Physical and mental health and well‐being

• Health and social care services

• Human rights

• Acceptability of measures

• Unfair differences in health outcomes

• Civil life, social interactions, and education

• Money and resources

• The environment

We summarised the available information on school measures and their side effects.

What did we find?

We included 60 studies from 25 countries. Regarding study design, 31 used numbers and statistics, 17 used text and descriptions, and 12 used both numbers and descriptions. Most studies looked at measures for staff (11 studies) or students (26 studies), or measures aimed at both staff and students or the whole school (20 studies). Only three studies looked at measures for parents or caregivers.

The studies investigated the following measures in schools.

• Individual protection (26 studies); for example, wearing masks

• Changes to the physical environment (20 studies); for example, improving ventilation

• Rules about social interactions (25 studies); for example, mixing online and in‐person teaching

• Changes to services (1 study); for example, cancelling extracurricular activities

• Rules about movement (3 studies); for example, staying in class at break time

• Surveillance measures (9 studies); for example, testing students and teachers

• Response measures (7 studies); for example, isolating people with COVID‐19

Nine studies looked at the joint effects of more than one measure (for example, joint effects of wearing masks and improving ventilation).

Most studies focused on the following unplanned effects.

• The influence of school measures on health and well‐being (29 studies)

• Whether school measures were acceptable or easy to carry out (31 studies)

• How school measures affected social interactions, school organisation, and education (23 studies)

Other unplanned effects included the following.

• Changes to unfair differences in health outcomes (2 studies)

• Effects on money and resources (7 studies)

• Effects on the environment (1 study)

What did we conclude?

Most of the studies in the review looked at the influence of school measures on health and well‐being; social interactions, school organisation, and education; and whether measures were acceptable and easy to carry out.

The most common school measures were wearing masks, cleaning and disinfecting rooms and surfaces, mixing online and in‐person learning, and ensuring physical distancing. Many studies also looked at combinations of several measures. This review identified the gaps, where more research is needed.

We need more studies to look at a wider variety of unplanned effects. We also need more studies looking at the unplanned effects of other school measures, such as testing and response measures.

Future studies should look at long‐term effects, too. We need more information about how these measures work in different types of schools and in different countries. This would help people make decisions about when and how to use different school measures.

How up to date is this evidence?

The evidence is current to January 2023.

Background

Despite global efforts to reduce morbidity and mortality from COVID‐19, and although the World Health Organization (WHO) declared that COVID‐19 was no longer a public health emergency of international concern in 2023 [1], the negative health consequences of the pandemic continue to be felt worldwide.

Throughout the pandemic, schools were a key setting for intervening with public health and social measures (PHSM) to help reduce transmission of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). In the beginning, more than 190 countries closed their schools completely or partially over an extended period of time [2]. A substantial body of evidence has documented the harmful unintended consequences of this measure. Moreover, there is mixed evidence regarding the effectiveness of school closures in containing the spread of the virus [3, 4]. In light of this, schools in many countries reopened relatively quickly with various PHSM in place. Prominent researchers in the field have called for schools to be the last place to close in the event of another pandemic due to the known harm of closure and the successful implementation of PHSM in this setting [5]. Relevant PHSM include measures to make contacts safer (e.g. mask wearing), measures to reduce contacts (e.g. hybrid teaching), and surveillance and response measures (e.g. testing) [6]. Studies have demonstrated the effectiveness of such measures for reducing the risk of infection in schools to levels below the general population risk [7].

Schools are conceptualised as complex adaptive systems, with multiple dynamic and interacting parts and agents that lead to system evolution or change within the school [8]. A school may be comprised of year groups, cohorts, and classes, as well as other groups, including friendship and peer groups, and groups for teachers or other school staff. These groups are nested within the school setting, and the school itself is a sub‐system nested within the broader educational system [8]. Decisions to implement PHSM in the school setting may be made by broad educational or political systems and groups, or by individual schools, and the level at which decisions are made affects how and to what extent system change is achieved. School or system functioning is an emergent property of its interacting parts and agents, which leads to behaviour change and disease control among the agents within the system [9]. This highlights the need to monitor the varied unintended consequences associated with PHSM implemented in the school setting.

Most research into PHSM implemented in the school setting has focused on their effectiveness [10, 6]. The original version of this scoping review found only 18 studies evaluating the unintended consequences of such measures, with a particular gap in evidence on psychosocial consequences and inequity and inequality. [11]. Understanding the unintended consequences of PHSM implemented in the school setting is important because gives decision makers a holistic picture of both the desired and the unintended (potentially negative) effects of measures. For example, Blanchard and colleagues described a school surveillance measure to test and isolate students in an attempt to reduce onward transmission, which had the intended consequence of days of absence from school [12]. However, one Cochrane review on the effectiveness of measures to reduce COVID‐19 transmission in the school setting found that different strategies could reduce the number of days of absence. For example, test‐to‐stay strategies allowed some asymptomatic students to continue attending school despite having been exposed to a COVID‐19 case [13]. The effectiveness review, and its included studies, did not assess the potential unintended consequences of absence from school, such as worse educational outcomes. Decision making about the well‐being of people affected by PHSM implemented in the school setting would benefit from a holistic evidence base, including evidence on effectiveness but also evidence that describes whether such measures have any positive or negative unintended consequences.

Many studies that have investigated unintended consequences of measures have relied on modelling. For example, lab‐based studies on mask wearing have found masks to influence sound propagation [14], impair individuals' ability to read the emotions of mask wearers [15], and impair individuals' ability to process faces holistically and obtain important social information (e.g. age, gender, ethnicity) at a glance [16]. Modelling techniques are highly dependent upon the model structure and assumptions, and the quality and validity of the data used to create the model [17]. In addition, models often have limited capacity to capture complex system dynamics [18], as in the school setting. Therefore, real‐world studies are necessary to identify the beneficial, harmful, or neutral unintended consequences of PHSM within the complex adaptive school system [19]. An overview of findings from such studies would help to inform decision makers and programme implementers.

Rationale for conducting a scoping review

A scoping review is defined as an exploratory but systematic review of the literature. The aim of a scoping review is to explore and describe the characteristics of a body of evidence, including the extent and variety of such evidence [20].

Why it is important to update this scoping review

This is an update of a Cochrane scoping review published in 2022, which found a paucity of empirical research on the unintended consequences of PHSM implemented in the school setting [11]. Because the original scoping review was conducted at an early stage of the pandemic, when there was scarce real‐world evidence available on the topic, there is a need for an updated overview of the evidence. This will help to determine the extent to which the empirical evidence base has grown between March 2021 and January 2023, and to highlight any persistent gaps in the literature [21].

Objectives

To comprehensively identify and summarise the published literature on the unintended consequences of public health and social measures implemented in the school setting to reduce the spread of SARS‐CoV‐2. This will serve to identify critical knowledge gaps to inform future primary research and systematic reviews [22]. It may also serve as a resource for future pandemic management.

Methods

The protocol was registered with the Open Science Framework (DOI 10.17605/OSF.IO/Y7TPB). We used the PRISMA extension guidance for scoping reviews to steer reporting [20] (see Supplementary material 5).

Differences between protocol and review

In the protocol, we defined three categories of measures (measures to make contacts safer, measures to reduce contacts, and surveillance and response measures). However, while conducting the published effectiveness review [13], it became apparent that some studies investigated multiple measures in combination. We therefore added a fourth category: multi‐component measures. We also modified the logic model of this scoping review, as described below ('Logic model').

This is an update of a previous review on the unintended consequences of measures implemented in the school setting to contain the COVID‐19 pandemic. The original review was published in 2022 [11], and this update covers the available evidence up to January 2023. For this update, we modified our eligibility criteria to exclude modelling studies and studies that investigated days lost due to quarantine. We also excluded studies that reported anticipated consequences. The lead author on the original scoping review, as well as many of the contributing authors, were not available to work on the update. This is clearly documented in the Acknowledgements section.

Logic model

The logic model (Figure 1), or pictorial representation of theory [23], for this scoping review has evolved since the original version [11]. This transformation was informed by the following three aspects.

1.

Logic model of measures implemented in the school setting to contain the COVID‐19 pandemic and their unintended consequences

• The findings of the original Cochrane scoping review [11]

• The findings of the update of the Cochrane effectiveness review on measures implemented in the school setting to contain the COVID‐19 pandemic [13]

• The conceptual framework of PHSM developed by Rehfuess and colleagues for WHO [24]

Specifically, we adapted the objectives, measures, and outcomes sections of the conceptual framework so that unintended consequences were comprehensively addressed and sufficiently prominent within the logic model.

The logic model describes measures that make contacts safer (e.g. individual protection and adapting the physical environment), measures that reduce contacts (e.g. related to social interactions, movement, and services), and surveillance and response measures (e.g. screening, testing, contact tracing, isolation, and quarantine).

The potential unintended consequences are categorised as follows.

• Health and well‐being

• Health system and social welfare services

• Human and fundamental rights

• Acceptability and adherence

• Equality and equity

• Social and institutional

• Economic and resource

• Ecological

We see logic model development as an iterative process, so new categories may be added as they emerge.

Criteria for considering studies for this review

Types of study

We included studies that empirically assessed the consequences of PHSM implemented in the school setting to reduce the spread of SARS‐CoV‐2. This update focused on real‐world evidence. Empirical quantitative and qualitative studies were included to increase the likelihood of capturing well‐documented consequences as well as emergent consequences. Although modelling studies were eligible for inclusion in the original review [11], they were excluded in this update because of their limited real‐world applicability and the much larger number of empirical studies now available. Work not published in scientific journals was excluded.

Settings

Schools are environments in which educational services are provided to children and adolescents with unique and diverse characteristics, abilities, and needs. Typically, the age of students in schools ranges from four to 18 years, depending on the country. In this review, we considered schools to be any setting with the primary purpose of providing education to children, and we distinguished between primary and secondary education. Primary, elementary, or basic school is often the first school type children attend as part of their compulsory education. Some countries refer to the first year of compulsory education as preschool, but we did not make this distinction in the present review. Compulsory primary education typically lasts six years, although the duration ranges from four to seven years. Students usually enter secondary school between the ages of 10 and 13 years. Other names for secondary school are middle school or high school [25].

We defined the school setting as the school, the school grounds, school vehicles, or any activity organised by or linked to the school. Relevant PHSM might impact activities anywhere within this setting, including in the classroom, hallways, cafeterias, playgrounds, or during transportation and movement around the school. As with previous reviews on this topic [11, 6], by 'in and around' the school we refer to activities such as public transportation to and from the school, and activities between students, staff, and other populations that take place before and after school, which would not have taken place if schools were closed (e.g. school breakfast programmes, after‐school clubs, and extracurricular activities). The school may have no formal control over these activities, but they are linked to the school and may influence SARS‐CoV‐2 transmission.

Measures

Many different measures may be implemented in the school setting. They can broadly be categorised as follows.

• Measures to make contacts safer: individual protection and adaptation to the physical environment

  • Individual protection: measures such as physical distancing, hygiene, masks, and other personal protective equipment

  • Physical environment: for example, improving ventilation or installing measures to facilitate the separation of individuals and groups

• Measures to reduce contacts (between individuals or cohorts): modification of social interactions, movement, and services

  • Social interactions: adapting the ways individuals and groups interact with each other (e.g. by limiting the number of people present in a room through hybrid teaching). School closure with 100% remote teaching was not an eligible measure in this review.

  • Movement: adapting the way students and school staff travel to and from school and move within school premises (e.g. changing modes of transportation)

  • Services: adapting, cancelling, or changing the timing of school‐related activities and services (e.g. ways of teaching, ways of delivering meals in schools, ways of delivering healthcare in schools).

• Surveillance and response measures: these are associated with measures to reduce contacts and measures to make contacts safer

  • Surveillance measures: aimed at testing, screening, and contact tracing symptomatic or asymptomatic individuals or clusters. Surveillance strategies make contacts safer by identifying potentially infectious individuals and reducing contacts with infected individuals and individuals at risk of being infected. One can differentiate between strategies to test symptomatic individuals or contact persons (diagnostic testing, e.g. using rapid tests or polymerase chain reaction (PCR) tests), to routinely screen an asymptomatic group of individuals within the school setting (screening, e.g. daily symptom screening of all students, PCR pool testing of all primary school children twice a week, rapid testing of all secondary school children three times a week), or to surveillance test a fraction of the school population to identify potential outbreaks of SARS‐CoV‐2 (surveillance testing, e.g. PCR testing of a representative sample of secondary schools).

  • Response measures: responding to a confirmed or suspected SARS‐CoV‐2 infection by isolating individuals or implementing quarantine measures. This specifically excludes periods of school closure. Response strategies reduce contacts by isolating or quarantining infected or exposed individuals. Test‐to‐stay response strategies make contacts safer by allowing exposed individuals to attend school and to come into contact with others while testing daily. These individuals then isolate if they test positive.

• Multi‐component measures: any combination of strategies from the above categories where individual components are assessed together, not separately.

We excluded studies concerned with the unintended consequences of vaccinations because children and adolescents were not usually vaccinated in the school setting, but through other avenues. Vaccination status, where available, was instead treated as a contextual factor.

Unintended consequences

We included studies that focused on unintended consequences. Studies concerned with intended consequences were excluded. This scoping review considered the following categories of unintended consequences, based on our logic model (Figure 1).

• Health and well‐being: direct and indirect physical health, psychosocial health, and health behaviour impacts not associated with COVID‐19 (e.g. hand eczema through frequent handwashing; weight gain through less physical education at school or on the way to school; anxiety; social skills; willingness or opportunity to seek health care).

• Health system and social welfare services: for example, access to, use of, or quality of health services and child protection

• Human and fundamental rights: for example, autonomy, self‐determination, and privacy, as well as (non‐)discrimination and stigmatisation.

• Acceptability and adherence: acceptability of, adherence to, and compliance with measures.

• Equality and equity: fair distribution of benefits and burdens in relation to health‐related, social, and economic consequences (e.g. disadvantage for children from lower income families, gender inequalities, impact on rural versus urban populations).

• Social and institutional: for example, civil life, sociocultural institutions, and participation; social cohesion and social well‐being; education and development; conditions of daily living; safety, security, and crime; and social norms, values, and practices.

• Economic and resource: consequences relating to both financial and non‐financial resources (e.g. household income, job opportunities for teachers and other staff, changes in workload and productivity, lost work days for parents or caregivers, national expenditures to implement measures in schools, personal expenditures to adhere to school policies).

• Ecological: consequences relating to the environment (e.g. energy consumption, noise, and greenhouse gas emissions).

In this update, we decided to exclude days of attendance lost/saved as an unintended consequence of school measures, although it was included in the original scoping review [11]. Our decision is in line with a recent PHSM conceptualisation framework that classifies days of attendance lost/saved due to measures as an intended consequence [24]. Another change from the original review was the exclusion of anticipated consequences (e.g. participants' perceptions of school reopening or specific measures to be taken in the future); for this update, we only considered consequences that were experienced or measured.

Search methods

The search strategies comprised keywords and controlled vocabulary (e.g. MeSH terms) and were built around combinations of four main components: SARS‐CoV‐2/COVID‐19, school settings, PHSM, and unintended consequences. The start date of the search was 2020, when publications about schools and the COVID‐19 pandemic began to appear. We applied no study design filters because a wide range of study types were potentially eligible for inclusion. An experienced information specialist (AN‐S) designed and executed the searches, which were verified by a content expert. Supplementary material 1 presents our search strategies in full.

We ran searches in the following electronic databases.

• MEDLINE via Ovid (2020 to 6 January 2023)

• Embase via Ovid (2020 to 2023, week 1)

• Cochrane Central Register of Controlled Trials (CENTRAL; 2023, Issue 3; searched 5 January 2023 via the Cochrane Register of Studies; crsweb.cochrane.org/)

• Ovid APA PsycINFO (2020 to 6 January 2023)

• Ovid ERIC (2020 to 6 January 2023)

• Web of Science via Clarivate (2020 to 6 January 2023)

We also searched the following COVID‐19‐specific databases.

• Cochrane COVID‐19 Study Register (covid‐19.cochrane.org/; searched 5 January 2023 via the Cochrane Register of Studies; crsweb.cochrane.org/)

• WHO COVID‐19 Global literature on coronavirus disease (search.bvsalud.org/global‐literature‐on‐novel‐coronavirus‐2019‐ncov/; searched 5 January 2023).

The scoping review can thus be considered current to 5 January 2023.

Finally, we reviewed the included studies of all relevant systematic reviews and guidelines identified through the searches.

Language

Within our review team, we were able to review studies written in English, French, German, Italian, and Spanish. Where necessary, we asked for help with translation from other languages (e.g. Chinese Mandarin) using our networks. However, we mainly found English studies that met our inclusion criteria at the title and abstract stage (there were also six Spanish studies).

Study selection

After deduplication, all review authors involved with screening (HL, SK, JH, EO, TH, BS, CJS) worked together to screen all titles and abstracts in duplicate, excluding all studies that were clearly irrelevant. TH was not involved in the screening/selection for Kretschmer 2022 [26], on which he is a co‐author. This study was screened by HL and SK.

The same review authors (HL, SK, JH, EO, TH, BS, CJS) assessed the full‐text articles of all studies deemed potentially relevant or unclear at the title/abstract screening stage. In case of any discrepancies, the two review authors who had screened the study in question discussed it further, involving a third review author (either HL or SK) or the larger review team to achieve consensus, where necessary. At this stage, we made a final decision regarding inclusion or exclusion.

We used Endnote to manage collection and deduplication of records [27]. For title and abstract screening, we used Rayyan, a web‐based application, designed for citation screening for systematic reviews [28]. We documented and reported reasons for excluding ineligible articles during full‐text review.

We developed screening guidance for the study selection process to ensure review authors applied the inclusion and exclusion criteria consistently. All review authors screened 10% of the search results and discussed discrepancies before starting to screen titles and abstracts. Before commencing full‐text screening, all review authors screened 10 full‐text articles and discussed discrepancies. Additionally, we collected and clarified all uncertainties in screening on a rolling basis and discussed these in regular online meetings.

Data extraction and charting

Four review authors (JH, SK, HL, EO) extracted and charted the characteristics and data from each study into the categories of the data extraction form in Microsoft Excel (Microsoft Corporation 2018). The form included categories that we had predefined based on our initial understanding of measures in and around schools and their consequences (Figure 1), as well as inductive subcategories that we developed as new concepts emerged during data extraction. Predefined categories included population (e.g. age group), setting (e.g. type of school), types of measures (e.g. measures making contacts safer), comparisons (where available), consequences of interest (e.g. health, economic, and social impact), and study designs (e.g. quantitative, qualitative, mixed method). We pilot tested the data extraction form on one study and subsequently revised it. Two experienced review authors (HL, SK) reviewed all extracted data between them. Supplementary material 6 presents the data extraction form.

Collation, summary, and reporting of results

We collated, summarised, and reported the extracted data. Specifically, we aimed to define, summarise, and present types of measures, consequences of interest, and study designs. We aimed to build upon findings from the previous review. We particularly looked for additional or more refined measure types, unintended consequences, resource and implementation aspects, and contextual factors that might impede or facilitate the implementation of the measures.

Results

Results of the search

The database searches identified 35,174 records, and the citation tracking of included studies within relevant systematic reviews identified 473 records. After deduplication and title and abstract screening, 426 full texts were assessed for eligibility, and 60 were included (Table 1; Supplementary material 2; Supplementary material 4). Of these 60 studies, three had been included in the original review and 57 were new to this update. Figure 2 presents brief reasons for exclusion of all studies at full‐text review, and Supplementary material 3 provides more detailed reasons for the exclusion of 10 studies that were very close to being included.

1. Overview of synthesis and included studies.

Study ID	Study Design	Setting	Measure Implemented	Outcomes
Ahmed 2022	Mixed method: concurrent design including primary and secondary sources	Primary and secondary schools in Ethiopia	INDIVIDUAL PROTECTION: masking, hand hygiene PHYSICAL ENVIRONMENT: ventilation, classroom layout SOCIAL INTERACTIONS: hybrid learning SURVEILLANCE: temperature checking	Acceptability and adherence
Amin‐Chowdury 2021	Quantitative: cross‐sectional online survey	Primary and secondary schools in England, UK	INDIVIDUAL PROTECTION: hand hygiene SOCIAL INTERACTIONS: physical distancing, hybrid learning SURVEILLANCE: temperature checking	Acceptability and adherence
Amman 2022	Quantitative: prospective cohort study	Primary and secondary schools in Switzerland	INDIVIDUAL PROTECTION: masking	Health and well‐being Acceptability and adherence
Assathiany 2021	Quantitative: survey	Primary schools in France	INDIVIDUAL PROTECTION: masking	Health and well‐being Acceptability and adherence Social and Institutional
Borch 2020	Quantitative: cross‐sectional survey	Schools in Denmark (school type not reported)	INDIVIDUAL PROTECTION: hand hygiene	Health and well‐being
Chadwick 2021	Quantitative: cross‐sectional survey	Primary and secondary schools in Ireland	INDIVIDUAL PROTECTION: masking PHYSICAL ENVIRONMENT: classroom layout, not sharing equipment SOCIAL INTERACTIONS: physical distancing SERVICES: cancellation of activities	Social and institutional
Coelho 2022	Mixed method: cross‐sectional study	K‐12 schools in Canada	INDIVIDUAL PROTECTION: masking	Health and well‐being
Collier 2022	Quantitative: feasibility and acceptability study	Fee‐paying schools in England, UK	SURVEILLANCE: PCR testing RESPONSE: isolation	Acceptability and adherence
Corneli 2022	Qualitative: in‐depth interviews	Primary and secondary schools in the USA	SURVEILLANCE: test to stay RESPONSE: quarantine	Health and well‐being Acceptability and adherence Social and Institutional
Denford 2022	Qualitative: in‐depth interviews	Schools in England, UK (school type not reported)	SURVEILLANCE: daily testing	Acceptability and adherence
Ding 2023	Mixed method: questionnaire and observation	Secondary schools in the Netherlands	PHYSICAL ENVIRONMENT: ventilation	Acceptability and adherence
Doron 2021	Quantitative: cross‐sectional	Middle and high school in the USA	SURVEILLANCE: asymptomatic screening RESPONSE: isolation	Acceptability and adherence Economic and resource
Duill 2021	Quantitative: experimental study	Primary schools in Germany	PHYSICAL ENVIRONMENT: air purification	Ecological
Ellis 2022	Mixed method: sequential explanatory design	Kindergarten and primary schools in the USA	INDIVIDUAL PROTECTION: masking, hand hygiene SOCIAL INTERACTION: physical distancing, cohorting, not sharing equipment MOVEMENT: restricted at lunchtime	Social and institutional
Falk 2022;	Quantitative: observational cohort study	Middle and high school in the USA	INDIVIDUAL PROTECTION: masking SOCIAL INTERACTION: physical distancing	Acceptability and adherence Economic and resource
Farhadi 2022;	Qualitative: focus group discussions	Secondary schools in Canada	SOCIAL INTERACTION: reduction in mixing across classrooms MOVEMENT: staying in classroom	Health and well‐being
Galasso 2022	Qualitative: in‐depth interviews	Schools in Italy and Ireland (school type not reported)	SOCIAL INTERACTION: reduction of mixing between classes MOVEMENT: staying in classroom RESPONSE: isolation	Health and well‐being Acceptability and adherence Economic and resource
Goldhaber 2022	Quantitative: cross‐sectional	Schools in the USA (school type not reported)	SOCIAL INTERACTION: hybrid learning	Social and institutional Equality and equity
Hanno 2022	Quantitative: longitudinal cohort study	Schools in the USA (school type not reported)	SOCIAL INTERACTION: hybrid learning	Social and institutional
Henderson 2021	Quantitative: cross‐sectional survey	Junior, middle, and high schools in the USA	SOCIAL INTERACTION: hybrid learning	Health and well‐being Acceptability and adherence Social and Institutional
Hilbert 2021	Mixed method: cross‐sectional survey with qualitative element	Schools in the USA (school type not reported)	PHYSICAL ENVIRONMENT: cleaning and disinfection	Health and well‐being
Jesser 2022	Qualitative: semi‐structured interviews	Secondary schools in Austria	SOCIAL INTERACTION: hybrid learning	Health and well‐being Social and institutional
Kaliyadan 2022	Qualitative: case study	Primary schools in India	INDIVIDUAL PROTECTION: masking	Health and well‐being
Kamoga 2022	Qualitative: semi‐structured interviews	Primary schools in Sweden	PHYSICAL ENVIRONMENT: increased cleaning regimens SOCIAL INTERACTION: no showering after physical education	Health and well‐being Economic and resource
Kavrayic 2021	Qualitative: case study	Primary and secondary schools in Turkey	INDIVIDUAL PROTECTION: masking PHYSICAL ENVIRONMENT: cleaning and disinfection	Acceptability and adherence Equality and equity
Kebede 2022	Quantitative: cross‐sectional	Secondary schools in Ethiopia	INDIVIDUAL PROTECTION: hand hygiene, messaging, mask wearing PHYSICAL ENVIRONMENT: classroom layout, ventilation, cleaning and disinfection SOCIAL INTERACTION: avoidance of sharing equipment, physical distancing	Acceptability and adherence
Kenworthy 2022	Qualitative: focus group discussions	Middle and secondary schools in the USA	SURVEILLANCE: testing	Acceptability and adherence
Khandewal 2022	Quantitative: cross‐sectional	K‐12 schools in the USA	INDIVIDUAL PROTECTION: masking	Social and institutional
Kretschmer 2022	Quantitative: feasibility and acceptability study	Primary, secondary, and special needs schools in Germany	SURVEILLANCE: testing	Acceptability and adherence
Levitt 2022	Quantitative: cross‐sectional survey	Elementary school in the USA	SOCIAL INTERACTION: hybrid learning	Health and well‐being Social and institutional
Li 2021	Quantitative: cross‐sectional	Schools in China (school type not reported)	INDIVIDUAL PROTECTION: masking	Health and well‐being
Mahmud 2022	Mixed method: cross‐sectional survey with qualitative element	Special schools in England, UK	SOCIAL INTERACTION: physical distancing	Acceptability and adherence
Marchant 2022	Mixed method: cross‐sectional survey with qualitative element	Primary schools in Wales, UK	INDIVIDUAL PROTECTION: masking	Health and well‐being Acceptability and adherence Social and Institutional
McDonald 2022	Qualitative: in‐depth interviews	Primary schools in England, UK	INDIVIDUAL PROTECTION: masking SOCIAL INTERACTION: physical distancing SURVEILLANCE: testing	Social and institutional
McKellar 2023	Quantitative: cross‐sectional	Schools in the USA (school type not reported)	SOCIAL INTERACTION: hybrid learning	Health and well‐being
Mendoza 2021	Quantitative: evaluation	K‐12 schools in the USA	SURVEILLANCE: testing	Economic and resource
Mickells 2021	Quantitative: prospective observational survey	Primary schools in the USA	INDIVIDUAL PROTECTION: masking	Health and well‐being Acceptability and adherence
Moliner 2022	Mixed method: exam results and interviews	Secondary schools in Spain	SOCIAL INTERACTION: hybrid learning	Social and institutional
Moliner and Alegre 2022	Mixed method: quantitative maths scores and focus groups	Secondary schools in Spain	PHYSICAL ENVIRONMENT: classroom layout	Social and institutional
Mori 2022	Quantitative	Elementary, junior high, and high schools in Japan	INDIVIDUAL PROTECTION: masking PHYSICAL ENVIRONMENT: ventilation	Health and well‐being Acceptability and adherence
Mudaly 2021	Qualitative: critical policy analysis	Schools in South Africa (school type not reported)	INDIVIDUAL PROTECTION: masking, hand hygiene PHYSICAL ENVIRONMENT: cleaning and disinfection SOCIAL INTERACTION: physical distancing	Social and institutional
Mwesige 2021	Quantitative: cross‐sectional	Primary and secondary schools in Uganda	INDIVIDUAL PROTECTION: masking	Acceptability and adherence
Nabe‐Nielsen 2021	Quantitative: cross‐sectional	Schools in Denmark (school type not reported)	INDIVIDUAL PROTECTION: hand hygiene PHYSICAL ENVIRONMENT: cleaning and disinfection SOCIAL INTERACTION: limiting contacts	Health and well‐being
Nelson 2021	Qualitative: in‐depth interviews	Schools in England, UK (school type not reported)	SOCIAL INTERACTION: physical distancing	Social and institutional
Obeng 2022	Qualitative: in‐depth interviews	Preschools in Ghana	INDIVIDUAL PROTECTION: masking, hand hygiene PHYSICAL ENVIRONMENT: classroom layout, cleaning and disinfection SOCIAL INTERACTION: physical distancing SURVEILLANCE: temperature checking	Health and well‐being
Özer 2022	Qualitative: descriptive approach	All school types in Turkey	PHYSICAL ENVIRONMENT: ventilation	Social and institutional
Pressley 2022	Mixed method: exploratory	Elementary school in the USA	SOCIAL INTERACTION: hybrid learning	Health and well‐being
Quinn 2022	Mixed method	Primary schools in England, UK	PHYSICAL ENVIRONMENT: adaptive design of shared spaces	Social and institutional
Schlegtendal 2022	Quantitative: RCT	Schools in Germany (school type not reported)	INDIVIDUAL PROTECTION: masking	Health and well‐being
Schwal 2022	Qualitative: semi‐structured interviews	Secondary schools in Argentina	INDIVIDUAL PROTECTION: masking, hand hygiene PHYSICAL ENVIRONMENT: ventilation SOCIAL INTERACTION: physical distancing, avoidance of sharing equipment, cohorting SURVEILLANCE: temperature checking	Acceptability and adherence
Scott 2021	Qualitative: diary entries	Schools in England, UK (school type not reported)	INDIVIDUAL PROTECTION: masking SOCIAL INTERACTION: physical distancing, hybrid teaching RESPONSE: isolation	Health and well‐being Acceptability and adherence Social and Institutional
Simpson 2022	Quantitative: longitudinal cohort study	Schools in Australia (school type not reported)	SOCIAL INTERACTION: hybrid learning	Health and well‐being
Srikrishna 2022	Quantitative: experimental study	Elementary school in the USA	PHYSICAL ENVIRONMENT: air purification	Acceptability and adherence
Sundaram 2022	Quantitative: observational cohort study	Primary and secondary schools in England, UK	INDIVIDUAL PROTECTION: masking and hand hygiene PHYSICAL ENVIRONMENT: cleaning and disinfection, classroom layout, not sharing equipment SOCIAL INTERACTION: physical distancing	Acceptability and adherence
Tsujimoto 2022	Quantitative: longitudinal cohort study	High schools in Canada	SOCIAL INTERACTION: hybrid learning	Health and well‐being
Virji 2021	Qualitative: semi‐structured interviews	Primary schools in Abu Dhabi	SURVEILLANCE: saliva testing	Acceptability and adherence
Walters 2021	Quantitative: cross‐sectional	Middle school in the USA	SOCIAL INTERACTION: hybrid learning	Health and well‐being
Wilson 2022	Quantitative: online survey	K‐5 schools in the USA	INDIVIDUAL PROTECTION: masking PHYSICAL ENVIRONMENT: ventilation SOCIAL INTERACTION: physical distancing RESPONSE: contact tracing	Acceptability and adherence
Worrell 2023	Mixed method: online survey	K‐12 schools in the USA	RESPONSE: quarantine	Health and well‐being Acceptability and adherence Social and Institutional Economic and resource
Xu 2022	Quantitative: cross‐sectional	Junior and senior high school in China	INDIVIDUAL PROTECTION: masking	Health and well‐being

RCT: randomised controlled trial.

2.

PRISMA flow diagram

Description of studies

Study type

There were 31 quantitative studies (Amin‐Chowdury 2021 [29]; Amman 2022 [30]; Assathiany 2021 [31]; Borch 2020 [32]; Chadwick 2021 [33]; Collier 2022 [34]; Doron 2021 [35]; Duill 2021 [36]; Falk 2022 [37]; Goldhaber 2022 [38]; Hanno 2022 [39]; Henderson 2021 [40]; Kebede 2022 [41]; Khandewal 2022 [42]; Kretschmer 2022; Levitt 2022 [43]; Li 2021 [44]; McKellar 2023 [45]; Mendoza 2021 [46]; Mickells 2021 [47]; Mori 2022 [48]; Mwesige 2021 [49]; Nabe‐Nielsen 2021 [50]; Schlegtendal 2022 [51]; Simpson 2022 [52]; Srikrishna 2022 [53]; Sundaram 2022 [54]; Tsujimoto 2022 [55]; Walters 2021 [56]; Wilson 2022 [57]; Xu 2022 [58]), 17 qualitative studies (Corneli 2022 [59]; Denford 2022 [60]; Farhadi 2022 [61]; Galasso 2022 [62]; Jesser 2022 [63]; Kaliyadan 2022 [64]; Kamoga 2022 [65]; Kavrayic 2021 [66]; Kenworthy 2022 [67]; McDonald 2022 [68]; Mudaly 2021 [69]; Nelson 2021 [70]; Obeng 2022 [71]; Özer 2022 [72]; Schwal 2022 [73]; Scott 2021 [74]; Virji 2021 [75]), and 12 mixed‐method studies (Ahmed 2022 [76]; Coelho 2022 [77]; Ding 2023 [78]; Ellis 2022 [79]; Hilbert 2021 [80]; Mahmud 2022 [81]; Marchant 2022 [82]; Moliner and Alegre 2022 [83]; Moliner 2022 [84]; Pressley 2022 [85]; Quinn 2022 [86]; Worrell 2023 [87]).

Study design

Quantitative studies

Of the 31 quantitative studies, 27 employed observational study designs (Amin‐Chowdury 2021; Amman 2022; Assathiany 2021; Borch 2020; Chadwick 2021; Collier 2022; Doron 2021; Falk 2022; Goldhaber 2022; Hanno 2022; Henderson 2021; Kebede 2022; Khandewal 2022; Kretschmer 2022; Levitt 2022; Li 2021; McKellar 2023; Mickells 2021; Mori 2022; Mwesige 2021; Nabe‐Nielsen 2021; Simpson 2022; Sundaram 2022; Tsujimoto 2022; Walters 2021; Wilson 2022; Xu 2022), three were quasi‐experimental (Duill 2021; Mendoza 2021; Srikrishna 2022), and one was a randomised controlled trial (Schlegtendal 2022).

Qualitative studies

Of the 17 qualitative studies, four used semi‐structured interviews (Jesser 2022; Kamoga 2022; Schwal 2022; Virji 2021), six used in‐depth interviews (Corneli 2022; Denford 2022; Galasso 2022; McDonald 2022; Nelson 2021; Obeng 2022), two used focus group discussions (Farhadi 2022; Kenworthy 2022), two used a case study analysis (Kaliyadan 2022; Kavrayic 2021), one was a critical policy analysis (Mudaly 2021), one looked at diary entries (Scott 2021), and one was a descriptive document analysis (Özer 2022).

Mixed‐method studies

Of the 12 mixed‐method studies, five included a quantitative survey with some open‐ended qualitative questions (Hilbert 2021; Mahmud 2022; Marchant 2022; Pressley 2022; Worrell 2023). Four studies consisted of a survey combined with interviews (Ellis 2022; Quinn 2022), focus groups (Coelho 2022), or both (Ahmed 2022). For one of these studies, only the qualitative data were relevant for inclusion (Quinn 2022). The last of these four studies combined a survey and interviews with a document analysis and observation checklists (Ahmed 2022). Two studies combined quantitative unintended consequences on educational outcomes with qualitative exploration: for one study, this consisted of maths scores combined with focus groups, where only the qualitative data were relevant for inclusion (Moliner and Alegre 2022); and the other study combined exam results with interviews, where both data types were relevant for inclusion (Moliner 2022). Finally, one study implemented a technical questionnaire, combined with an observation sheet of classroom ventilation and a classroom checklist (Ding 2023).

Population

Quantitative studies

The 31 quantitative studies evaluated measures that targeted the following populations.

• Students (15 studies: Amman 2022; Assathiany 2021; Borch 2020; Goldhaber 2022; Hanno 2022; Henderson 2021; Kebede 2022; Levitt 2022; McKellar 2023; Mickells 2021; Mwesige 2021; Schlegtendal 2022; Tsujimoto 2022; Walters 2021; Xu 2022)

• Teachers (5 studies: Chadwick 2021; Khandewal 2022; Kretschmer 2022; Li 2021; Nabe‐Nielsen 2021)

• Parents (Simpson 2022)

• Students and staff (6 studies: Amin‐Chowdury 2021; Collier 2022; Falk 2022; Mori 2022; Sundaram 2022; Wilson 2022)

• Students and teachers (Duill 2021; Srikrishna 2022)

• Whole school (Doron 2021; Mendoza 2021)

Nine studies investigated unintended consequences for students by proxy, collecting data from parents (Assathiany 2021; Hanno 2022; Kebede 2022; Levitt 2022), teachers (Mickells 2021; Wilson 2022), parents and staff (Amin‐Chowdury 2021), head teachers and staff (Sundaram 2022), and school nurses (Collier 2022).

Qualitative studies

The 17 qualitative studies evaluated measures that targeted the following populations.

• Students (6 studies: Corneli 2022; Jesser 2022; Kaliyadan 2022; Nelson 2021; Scott 2021; Virji 2021)

• Teachers (3 studies: Farhadi 2022; Kavrayic 2021; Obeng 2022)

• Parents (Galasso 2022)

• Students and staff (Özer 2022; Schwal 2022)

• Students and teachers (Kamoga 2022)

• Staff and parents (McDonald 2022)

• Whole school (3 studies: Denford 2022; Kenworthy 2022; Mudaly 2021)

Mixed‐method studies

The 12 mixed‐method studies evaluated measures that targeted the following populations.

• Students (5 studies: Coelho 2022; Moliner 2022; Moliner and Alegre 2022; Quinn 2022; Worrell 2023)

• Teachers (Ellis 2022; Pressley 2022)

• Staff (Hilbert 2021)

• Students and staff (4 studies: Ahmed 2022; Ding 2023; Mahmud 2022; Marchant 2022)

Of the studies focused on students, one investigated unintended consequences for students by proxy through data collection with school staff (Quinn 2022).

Setting

Quantitative studies

The 31 quantitative studies took place in the following continents and countries.

• North America (14 studies)

  • USA (Doron 2021; Falk 2022; Goldhaber 2022; Hanno 2022; Henderson 2021; Khandewal 2022; Levitt 2022; McKellar 2023; Mendoza 2021; Mickells 2021; Srikrishna 2022; Walters 2021; Wilson 2022)

  • Canada (Tsujimoto 2022)

• Europe (11 studies)

  • Germany (Duill 2021; Kretschmer 2022; Schlegtendal 2022)

  • England, UK (Amin‐Chowdury 2021; Collier 2022; Sundaram 2022)

  • Denmark (Borch 2020; Nabe‐Nielsen 2021)

  • France (Assathiany 2021)

  • Switzerland (Amman 2022)

  • Ireland (Chadwick 2021)

• Asia (3 studies)

  • China (Li 2021; Xu 2022)

  • Japan (Mori 2022)

• Africa (2 studies)

  • Ethiopia (Kebede 2022)

  • Uganda (Mwesige 2021)

• Australasia (1 study)

  • Australia (Simpson 2022)

Figure 3 illustrates the geographical locations of included studies.

3.

A world map to visualise the location of included studies

The types of school in the quantitative studies were as follows.

• Combination of school types (either primary and secondary; all school types; or primary, secondary, and special schools) in 10 studies (Amin‐Chowdury 2021; Amman 2022; Chadwick 2021; Henderson 2021; Khandewal 2022; Kretschmer 2022; Mendoza 2021; Mwesige 2021; Mori 2022; Sundaram 2022)

• Primary schools (6 studies: Assathiany 2021; Duill 2021; Levitt 2022; Mickells 2021; Srikrishna 2022; Wilson 2022)

• Secondary schools (6 studies: Doron 2021; Falk 2022; Kebede 2022; Tsujimoto 2022; Walters 2021; Xu 2022)

• School type not reported (9 studies: Borch 2020; Collier 2022; Goldhaber 2022; Hanno 2022; Li 2021; McKellar 2023; Nabe‐Nielsen 2021; Schlegtendal 2022; Simpson 2022)

Qualitative studies

The 17 qualitative studies took place in the following continents and countries.

• North America (3 studies)

  • USA (Corneli 2022; Kenworthy 2022)

  • Canada (Farhadi 2022)

• South America (1 study)

  • Argentina (Schwal 2022)

• Europe (9 studies)

  • Sweden (Kamoga 2022)

  • Turkey (Kavrayic 2021; Özer 2022)

  • England, UK (Denford 2022; McDonald 2022; Nelson 2021; Scott 2021)

  • Austria (Jesser 2022)

  • Italy (Galasso 2022)

  • Ireland (Galasso 2022)

• Asia (2 studies)

  • India (Kaliyadan 2022)

  • Abu Dhabi (Virji 2021)

• Africa (2 studies)

  • Ghana (Obeng 2022)

  • South Africa (Mudaly 2021)

The types of school in the qualitative studies were as follows.

• Combination of school types (primary and secondary) in three studies (Corneli 2022; Kavrayic 2021; Kenworthy 2022)

• Primary schools (4 studies: Kaliyadan 2022; Kamoga 2022; McDonald 2022; Virji 2021)

• Secondary schools (4 studies: Farhadi 2022; Jesser 2022; Özer 2022; Schwal 2022)

• School type not reported (6 studies: Denford 2022; Galasso 2022; Mudaly 2021; Nelson 2021; Obeng 2022; Scott 2021)

Mixed‐methods studies

The 12 mixed‐method studies took place in the following continents and countries.

• North America (5 studies)

  • USA (Ellis 2022; Hilbert 2021; Pressley 2022; Worrell 2023)

  • Canada (Coelho 2022)

• Europe (6 studies)

  • England, UK (Mahmud 2022; Quinn 2022)

  • Spain (Moliner 2022; Moliner and Alegre 2022)

  • Wales, UK (Marchant 2022)

  • The Netherlands (Ding 2023)

• Africa (1 study)

  • Ethiopia (Ahmed 2022)

The types of school in the mixed‐method studies were as follows.

• Combination of school types (primary and secondary) in three studies (Ahmed 2022; Coelho 2022; Worrell 2023)

• Primary schools (4 studies: Ellis 2022; Marchant 2022; Pressley 2022; Quinn 2022)

• Secondary schools (3 studies: Ding 2023; Moliner 2022; Moliner and Alegre 2022)

• Special schools (Mahmud 2022)

• School type not reported (Hilbert 2021)

Measures

Quantitative studies

Individual protection

Seventeen quantitative studies assessed the following measures within the individual protection category.

• Masks (12 studies: Amman 2022; Assathiany 2021; Chadwick 2021; Falk 2022; Khandewal 2022; Li 2021; Mickells 2021; Mori 2022; Mwesige 2021; Schlegtendal 2022; Wilson 2022; Xu 2022)

• Hand hygiene (3 studies: Amin‐Chowdury 2021; Borch 2020; Nabe‐Nielsen 2021)

• Both hand hygiene and messaging (Kebede 2022)

• Masks, hand hygiene, and hygiene relating to coughs and sneezes (Sundaram 2022)

Physical environment

Twelve quantitative studies assessed the following measures within the physical environment category.

• Ventilation (3 studies: Kebede 2022; Mori 2022; Wilson 2022)

• Air purification (Duill 2021; Srikrishna 2022)

• Physically distanced classroom layout (3 studies: Chadwick 2021; Kebede 2022; Sundaram 2022)

• Equipment rules, such as no sharing of equipment (Chadwick 2021; Sundaram 2022)

• Surface cleaning/disinfection (Nabe‐Nielsen 2021; Sundaram 2022)

Social interactions

Fourteen quantitative studies assessed the following measures within the social interactions category.

• Some form of hybrid teaching (8 studies: Goldhaber 2022; Hanno 2022; Henderson 2021; Levitt 2022; McKellar 2023; Simpson 2022; Tsujimoto 2022; Walters 2021)

• Physical distancing rules (4 studies: Chadwick 2021; Falk 2022; Sundaram 2022; Wilson 2022)

• Both hybrid teaching and physical distancing (Amin‐Chowdury 2021)

• Limitation of the number of social contacts (Nabe‐Nielsen 2021).

Services

One quantitative study assessed measures within the services category, which involved the cancellation of activities, such as practical science activities (Chadwick 2021).

Movement

No quantitative studies assessed measures in the movement category.

Surveillance

Five quantitative studies assessed the following measures within the surveillance category.

• Some form of testing (4 studies)

  • Lolli method PCR (Kretschmer 2022)

  • PCR‐based rapid testing (Collier 2022)

  • Pooled PCR testing (Doron 2021)

  • Pooled saliva PCR testing (Mendoza 2021)

• Daily temperature checking (Amin‐Chowdury 2021)

Response

Three quantitative studies assessed the following measures within the response category.

• Isolation as a consequence of surveillance (Collier 2022; Doron 2021)

• Contact tracing (Wilson 2022)

Multi‐component measures

Four quantitative studies assessed the following multi‐component measures.

• Mask wearing, hand hygiene, physically distanced classroom layout, cleaning and disinfection, avoidance of sharing equipment, and physical distancing (Kebede 2022)

• Mask wearing and ventilation (Mori 2022)

• Mask wearing, a physically distanced classroom layout, avoidance of sharing equipment, cancellation of practical activities, and physical distancing (Chadwick 2021)

• Hand hygiene, cleaning and disinfection, a physically distanced classroom layout, and physical distancing (Nabe‐Nielsen 2021)

Qualitative studies

Individual protection

Five qualitative studies assessed the following measures within the individual protection category.

• Masks (4 studies: Kaliyadan 2022; Kavrayic 2021; Schwal 2022; Scott 2021)

• Hand hygiene intervention (Obeng 2022)

Physical environment

Three qualitative studies assessed the following measures within the physical environment category.

• Ventilation (Özer 2022)

• Increased cleaning regimens (Kamoga 2022; Kavrayic 2021)

Social interactions

Six qualitative studies assessed the following measures within the social interactions category.

• Reduction of mixing between classes (Farhadi 2022; Galasso 2022)

• Hybrid teaching (Jesser 2022)

• Physical distancing (Nelson 2021)

• Both hybrid teaching and physical distancing (Scott 2021)

• No shower policy after physical education (Kamoga 2022)

Services

No qualitative studies assessed measures in the services category.

Movement

Two qualitative studies assessed the following measures in the movement category.

• Students not moving from their seats during break times (Galasso 2022)

• Teachers staying in their own classrooms (Farhadi 2022)

Surveillance

Three qualitative studies assessed the following surveillance measures.

• Daily lateral flow testing (Denford 2022)

• Saliva‐based PCR testing (Virji 2021)

• Unspecified form of testing (Kenworthy 2022)

Response

Three qualitative studies assessed the following measures within the response category.

• Isolating teachers infected with SARS‐CoV‐2 (Galasso 2022)

• Isolation (Scott 2021)

• Different test‐to‐stay and quarantine strategies (Corneli 2022)

Multi‐component measures

Five qualitative studies described the following multi‐component measures.

• Mask wearing, physical distancing, and testing (McDonald 2022)

• Mask wearing, hand hygiene, physical distancing, cleaning and disinfection, a physically distanced classroom layout, and temperature checking (Obeng 2022)

• Mask wearing, hand hygiene, cleaning and disinfection, and physical distancing (Mudaly 2021)

• Mask wearing, hand hygiene, physical distancing, ventilation, avoidance of sharing equipment, cohorting, and temperature checking (Schwal 2022)

• Hybrid learning and physical distancing (Jesser 2022)

Mixed‐methods studies

Individual protection

Four mixed‐method studies assessed the following measures within the individual protection category.

• Mask wearing (4 studies: Ahmed 2022; Coelho 2022; Ellis 2022; Marchant 2022)

• Hand hygiene (Ahmed 2022; Ellis 2022).

Physical environment

Five mixed‐method studies assessed the following measures to modify the physical environment.

• Modified seating plan (Ahmed 2022; Moliner and Alegre 2022)

• Improved ventilation (Ahmed 2022; Ding 2023)

• Adaptive design for shared spaces (Quinn 2022)

• Cleaning and disinfection (Hilbert 2021)

Social interactions

Five mixed‐method studies assessed the following measures within the social interactions category.

• Cohorting (Ellis 2022)

• Physical distancing (Ellis 2022; Mahmud 2022)

• Restrictions on sharing materials (Ellis 2022)

• Hybrid teaching (3 studies: Ahmed 2022; Moliner 2022; Pressley 2022).

Services

No mixed‐method studies assessed measures in the services category.

Movement

One mixed‐method study assessed a measure in the movement category, namely restricting the movement of students during lunch breaks (Ellis 2022).

Surveillance

One mixed‐method study assessed a measure within the surveillance category, namely temperature checking at the school gate (Ahmed 2022).

Response

One mixed‐method study assessed a measure within the response category, namely home quarantine versus modified quarantine (attending school with measures in place, such as testing and no extracurricular activities; Worrell 2023).

Multi‐component measures

No mixed‐method studies assessed the combined effect of multi‐component measures.

Unintended consequences

The evidence maps present the clusters of published literature relating to measure‐unintended consequence pairs and highlight relevant gaps in the available evidence (see Figure 4 and Figure 5). The best‐populated categories are health and well‐being, acceptability and adherence, and social and institutional consequences. Within these three categories, consequences are most frequently paired with individual protection measures (e.g. mask wearing) and measures targeting social interactions (e.g. hybrid teaching and physical distancing). Within the acceptability and adherence category, consequences were also frequently paired with measures targeting the physical environment (e.g. ventilation) and surveillance measures (e.g. screening through PCR or rapid antigen tests). Assessing unintended consequences was the primary aim for 26/31 quantitative studies, 16/17 qualitative studies, and 9/12 mixed‐method studies.

4.

Evidence map of unintended consequences of measures implemented in the school setting to contain the COVID‐19 pandemic

5.

Detailed evidence map

Most published literature is derived from quantitative studies, which have assessed consequences for health and well‐being, acceptability and adherence, equality and equity, social and institutional aspects, economic and resource‐related aspects, and ecological aspects. Qualitative studies have mainly investigated unintended consequences in the categories health and well‐being, acceptability and adherence, equality and equity, social and institutional, and economic and resource. Mixed‐method studies were relatively rare and only reported consequences for health and well‐being, acceptability and adherence, and social and institutional aspects. We found no studies that explored consequences for health system and social welfare services or human and fundamental rights. There was sparse evidence within the equality and equity category (2 studies) and ecological category (1 study).

The following section describes the published literature available for distinct measure‐unintended consequences pairs, according to the unintended consequence categories of the logic model (Figure 1). Some studies cover multiple measures and consequences so appear in several sections.

Health and well‐being

Overall, 29 studies addressed consequences for health and well‐being. Within this category, there were 15 quantitative studies (Amman 2022; Assathiany 2021; Borch 2020; Henderson 2021; Levitt 2022; Li 2021; McKellar 2023; Mickells 2021; Mori 2022; Nabe‐Nielsen 2021; Schlegtendal 2022; Simpson 2022; Tsujimoto 2022; Walters 2021; Xu 2022), 10 qualitative studies (Corneli 2022; Farhadi 2022; Galasso 2022; Jesser 2022; Kaliyadan 2022; Kamoga 2022; McDonald 2022; Obeng 2022; Pressley 2022; Scott 2021), and four mixed‐method studies (Coelho 2022; Hilbert 2021; Marchant 2022; Worrell 2023).

Physical health

Ten studies focused on physical health (Amman 2022; Assathiany 2021; Borch 2020; Henderson 2021; Hilbert 2021; Kaliyadan 2022; Kamoga 2022; Levitt 2022; Mickells 2021; Mori 2022). They investigated individual protection, physical environment, social interaction, and multi‐component measures. None addressed modifications to services, movement, surveillance, or response.

The vast majority of studies concerned with individual protection measures explored the physical consequences of mask wearing, such as discomfort (Amman 2022), headaches (Assathiany 2021), breathing difficulties (Assathiany 2021), cutaneous disorders (Assathiany 2021), accidents (Mickells 2021), facial fungal skin infections (Kaliyadan 2022), and vision (Marchant 2022). One study assessed irritant contact dermatitis as a result of hand hygiene (Borch 2020).

In relation to measures modifying the physical environment, one study assessing ventilation measured CO₂ concentrations within classrooms, as well as the thermal sensation within ventilated classrooms with and without mask wearing (Mori 2022). Another study focused on the physical health effects of cleaning and disinfection, including burning eyes, respiratory difficulties, headache, nausea, and skin irritation (Hilbert 2021).

Regarding measures related to social interactions, studies focused on the unintended physical health consequences of hybrid teaching, such as effects on physical fitness (Henderson 2021) and sleep patterns (Levitt 2022). One study investigated how a no‐shower policy may have affected the level of physical activity during physical education lessons (Kamoga 2022).

One study assessed the combined effect of masks and ventilation (multi‐component measure) on discomfort (Mori 2022).

Psychosocial health and well‐being

Twenty‐three studies focused on psychosocial health and well‐being (Assathiany 2021; Coelho 2022; Corneli 2022; Farhadi 2022; Galasso 2022; Henderson 2021; Jesser 2022; Kamoga 2022; Li 2021; Marchant 2022; McKellar 2023; McDonald 2022; Mickells 2021; Nabe‐Nielsen 2021; Obeng 2022; Pressley 2022; Schlegtendal 2022; Scott 2021; Simpson 2022; Tsujimoto 2022; Walters 2021; Worrell 2023; Xu 2022). Most published literature on psychosocial health effects related to measures targeting individual protection, the physical environment, and social interactions. Measures concerned with movement and response were also examined. No studies addressed modifications to services or surveillance.

Of the studies investigating individual protection measures, two evaluated the anxiety levels of teachers (Li 2021) and students (Xu 2022) in relation to mask wearing. Others looked at the effect of mask wearing on students' mental health (Assathiany 2021; Coelho 2022; Marchant 2022; Mickells 2021; Schlegtendal 2022) and teachers' stress levels (Obeng 2022).

In the physical environment category, one study focused on cleaning and disinfection and its effect on the workload of teachers and their subsequent stress (Kamoga 2022).

Among studies concerned with social interactions, one measured mental and emotional well‐being and loneliness of students as a result of physical distancing (Scott 2021), and two measured emotional consequences among students in relation to maintaining a physical distance (Marchant 2022; Simpson 2022). Other studies investigated the effect of hybrid teaching on students' emotional well‐being (Henderson 2021); school‐related stress, fear, and anxiety (Jesser 2022); mental and emotional well‐being (Simpson 2022); connectedness and well‐being (McKellar 2023); and mental health and depression (Tsujimoto 2022; Walters 2021). One study explored the effect of hybrid teaching on teachers, in terms of changes in workload, work‐life balance, and stress (Pressley 2022).

Within the movement category, two studies measured mental and emotional well‐being and loneliness of students as a result of not being able to move from their classroom during breaks (Farhadi 2022; Galasso 2022).

In relation to response measures, one study focused on students' well‐being in relation to different test‐to‐stay strategies (Corneli 2022), two studies measured the perceived effect of isolation on students' well‐being (Galasso 2022; Scott 2021), and one study assessed the effect of teaching staff absences on head teachers' well‐being (Galasso 2022). Another study measured social isolation, anxiety, and frustration among students, as well as parental stress, in relation to test‐to‐stay and quarantine strategies (Worrell 2023).

Three studies assessed multi‐component measures, focusing on students' overall concern about school measures (Jesser 2022), anxiety in relation to school measures in general (McDonald 2022), and general worries among teachers about going to work and fear of infection/transmission in relation to school measures (Nabe‐Nielsen 2021).

Health system and social welfare services

No studies reported unintended consequences related to health system and social welfare services.

Human and fundamental rights

No studies reported unintended consequences related to human and fundamental rights.

Acceptability and adherence

Overall, 31 studies focused on acceptability and adherence. There were 16 quantitative studies (Amin‐Chowdury 2021; Amman 2022; Assathiany 2021; Collier 2022; Doron 2021; Falk 2022; Henderson 2021; Kebede 2022; Kretschmer 2022; Mickells 2021; Mori 2022; Mwesige 2021; Nabe‐Nielsen 2021; Srikrishna 2022; Sundaram 2022; Wilson 2022), 10 qualitative studies (Denford 2022; Galasso 2022; Kavrayic 2021; Kenworthy 2022; McDonald 2022; Mudaly 2021; Obeng 2022; Schwal 2022; Scott 2021; Virji 2021), and five mixed‐method studies (Ahmed 2022; Ding 2023; Mahmud 2022; Marchant 2022; Worrell 2023).

Acceptability

Nineteen studies focused on the acceptability of the measures implemented (Ahmed 2022; Amman 2022; Assathiany 2021; Collier 2022; Denford 2022; Ding 2023; Doron 2021; Galasso 2022; Henderson 2021; Kebede 2022; Kenworthy 2022; Kretschmer 2022; McDonald 2022; Mickells 2021; Schwal 2022; Scott 2021; Srikrishna 2022; Virji 2021; Worrell 2023). These covered measures across the following categories: individual protection, physical environment, social interactions, movement, surveillance, and response. No studies addressed modifications to services.

Studies concerned with individual protection measures assessed consequences such as the general acceptability and perceived usefulness of student masking, according to students (Amman 2022; Scott 2021), staff (Mickells 2021; Schwal 2022), and parents (Assathiany 2021); as well as general acceptability and perceived usefulness of staff masking according to staff (Schwal 2022). Another study assessed the attitudes of students towards student hand hygiene and messaging around interventions (Kebede 2022).

Three studies focused on the physical environment, looking at the acceptability of classroom temperature and CO₂ level as a result of ventilation, measured quantitatively during classroom visits (Ding 2023); staff and students’ perceived acceptability of the avoidance of sharing equipment and cleaning and disinfection (Ahmed 2022); and the acceptability of the noise generated by air purifiers according to teachers (Srikrishna 2022).

Studies that focused on social interactions assessed the acceptability of hybrid teaching for staff and students (Ahmed 2022), acceptability of physical distancing for students (Scott 2021), and acceptability of students staying in class according to adults (Galasso 2022). One study assessed the acceptability of hybrid teaching for parents (Henderson 2021).

Surveillance measures were examined with regard to acceptability of asymptomatic testing for staff (Collier 2022; Doron 2021; Kenworthy 2022; Virji 2021), parents (Doron 2021; Kenworthy 2022; Kretschmer 2022), students (Kenworthy 2022; Virji 2021), and lab workers (Virji 2021). One study reviewed the acceptability of temperature checking for students and staff (Ahmed 2022).

Studies exploring response measures focused on the acceptability of test‐to‐stay strategies for parents (Denford 2022; Worrell 2023), and staff and students (Denford 2022); as well as the acceptability of isolation for staff (Collier 2022; Doron 2021) and parents (Doron 2021).

Two studies assessed the general acceptability of multi‐component school‐based measures according to students (Kebede 2022) and parents and staff members (McDonald 2022).

Adherence and compliance

Twenty‐one studies focused on adherence to and compliance with PHSM within the following categories: individual protection, physical environment, social interactions, surveillance, and response (Ahmed 2022; Amin‐Chowdury 2021; Falk 2022; Kavrayic 2021; Kebede 2022; Kenworthy 2022; Mahmud 2022; Marchant 2022; Mickells 2021; Mori 2022; Mudaly 2021; Mwesige 2021; Nabe‐Nielsen 2021; Obeng 2022; Schwal 2022; Scott 2021; Srikrishna 2022; Sundaram 2022; Virji 2021; Wilson 2022; Worrell 2023). No studies addressed modifications to services or movement.

In relation to individual protection measures, 10 studies explored adherence to mask wearing. One study focused on adherence to mask wearing among students (Schwal 2022), one study focused on staff (Kavrayic 2021), and eight studies measured adherence among both students and staff (Ahmed 2022; Falk 2022; Marchant 2022; Mickells 2021; Mori 2022; Mwesige 2021; Sundaram 2022; Wilson 2022). Two studies looked at school adherence to messaging through posters (Kavrayic 2021; Kebede 2022), whilst others focused on adherence to hand hygiene among students (Ahmed 2022; Amin‐Chowdury 2021; Kebede 2022; Sundaram 2022) and staff (Ahmed 2022; Amin‐Chowdury 2021; Sundaram 2022); and adherence to hygiene measures relating to coughs and sneezes among students and staff (Sundaram 2022).

Among studies that assessed adherence to measures targeting the physical environment, two focused on cleaning and disinfecting among students (Wilson 2022) and staff (Kavrayic 2021; Wilson 2022), six assessed adherence to ventilation (Ahmed 2022; Kebede 2022; Mickells 2021; Mori 2022; Srikrishna 2022; Wilson 2022), one assessed use of air purifiers (Srikrishna 2022), one focused on physically distanced classroom seating plans among students and staff (Ahmed 2022), and one assessed equipment rules such as avoidance of sharing equipment among staff and students (Sundaram 2022).

For social interaction measures, studies investigated adherence to physical distancing among students (Amin‐Chowdury 2021; Falk 2022; Kebede 2022; Mahmud 2022; Marchant 2022; Scott 2021; Sundaram 2022; Wilson 2022) and staff (Amin‐Chowdury 2021; Falk 2022; Mahmud 2022; Marchant 2022; Sundaram 2022; Wilson 2022); student adherence to cohort grouping (Ahmed 2022; Scott 2021) and hybrid teaching (Amin‐Chowdury 2021); and student and staff adherence to physically distanced classroom layout (Sundaram 2022).

Four studies that focused on surveillance measures assessed adherence to asymptomatic testing among students (Kenworthy 2022; Virji 2021) and temperature checking among students (Ahmed 2022; Amin‐Chowdury 2021) and staff (Amin‐Chowdury 2021).

In addition, two studies that focused on response measures assessed adherence to test‐to‐stay strategies among students (Worrell 2023), and adherence to contact tracing among students and staff (Wilson 2022).

Five studies examined adherence to multi‐component measures among students (Kebede 2022; Mudaly 2021; Schwal 2022) and staff (Mudaly 2021; Nabe‐Nielsen 2021; Obeng 2022; Schwal 2022). The combination of measures always included hand hygiene and physical distancing but also covered: ventilation (Kebede 2022; Schwal 2022); mask wearing (Mudaly 2021; Obeng 2022; Schwal 2022); cohorting, temperature checking, and avoidance of sharing equipment (Schwal 2022); cleaning and disinfection (Mudaly 2021; Nabe‐Nielsen 2021); and messaging through posters (Kebede 2022).

Equality and equity

Two studies assessed unintended consequences relating to equality and equity. One study employed a quantitative design (Goldhaber 2022), and one study employed a qualitative design (Kavrayic 2021).

Social and economic equality and equity

Two studies focused on consequences for social and economic equality and equity in the categories of individual protection, physical environment, and social interactions. No studies addressed modifications to services, movement, surveillance, or response.

One study focused on individual protection and physical environment measures, assessing the perceived impact of inequality in school funding levels on head teachers' ability to adhere to masking, messaging, and cleaning and disinfection protocols (Kavrayic 2021).

In relation to social interactions, another study focused on hybrid teaching and assessed how individuals' educational outcomes were disproportionately affected according to their socioeconomic status (Goldhaber 2022).

Social and institutional

Overall, 23 studies assessed social and institutional unintended consequences. There were 10 quantitative studies (Assathiany 2021; Chadwick 2021; Goldhaber 2022; Hanno 2022; Henderson 2021; Khandewal 2022; Levitt 2022; McKellar 2023; Schlegtendal 2022; Walters 2021), seven qualitative studies (Corneli 2022; Jesser 2022; McDonald 2022; Mudaly 2021, Nelson 2021; Obeng 2022; Scott 2021), and six mixed‐method studies (Ellis 2022; Marchant 2022; Moliner 2022; Moliner and Alegre 2022; Quinn 2022; Worrell 2023).

Civil life, sociocultural institutions, and participation

Three studies focused on unintended consequences relating to civil life, sociocultural institutions, and participation in the categories of physical environment and response (Corneli 2022; Quinn 2022; Worrell 2023). No studies addressed individual protection, modifications to services, movement, social interactions, or surveillance.

Considering physical environment measures, one study assessed the adaptive design of shared spaces in relation to the value of play and learning in the organisation (Quinn 2022).

Within the response category, studies looked at test‐to‐stay strategies in relation to organisational burden (Corneli 2022) and activity disruptions, such as extracurricular activities (Worrell 2023).

Social cohesion and social well‐being

Seventeen studies focused on unintended consequences relating to social cohesion and social well‐being in the categories of individual protection, social interactions, and response (Assathiany 2021; Corneli 2022; Ellis 2022; Hanno 2022; Henderson 2021; Khandewal 2022; Jesser 2022; Levitt 2022; McDonald 2022; McKellar 2023; Moliner 2022; Mudaly 2021; Nelson 2021; Obeng 2022; Scott 2021; Walters 2021; Worrell 2023). No studies addressed physical environment, modifications to services, movement, or surveillance.

Within the individual protection category, studies assessed the effects of mask wearing on social consequences such as communication (Assathiany 2021; Khandewal 2022).

Regarding social interactions, several studies evaluated the effect of hybrid teaching on social cohesion and well‐being in terms of behavioural consequences, namely general behaviour (Hanno 2022; Levitt 2022), maladaptive behavioural changes (Hanno 2022), frequency of dysregulated behaviours (Hanno 2022), mood changes (Assathiany 2021), and delinquency or impulsivity and deviance (Walters 2021). Other studies investigated the effect of hybrid teaching on social relationships (Henderson 2021); connectedness and social well‐being (McKellar 2023); communication, including speaking difficulties (Assathiany 2021); parental knowledge and support (Walters 2021); and bullying victimisation and perpetration (Walters 2021). Two studies looked at the perceived effect of hybrid teaching on educational outcomes (Jesser 2022; Moliner 2022). Three studies examined the effect of physical distancing on teaching and learning quality (Nelson 2021), educational outcomes (Jesser 2022), and the feasibility of implementing play‐based learning (Ellis 2022). Cohort grouping and staying in the classroom were also assessed in relation to the feasibility of implementing play‐based learning (Ellis 2022).

Considering response measures, two studies investigated the effect of test‐to‐stay strategies on educational outcomes (Corneli 2022; Worrell 2023), one assessed the perceived effects of test‐to‐stay strategies on isolation and learning (Scott 2021), and one evaluated the effect of the length of quarantine on educational outcomes (Corneli 2022).

In relation to multi‐component measures, one study assessed the combined effect of mask wearing, physical distancing, restricted entry for parents and caregivers, testing, isolation, and cohort grouping on attendance and schools' relationship with parents and caregivers (McDonald 2022). Another study assessed the combined effect of mask wearing, physical distancing, hand hygiene, and cleaning and disinfection on participation in practical activities in classes (Mudaly 2021). A third multi‐component study assessed the combined effect of masks and physical distancing on teachers' interactions with students (Obeng 2022).

Education and development

Ten studies focused on the educational impact of measures in the categories of individual protection, physical environment, social interactions, and movement (Chadwick 2021; Ellis 2022; Goldhaber 2022; Henderson 2021; Khandewal 2022; Marchant 2022; Moliner 2022; Moliner and Alegre 2022; Mudaly 2021; Schlegtendal 2022). No studies assessed the consequences on education and development of measures related to services, surveillance, or response.

Among studies investigating individual protection measures, two assessed the impact of masks on educational outcomes (Marchant 2022) and cognitive performance in terms of attention and executive functions (Schlegtendal 2022). Three studies focused on the impact of masks on the classroom, specifically the feasibility of implementing play‐based learning (Ellis 2022) and teaching skills/classroom management (Chadwick 2021; Khandewal 2022). The perceived effect of hand hygiene was also explored in relation to the feasibility of implementing play‐based learning (Ellis 2022).

In relation to the physical environment, one study investigated the feasibility of implementing play‐based learning in classrooms where students could not share equipment (Ellis 2022). Two studies assessed educational outcomes in relation to modified classroom seating plans (Moliner 2022; Moliner and Alegre 2022).

For social interactions, three studies focused on the effect of hybrid teaching on educational outcomes and skills (Goldhaber 2022; Henderson 2021, Moliner 2022). Another study looked at the effect of physical distancing alongside cohorting on the feasibility of play‐based learning (Ellis 2022).

In relation to movement, one study assessed the effect of not moving from the classroom on the ability to implement play‐based learning (Ellis 2022).

Two studies assessed the effect of multi‐component measures on practical learning. The measures consisted of: masks, physical distancing, a physically distanced classroom layout, cancelling activities, and avoidance of sharing equipment (Chadwick 2021); and mask wearing, physical distancing, hand hygiene, and cleaning and disinfection (Mudaly 2021).

Social norms, values, and practices

One study focused on social norms, values, and practices in the category of individual protection (Obeng 2022). It assessed hand hygiene measures in relation to awareness of hygiene behaviour in children.

Economic and resource

Overall, seven studies assessed unintended consequences relating to economic aspects and resources. There were three quantitative studies (Doron 2021; Falk 2022; Mendoza 2021), four qualitative studies (Galasso 2022; Kamoga 2022; Obeng 2022; Özer 2022), and no mixed‐method studies.

Financial

In relation to financial consequences, three studies looked at surveillance and one at multi‐component measures. No studies addressed modifications to services, movement, or response.

Three studies focused on costs of surveillance testing (Doron 2021; Falk 2022) and costs averted due to pooled testing strategies (Mendoza 2021).

Another study assessed the consequences of a multi‐component measure (hand hygiene, masking, cleaning and disinfection, and physical distancing) on school financial challenges and teachers' salaries (Obeng 2022).

Non‐financial resources

Five studies looked at consequences for non‐financial resources in the categories of individual protection, physical environment, surveillance, and response. No studies addressed social interactions, modifications to services, or movement.

One study focused on individual protection, assessing teachers' workload in relation to hand hygiene, mask wearing, physical distancing, and cleaning and disinfection (Obeng 2022).

In relation to measures in the physical environment, studies focused on teachers' workload due to cleaning/disinfection (Kamoga 2022) and new staff recruited for ventilation and cleaning (Özer 2022).

One study focused on surveillance, assessing the additional working hours required for surveillance testing (Falk 2022), whilst another study assessed staff absence due to isolation (Galasso 2022).

Ecological

One quantitative study assessed ecological consequences (Duill 2021).

Noise level

One study focused on ecological consequences in the physical environment category, specifically the noise level generated by air purifiers (Duill 2021).

Discussion

Summary of published literature

In this scoping review, we mapped the published literature on the unintended consequences of PHSM implemented in the school setting to reduce the spread of SARS‐CoV‐2. We identified 60 studies across 25 countries that provided empirical insights on at least one unintended consequence. The original review identified 18 studies [11].

Consequences identified in this update were mostly reported for students (26 studies), teachers and school staff (11 studies), and students and school staff (12 studies). Other studies reported consequences for parents (2 studies), staff and parents (1 study), students and teachers (3 studies), or the whole school (5 studies).

The included studies examined a broad range of measures, including individual protection (26 studies, mainly focused on mask wearing), changes to the physical environment (20 studies, mainly focused on ventilation and a physically distanced classroom layout), social interactions (25 studies, mainly focused on hybrid teaching), services (1 study involving cancellation of activities), movement (3 studies, mainly focused on not moving from classrooms during break times), surveillance (9 studies, mainly focused on asymptomatic testing), response (7 studies, mainly focused on isolation and test‐to‐stay strategies to replace quarantine), and multi‐component measures (9 studies, mainly focused on mask wearing, hand hygiene, and physical distancing).

The studies assessed consequences for health and well‐being (29 studies), acceptability and adherence (31 studies), equality and equity (2 studies), social and institutional aspects (23 studies), economic and resource‐related aspects (7 studies), and ecological aspects (1 study). No studies assessed consequences for health system and social welfare services or for human and fundamental rights. For most studies, assessing unintended consequences was a primary aim. Unintended consequences, whether assessed quantitatively or qualitatively, were mostly presented as negative/harmful (e.g. increased teacher workload due to hand hygiene (Obeng 2022) and anxiety due to a positive test (Corneli 2022)), although some positive consequences were also reported (e.g. increased health and safety with hand hygiene as the new normal (Obeng 2022) and reassurance provided by testing (Corneli 2022)).

This review builds on the findings of the previous scoping review of unintended consequences [11], as well as on reviews that have assessed the effectiveness of school measures during the COVID‐19 pandemic [10, 6, 13]. Together, these reviews describe published literature that should inform and facilitate decisions about whether, how, and when to implement or terminate COVID‐19 risk mitigation measures in school settings. Balancing the likely benefits and possible harms of such measures from a broad societal perspective is important.

We conducted the search for this update 22 months after the original search [11]. We included 31 quantitative studies, 17 qualitative studies, and 12 mixed‐method studies. Similarly, the original review included mostly quantitative studies (13/18 included studies), although five of these were modelling studies, which were not eligible for inclusion in this update. Due to changes in inclusion criteria, only three studies were included in both the original and updated review.

As in the original review, most studies included in this update focused on directly affected populations within schools (e.g. students, teachers, and school staff), and only a few focused on indirectly affected populations (e.g. parents or caregivers, families, and the local community). Compared with the original review, this update identified a wider range of empirical published literature in terms of types of measures, outcomes, and geographical locations. Notably, 22 studies investigated the unintended consequences of measures to reduce contacts (Ahmed 2022; Amin‐Chowdury 2021; Collier 2022; Doron 2021; Ellis 2022; Farhadi 2022; Galasso 2022; Goldhaber 2022; Hanno 2022; Henderson 2021; Jesser 2022; Levitt 2022; McKellar 2023; Moliner 2022; Nabe‐Nielsen 2021; Pressley 2022; Schwal 2022; Scott 2021; Simpson 2022; Tsujimoto 2022; Walters 2021; Wilson 2022), compared with only four studies in the original review [11]. Both versions of this scoping review included many studies relating to measures aimed at making contacts safer [11]. As in the original review, most studies included in this update were concerned with educational (social) or health consequences; in addition, several studies reported acceptability and adherence and economic consequences. Whilst the original review included only one study outside Europe and North America, this update covers six continents. In addition, the 2023 search returned more eligible studies commenting on context or implementation factors. However, even where context was mentioned, it was rarely addressed in depth, making it difficult to explore how differences across countries and regions may have influenced the unintended consequences of measures or the pathways leading to these. Furthermore, most studies either did not report the duration of the intervention, or the reporting was unclear. Future studies should clearly report this information to facilitate decision making around implementation of such measures.

We identified gaps in the published literature regarding consequences for the health system and social welfare services, and for human and fundamental rights. In addition, very few studies examined equality and equity or ecological consequences. Whilst all remaining categories of unintended consequences were populated with published literature, we identified no studies for several sub‐categories. No studies assessed health behaviour (within the health and well‐being category), health‐related equality and equity (within the equality and equity category), safety, security, and crime and conditions of daily living (within the social and institutional category), or energy consumption and greenhouse gas emissions (within the ecological category). In some cases, these gaps in the published literature are likely due to lack of research rather than absence of unintended consequences. For example, mask wearing does have ecological consequences in terms of waste generation and the production and transportation of materials. On the other hand, some of the measures identified in this review are unlikely to have human rights or equity consequences. For example, a physically distanced classroom layout or seating plan will most likely have no implications for equality and equity. We also identified an emergent outcome sub‐category within the ecological category (noise level), which was assessed as an unintended consequence of air purifiers in one study (Duill 2021). This consequence warrants further investigation as it could impact upon outcomes such as concentration and communication in class, as well as the acceptability of the measure. We found no published literature that explored the impact of COVID‐19 containment measures in school settings on extremely vulnerable children or households, although we appreciate that school closures and measures to reduce the number of contacts in schools might have negatively impacted this population disproportionately.

Regarding gaps in the published literature related to measures, all categories of measures to make contacts safer, measures to reduce contacts, surveillance and response measures, and multi‐component measures were populated with published literature. However, within the categories of surveillance and response measures, few studies investigated unintended consequences related to provision of messages, air purification, temperature checking, contact tracing, and quarantine. The reasons for these gaps are unclear, but one explanation may be the greater prevalence of measures such as mask wearing and hand hygiene, as well as their immediately apparent unintended consequences, such as hindered communication and skin problems. Surprisingly, whilst many studies have focused on the reduction in school attendance associated with measures such as quarantine [88, 89], the potential unintended consequences of this, such as worse educational outcomes, have largely been assumed rather than empirically assessed. However, evaluation of some consequences, including the impact of various measures on educational outcomes, may require a longitudinal study design, and it is possible these studies have yet to be published.

Strengths and limitations

This review has several methodological strengths. Whilst only quantitative studies were eligible for inclusion in the effectiveness reviews [6, 13], this scoping review also included qualitative studies, which may be better for identifying and describing emergent unintended consequences. Mixed‐methods studies, also included in this scoping review, may be particularly helpful for capturing unintended consequences from multiple perspectives and at different points in time.

We used a logic model to inform our eligibility criteria and search strategy, and followed rigorous methods to search for, identify, and summarise the available published literature. We deliberately took a holistic view of unintended consequences, considering consequences across multiple population groups and sectors, regardless of whether these might be reported as positive/beneficial or negative/harmful. A multi‐pronged search strategy, developed and implemented with the support of an information specialist, ensured that we searched databases covering a range of disciplines and thus consequences beyond health. This was supplemented with forward and backward citation tracking. The review team carried out rigorous screening and data extraction procedures, where regular meetings and calibration exercises took place to ensure consistency across screening decisions and to address queries and uncertainties early. All screening was done in duplicate, and any discrepancies were resolved through discussion with a senior author. Both lead authors checked the extracted data to ensure consistency.

However, it may be that our methodological approach did not fully capture the types of studies where gaps were observed in the published literature. For example, although we searched COVID‐19‐specific databases, we did not separately search environmental studies databases, so could have missed studies concerned with ecological consequences (e.g. waste generation due to masking policies). Similarly, isolation, quarantine and some other measures reduce individual freedom and thus impede a fundamental human right. However, we found no published literature on this aspect, possibly because we searched no legal literature databases.

This update built upon the original scoping review and therefore had a guide for which types of consequences to include. However, we adapted some eligibility criteria in line with our understanding of the school setting and of the likely impact of school measures. For example, some studies evaluated perceived or anticipated consequences prior to school reopening, rather than consequences that actually occurred (e.g. [90]). The original review included these studies and viewed the perceived or anticipated consequences as outcomes [11], whereas this update excluded them and focused on measured or experienced consequences. In some cases, there were uncertainties regarding whether consequences should be treated as intended or unintended, notably in relation to the number of school days lost. Our data extraction form included a field for explaining whether study authors had described the pathway by which the measure may have led to the consequence in question. This approach helped to finalise judgements when the classification of consequences as intended or unintended was unclear.

Other challenges included the implementation of multiple measures in many studies, which made it difficult to ascertain whether unintended consequences were due to one particular measure or a combination of measures. Additionally, many quantitative studies were cross‐sectional; very few employed a controlled analysis. This means we cannot rule out the effect of extraneous variables, such as local or regional COVID‐19 incidence and other control measures implemented in the community.

Previous reviews on the effectiveness of school measures identified many modelling studies, which were warranted in the early stages of the pandemic because there were few empirical data available [10, 6]. Since then, the volume and nature of research on the topic has changed considerably, and there is now a large body of empirical published literature studying measures implemented in the school setting, both in terms of their effectiveness [6, 13], and their unintended consequences. For these reasons, and due to the limited real‐world relevance of modelling studies, we decided to exclude this study design and focus on empirical literature in the review update. Nonetheless, modelling is used widely in disciplines such as economics, so we may have missed some relevant studies [91].

Another key challenge of this review was the lack of data on long‐term unintended consequences. Indeed, many of the long‐term consequences of measures implemented in the school setting, such as impaired mental health, may not be evident for several years. The published literature has continued to evolve over the course of the review timeline, and new studies may have been published since the last searches were run in January 2023.

Because this is a scoping review, it lacks the depth of a full systematic review, which would assess the impact of unintended consequences through a quantitative analysis or some form of qualitative evidence synthesis. However, this scoping review does provide an up‐to‐date overview of the available published literature on the wide‐ranging, cross‐sectoral unintended consequences of school measures. It also provides an evidence map, showing significant gaps in published literature relating to different types of unintended consequences (e.g. human and fundamental rights, ecological consequences) as well as relating to different types of measures (e.g. air purification, cancellation of activities).

It is important to evaluate the unintended consequences of school measures from a holistic whole‐of‐society perspective. Future research should seek to fill the above‐described gaps and explore unintended consequences already identified in more depth, over longer periods of time, across different population groups (age, sex, socioeconomic status), and across different contexts and geographical regions. Importantly, this research agenda should seek to identify all consequences, whether positive/beneficial or negative/harmful. Regarding potential harms, their severity can vary and may also disproportionally affect sub‐groups of the population. For example, unintended consequences of mask wearing range from discomfort to more severe outcomes such as breathing difficulties and fungal skin infections, with schools located in areas of lower socioeconomic status reporting more difficulties with adherence to such measures.

Authors' conclusions

This scoping review provides an overview of the evidence on the unintended consequences of public health and social measures implemented in the school setting to reduce the spread of severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). The 60 included studies described a range of different school measures and unintended consequences, mainly related to health and well‐being, acceptability of and adherence to measures, social and institutional aspects, and economic aspects.

Because the influence of the COVID‐19 pandemic on daily life has subsided, and because SARS‐CoV‐2 containment measures in the school setting have evolved and adapted, the utility of this review lies mainly in the identification of gaps in the published literature. In particular, there was a lack of research on the consequences of school measures for health system and social welfare services, human and fundamental rights, equality and equity, and ecological aspects.

Further studies are needed to explore these potential unintended consequences in relation to different types of measures and across different population groups. Investigators should make use of diverse methodological approaches to fill the gaps in the literature and, in particular, to provide evidence on how context influences the unintended consequences of school measures. This is especially important in low‐ and middle‐income countries and other regions where evidence is scarce. Studies assessing unintended consequences should focus on long‐term follow‐up of participants to reflect changes over time. This will be particularly important for social and educational consequences in younger cohorts of students whose early school years have been dominated by SARS‐CoV‐2 containment measures. Researchers could also consider exploratory methods to identify unanticipated consequences. This aligns with complex systems thinking, which states that emergent consequences can be unintended, non‐linear, and long‐term [18].

Overall, further comprehensive, high‐quality primary research will ultimately contribute to a more robust evidence base. This can facilitate decision‐making that goes well beyond whether to implement school measures during an epidemic or pandemic. It can shed light on when and how such measures should be implemented, and which further measures might be taken to counter anticipated unintended consequences.

Supporting Information

Supplementary materials are available with the online version of this article: 10.1002/14651858.CD015397.

Supplementary materials are published alongside the article and contain additional data and information that support or enhance the article. Supplementary materials may not be subject to the same editorial scrutiny as the content of the article and Cochrane has not copyedited, typeset or proofread these materials. The material in these sections has been supplied by the author(s) for publication under a Licence for Publication and the author(s) are solely responsible for the material. Cochrane accordingly gives no representations or warranties of any kind in relation to, and accepts no liability for any reliance on or use of, such material.

Supplementary material 1 Search strategies

Supplementary material 2 Characteristics of included studies

Supplementary material 3 Characteristics of excluded studies

Supplementary material 4 Data package

Supplementary material 5 Preferred Reporting Items for Systematic reviews and Meta‐Analyses extension for Scoping Reviews (PRISMA‐ScR) Checklist

Supplementary material 6 Data Extraction Categories

New search for studies and content updated (no change to conclusions)

Additional information

Contributions of authors

HL, SK, ER: defined the study scope and developed the study protocol with significant intellectual input from all review authors.
HL and SK: co‐ordinated the entire study process.
ANS: contributed to the protocol, designed the search strategies, and conducted the search.
HL, SK, EO, ANS, JH, TH, BS, CJS: conducted title and abstract screening.
HL, SK, EO, ANS, JH, TH: conducted full‐text screening.
SK, HL, JH, and EO: extracted data.
HL and SK: conducted the mapping.
HL, SK and ER: drafted the manuscript.
ER and URS: obtained funding.
All review authors read, revised, and approved the manuscript.

Declarations of interest

HL declares no conflict of interest.

SK declares no conflict of interest.

JH declares no conflict of interest.

EO declares no conflict of interest.

TH is a co‐author of the following included paper: Kretschmer AC, Junker L, Dewald F, Linne V, Hennen L, Horemheb‐Rubio G, et al. Implementing the Lolli‐Method and pooled RT‐qPCR testing for SARS‐CoV‐2 surveillance in schools: a pilot project. Infection 2022;51(2):459‐64. DOI: 10.1007/s15010‐022‐01865‐0. TH was not involved with the screening or data extraction of this study. TH declares first authorship of an article related to the topic of this review, but with a different focus: Heinsohn T, Lange B, Vanella P, Rodiah I, Glöckner S, Joachim A, et al. Infection and transmission risks of COVID‐19 in schools and their contribution to population infections in Germany: a retrospective observational study using nationwide and regional health and education agency notification data. PLOS Med. 2022;19(12):e1003913. DOI: 10.1371/journal.pmed.1003913.

AHN is a Cochrane editor but was not involved in the editorial process for this review.

BS declares being part of the scientific secretariat that supports the development of a living interdisciplinary, evidence‐based, and consensus‐based guideline on measures to prevent and control SARS‐CoV‐2 transmission in schools, registered with the Association of the Scientific Medical Societies (AWMF) in Germany (register.awmf.org/de/leitlinien/detail/027‐076). BS is involved in an ongoing study focusing on the implementation of school‐related measures to contain the COVID‐19 pandemic in the German state of Bavaria (INSIDE: ImplementieruNg Schulmaβnahmen covID‐19 bayErn: "Corona Forschungsprogramm 2021/22" – Corona research program 2021/22). BS declares having the following grants/contracts: CEOsys (9/1/2020‐6/30/2021) funded by the German Federal Ministry of Education and Research (9/1/2020‐6/30/21). INSIDE is in the early stages of being conducted and therefore was not screened as part of the review process. After completion, this study would likely be eligible for inclusion in the scoping review.

CJS declares involvement in the conduct of COVID Kids Bavaria Project funded by the Bavarian Ministry of Research and Art (7/1/2020‐12/31/2021). CJS declares having the following grants/contracts: COVERChild project (7/1/2022‐12/31/2023) funded by the German Federal Ministry of Education and Research.

URS declares being part of the scientific secretariat that supports the development of a living interdisciplinary, evidence‐based, and consensus‐based guideline on measures to prevent and control SARS‐CoV‐2 transmission in schools, registered with the Association of the Scientific Medical Societies (AWMF) in Germany (register.awmf.org/de/leitlinien/detail/027‐076). URS declares having the following grants/contracts: COVERChild project (7/1/2022‐12/31/2023) funded by the German Federal Ministry of Education and Research.

ER declares being a founding member and methods editor for Cochrane Public Health Europe, but was not involved in the editorial process for the current review. ER declares being a member of the scientific advisory board of the Robert Koch Institute that has issued guidance on schooling during COVID‐19, but not having been involved with developing this guidance; and being a member of the World Health Organization (WHO) Regional Office for Europe's Technical Advisory Group on Schooling during COVID‐19 and, in this role, having been involved with advising the WHO Regional Office for Europe on the issue. ER declares being part of the scientific secretariat that supports the development of a living interdisciplinary, evidence‐based, and consensus‐based guideline on measures to prevent and control SARS‐CoV‐2 transmission in schools, registered with the Association of the Scientific Medical Societies (AWMF) in Germany (register.awmf.org/de/leitlinien/detail/027‐076). ER is involved in an ongoing study focusing on the implementation of school‐related measures to contain the COVID‐19 pandemic in the German state of Bavaria (INSIDE: ImplementieruNg Schulmaβnahmen covID‐19 bayErn: "Corona Forschungsprogramm 2021/22" – Corona research program 2021/22). This work is funded by the Bavarian Ministry of Health. ER declares having the following grants/contracts: COVID Kids Bavaria Project funded by the Bavarian Ministry of Research and Art (7/1/2020‐12/31/2021), as well as the PREPARED project (1/1/2023‐12/31/2023), COVERChild project (7/1/2022‐12/31/2023), and CEOsys project (9/1/2020‐6/30/2021), all funded by the German Federal Ministry of Education and Research. INSIDE and COVID Kids Bavaria were in the early stages of being conducted and therefore were not screened as part of the review process. After completion, these studies would likely be eligible for inclusion in the review.

Sources of support

Internal sources

• No sources of support provided, Other

  N/A

External sources

• Ministry of Education and Research, Germany, Other

  This publication was partially funded by the German Federal Ministry of Education and Research (BMBF) within the Network of University Medicine (NUM) 1.0, Grant No. 01KX2021 in the context of the project CEOsys, and NUM 2.0, Grant No. 01KX2121 in the context of the projects PREPARED and coverCHILD.

Registration and protocol

osf.io/bsxh8

Data, code and other materials

See Cochrane's editorial policy on data sharing for more information and what is automatically shared through a published Cochrane review.

As part of the published Cochrane review, the following is made available for download for users of the Cochrane Library.

• Full search strategies for each database

• Full citations of each unique report for all studies included, ongoing, awaiting classification, or excluded at the full text screen, in the final review

• Study data, including study information and study results

• Analysis data, including settings

Appropriate permissions have been obtained for such use. Analyses and data management were conducted with Cochrane's authoring tool, RevMan, using the inbuilt computation methods [92]. Template data extraction forms from Covidence and Excel are available from the review authors on reasonable request.

What's new

Date	Event	Description
12 December 2024	New citation required but conclusions have not changed	This is an update of a previous review [11]
12 December 2024	New search has been performed	In this new review, the inclusion criteria have been updated, there have been authorship changes, the literature was searched up until January 2023, and a more advanced logic model has been used.

History

Review first published: Issue 6, 2022