Primary prevention


  • The World Health Organization (WHO) has authorised the use of the following monovalent vaccines (based on the original wild-type virus) for global use:

    • mRNA vaccines: Comirnaty® (Pfizer/BioNTech); Spikevax® (Moderna)

    • Adenovirus vector vaccines: Vaxzevria® (AstraZeneca); Jcovden® (Janssen); Convidecia® (Cansino)

    • Protein subunit vaccines: Nuvaxoid® (Novavax); Covovax® (Serum Institute of India)

    • Inactivated virus vaccines: Covilo® (Sinopharm); CoronaVac® (Sinovac); VLA2001 (Valneva)

    • WHO: COVID-19 vaccines technical documents Opens in new window

  • Other monovalent vaccines may be available in some countries, for example:

    • VidPrevtyn Beta®: based on the SARS-CoV-2 Beta variant

    • Bimervax®: based on a protein that consists of part of the SARS-CoV-2 spike protein from the Alpha and Beta variants.

  • Second-generation, bivalent mRNA vaccines are available and are recommended as part of immunisation programmes.

    • Bivalent vaccines target two SARS-CoV-2 variants, the original wild-type virus and the Omicron variant (either the BA.1 subvariant or the BA.4/BA.5 subvariants).

  • In some countries, monovalent vaccines may no longer be authorised for use due to the availability of bivalent vaccines. Vaccine availability differs between countries and you should consult your local public health authority for more information.

  • WHO: draft landscape and tracker of COVID-19 candidate vaccines Opens in new window

Vaccines: efficacy

  • Vaccine efficacy for monovalent vaccines depends on the vaccine used, the predominant circulating severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variant, and time since vaccination.

    • Initial authorization of vaccines was based on interim analyses of ongoing phase 3 clinical trials with a median follow-up of 2 months. Overall initial vaccine efficacy for preventing symptomatic infection was reported as 95% (Pfizer/BioNTech), 94.1% (Moderna), 74% (AstraZeneca), and 66.9% (Janssen).[366][367][368][369]

    • Observational evidence from the initial global vaccine rollout suggested real-world efficacy in reducing the rate of symptomatic or asymptomatic infection, disease severity, hospitalisation, death, and possibly even reinfection. However, evidence indicated a minimal to modest reduction of vaccine protection against severe disease over the 6 months after the primary series, while waning efficacy against all clinical disease and infection was more pronounced. Vaccine efficacy against severe disease decreased by about 8% over a 6-month period in all age groups (10% in those aged >50 years), and vaccine efficacy against symptomatic disease decreased by 32% in those aged >50 years.[370] Waning of immunity after vaccination began as early as the first month and continued to decline until the sixth month, where the level of immunity may not have provided adequate protection.[371]

    • Efficacy was highest for the Alpha variant, with lower efficacy reported for Beta, Gamma, and Delta variants.[372] Efficacy against severe outcomes (e.g., hospitalisation) were lower for the original Omicron variant compared with Delta, but mostly remained >50% after the primary series and improved with a booster dose to >80%.[373]

    • A systematic review and meta-analysis found that the pooled vaccine efficacy against any Omicron infection was 20.4% (23.4% against symptomatic infection) after the primary vaccination series, and this decreased to 4% (10.6% against symptomatic infection) within 6 months. The waning was less pronounced against severe disease (decreasing from 64% at 3 months to 49% after 6 months). A booster dose restored protection for all outcomes (51% for any infection, 57% for symptomatic infection, 86% for severe disease), and this waned to 33% at 6 months. This varied depending on the age group and vaccine type.[374]​ Some immune evasion/escape has been observed in the context of currently circulating Omicron subvariants (e.g., BA.2.75, BA.4, BA.5, XBB).[375][376][377]​​​

    • A Cochrane review found that most vaccines reduce (or likely reduce) the proportion of people with symptomatic disease. There is high-certainty evidence that some vaccines may reduce severe or critical disease. There is insufficient evidence to determine whether there was a reduction in mortality compared with placebo. Most studies assessed the vaccine for a short period after injection (2 months); therefore, it is unclear whether protection wanes over time. Most studies were conducted before the emergence of variants of concern.[378]

  • Evidence for the use of bivalent mRNA vaccines is limited.

    • Studies indicate that BA.1-based bivalent vaccines produce a marginally higher immune response against Omicron variants compared with the original wild-type vaccine.[379] However, the clinical relevance of these small differences is uncertain, and there is a lack of published safety and efficacy data available.[380][381] There are no human data available for the BA.4/BA.5 vaccine, and approval was based on studies in mice.[382]

    • Peak serum neutralising antibody titres against SARS-CoV-2 variants following a bivalent vaccine booster have been found to be similar to the peak titres following a monovalent booster.[383]

    • A retrospective cohort study in Israel found that adults aged ≥65 years who were eligible and received a bivalent booster vaccine had lower rates of hospitalisation compared with those who did not receive a bivalent booster for up to 120 days after vaccination.[384]

Vaccines: dose schedules

  • Administer the vaccination series according to local public health authority recommendations.

    • The schedule generally consists of a primary vaccination series (additional doses may be recommended for moderately to severely immunocompromised people), and booster doses in certain populations. Heterologous vaccination schedules may be recommended.

    • The WHO currently recommends a primary and booster vaccination series based on priority groups (high, medium, low) that reflect the risk of severe disease and death. This approach takes into account the impact of the Omicron variant and high population-level immunity due to infection and/or vaccination.[385]

  • Consult your local immunisation schedule for detailed information on choice of vaccine, dose schedule, contraindications, warnings, and cautions. Vaccines may now be included in routine immunisation schedules in some countries.

Vaccines: special patient populations

  • Pregnancy

    • There are limited safety and efficacy data available in pregnant women. However, available data do not support an increased risk of adverse outcomes following vaccination in pregnancy.[386]

    • Systematic reviews and meta-analyses have found no evidence of a higher risk of adverse outcomes in pregnant women including miscarriage, earlier gestation at birth, placental abruption, pulmonary embolism, postnatal haemorrhage, maternal death, intensive care unit admission, lower birthweight Z-score, Apgar score ≤7 at 5 minutes, or neonatal intensive care unit admission with mRNA vaccines.[387][388][389][390]​​​​​ However, the evidence is of low certainty, the data have limitations, and continued monitoring is needed to further assess the risk.[391]​ Additional data on vaccination early in pregnancy, non-mRNA vaccines, and longer-term infant outcomes are needed.

    • Observational evidence suggests that vaccination during pregnancy may protect the infant against infection during the first 4 months of life, and reduce the risk of hospitalisation among infants <6 months of age.[392][393]​ However, the optimal timing of vaccination in pregnancy for neonatal benefit remains uncertain and further research is required.[386]

  • Breastfeeding

    • There are limited safety and efficacy data available in breastfeeding women. Vaccine-derived mRNA has been detected in the breast milk of women who received vaccination within 6 months of delivery. The implications of this are currently unknown, and further research is required.[394]

    • Studies have found robust secretion of SARS-CoV-2 specific immunoglobulin A (IgA) and IgG antibodies in breast milk after vaccination.[395] However, it is unclear how long antibodies persist in the breast milk after vaccination, and their impact on the prevention of infection in infants is also unclear. 

  • Children and adolescents

    • Available evidence suggests that the safety and efficacy of vaccines are acceptable in children and adolescents. Older children and adolescents were at significantly increased risk of adverse reactions after vaccination compared with younger children. There is a need for additional multicentre, large-sample studies and long-term follow-up data.[396]

    • Due to the limited number of children included in the original clinical trials, studies could not have detected rare adverse effects such as myocarditis. However, safety monitoring of the Vaccine Adverse Event Reporting System (VAERS) noted over 9000 reports of adverse events post-vaccination in adolescents aged 12 to 17 years (as of 16 July 2021), 9.3% of which were for serious adverse events including myocarditis (4.3%).[397] Preliminary real world data has not picked up an increased risk of myocarditis in children aged 5 to 11 years as yet.[398][399]​ 

Vaccines: breakthrough infections

  • Breakthrough infections that may potentially result in hospitalisation or death are possible after vaccination.

    • One observational study found that 46% of fully vaccinated people with breakthrough infection were asymptomatic, while 26% had severe or critical disease, 20% had moderate disease, and 7% had mild disease.[400] In another study, the rate of severe disease or death per 1000 person-days was 4.08 among those with breakthrough infections and 3.6 among unvaccinated matched controls with infection.[401]

    • One systematic review and meta-analysis found that there were no statistically significant differences in the risk of hospitalisation, invasive mechanical ventilation, or mortality between unvaccinated people and fully vaccinated people with breakthrough infections (during the Delta variant-dominant period). However, unvaccinated people showed an increased need for oxygen supplementation. There was a limited number of studies included in the meta-analysis and a high level of heterogeneity across studies; therefore, these results should be interpreted with caution. Further prospective studies that adjust for the baseline characteristics of patients are necessary to evaluate vaccine efficacy more precisely.[402]

    • While breakthrough infections with the Omicron variant are more frequent compared with the Delta variant, hospital admissions were less frequent with Omicron.[403][404]

  • Vaccinated people should be considered a possible source of transmission and continue to follow local public health recommendations.

    • Limited evidence suggests that fully vaccinated people with breakthrough infections have similar viral loads compared with unvaccinated people, and therefore may be equally likely to transmit the infection, including to fully vaccinated contacts.[405][406]

    • Secondary attack rates among household contacts exposed to fully vaccinated index cases were similar to household contacts exposed to unvaccinated index cases (25% for vaccinated versus 23% for unvaccinated).[407]

Vaccines: adverse events

  • Consult the prescribing information for detailed information about the adverse events associated with a specific vaccine. Adverse events may vary depending on the type of vaccine used and include, but are not limited to, the following:

    • Common: injection-site reactions, fatigue, headache, myalgia, arthralgia, chills, fever, rash, nausea/vomiting, and diarrhoea; these are usually mild or moderate, and generally resolve a few days after vaccination

    • Uncommon: lymphadenopathy, hypersensitivity reactions including anaphylaxis, hyperhidrosis, night sweats, insomnia, dizziness, lethargy, asthenia, malaise, abdominal pain, pain in vaccinated arm, extensive swelling of vaccinated limb, heavy menstrual bleeding, tinnitus, tremor

    • Rare: transverse myelitis (mainly adenovirus-vector vaccines), Guillain-Barre syndrome (mainly adenovirus-vector vaccines), acute peripheral facial paralysis, myocarditis/pericarditis, vaccine-induced immune thrombocytopenia and thrombosis and other thromboembolic events (mainly adenovirus-vector vaccines), immune thrombocytopenia (mainly adenovirus-vector vaccines), paraesthesia/hypoaesthesia, erythema multiforme, extensive swelling of the vaccinated limb, facial swelling (in people with a history of dermatological fillers) 

    • For information on the diagnosis and management of myocarditis/pericarditis and vaccine-induced immune thrombocytopenia and thrombosis, see Complications.

  • Serious adverse events, including fatal adverse events, have been reported in clinical trials, case reports, case studies, and observational studies (e.g., neurological disorders, cutaneous manifestations, varicella zoster virus reactivation, autoimmune disorders, postural orthostatic tachycardia syndrome, myocardial infarction, ocular vascular events).[408][409][410][411][412][413][414]​​​​ However, a causal link may not have been confirmed for some. A detailed discussion of adverse events is beyond the scope of this topic.

    • A secondary analysis of phase 3 randomised clinical trials of mRNA vaccines in adults found an excess risk of serious adverse events of special interest of 12.5 per 10,000 vaccinated, which suggests a higher risk than initially estimated at the time of emergency authorisation.[415]

    • A self-controlled case series analysis found minimal evidence of an increased incidence of cardiac or all-cause mortality overall in the 12 weeks following vaccination (for all vaccines). However, there is evidence of a smaller increased incidence of cardiac or all-cause mortality after a second dose of an mRNA vaccine in males, and after a first dose of a non-mRNA-based vaccine among females.[416]

    • The US Centers for Disease Control and Prevention is evaluating a safety signal with the Pfizer/BioNTech bivalent vaccine to determine whether people ≥65 years of age are at increased risk of ischaemic stroke in the 21 days following vaccination. At this stage, analyses have not validated this signal and no change is recommended to vaccination practice at this time.[417][418] 

    • Medicines and Healthcare products Regulatory Agency: coronavirus vaccine - summary of Yellow card reporting Opens in new window

  • Report all suspected adverse events after vaccination via your local reporting system. This is mandatory in some countries. Surveillance of adverse events is extremely important, and may reveal additional, less frequent serious adverse events not detected in clinical trials. The mRNA vaccines have not been authorised for use in humans previously, so there is no long-term safety and efficacy data available for these types of vaccines.

Pre-exposure prophylaxis

  • Tixagevimab/cilgavimab, a long-acting neutralising monoclonal antibody combination with activity against SARS-CoV-2, may be authorised in some countries for pre-exposure prophylaxis.

    • The WHO, the UK National Institute for Health and Care Excellence, and the US National Institutes of Health do not currently recommend tixagevimab/cilgavimab for pre-exposure prophylaxis, as currently circulating SARS-CoV-2 subvariants are unlikely to be susceptible.[419][420]​​[421][422]​​​​​​​

    • Evidence for the use of tixagevimab/cilgavimab is limited. A systematic review and meta-analysis found that tixagevimab/cilgavimab was associated with a reduction in RT-PCR positivity, symptomatic disease, severe disease, hospitalisation, intensive care unit admission, need for oxygen, and mortality compared with no treatment or another alternative treatment. Its activity may be reduced for the Omicron BA.1 and BA.2 subvariants, but it likely maintains most of its activity against the BA.4 and BA.5 subvariants. The analysis did not include currently circulating recombinant SARS-CoV-2 variants (e.g., XBB).[423] 

  • No drugs used for prophylaxis (e.g., ivermectin, hydroxychloroquine, lopinavir/ritonavir) have provided convincing evidence for a reduction in the risk of infection. However, much of the evidence remains very low-certainty.[424][425]

    • The WHO strongly recommends against administering hydroxychloroquine prophylaxis to people who do not have COVID-19, based on high-certainty evidence.[420]​​

​Infection prevention and control for healthcare professionals

  • Consult local infection prevention and control protocols; only basic principles from the World Health Organization (WHO) guidelines are detailed here.[426][427]

  • Screen all people, including patients, visitors, and others entering the facility, for COVID-19 at the first point of contact with the health facility to allow for early recognition.

  • Immediately isolate all suspected or confirmed cases in a well-ventilated area that is separate from other patients. Place patients in adequately ventilated single rooms if possible. When single rooms are not available, place all cases together in the same adequately ventilated room and ensure there is at least 1 metre (3 feet) between patients.

  • Implement standard precautions at all times:

    • Practice hand and respiratory hygiene

    • Give patients a medical mask to wear

    • Wear appropriate personal protective equipment

    • Practice safe waste management and environmental cleaning.

  • Implement additional contact and droplet precautions before entering a room where suspected or confirmed cases are admitted.

    • A respirator or medical mask should be worn along with other personal protective equipment (i.e., gown, gloves, eye protection) before entering a room with a suspected or confirmed case.

    • A respirator should be worn in the following situations: in care settings where ventilation is known to be poor or cannot be assessed, or the ventilation system is not properly maintained; based on the worker’s values and preferences and on their perception of what offers the highest protection possible to prevent infection.

    • Appropriate mask fitting should always be ensured, as should compliance with appropriate use of personal protective equipment and other precautions.

    • Universal masking is strongly recommended in health facilities in areas of known or suspected community or cluster transmission.

  • Implement airborne precautions when performing aerosol-generating procedures, including placing patients in a negative pressure room and wearing a particulate respirator.

    • A respirator should always be worn along with other personal protective equipment while performing aerosol-generating procedures, and in settings where these procedures are regularly performed on patients with suspected or confirmed disease (e.g., intensive care units, emergency departments).

    • Some countries and organisations recommend airborne precautions for any situation involving the care of a COVID-19 patient.

  • All specimens collected for laboratory investigations should be regarded as potentially infectious.

  • Appropriate personal protective equipment gives healthcare workers a high level of protection.

    • A rapid review and meta-analysis found that wearing personal protective equipment conferred significant protection against infection compared with not wearing it. High-certainty evidence indicates that using N95 masks significantly reduces the risk of infection. No effect was found for wearing gloves and gowns. There is a lack of evidence for different combinations of personal protective equipment.[428]

    • An expert panel statement informed by review-level evidence (as of May 2022) concluded that N95 respirators (or their equivalent) may be more effective than surgical masks in reducing the risk of infection in the mask wearer (low confidence) based on epidemiological evidence (usually of low or very low certainty) from SARS-CoV-2 and other coronaviruses.[429]

  • Patients can be managed remotely by telephone or video consultations depending on local protocols.[430]

  • Detailed infection prevention and control guidance is available:

Infection prevention and control for the general public

  • Public health recommendations vary between countries and you should consult your local guidance.

    • It is generally recommended that people stay at least 1 to 2 metres (3-6 feet) away from others (recommendations on distance vary between countries), wash their hands often with soap and water (or hand sanitiser that contains at least 60% alcohol), cover coughs and sneezes, avoid crowds and poorly ventilated spaces, clean and disinfect high touch surfaces, monitor their health and self-isolate or seek medical attention if necessary, and get vaccinated.[431][432]

    • WHO: advice for the public – coronavirus disease (COVID-19) Opens in new window

  • The WHO strongly recommends mask use in community settings in higher risk situations regardless of the local epidemiological situation, and conditionally recommends a risk-based approach for mask use in community settings in other situations.[427]​ Public health recommendations vary between countries and you should consult your local guidance.

    • There is no high-quality or direct scientific evidence to support the widespread use of masks by healthy people in the community setting.

    • A Cochrane review, which included results of studies from during the COVID-19 pandemic, found that the pooled results of randomised controlled trials did not show a clear reduction in respiratory viral infection with the use of surgical masks. Mask use in the community setting probably makes little or no difference to the outcome of laboratory-confirmed influenza or COVID‐19 compared with not wearing masks (moderate-certainty evidence). There were no clear differences between the use of N95/P2 respirators and surgical masks in healthcare workers when used in routine care to reduce the risk of respiratory viral infection, and the evidence was very uncertain.[433]

    • A living rapid review found that the evidence for mask effectiveness for respiratory tract infection prevention is stronger in healthcare settings compared with community settings; however, direct evidence on comparative effectiveness in SARS-CoV-2 infection is insufficient. The strength of evidence for any mask use versus non-use in community settings is low-moderate and is based mainly on observational studies with methodological limitations.[434][435]​​

    • The only randomised controlled trial to investigate the efficacy of masks in the community during the pandemic found that the recommendation to wear surgical masks when outside the home did not reduce infection compared with a no mask recommendation. However, the study did not assess whether masks could decrease disease transmission from mask wearers to others (source control).[436]

    • Cloth masks have limited efficacy in preventing viral transmission compared with medical-grade masks and the efficacy is dependent on numerous factors (e.g., material type, number of layers, fitting, moisture level), and may result in increased risk of infection.[437][438]​​

    • Harms and disadvantages of wearing masks include headache, breathing difficulties, facial skin lesions, irritant dermatitis, worsening acne, difficulty wearing masks by certain members of the population (e.g., children, people with learning disabilities, mental illness or cognitive impairment, asthma, chronic respiratory or breathing problems, facial trauma or recent oral maxillofacial surgery, living in hot and humid environments), psychological issues, difficulty communicating, poor compliance, waste disposal issues, and increased viral load. There are insufficient data to quantify all of the adverse effects that might reduce the acceptability, adherence, and effectiveness of face masks.[427][439][440]​​​

  • Many countries implemented non-pharmaceutical interventions during the pandemic in order to reduce and delay viral transmission (e.g., social distancing, city lockdowns, stay-at-home orders, curfews, non-essential business closures, bans on gatherings, school and university closures, remote working, quarantine of exposed people). However, there was no clear, significant beneficial effect of more restrictive non-pharmaceutical interventions compared with less restrictive non-pharmaceutical interventions in any of the countries studied.[441]

  • Many countries implemented travel-control measures during the pandemic including complete or partial closure of borders, entry or exit screening, and/or quarantine of travellers. However, evidence to support the use of these measures was of low to very low certainty.[442][443][444]​​​​

​Lifestyle modifications

  • Lifestyle modifications (e.g., smoking cessation, weight loss) may help to reduce the risk of infection, and may be a useful adjunct to other interventions.[445]

  • The WHO recommends that tobacco users stop using tobacco given the well-established harms associated with tobacco use and second-hand smoke exposure.[227]

Use of this content is subject to our disclaimer