Consultation outcome

Social Care Working Group consensus statement, March 2021

Updated 9 November 2021

Applies to England

Estimating the minimum level of vaccine coverage in care home settings

Care homes for older adults have been significantly impacted by the COVID-19 pandemic to date due to a combination of vulnerability of the resident population to severe outcome following COVID-19 infection, the close contact nature of mixing within such settings and the closed nature of these settings providing a risk of outbreaks. Ensuring very high levels of vaccination in vulnerable residents and the people who care for them in these settings is an appropriate public health intervention for a serious vaccine-preventable disease.

Modelling analysis, published in the SPI-M consensus statement of 11 March 2021 estimated that 75% of staff (given that 90% of residents in each individual care home had been vaccinated) provided a level of protection sufficient to limit outbreaks assuming other mitigations are in place. During March this analysis was updated to 80% coverage in staff and 90% in residents reflecting a slight change in evidence for efficacy of vaccination.

These values are minimum rates that would achieve a level of protection within homes against widespread outbreaks at time of the advice (March 2021) given the epidemic situation. Critically the 80% coverage for staff is derived given that 90% of residents are vaccinated and both groups have had a single dose of vaccine (as that was the situation at the time) as an illustration of necessary further effort earlier in the campaign. These are unlikely to be meaningful ‘critical thresholds’ after second doses of current vaccines against currently dominantly circulating variants (higher efficacy is likely with a second dose).

The mathematics that underpin this is relatively simplistic though justifiable as a way of informing policy but data on contact within the care homes both between staff and residents and within staff and resident groups is lacking and would be required for more detailed analysis. Projects such as CONTACT (University of Leeds) intend to resolve this gap but will not deliver data until summer 2021. Care homes are a close-knit network and the transmission dynamics are very different within different homes. This is particularly so with carer to resident interaction, even for residents kept in rooms. Resident to resident interaction is also impractical to stop in homes especially with ambulatory patients with dementia.

Any benefits of achieving high staff and resident uptake from any future policy will need to be balanced alongside potential scientific advice on the negative impacts on care provision. Workplace related transmission of COVID-19, whether in hospitals, other health settings and care homes, occurs both through worker networks, and links between workers and clients and patients. There is a strong scientific case for parity of approaches to vaccination offer and support between NHS inpatient settings and care homes, given the similarly close and overlapping networks between residents or patients and workers of all kinds in both. There is no body of evidence to suggest this would be different in social care from NHS settings in general other than the risk of outbreaks.

Stochastic modelling from The London School of Hygiene & Tropical Medicine (LSHTM) and University of Manchester indicates that the simple deterministic model predictions of coverage levels are pessimistic of likely vaccine impact due to smaller populations at risk in discrete care homes but variation in assumptions may make this less clear (such as inclusion in model of vaccine causing a reduction in transmission once infected rather than just making individuals less susceptible to infection). Both stochastic models suggest that vaccination of staff has higher relative impact than vaccinating residents likely due to the additional reduction in ingress as a consequence of protection afforded by vaccination.

However, this analysis does not and cannot take into account the impact of variants. Given the reports that the AstraZeneca (AZ) vaccine is sensitive to the South African variant then using minimum threshold values may not be credible as other variants emerge. The advice for the original wildtype (that vaccine was developed against) or even current ‘Kent’ variant that is dominant and single or 2 dose regimes is then superseded by the fact that vaccine may be rendered ineffective by mutation and so this will be a powerful argument for agile vaccination delivery with simple access to vaccine stocks for social care staff to ensure protection is against any dominant virus.

The calculations on recommended coverage should therefore be taken as the best estimate at the time of writing (March 2021). Given the changing epidemiological situation, they should be continually reviewed as evidence emerges. There is no certain threshold for protective vaccine coverage levels, the 80% to 90% coverage values previously calculated were based on single dose reported AZ efficacy rates. Vaccine is not a silver bullet, just part of our armoury against COVID-19. There is a risk that vaccination may lead to a reduced use of testing, PPE and IPC at a time that vigilance is needed against new variants with poorer vaccine efficacy.