It is imperative that local residents' views are heard, particularly in underserved
communities, when trying to address poor vaccine uptake. We are interested to elicit a
wide range of views, and channel these to co-design a feasible and acceptable
intervention with community groups.

This intervention will be evaluated in this randomised pilot study to assess feasibility,
practicality and acceptability. This will be in preparation of a fully powered randomised
controlled trial to evaluate the efficacy afterwards.

Eligible patients at all n=6 practices will be enrolled into the study at two time points
(see below). Patients will be individually randomised 1:1 to receive the PET or routine
care. We will use the Appt-Health workflow tool for both groups, although patients in the
control group will be monitored but will not receive additional interventions than
routine care. The comparison between those randomised to the intervention or control
group is envisaged to be the primary comparison in a subsequent trial to evaluate
efficacy (vaccination uptake). Individual randomisation will be stratified by GP, using a
random block allocation list implemented into the software used for the study.

Participating centres All practices will be selected based on list size (larger practices
with >10,000 patients), technological and practice infrastructure and procedures (to
ensure smooth operability of the PET), demography of the practice population (to ensure
good representation of underserved population groups) and capability and capacity
(decided by the practice and network leads).

Randomisation The practices will be randomised by the study statistician. Individuals
will be randomised, stratified by centre, using randomised block randomisation list
allocations in the practice software. This will be done on the same day or shortly after
the eligible list is determined. The randomised allocation will be visible to staff, but
we will use the feasibility study to explore ways of blinding staff and patients for the
definitive study.

Data collection and Analysis Eligible patients will be identified through the practice IT
system (EMISWeb). We will provide some initial support and training to enable practices
to complete missing ethnicity data. However, patients with unknown ethnicity resident in
areas with index of multiple deprivation greater than the bottom quintile would not be
included. The database associated with the software will record randomisation group, and
record of the intervention to each participant, as well as engagement with different
aspects of the tool (see secondary outcomes).

For analysis, we will extract pseudo-anonymised data and storage in the CASTOR database
and analyse this on a secure research database and environment within the Barts Cancer
Centre network at QMUL. Ethics and information governance approval will be sought for
this.