JMIR Publications

ABSTRACT

Background:

There are considerable inconsistencies in how immunosuppression is characterised and subdivided as a clinical risk group. This is detrimental to both the precision and comparability of international disease surveillance efforts – negatively implicating immunosuppressed health outcomes. Clinical consensus must therefore be built around which conditions and medications would constitute immunosuppression, their gradations of severity, and how to formalise both in a definitive phenotype for ongoing use in surveillance data flows.

Objective:

This protocol outlines e-Delphi objectives, methodology and statistical approaches that will help address this lack of consensus and construct an adult ‘immunosuppression’ phenotype.

Methods:

Leveraging existing evidence for heterogeneous COVID-19 outcomes in immunosuppressed adults, this work will recruit between 10 and 50 clinical or policy experts in the remit of vaccine prioritisation. Subsequent to two rounds of clinical consensus building and one round of concluding debate, these panellists will suggest the conditions, dependencies and clinical coding languages that should be incorporated into a phenotype for ‘immunosuppression’ in adults. This work will be conducted iteratively, with opportunities for panellists to ask clarifying questions between rounds and provide ongoing feedback to improve questionnaire items. Statistical analysis will focus on levels of agreement between responses and rankings.

Results:

This protocol outlines a robust method for generating consensus around best defining and subdividing adult immunosuppression for the benefit of disease surveillance.

Conclusions:

A universally acceptable, clinically relevant, and computerised medical record compatible phenotype for adult immunosuppression will have immediate value for vaccine distribution and the prioritisation of scarce or expensive medical supplies amongst affected adults.