OBJECTIVES:

In the United States, vaccination with recombinant zoster vaccine is recommended for herpes zoster (HZ) prevention in adults aged ≥19 years who are or will be immunodeficient or immunosuppressed because of disease or therapy. A targeted literature review was conducted to identify algorithms which can be applied to claims databases to identify immunocompromised populations for research or to target for HZ prevention.

METHODS:

A search of publications from January 2016 to August 2022 was conducted in the PubMed and Cochrane Review databases and supplemented with a review of publications cited by and citing 8 key pre-specified publications. Publications were screened against pre-specified criteria.

RESULTS:

Of 1,011 publications identified, 41 were eligible for full-text screening, of which 25 priority publications were selected. Algorithms identifying immunocompromised status were used to stratify analyses or as covariates in 12 publications, as an exclusion criterion in 9 publications, and as an inclusion criterion in 4 publications. All 25 publications used diagnoses from claims and/or clinical data to identify immunocompromised patients, and 20 also used immunosuppressive medications. Diagnoses and medications used varied across publications. For example, one publication identified immunocompromised populations using diagnoses for HIV/AIDS or leukemia/lymphoma, or use of antineoplastic, antiarthritic, or other immunosuppressant medications. In contrast, another publication identified immunocompromised populations using diagnoses for HIV/AIDS, solid or hematologic malignancy, bone marrow or organ transplant, rheumatologic or inflammatory condition, primary immunodeficiency or other immune conditions, chronic kidney disease or end stage renal disease, or use of medications including chemotherapeutic agents, antimetabolites, systemic steroids, or immunomodulators.

CONCLUSIONS:

There are several published algorithms for identifying immunocompromised populations that can be applied in administrative claims data. Diagnoses and medications for determining immunocompromised status varied across algorithms. Further research comparing algorithms is needed to support future use in research or to identify immunocompromised populations for HZ prevention.