(PO-2528) Selecting An Appropriate Comparator Group In Vaccine Cohort Studies Of Older Adults: Recombinant Zoster Vaccine As An Example

Jennifer H Han, Carla A Talarico, Soowoo Back, Susan dosReis, Aziza Jamal-Allial, Cheryl N McMahill-Walraven, Talia J Menzen, Vinit P Nair, Young Hee Nam, Josephine C Ocran-Appiah, Jeffrey S Brown and Katherine W Yih

Background: Observational studies are critical for assessing effectiveness and safety of vaccines in the real-world setting. However, substantive differences between individuals who do and do not receive the vaccine can lead to a number of biases. Selection of an appropriate comparator group is an important study design consideration, especially in studies of older adults with multiple comorbidities.

Objectives: To identify the most appropriate comparator group for a cohort study assessing safety outcomes in recipients of the two-dose adjuvanted Recombinant Zoster Vaccine (RZV) aged ≥ 50 years.

Methods: Curated administrative data in the Sentinel common data model format from 4 health plan participants in the US Food and Drug Administration’s Sentinel Initiative were used to select commercially-insured adults ≥ 50 years of age with medical and drug coverage from 1/2018-1/2019. The exposed group was individuals who received ≥ 1 RZV dose, an adjuvanted subunit vaccine approved in the US for the prevention of herpes zoster in people aged ≥ 50 years. Six different comparator selection algorithms were evaluated, using various combinations of receiving ≥ 1 healthcare utilization service: annual physical exam; age-appropriate screening; annual exam or screening; age-appropriate vaccination; annual exam and influenza vaccination; and outpatient encounter for chronic condition management. Demographics, healthcare-seeking behaviors, and the prevalence of comorbidities and immunosuppressive conditions for the RZV recipient group and the comparator groups were descriptively compared.

Results: 261,050 and 133,256 individuals received ≥ 1 and two doses of RZV, respectively, during the study period. The number of individuals with ≥ 1 comparator visit ranged from 1,666,326 to 6,698,184, depending on the definition. The mean age for each of the comparator groups was similar to that of RZV recipients (68 years, standard deviation, 9). However, there was substantial variation in the prevalence of some comorbidities across the six comparator groups compared to RZV recipients. Adults who received a routine annual physical exam or an age-appropriate preventive screening measure were one of the most comparable unexposed groups compared to RZV recipients.

Conclusions: Differences among potential comparator groups highlight the importance of the rigorous selection of a comparator group in cohort studies evaluating vaccine effectiveness or safety, especially in older adults who have multiple comorbidities.