At risk population

Study Design

• A retrospective cohort study in US to determine the incidence of pneumococcal disease—all-cause pneumonia, IPD, and pneumococcal pneumonia—by age, risk profile, and individual medical condition among adults.³
• The study cohort was identified using de-identified claims data from Optum’s de-identified Clinformatics Data Mart Database (CDM) between January 1, 2016, and December 31, 2019. The study period was chosen to exclude the coronavirus disease 2019 (COVID-19) era and to include the time period immediately before PCV20 review/recommendation.^3,b
• Patients ages were variable (18–49;n= 15.2 M) , (50–64; n=8.5 M), and (≥65 years n= 14.2M) person–calendar years of observation.³

Included Criteria

• Geographically diverse from all 50 states were included.³
• Patients were enrolled if they were: in a participating health plan on January 1st of each calendar year from 2016 to 2019 (index date), ≥18 years of age on the index date, and at least 1 year of continuous enrollment before the index date of each corresponding calendar year.³
• Patients were also included if they had episode of disease (all-cause pneumonia, invasive pneumococcal disease (IPD), pneumococcal pneumonia within a calendar year.³
• Observation period spanned January 1 to December 31 of each year, or until: Date of death, or date of disenrollment.³
• Patients meeting the criteria in multiple calendar years were included in the pooled analysis.³

Exclusion Criteria

• Patients with missing gender data.³
• Patients with a recorded death date before January 1 of the index year.³
• Patients with overlapping pneumonia inpatient admissions during the observation period.³

Outcomes Assessed

• Crude disease rates were calculated by episode care setting, within each age group and by risk profile, and among adults with a chronic medical condition (CMC) or an immunocompromising condition (IC), by number of medical conditions.³
• Within an age group and risk profile, disease rates of persons with select individual medical conditions were also calculated.³
• Disease rates were reported as the number of events per 100,000 person-years.³
• To maintain patient de-identification, disease rates were only presented when there were ≥10 episodes in that group.³

^bCDM is a database of administrative health claims for members of large commercial and Medicare Advantage health plans.