Effectiveness of Fostair® (ICS/LABA) vs. dual bronchodilation (LABA/LAMA) in COPD (FELICITI)

Study type

• Study type:
  Non-interventional study
• Scope of the study:
  Assessment of risk minimisation measure implementation or effectiveness

  Effectiveness study (incl. comparative)

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Name of medicine, other:
  Fostair
• Anatomical Therapeutic Chemical (ATC) code:
  (R03A) ADRENERGICS, INHALANTS
• Medical condition to be studied:
  Chronic obstructive pulmonary disease

Population studied

• Age groups:
  Adults (18 to < 46 years)

  Adults (46 to < 65 years)

  Adults (65 to < 75 years)

  Adults (75 to < 85 years)

  Adults (85 years and over)
• Estimated number of subjects:
  4000

Study design details

• Main study objective:
  Determine if treatment with beclometasone/formoterol (Fostair®) is non-inferior in terms of effectiveness, compared to LABA/LAMA treatment in frequently exacerbating COPD patients. Also, differential responses to Fostair® treatment across patient subgroup will be explored.
• Outcomes:
  The rate of moderate/severe COPD exacerbations during the entire follow-up. a) Time to treatment failure, b) rate of acute respiratory events, c) rate of acute OCS course, d) rate of antibiotic courses for COPD, e) mMRC score within 18 months, f) time until first pneumonia
• Data analysis plan:
  Patients initiating Fostair or LAMA/LAMA will be identified, and the rate of exacerbations (plus other outcomes) will be determined during the follow-up period. Patients will be right-censored when an important treatment change occurs, or at the end of data availability. A set of confounding handling approaches will be evaluated, and the best one with regard to residual bias will be chosen. Direct matching, propensity score matching, direct+propensity score matching and inverse probability of treatment weighting are the candidate approaches. Bias potential as well as restriction of study population will be used to determine the optimal approach. Event rates during follow-up will be analysed using negative binomial regression, taking the follow-up duration into account.