Lecture 2 Flashcards
Comparative Effectiveness Research
A rigorous evaluation of the impact of different options that are available for treating a given medical condition for a particular set of patients
Drug Efficacy
whether a drug has the ability to bring about the intended effect
- ideal situation
Drug effectiveness
whether a drug achieves its desired effect in a real world situation
What data can we use from non-experimental studies
Large cohort studies (prospective or ongoing)
Healthcare and clinical databases
Disease registries
Drug registries
Publicly available survey data
HMO/health plans
Commercial insurance databases
US Government Claims database
EHRs
Desired qualities of a database
representative
large
up to date
continuity
linkage on unique identifier
accessible
US HMO/Health Plans as an information source
administrative and clinical datasets maintained by HMOs are used for clinical care, payment, and operational purposes
- used for many epidemiological studies
Strengths of HMO database
- large, diverse, defined populations
- automated claims, EMR data, access to providers
- varied delivery models and practice patterns
- large cohorts can be identified to measure incidence of rare events or with specific exposures
- research centers’ ability to contact health plan enrollee for participation in clinical research
Weaknesses of HMO database
- absence of populations that are uninsured
- smaller fraction of elderly than general populations
- prescription meds filled out-of-plan, nonprescription medications, and inpatient drug dispensing not routinely captured
US Government Claims Databases
Medicaid
Medicare
US Gov Claims Databases Strengths
- population size and length of follow-up
- accuracy of pharmacy claims
- validity of procedure claims
- over representation of underserved populations
- ability to validate outcomes
- ability to link to external data
US Gov Claims Database Weaknesses
- non-representativeness
- unavailable info
- limitations in Rx coverage
- Eligibility
- Data validity/access to medical records
- out of plan care
In-Hospital Databases Strengths
- sample size
- versatile data source
- data quality
- efficient
- data longevity
In-Hospital Databases Weaknesses
- generalizability of study findings can be questioned
- possible misclassification
- Ascertainment bias: hospitalized patients have different lengths of stay, vary across hospitals -> different durations of observation for exposures and outcomes
