Course description
Certain: Observational Studies and Registries
While randomized controlled trials are considered to be the "gold standard" in health research, they cannot always be performed, for ethical or practical reasons. Observational studies gather information from data that has already been collected, or by observing and measuring patients' changes in health status and their response to interventions outside of a clinical trial. In this course, you will learn to identify the characteristics of observational studies, to interpret the results of observational studies, and to describe the use of health registries in comparative effectiveness research (CER).
This course includes the following 11 lectures:
- Overview of Using Observational Data in Comparative Effectiveness Research (CER)
- Cancer Registries and Data Linkage
- SEER-Medicare and Other Data Sources
- Overview of Analytic Methods I
- Overview of Analytic Methods II
- Longitudinal Data Analysis
- Advanced Methods in CER I
- Advanced Methods in CER II
- Survival Analysis
- Analysis of Medical Cost Data in Observational Studies
- Healthcare Policy Research
This course is intended for anyone interested in comparative effectiveness research (CER) and patient-centered outcomes research (PCOR) methods.
This course is supported by grant number R25HS023214 from the Agency for Healthcare Research and Quality.
Upcoming start dates
Suitability - Who should attend?
Prerequisites
Introduction course (Certain.1x) recommended, but not required.
Outcome / Qualification etc.
What you'll learn
- Define several types of observational studies and describe their characteristics
- Interpret observational study results
- Describe how health registries can be used for CER
Training Course Content
Overview of Using Observational Data in Comparative Effectiveness Research (CER)
- When to use observational studies in CER
- Observational study biases
- Observational study data sources
Cancer Registries and Data Linkage
- Data linkage definition
- Why cancer registries link their data
- Basics of data linkage methods
SEER-Medicare and Other Data Sources
- Uses of cancer registry data
Overview of Analytic Methods I
- Correlation definition
- Statistical methods for continuous outcome variables
- Simple linear regression and multiple linear regression
Overview of Analytic Methods II
- Statistical methods for categorical data
- Mantel Haenszel Method
- Logistic regression
Longitudinal Data Analysis
- Longitudinal data situations and the problems of repeated measures
- Common approaches to measure change over time
- Understanding basic results obtained from longitudinal analysis of linear and logistic regression
Advanced Methods in CER I
- Limitations of randomized clinical trials and advantages of observational studies
- Propensity score definition and estimation
- Checking the proper use of propensity scores
Advanced Methods in CER II
- Endogeneity bias definition
- Conditions and properties of instrumental variable estimation and models
Survival Analysis
- Censoring and person-time
- Application of life tables
- Kaplan Meier estimator
- Using the log-rank test to compare survival curves
- Hazard function definition and hazard ratio computation
- Applying the Cox proportional hazards model
Analysis of Medical Cost Data in Observational Studies
- Importance of medical cost analysis
- Basic elements of medical cost data
- Types of medical cost studies and their analytical methods
Healthcare Policy Research
- Objectives and stages of the policy-making process
- Comparison and contrast of CER and policy research
- Policy changes and evaluation results of a case study
Course delivery details
This course is offered through The University of Texas MD Anderson Cancer Center in Houston, a partner institute of EdX.
4-5 hours per week
Expenses
- Verified Track -$249
- Audit Track - Free