Related Experiment Video
Updated: Mar 3, 2026

Using Cholesky Decomposition to Explore Individual Differences in Longitudinal Relations between Reading Skills
Published on: September 17, 2019
Adjusting for unmeasured confounding in nonrandomized longitudinal studies: a methodological review
Adam J Streeter1, Nan Xuan Lin2, Louise Crathorne3
1Health Statistics Group, Institute of Health Research, University of Exeter Medical School, University of Exeter, St. Luke's Campus, Exeter EX1 2LU, United Kingdom; Medical Statistics, Institute of Translational and Stratified Medicine, Plymouth University Peninsula School of Medicine & Dentistry, University of Plymouth, Plymouth Science Park, Derriford, Plymouth PL6 8BX, United Kingdom.
Researchers reviewed methods for unmeasured confounding in longitudinal data using electronic health records. Established techniques like instrumental variable analysis (IVA) and difference-in-differences (DiD) are common, but new methods show promise.
Area of Science:
- Epidemiology
- Biostatistics
- Health Informatics
Background:
- Electronic health records (EHRs) offer valuable data for research.
- Unmeasured confounding poses a significant threat to causal inference in longitudinal studies.
- Addressing bias in observational health data is crucial for reliable research findings.
Purpose of the Study:
- To review methods for addressing unmeasured confounding bias in longitudinal data research.
- To assess the application and development of these methods, particularly using EHRs.
- To identify common and emerging strategies for causal inference in health research.
Main Methods:
- Methodological literature review.
- Searched MEDLINE and EMBASE databases.
- Focused on quasi-experimental analyses for causal inference in nonrandomized longitudinal health data.
Main Results:
- 121 studies were reviewed, with instrumental variable analysis (IVA) used in 84.
- Difference-in-differences (DiD) and fixed effects (FE) models were employed in 29 studies.
- Emerging methods like propensity score calibration and negative control outcomes were also identified.
Conclusions:
- Difference-in-differences (DiD) and instrumental variable analysis (IVA) are frequently used for unmeasured confounding.
- Researchers can better leverage longitudinal data information.
- New methods show potential but require further validation for widespread adoption.
More Related Videos
06:55Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
04:35Development of an Individual-Tree Basal Area Increment Model using a Linear Mixed-Effects Approach
Published on: July 3, 2020
Related Concept Videos
Strategies for Assessing and Addressing Confounding
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
Confounding in Epidemiological Studies
Randomized Experiments
Simple randomization
Simple...
Longitudinal Research
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
Bioequivalence Experimental Study Designs: Repeated Measures, Cross-Over, Carry-Over, and Latin Square Designs