Two stage least squares with time-varying instruments : an application to an evaluation of treatment intensification for type-2 diabetes
- Author
- Daniel Tompsett, Stijn Vansteelandt (UGent) , Richard Grieve, John Robson and Manuel Gomes
- Organization
- Abstract
- As routinely collected longitudinal data becomes more available in many settings, policy makers are increasingly interested in the effect of time-varying treatments (sustained treatment strategies). In settings such as this, many commonly used statistical approaches for estimating treatment effects, such as g-methods, often adopt the 'no unmeasured confounding' assumption. Instrumental variable (IV) methods aim to reduce biases due to unmeasured confounding, but have received limited attention in settings with time-varying treatments. This paper extends and critically evaluates a commonly used IV estimating approach, Two Stage Least Squares (2SLS), for evaluating time-varying treatments. Using a simulation study, we found that, unlike standard 2SLS, the extended 2SLS performs relatively well across a wide range of circumstances, including certain model misspecifications. We illustrate the methods in an evaluation of treatment intensification for Type-2 Diabetes Mellitus, exploring the exogeneity in prescribing preferences to operationalise a time-varying instrument.
- Keywords
- Instrumental variable, time-varying, two stage least squares, physician preference, diabetes, VARIABLE METHODS, METFORMIN, MODELS, SULFONYLUREA, EFFICACY, SAFETY, BIAS
Downloads
-
publisher version.pdf
- full text (Published version)
- |
- open access
- |
- |
- 517.63 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01KEWRFH8RERX9943A86FK92BW
- MLA
- Tompsett, Daniel, et al. “Two Stage Least Squares with Time-Varying Instruments : An Application to an Evaluation of Treatment Intensification for Type-2 Diabetes.” STATISTICAL METHODS IN MEDICAL RESEARCH, vol. 35, no. 2, 2026, pp. 346–69, doi:10.1177/09622802251404064.
- APA
- Tompsett, D., Vansteelandt, S., Grieve, R., Robson, J., & Gomes, M. (2026). Two stage least squares with time-varying instruments : an application to an evaluation of treatment intensification for type-2 diabetes. STATISTICAL METHODS IN MEDICAL RESEARCH, 35(2), 346–369. https://doi.org/10.1177/09622802251404064
- Chicago author-date
- Tompsett, Daniel, Stijn Vansteelandt, Richard Grieve, John Robson, and Manuel Gomes. 2026. “Two Stage Least Squares with Time-Varying Instruments : An Application to an Evaluation of Treatment Intensification for Type-2 Diabetes.” STATISTICAL METHODS IN MEDICAL RESEARCH 35 (2): 346–69. https://doi.org/10.1177/09622802251404064.
- Chicago author-date (all authors)
- Tompsett, Daniel, Stijn Vansteelandt, Richard Grieve, John Robson, and Manuel Gomes. 2026. “Two Stage Least Squares with Time-Varying Instruments : An Application to an Evaluation of Treatment Intensification for Type-2 Diabetes.” STATISTICAL METHODS IN MEDICAL RESEARCH 35 (2): 346–369. doi:10.1177/09622802251404064.
- Vancouver
- 1.Tompsett D, Vansteelandt S, Grieve R, Robson J, Gomes M. Two stage least squares with time-varying instruments : an application to an evaluation of treatment intensification for type-2 diabetes. STATISTICAL METHODS IN MEDICAL RESEARCH. 2026;35(2):346–69.
- IEEE
- [1]D. Tompsett, S. Vansteelandt, R. Grieve, J. Robson, and M. Gomes, “Two stage least squares with time-varying instruments : an application to an evaluation of treatment intensification for type-2 diabetes,” STATISTICAL METHODS IN MEDICAL RESEARCH, vol. 35, no. 2, pp. 346–369, 2026.
@article{01KEWRFH8RERX9943A86FK92BW,
abstract = {{As routinely collected longitudinal data becomes more available in many settings, policy makers are increasingly interested in the effect of time-varying treatments (sustained treatment strategies). In settings such as this, many commonly used statistical approaches for estimating treatment effects, such as g-methods, often adopt the 'no unmeasured confounding' assumption. Instrumental variable (IV) methods aim to reduce biases due to unmeasured confounding, but have received limited attention in settings with time-varying treatments. This paper extends and critically evaluates a commonly used IV estimating approach, Two Stage Least Squares (2SLS), for evaluating time-varying treatments. Using a simulation study, we found that, unlike standard 2SLS, the extended 2SLS performs relatively well across a wide range of circumstances, including certain model misspecifications. We illustrate the methods in an evaluation of treatment intensification for Type-2 Diabetes Mellitus, exploring the exogeneity in prescribing preferences to operationalise a time-varying instrument.}},
author = {{Tompsett, Daniel and Vansteelandt, Stijn and Grieve, Richard and Robson, John and Gomes, Manuel}},
issn = {{0962-2802}},
journal = {{STATISTICAL METHODS IN MEDICAL RESEARCH}},
keywords = {{Instrumental variable,time-varying,two stage least squares,physician preference,diabetes,VARIABLE METHODS,METFORMIN,MODELS,SULFONYLUREA,EFFICACY,SAFETY,BIAS}},
language = {{eng}},
number = {{2}},
pages = {{346--369}},
title = {{Two stage least squares with time-varying instruments : an application to an evaluation of treatment intensification for type-2 diabetes}},
url = {{http://doi.org/10.1177/09622802251404064}},
volume = {{35}},
year = {{2026}},
}
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: