Adjusting for unmeasured confounding using validation data: simplified two-stage calibration for survival and dichotomous outcomes

Vidar Hjellvik; Marie L De Bruin; Sven O Samuelsen; Øystein Karlstad; Morten Andersen; Jari Haukka; Peter Vestergaard; Frank de Vries; Kari Furu

doi:10.1002/sim.8131

Adjusting for unmeasured confounding using validation data: simplified two-stage calibration for survival and dichotomous outcomes

Vidar Hjellvik, Marie L De Bruin, Sven O Samuelsen, Øystein Karlstad, Morten Andersen, Jari Haukka, Peter Vestergaard, Frank de Vries, Kari Furu

1 Citation (Scopus)

Abstract

In epidemiology, one typically wants to estimate the risk of an outcome associated with an exposure after adjusting for confounders. Sometimes, outcome and exposure and maybe some confounders are available in a large data set, whereas some important confounders are only available in a validation data set that is typically a subset of the main data set. A generally applicable method in this situation is the two-stage calibration (TSC) method. We present a simplified easy-to-implement version of the TSC for the case where the validation data are a subset of the main data. We compared the simplified version to the standard TSC version for incidence rate ratios, odds ratios, relative risks, and hazard ratios using simulated data, and the simplified version performed better than our implementation of the standard version. The simplified version was also tested on real data and performed well.

Original language	English
Journal	Statistics in Medicine
Volume	38
Issue number	15
Pages (from-to)	2719-2734
ISSN	0277-6715
DOIs	https://doi.org/10.1002/sim.8131
Publication status	Published - 10 Jul 2019

Access to Document

10.1002/sim.8131

Cite this

@article{33a78c3fbff84844a9d7868de127e433,

title = "Adjusting for unmeasured confounding using validation data: simplified two-stage calibration for survival and dichotomous outcomes",

abstract = "In epidemiology, one typically wants to estimate the risk of an outcome associated with an exposure after adjusting for confounders. Sometimes, outcome and exposure and maybe some confounders are available in a large data set, whereas some important confounders are only available in a validation data set that is typically a subset of the main data set. A generally applicable method in this situation is the two-stage calibration (TSC) method. We present a simplified easy-to-implement version of the TSC for the case where the validation data are a subset of the main data. We compared the simplified version to the standard TSC version for incidence rate ratios, odds ratios, relative risks, and hazard ratios using simulated data, and the simplified version performed better than our implementation of the standard version. The simplified version was also tested on real data and performed well.",

author = "Vidar Hjellvik and {De Bruin}, {Marie L} and Samuelsen, {Sven O} and {\O}ystein Karlstad and Morten Andersen and Jari Haukka and Peter Vestergaard and {de Vries}, Frank and Kari Furu",

note = "{\textcopyright} 2019 John Wiley & Sons, Ltd.",

year = "2019",

month = jul,

day = "10",

doi = "10.1002/sim.8131",

language = "English",

volume = "38",

pages = "2719--2734",

journal = "Statistics in Medicine",

issn = "0277-6715",

publisher = "JohnWiley & Sons Ltd",

number = "15",

}

TY - JOUR

T1 - Adjusting for unmeasured confounding using validation data

T2 - simplified two-stage calibration for survival and dichotomous outcomes

AU - Hjellvik, Vidar

AU - De Bruin, Marie L

AU - Samuelsen, Sven O

AU - Karlstad, Øystein

AU - Andersen, Morten

AU - Haukka, Jari

AU - Vestergaard, Peter

AU - de Vries, Frank

AU - Furu, Kari

PY - 2019/7/10

Y1 - 2019/7/10

N2 - In epidemiology, one typically wants to estimate the risk of an outcome associated with an exposure after adjusting for confounders. Sometimes, outcome and exposure and maybe some confounders are available in a large data set, whereas some important confounders are only available in a validation data set that is typically a subset of the main data set. A generally applicable method in this situation is the two-stage calibration (TSC) method. We present a simplified easy-to-implement version of the TSC for the case where the validation data are a subset of the main data. We compared the simplified version to the standard TSC version for incidence rate ratios, odds ratios, relative risks, and hazard ratios using simulated data, and the simplified version performed better than our implementation of the standard version. The simplified version was also tested on real data and performed well.

AB - In epidemiology, one typically wants to estimate the risk of an outcome associated with an exposure after adjusting for confounders. Sometimes, outcome and exposure and maybe some confounders are available in a large data set, whereas some important confounders are only available in a validation data set that is typically a subset of the main data set. A generally applicable method in this situation is the two-stage calibration (TSC) method. We present a simplified easy-to-implement version of the TSC for the case where the validation data are a subset of the main data. We compared the simplified version to the standard TSC version for incidence rate ratios, odds ratios, relative risks, and hazard ratios using simulated data, and the simplified version performed better than our implementation of the standard version. The simplified version was also tested on real data and performed well.

U2 - 10.1002/sim.8131

DO - 10.1002/sim.8131

M3 - Journal article

C2 - 30828842

SN - 0277-6715

VL - 38

SP - 2719

EP - 2734

JO - Statistics in Medicine

JF - Statistics in Medicine

IS - 15

ER -

Adjusting for unmeasured confounding using validation data: simplified two-stage calibration for survival and dichotomous outcomes

Abstract

Access to Document

Fingerprint

Cite this