Nonparametric Bounds for the Risk Function

Stephen R Cole; Michael G Hudgens; Jessie K Edwards; M Alan Brookhart; David B Richardson; Daniel Westreich; Adaora A Adimora

doi:10.1093/aje/kwz013

Nonparametric Bounds for the Risk Function

Am J Epidemiol. 2019 Apr 1;188(4):632-636. doi: 10.1093/aje/kwz013.

Authors

Stephen R Cole¹, Michael G Hudgens², Jessie K Edwards¹, M Alan Brookhart^{1

3}, David B Richardson¹, Daniel Westreich¹, Adaora A Adimora^{1

4}

Affiliations

¹ Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina.
² Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina.
³ NoviSci LLC, Durham, North Carolina.
⁴ Department of Medicine, School of Medicine, University of North Carolina, Chapel Hill, North Carolina.

Abstract

Nonparametric bounds for the risk difference are straightforward to calculate and make no untestable assumptions about unmeasured confounding or selection bias due to missing data (e.g., dropout). These bounds are often wide and communicate uncertainty due to possible systemic errors. An illustrative example is provided.

Keywords: bias; bounds; inference; missing data.

Publication types

Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

MeSH terms

Confounding Factors, Epidemiologic
Epidemiologic Research Design*
Humans
Risk Assessment / methods*
Selection Bias
Statistics, Nonparametric*

Abstract

Publication types

MeSH terms

Grants and funding