Invited commentary: the perils of birth weight--a lesson from directed acyclic graphs.

• Published in 2006
• Added on 2013-01-02

The strong association of birth weight with infant mortality is complicated by a paradoxical finding: Small babies in high-risk populations usually have lower risk than small babies in low-risk populations. In this issue of the Journal, Hernández-Díaz et al. (Am J Epidemiol 2006;164:1115-20) address this "birth weight paradox" using directed acyclic graphs (DAGs). They conclude that the paradox is the result of bias created by adjustment for a factor (birth weight) that is affected by the exposure of interest and at the same time shares causes with the outcome (mortality). While this bias has been discussed before, the DAGs presented by Hernández-Díaz et al. provide more firmly grounded criticism. The DAGs demonstrate (as do many other examples) that seemingly reasonable adjustments can distort epidemiologic results. In this commentary, the birth weight paradox is shown to be an illustration of Simpson's Paradox. It is possible for a factor to be protective within every stratum of a variable and yet be damaging overall. Questions remain as to the causal role of birth weight.

Links

• http://www.ncbi.nlm.nih.gov/pubmed/16931545
• http://aje.oxfordjournals.org/content/164/11/1121.long

Other information

BibTeX entry

@article{Wilcox2006,
	key = {Wilcox2006},
	type = {article},
	title = {Invited commentary: the perils of birth weight--a lesson from directed acyclic graphs.},
	author = {Wilcox, Allen J},
	abstract = {The strong association of birth weight with infant mortality is complicated by a paradoxical finding: Small babies in high-risk populations usually have lower risk than small babies in low-risk populations. In this issue of the Journal, Hern{\'{a}}ndez-D{\'{i}}az et al. (Am J Epidemiol 2006;164:1115-20) address this "birth weight paradox" using directed acyclic graphs (DAGs). They conclude that the paradox is the result of bias created by adjustment for a factor (birth weight) that is affected by the exposure of interest and at the same time shares causes with the outcome (mortality). While this bias has been discussed before, the DAGs presented by Hern{\'{a}}ndez-D{\'{i}}az et al. provide more firmly grounded criticism. The DAGs demonstrate (as do many other examples) that seemingly reasonable adjustments can distort epidemiologic results. In this commentary, the birth weight paradox is shown to be an illustration of Simpson's Paradox. It is possible for a factor to be protective within every stratum of a variable and yet be damaging overall. Questions remain as to the causal role of birth weight.},
	comment = {},
	date_added = {2013-01-02},
	date_published = {2006-12-01},
	urls = {http://www.ncbi.nlm.nih.gov/pubmed/16931545,http://aje.oxfordjournals.org/content/164/11/1121.long},
	collections = {Easily explained,Probability and statistics},
	doi = {10.1093/aje/kwj276},
	issn = {0002-9262},
	journal = {American journal of epidemiology},
	keywords = {Causality,Confounding Factors (Epidemiology),Female,Humans,Infant,Infant Mortality,Low Birth Weight,Newborn,Prevalence,Risk Assessment,Risk Factors,Smoking,Smoking: adverse effects,Smoking: epidemiology,United States,United States: epidemiology},
	month = {dec},
	number = 11,
	pages = {1121--3; discussion 1124--5},
	pmid = 16931545,
	url = {http://www.ncbi.nlm.nih.gov/pubmed/16931545 http://aje.oxfordjournals.org/content/164/11/1121.long},
	volume = 164,
	year = 2006,
	urldate = {2013-01-02}
}