0
The Pediatric Forum |

Autism Prevalence and Precipitation: The Potential for Cross-Level Bias

Joe M. Braun, MSPH, RN; Amy Kalkbrenner, MPH
[+] Author Affiliations

Copyright 2009 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.

More Author Information
Arch Pediatr Adolesc Med. 2009;163(5):492-493. doi:10.1001/archpediatrics.2009.83
Text Size: A A A
Published online

In their article, Waldman et al1 present a novel hypothesis that examines the association between autism and environmental factors using ecologic data from California, Oregon, and Washington. Their findings suggest a positive association between mean annual precipitation and autism prevalence at the county level. Although the authors address confounding, we believe they should have better addressed the potential effect of confounding by county-level urbanicity and the potential for cross-level bias (ecologic fallacy) when interpreting their results.

First, inadequate control for urbanicity may have confounded the county-level estimates. Most of the counties with the highest prevalence of autism and annual precipitation are also the most populated counties. Weiss2 has noted that agencies in the western parts of Oregon and Washington may have identified a greater proportion of children with autism than their eastern counterparts. Urbanicity has a strong association with autism at the ecologic level; urban school districts exhibit an autism prevalence 5 times as high as that in rural districts.3 High correlations between urbanicity, autism, and rainfall in these data warrant careful control and specification of county-level urban status. Greenland and Robins4 demonstrated that misspecification of confounders in ecologic studies can result in substantial changes in the effect estimate of interest. Waldman et al used log-transformed county population as a proxy for county-level urbanicity. It is unlikely that county population completely characterizes the degree of urbanicity and access that residents have to diagnostic and medical services. Given the relatively small effect size that annual precipitation had on autism prevalence, it is possible that better control for urbanicity would have eliminated any observed association.

Second, the authors gave individual-level interpretation to their county-level estimates without addressing the ecologic fallacy. Even with perfect control of county-level urbanicity, the potential for cross-level bias still exists. Within-group heterogeneity of county-level urbanicity and annual precipitation will not be captured solely by county-level information.5 Urbanicity likely acts as a proxy for neighborhood and individual-level factors, including better access to diagnostic and medical services. Only studies that can collect exposure information and control for confounders, such as accessibility to diagnostic services, at the individual level can lead to causal conclusions. Thus, we are concerned about the authors' conclusion that autism prevalence would be reduced by 30% to 40% if the mean annual precipitation were reduced by 35.6 inches. Such interpretation implies that causation at the individual level has been established, which is certainly outside the realm of this ecologic, hypothesis-generating study.

AUTHOR INFORMATION

Correspondence: Mr Braun, Department of Epidemiology, University of North Carolina–Chapel Hill, Campus Box 7435, McGavran-Greenberg Hall, Chapel Hill, NC 27514 (jmbraun@unc.edu).

Author Contributions:Study concept and design: Braun. Analysis and interpretation of data: Kalkbrenner. Drafting of the manuscript: Braun. Critical revision of the manuscript for important intellectual content: Braun and Kalkbrenner.

Financial Disclosure: None reported.

Waldman  M, Nicholson  S, Adilov  N, Williams  J. Autism prevalence and precipitation rates in California, Oregon, and Washington counties. Arch Pediatr Adolesc Med 2008;162 (11) 1026- 1034
PubMed
Weiss  NS. Precipitation and autism: do these results warrant publication? Arch Pediatr Adolesc Med 2008;162 (11) 1095- 1096
PubMed
Palmer  RF, Blanchard  S, Stein  Z, Mandell  D, Miller  C. Environmental mercury release, special education rates, and autism disorder: an ecological study of Texas. Health Place 2006;12 (2) 203- 209
PubMed
Greenland  S, Robins  J. Invited commentary: ecologic studies–biases, misconceptions, and counterexamples. Am J Epidemiol 1994;139 (8) 747- 760
PubMed
Morgenstern  H, Thomas  D. Principles of study design in environmental epidemiology. Environ Health Perspect 1993;101 ((suppl 4)) 23- 38
PubMed

First Page Preview

First page PDF preview

Figures

Tables

Interactive Graphics

Video

Country-Specific Mortality and Growth Failure in Infancy and Yound Children and Association With Material Stature

Use interactive graphics and maps to view and sort country-specific infant and early dhildhood mortality and growth failure data and their association with maternal

Waldman  M, Nicholson  S, Adilov  N, Williams  J. Autism prevalence and precipitation rates in California, Oregon, and Washington counties. Arch Pediatr Adolesc Med 2008;162 (11) 1026- 1034
PubMed
Weiss  NS. Precipitation and autism: do these results warrant publication? Arch Pediatr Adolesc Med 2008;162 (11) 1095- 1096
PubMed
Palmer  RF, Blanchard  S, Stein  Z, Mandell  D, Miller  C. Environmental mercury release, special education rates, and autism disorder: an ecological study of Texas. Health Place 2006;12 (2) 203- 209
PubMed
Greenland  S, Robins  J. Invited commentary: ecologic studies–biases, misconceptions, and counterexamples. Am J Epidemiol 1994;139 (8) 747- 760
PubMed
Morgenstern  H, Thomas  D. Principles of study design in environmental epidemiology. Environ Health Perspect 1993;101 ((suppl 4)) 23- 38
PubMed

Correspondence

CME Course for:


You need to register in order to view this quiz.


To understand the clinical management of acute heart failure syndromes.
Accreditation Information The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
To view and print your certificate and access a summary of your CME courses go to My CME.
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s “Cited By” API will populate this tab (http://www.crossref.org/citedby.html).
Submit a Response

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Topics
PubMed Articles