CHILD RESTRAINT SYSTEMS FOR YOUNG CHILDREN DURING MOTOR VEHICLE COLLISIONS
Michael R. Elliott, PhD,
Michael J. Kallan, MS and
Dennis R. Durbin, MD, MSCE
Michael R. Elliott is with the Department of Biostatistics and the Survey Methodology Program at the University of Michigan, Ann Arbor. Michael J. Kallan is with the Center for Epidemiology and Biostatistics at the University of Pennsylvania, Philadelphia. Dennis R. Durbin is with the Center for Epidemiology and Biostatistics and the Division of Pediatric Emergency Medicine at the University of Pennsylvania, Philadelphia.
Correspondence: Correspondence should be sent to Michael R. Elliott, PhD, Department of Biostatistics, University of Michigan School of Public Health, M4041 SPHII, 1420 Washington Heights, Ann Arbor, MI 48109 (e-mail: mrelliot@umich.edu). Reprints can be ordered at http://www.ajph.org by clicking on the "Reprints/Eprints" link.
Because this article has no abstract, we have provided an extract of the first 100 words of the full text and any section headings.
Dew Drops on Cobweb.Printed with permission of Getty.
Rice and Anderson's matchedcohort study1 assesses the effect of seat belt and booster seatuse on fatality risk among 0–3 year olds using Fatal AccidentReporting System data.2 We are concerned that their implementationof the method extrapolates far beyond available data, to thepoint of making comparisons among adults and young childrenthat are not meaningful. The authors use conditional Poissonregression to estimate the effects of different restraint typeson risk of death. This model uses a log link to model risk ofdeath Y: