Social risk factors are often less vigorously pursued in clinical assessments of infant risk than are biologic risk factors. We examined the relative importance of early social and biologic risk factors in predicting poor health and educational outcomes in children.
The study was composed of all infants born in Winnipeg, Canada, during April–December 1984, who were followed up until age 19 years (n = 4667). Predictors were 3 routinely assessed biologic risks (birth weight, gestational age, and Apgar score) and 3 prominent social factors (mother's age, parent marital status, and socioeconomic status). Outcomes were childhood hospitalization and passage of a required high school examination. Analyses included logistic regression, measures of accuracy, and population attributable risk percent (PAR%).
Biologic and social risk factors were associated with similarly steep poor outcomes gradients. Social risk factors had similar, and in some cases stronger, measures of association and accuracy. Using biologic risk criteria alone misclassified as low-risk 65% of cohort children who had high rates of later hospitalization and examination failure. PAR% associated with social risk factors exceeded biologic risk factors in most cases (eg, hospitalization PAR% = 4.4 for offspring of teen mothers vs. 1.7 for low birth weight).
In a population-based sample of infants followed-up through adolescence, early social risk factors were as threatening as, and more common than, routinely documented biologic risks—frequently identifying otherwise-unrecognized at-risk children. These findings together suggest that rigorous evaluation of social factors should be made a routine part of clinical assessment to more comprehensively and accurately identify infants at risk for later serious health problems and academic failure.
From the aSchool of Public Health, Division of Community Health & Human Development, University of California, Berkeley, CA; bManitoba Centre for Health Policy, University of Manitoba, Winnipeg, Manitoba, Canada; and cCollege for Interdisciplinary Studies and Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada.
Submitted 29 January 2008; accepted 23 October 2009.
Supported by an RBC Financial Group Child Health Award (to D.P.J., N.P.R., and M.B.), the Robert Wood Johnson Health & Society Scholars program (to D.P.J. and W.T.B.), the Canadian Population Health Initiative (to M.B. and N.P.R.), a Canadian Institutes of Health Research New Investigator Award (to M.B.), and a Canada Research Chair (to N.P.R.).
Correspondence: Douglas P. Jutte, School of Public Health, UC Berkeley-UCSF Joint Medical Program, University of California, 570 University Hall, Berkeley, CA 94720–1190. E-mail: email@example.com.