Demonstrating the utility of a public school administrative database for epidemiologic research: A pilot study of factors associated with absenteeism

Childhood obesity is associated with poor school performance. We used administrative data from a large urban school district to investigate the association of obesity and other student characteristics with absenteeism, since the latter may impact school performance.

METHODS: We analyzed data for students 4 to 20 years of age, enrolled in Philadelphia public schools during three recent academic years. Students' obesity status was classified according to their body mass index (BMI)-for-age values. Negative binomial regression analyses of absence rates (number of days absent, divided by number of days enrolled) accounted for the correlation of a student's data over time.

RESULTS: We analyzed records of 178,149 students, including 61% African Americans, 15% Hispanics, and 8% Asians or students of other minority ethnic groups. The rates of overweight and obesity were 17% and 20%, respectively. Overall, the estimated absence rate was 17 absences per 180 school days enrolled. Obesity was weakly associated with increased school absences. Only extremely obese children had absence rates higher, by about 10%, than normal-weight children. Compared to white students, Hispanics and African Americans had higher absence rates (by 15% and 10%, respectively), and Asians had lower absence rates (by 42%). Students eligible for free or reduced-cost meals had 20% higher absence rates than those who were not eligible.

CONCLUSIONS: Routinely collected student data can be used to conduct epidemiologic studies of factors that affect school performance and health. In our analyses, we found that only very severe obesity is associated with increased school absenteeism