Gender Bias in U.S. Pediatric Growth Hormone Treatment.

Growth hormone (GH) treatment of idiopathic short stature (ISS), defined as height <-2.25 standard deviations (SD), is approved by U.S. FDA. This study determined the gender-specific prevalence of height <-2.25 SD in a pediatric primary care population, and compared it to demographics of U.S. pediatric GH recipients. Data were extracted from health records of all patients age 0.5-20 years with ≥ 1 recorded height measurement in 28 regional primary care practices and from the four U.S. GH registries. Height <-2.25 SD was modeled by multivariable logistic regression against gender and other characteristics. Of the 189,280 subjects, 2073 (1.1%) had height <-2.25 SD. No gender differences in prevalence of height <-2.25 SD or distribution of height Z-scores were found. In contrast, males comprised 74% of GH recipients for ISS and 66% for all indications. Short stature was associated (P < 0.0001) with history of prematurity, race/ethnicity, age and Medicaid insurance, and inversely related (P < 0.0001) with BMI Z-score. In conclusion, males outnumbered females almost 3:1 for ISS and 2:1 for all indications in U.S. pediatric GH registries despite no gender difference in height <-2.25 SD in a large primary care population. Treatment and/or referral bias was the likely cause of male predominance among GH recipients.

Imputing Missing Race/Ethnicity in Pediatric Electronic Health Records: Reducing Bias with Use of U.S. Census Location and Surname Data.

OBJECTIVE: To assess the utility of imputing race/ethnicity using U.S. Census race/ethnicity, residential address, and surname information compared to standard missing data methods in a pediatric cohort. DATA SOURCES/STUDY SETTING: Electronic health record data from 30 pediatric practices with known race/ethnicity. STUDY DESIGN: In a simulation experiment, we constructed dichotomous and continuous outcomes with pre-specified associations with known race/ethnicity. Bias was introduced by nonrandomly setting race/ethnicity to missing. We compared typical methods for handling missing race/ethnicity (multiple imputation alone with clinical factors, complete case analysis, indicator variables) to multiple imputation incorporating surname and address information. PRINCIPAL FINDINGS: Imputation using U.S. Census information reduced bias for both continuous and dichotomous outcomes. CONCLUSIONS: The new method reduces bias when race/ethnicity is partially, nonrandomly missing.

Sex-based prevalence of growth faltering in an urban pediatric population.

OBJECTIVE: To determine the sex-based prevalence of growth faltering in a pediatric primary care setting. STUDY DESIGN: A total of 33 476 children attending 4 urban pediatric primary care practices affiliated with a tertiary pediatric hospital between July 2002 and June 2005 were studied. Growth faltering was defined as height <5th percentile or a drop in height z-score by >or= 1.5 standard deviations (SD) before age 18 months or by >or= 1 SD thereafter. The growth-faltering and nonfaltering groups were compared in terms of sex, race, age, number of clinic visits, and insurance, and by US census tract, socioeconomic status and parental education. Similar comparisons were made for children with height z-scores below -2.25 SD. RESULTS: Growth faltering was present in 3007 of the children studied (9%). Univariate and multivariate logistic regression analyses identified significant associations between growth faltering and younger age (P< .0001), Caucasian race (P< .0001), fewer clinic visits (P< .0001), and Medicaid insurance (P< .005), but not with sex nor by residential census tract, median income or proportion with less than high school education. Height below -2.25 SD was associated with male sex (P< .01), Medicaid insurance (P< .01), and more primary care visits (P< .0005). CONCLUSIONS: The sex disparity in subspecialty growth center referrals (2:1 male:female) is not due to male predominance in growth faltering among children in the urban primary care setting.