155 - Examining variability in clinical measurements of height and weight data within the first year of pediatric cancer treatment

Poster Number: 155
Publication Number: 155.22

Alexandra R. Roeder, University of Kansas Medical Center, Mission, KS, United States; Carolyn R. Bates, University of Kansas Medical Center, Kansas City, KS, United States; Billie Anderson, University of Missouri-Kansas City School of Medicine, Kansas City, MO, United States; Vincent Staggs, Children's Mercy KC / University of Missouri-Kansas City, Kansas City, MO, United States; Meredith Dreyer Gillette, Children's Mercy Hospitals and Clinics, Kansas City, MO, United States; Christie Befort, University of Kansas Medical Center, Kansas City, KS, United States; Keith August, Children;s Mercy Kansas City, Kansas CIty, MO, United States

Background: Valid height and weight measurements are critical for pediatric cancer treatment as body mass directly informs chemotherapy dosage. Large variability in child height and weight trajectories have been observed highlighting the importance of early detection of excessive weight gain or loss. Studies suggest that clinical measurements of child height and weight are susceptible to human error (Conkle et al., 2014), but this has not been examined in the pediatric oncology setting. Clinical observations indicate a need for innovative solutions for collecting valid and reliable height and weight data across and various settings and populations.

Objective: To examine this issue in pediatric oncology, we examined variability in pediatric height and weight measurements within the first year of cancer treatment. Our goal is to explore demographic characteristics that could be associated with greater measurement variability.

Design/Methods: Participants included 61 children (M age=7.9, SD=3.9) who were part of a larger study examining family functioning and child outcomes during the first year of cancer treatment. We extracted all height (cm) and weight (kg) data from the electronic medical record growth chart in the first year following diagnosis and location of each measurement. Patient demographics, including age, race, ethnicity, and diagnosis, were compiled via validated caregiver surveys (PAT 3.0; Kazak et al., 2018) and dichotomized for analysis: Age (2-8 vs 9-14 years); Race/Ethnicity (White, Non-Hispanic vs Black, Asian, Hispanic, or Other); Diagnosis (Leukemia and Lymphomas vs Brain, Central Nervous System, Solid, or Other Tumors).

Results: Potential correlates were analyzed across 2,140 height and 4,111 weight datapoints. Independent sample t-tests examined demographic predictors of between-measurement variability (SD) in height and weight. Age group was associated with variability in weight (t(31)= -5.00, p< .001) with older participants (9-14 years), evidencing greater between measurement variability in weight compared to the younger participants. Variability in weight was comparable across race/ethnicity groups and diagnosis types (p > .05). No demographic characteristics were associated with significant differences in between-measurement variability in height (p > .05).

Conclusion(s):

Demographic characteristics were not consistently associated with differences in between-measurement variability. A more in-depth analysis of the impact of age on weight variability is needed to account for higher baseline weight and normative growth rates in older participants.