ASSESSING INTER-RATER RELIABILITY (IRR) OF TANNER STAGING AND ORCHIDOMETER USE IN BOYS: A STUDY FROM PROS

Eric J. Slora, Alison B. Bocian, Marcia E. Herman-Giddens, Steven E. Pedlow, Steven Dowshen, Richard C. Wasserman.. PROS, Dept of Research, American Academy of Pediatrics, Elk Grove Village, IL; UNC School of Public Health, Chapel Hill, NC; NORC, Chicago, IL; AI DuPont Hosp for Children, Wilmington, DE; Pediatrics, Univ of Vermont, Burlington, VT.

BACKGROUND: Few data exist on IRR of Tanner staging in boys.

OBJECTIVE: To assess staging reliability for selected key markers of male puberty.

DESIGN/METHODS: Two raters in each of 8 PROS practices independently rated consecutive boys aged 8-14 years presenting for complete physical exams on key pubertal markers: pubic hair and genitalia (both on 5-point ordinal Tanner scale), left and right testicular size (=1, 2, 3, =4 ml with modified orchidometer), and axillary hair (3-point ordinal scale: none, sparse, mature). Descriptive analyses were done to characterize the study sample. Intraclass correlation coefficients (ICCs) were generated to assess agreement in assigning stages. Kappas were calculated to assess agreement regarding stages representing initiation of pubertal maturation.

RESULTS: The sample comprised 79 boys (mean age 10.4 years) from practices in 7 states. Regarding race, 54.5% were white, 18.2% black, 11.7% Asian, 2.6% Hawaiian/Pacific Islander, and 13.0% multiracial, with 16.9% of Hispanic ethnicity. Twenty-eight percent reported a chronic disease (46% of those had asthma), and none of the 79 presented for growth or pubertal concerns. ICCs were all significant at the p<.001 level, including .61 (left testicle size), .69 (right testicle size), .77 (axillary hair), .83 (genitalia), and .94 (pubic hair development). Kappas ranged from good to excellent, including .49 (genitalia), .66 (left testicle size), .75 (right testicle size), .79 (pubic hair development), and .78 (axillary hair).

CONCLUSIONS: Practitioners were able to reliably stage key markers of puberty. Caution is advised regarding genitalia rating, as examiners were not as reliable in identifying signs of puberty for this dimension. This is consistent with the subjective nature of genital inspection for staging sexual maturation as opposed to more objective measures, such as testicular volume measurement. Nevertheless, establishment of IRR by pediatricians of Tanner staging and testicular size provides a sound foundation for future studies of boys' secondary sexual development.