Validity and reliability of the neonatal skin condition score

Objective: To demonstrate the validity and reliability of the Neonatal Skin Condition Scale (NSCS) used in the Association of Women's Health, Obstetric and Neonatal Nurses (AWHONN) and the National Association of Neonatal Nurses (NANN) neonatal skin care evidence-based practice project. Setting: NICU and well-baby units in 27 hospitals located throughout the United States. Participants: Site coordinators (N = 27) and neonates (N = 1,006) observed during both the pre-and postimplementation phases of the original neonatal skin care project. Method: To assess reliability two consecutive NSCS assessments on a single infant were analyzed. Site coordinators were contacted after the original project was concluded. Sites indicating that a single nurse scored all infant skin observations provided data that were used to evaluate intrarater reliability. Sites using more than one nurse to score skin observations provided data that were used to assess interrater reliability. To assess validity the following variables were used from the original data set: the Neonatal Skin Condition Scale (NSCS), with three subscales for dryness, erythema, and breakdown; birth weight in grams; number of skin score observations for each infant; and the prevalence of infection, defined as a positive blood culture. Results: For intrarater reliability, 16 sites used a single nurse for all NSCS assessments; total NSCS assessments 475. For interrater reliability 11 sites used multiple raters; total assessments 531. The NSCS demonstrated adequate reliability for each of the three subscales and for the total score, with the percent agreement between scores ranging from 68.7% to 85.4% (intrarater) and 65.9% to 89% (interrater); all Kappas were significant at p <.001 and were in the moderate range for reliability. The validity of the NSCS was demonstrated by the findings that smaller infants were 6 times more likely to have erythema (chi((6))(2) = 109 55 p <.0001), and approximately twice as likely to have the most severe breakdown (chi((6))(2) 108.01, P <.0001). infants with more observations (longer length of stay) had higher skin scores (odds ratio = 1.21, p <.0001), and an increased probability of infection was noted for infants with higher skin scores (odds ratio = 2.25, p <.0001). Conclusions: The Neonatal Skin Condition Score (NSCS) is reliable when used by single and multiple raters to assess neonatal skin condition, even across weight groups and racial groups. Validity of the NSCS was demonstrated by confirmation of the relationship of the skin condition scores with birth weight, number of observations, and prevalence of infection. JOGNN, 33, 320-327; 2004.