Objective: Pressure ulcer prevention begins with an accurate assessment of patient risk. While there are several valid and reliable pressure ulcer risk assessment tools for adults, none exist for infants and children. To remedy this, Curley and Quigley (J Soc Pediatr Nurses. 1(1):7-18, 1996) adapted the Braden Scale for use in pediatrics, calling it the Braden Q. The Braden Q is composed of seven subscales - the Braden Scale's original six (mobility, activity, sensory perception, moisture, friction/sheer, nutrition) and a tissue perfusion/oxygenation subscale. All seven subscales are rated from 1 to 4 (lower numbers least favorable). Each level is mutually exclusive, with only one choice per subscale. The range of possible scores for the Braden Q is 7 to 28 (lower numbers increased risk).The purpose of this study was to establish the predictive validity of the Braden Q in a critically ill pediatric population.
Design: Multisite prospective cohort descriptive study.
Population, Sample, Setting, Years: From 10/1998 to 7/2000, a convenience sample of 327 patients was enrolled from three separate pediatric intensive care units (PICU). To ensure an equal distribution of age within the sample, each site sequentially enrolled 25 patients in each of four age groups: infant, 21 days to 12 months; toddler, 12 to 36 months; preschool, 3 to 5 years; young school, 5 to 8 years. Inclusion criteria were PICU admission and bedrest for at least 24 hours. Exclusion criteria were patients with pre-existing pressure ulcers or congenital heart disease.
Intervention and Outcome Variable: The intervention was assessment of pressure ulcer risk using the Braden Q. The outcome variable was the development of a pressure ulcer staged according to the American National Pressure Ulcer Advisory Panel Consensus Development Conference recommendations.
Methods: Two nurses (Nurse I and Nurse II) observed patients up to 3 times/week for 2 weeks then once/week until PICU discharge. Both nurses were blind to the others' assessment and scores. Nurse I screened and enrolled patients, completed the nursing intervention checklist then scored patients using the Braden Q. Nurse II completed a head to toe skin assessment for pressure ulcers. If a pressure ulcer was identified, treatment was instituted per standard of care. Inter-rater reliability was established on the Braden Q and staging of pressure ulcers at the onset of the study and every other month thereafter. A minimum reliability coefficient of .80 was maintained throughout the study.
Findings: 327 patients were evaluated for a median of 2 (IQR: 1-4) observation periods reflecting 889 separate skin assessments. Subjects were 61% male, 67% white, developmentally appropriate for age, of low predictive risk of mortality (PRISM: 5±6). Eighty-seven patients (27%) developed 201 pressure ulcers; 142 (71%) were Stage I, 53 (26%) were Stage II and 6 (3%) were Stage III. Of the 59 Stage II/III, 34% (20) involved the head. Stage III pressure ulcers involved the occiput, ear, chest and coccyx. Most pressure ulcers (58%) were present at the first observation period (Day 2).
Using Day 2 Stage II+ pressure ulcer data, a Receiver Operator Characteristic (ROC) curve plotting sensitivity and 1-specificity for each possible score of the Braden Q was constructed. The area under the curve (AUC) was 0.83. A cutoff score of 16 provided a high sensitivity and adequate specificity. At a score of 16 the sensitivity was 88% and the specificity was 58% producing a Likelihood Ratio (LR=sensitivity/1-specificity) of 2.08.
The Braden Q was then modified to eliminate 4 subscales with an AUC <0.7 (activity, moisture, friction/shear, and nutrition).With 3 subscales (mobility, sensory perception, tissue perfusion/oxygenation) the AUC was maintained 0.84. A cutoff score of 7 provided a high sensitivity and adequate specificity. At a score of 7 the sensitivity was 92% and the specificity was 58% producing a LR of 2.21.
Conclusions: We report a 27% incidence of pressure ulcers in critically ill infants and children. Most pressure ulcers were present by the second PICU day. The performance of the Braden Q in a pediatric population is similar to that consistently reported for the original Braden Scale in adults. The modified Braden Q, with 3 subscales (mobility, sensory perception, tissue perfusion/oxygenation) provides a shorter yet comparable tool.
Implications: Risk assessment for pressure ulcers should start within 24 hours of PICU admission. The modified Braden Q with a cutoff score of 7 provides a parsimonious tool that predicts pressure ulcer risk in a vulnerable pediatric patient population. Improved assessment of patient risk may help reduce unnecessary variation in the prevention and management of pressure ulcers in critically ill infants and children.
Back to Intervention to Minimize Risk to Children
Back to The Advancing Nursing Practice Excellence: State of the Science