Can a decision tree help nurses to grade and treat pressure ulcers?

This study investigated whether a decision tree can aid pressure ulcer management. It hypothesised that nurses who use a decision tree in pressure ulcer care assess wound grade more accurately and select more appropriate dressings.

Sixty-six nurses were randomly assigned to two groups. Each nurse reviewed written information and pictures of three cases of pressure ulceration, and then graded and selected a dressing for the lesions. The absence or presence of a decision tree was used as an independent variable. Measurements included: mean and standard deviation; confidence intervals; percentages; Chi-square test, Pearson's correlation and Kruskal-Wallis test.

Baseline characteristics were comparable for both groups. There was no statically significant difference between the grades selected by the two groups. But significantly more of the decision tree users selected an appropriate dressing (p < 0.02). Joint scores (ranging from 0 to 6) were assigned for each accurate decision. The mean score for the experimental group was 4.47 +/- 1.14 (95% CI: 0.95 [4.06-4.88]) and for the control group, 3.35 +/- 1.30 (95% CI: 0.95 [2.90-3.81]).

Decision trees could help nurses to make complex clinical decisions. Further studies undertaken in a clinical setting are needed.