Communication board decreased nurses' perceived time to interpret nursing tasks (2.13 to 3.29, p=0.002)

Description #

After implementing the communication board, nurses reported a perceived improvement in having adequate time to interpret peri-operative nursing tasks, with the mean 5-point Likert rating rising from 2.13 (± 0.83) pre-intervention to 3.29 (± 0.83) post-intervention (p = 0.002) (Figure 3). The change was statistically significant by Wilcoxon rank-sum test. (Note: the survey item is framed as having "adequate time," so the higher post score reflects a perceived reduction in the time burden of interpreting.)

"Lastly, nurses reported a perceived decrease in the time required to interpret nursing tasks (2.13 ± 0.83 pre; 3.29 ± 0.83 post; p = 0.002) (Figure 3)." (Kiblinger, 2022, p. 22)

Methods Context #

What? #ⓘ

The observable: nurse-reported adequacy of time to communicate effectively about pre- and post-operative nursing tasks for LEP patients, a single 5-point Likert survey item.

"Sample pre-survey items included the following:... (3) I have adequate time to communicate effectively regarding pre- and post-operative nursing tasks for LEP patients/families." (Kiblinger, 2022, p. 20)

How? #ⓘ

Single-group pre-post survey design evaluating the A-0014ArtifactA-0014Initial AI draftInteractive pictorial communication board (pediatric surgery center)An interactive pictorial communication board deployed in a pediatric ambulatory surgery center to let bedside nurses convey routine peri-operative nursing tasks to limited-English-proficiency (LEP) patients and caregiver…; pre- vs post-implementation nurse responses compared with Wilcoxon rank-sum tests in R.

"Wilcoxon rank-sum tests were performed to determine statistical significance between pre- and post-survey staff nurse responses." (Kiblinger, 2022, p. 22)

Who? #ⓘ

Staff nurses at a single large, urban pediatric ambulatory surgery center; 15 completed the pre-survey and 14 the post-survey (of 17 surveyed).

"Surveys were distributed to all surgery center staff nurses (n = 17). Pre-surveys were completed by 88% of nurses (n = 15), and 82% of nurses completed the post-survey (n = 14)." (Kiblinger, 2022, p. 22)

Other Notes #

Plotted in the Figure 3 pre/post comparison. On a post-only item, 64% of nurses agreed the tool decreased the time spent interpreting nursing tasks (3.86 ± 0.77) — captured in the acceptability EVD.

Caveats #

• Uncontrolled pre-post design with concurrent bilingual-staff availability confounds attribution (Kiblinger board) The single-group pre-post design has no concurrent control arm, so improvements cannot be cleanly attributed to the board rather than to co-occurring factors. In particular, a bilingual staff member was available at the surgery center and was viewed by nurses as the most useful interpreting resource both before and after the intervention, providing an alternative explanation for perceived communication gains and for the low board utilization. [Inferred: the confounding/attribution weakness of the uncontrolled design is not stated by the authors as a formal limitation; the authors do note the bilingual-staff availability as a factor in the satisfaction–utilization contrast.]
• Small, homogenous, single-center nurse sample limits generalizability (Kiblinger board) The project drew on a very small, homogenous nurse sample (15 pre / 14 post of 17 nurses) at a single ambulatory surgery center within one large hospital system, so the effect estimates are imprecise and the findings may not transfer to other practice settings, patient populations, or non-surgical environments.
• Self-reported outcomes vulnerable to social-desirability bias (Kiblinger board) All effectiveness outcomes are nurse self-reports on Likert surveys and are subject to social-desirability bias: nurses may have felt obligated to report improvement after the intervention. The authors themselves flag the striking contrast between the highly positive survey responses and the low actual board utilization as a signal that the satisfaction gains may be inflated.