Background: Interdisciplinary communication is essential for safe and high-quality patient care. Outside of bedside rounds, paging is the most used means for communication to clinicians (physicians and advanced practice providers (APPs)). Quality improvement (QI) initiatives have aimed to improve patient safety by expanding technology and standardizing paging communications as well as to characterize what types of information are being expressed through text-pages. However, studies examining quantitative aspects of paging and linking them to QI in interdisciplinary communication are lacking. We hypothesized that using a quantitative analysis of paging records to identify targeted improvement interventions would improve adherence to best practices in paging communications.

Methods: We conducted a retrospective analysis of all pages sent to hospitalist clinicians over a 7-day period to characterize volume, content, and effectiveness both pre- and post-QI intervention. Pages were excluded if they were automatic pages (for lab results, cardiac arrests), personal pages from colleagues, pages for admission, from radiology, or consults. Content of page was categorized into seven subjects: labs, medications, vital signs, diet, patient assessment/clinical change, pain or miscellaneous/other. Pages were also analyzed for five critical elements: 1) two patient identifiers with patient name and medical record number (MRN), 2) the sender’s name, 3) sender’s callback number, 4) priority or acuity of the page (our hospital has four – FYI, please advise, please call, and please come), and 5) the patient-care concern. Pages were considered successful if they contained all five essential elements. A Lean Six Sigma methodology was applied to this QI project using the pre-intervention results to guide targeted interventions to increase adherence to best practices. Three months post-intervention, the post-QI paging records were obtained and analyzed. The data were summarized using frequency and percentages. Differences between pre-and post-intervention variables were calculated using Pearson’s chi-square test. To examine the effect of the intervention, we used negative binomial analysis with robust standard errors. A p-value of 0.05 was considered significant. Analyses were performed using Stata 17.0 (College Station, Texas).

Results: Of 3483 included pages, 1806 and 1677 were sent during the pre- and post-intervention periods, respectively. During both periods, the percentage of pages related to diet remained higher than other subjects (Figure). Compared to the pre-intervention period, the post-intervention period had significantly fewer pages related to all subjects except for pages related to pain (3% vs. 7%; p< 0.001). We found a statistically significant increase in the presence of three paging elements out of five (Table). Post-intervention, there was a 42% increase in the presence of the name and MRN (IRR=1.42, 95%CI: 1.34 to 1.50, p< 0.001), 64% increase in the presence of priority (IRR=1.64, 95%CI: 1.51 to 1.78, p< 0.001), and 1.5% increase in the presence of concern (IRR=1.015, 95%CI: 1.00 to 1.02, p=0.007). Overall adherence pre-intervention was 15.2% and increased significantly to 40% post-intervention (Table).

Conclusions: A rigorous, interdisciplinary quantitative analysis of paging records for a hospitalist service effectively guided quality improvement interventions to improve adherence to best practices.

IMAGE 1: Summarized Findings for Critical Page Elements Pre- and Post-Intervention Periods

IMAGE 2: Kernel Density Plot of Pages Organized by Subject and Hours for the Pre- and Post-Intervention Periods