QI Journal Club

Analysis/Journal Club of QI article

Facilitating the Timely Discharge of Well Newborns by Using Quality Improvement Methods

Manuscript citation: Rochester NT, Banach LP, Hoffner W, et al. Facilitating the Timely Discharge of Well Newborns by Using Quality Improvement Methods. Pediatrics. 2018;141(5): e20170872

Type of article

Quality improvement

Methods

Design: Retrospective and prospective study to increase the percentage of newborn discharge order by 11am.

Blinding: N/A

Setting: single center

Patients: well newborns

Intervention: implementation of a discharge checklist

Primary measure: Increase placement of discharge order from 24% to 36% by 11am.

Commentary:

General overview, this is a well-written and well-designed QI initiative with sound methodology including team composition, aims and data analysis. The SQUIRE guidelines should be followed in order to publish quality improvement work.[1]

The study is generalizable for other academic institutions as the workflow is similar. The main limitation to this paper is the lack of balancing measures. In the case of newborn discharges, routine screening includes the following to be done prior to discharge: hearing screen, hepatitis b vaccine administration, inborn metabolic newborn screening, critical congenital heart disease screening and in some cases car seat challenges. It is important to determine whether any of these were missed due to earlier discharges.  

Introduction to define the purpose and importance of the project and to include the principles of models of improvement. It is important to include 5 key principles to each project. These include the following: is it realistic and relevant, specificity, measurable outcomes, is it achievable in the time allotted, and there should be a designated timeline. The authors described the importance of timely newborn discharges as it relates to hospital throughput.

Ideally, at least 12 data points are used for a fair assessment of a baseline. The authors described baseline data with 24 data points before the initiation of the study period. This allowed the reader to understand their initiation percentage was a steady state. The aim of this initiative was to increase the percentage of newborns with a discharge order by 11am by 50% compared with the baseline in 18 months. The timeline is stated in Figure 1, not specifically in the manuscript.

Figure 1, the key diagram, is an excellent visualization to display the different variables that affect the project aim. The use of a pareto chart is an opportunity to determine which factors make the most effects on the outcome. It is also an easy visual display to discuss with the team and stakeholders. Two out of the first three most common causes were “missing data” and “other” that made up almost 40% of the delays. It should be explained in better detail and possibly broken down to be addressed individually. More information may have offered an improved workflow method.

Interventions:

In this initiative, it is described that checklist education was the 1^st PDSA cycle and the 2^nd PDSA cycle was improved attending workflow. Each of these initiatives could be broken down in further details to more cycles. More description is needed for another team to replicate this initiative. The discharge checklist show in the supplement is a helpful tool. It is unclear how the documentation was shared. The description of the improved workflow could be described in more detail.

The authors described the local context and its impact on the initiative very well. This included per diem attending staff and monthly new interns and residents.

Results:

The baseline data shows a good representation of multiple data points to assure an initiative needed to be done. The statistical process control (SPC) chart displayed in Figure 2 is a good measure to determine what interventions affect the outcome. The paper describes the baseline data at 24%; however, the initial central line in the SPC chart is 25%. In the SPC chart, it is unclear why some data points are blue squares and others are in red diamonds. The authors described why the mean shifted after 8 consecutive data points were above the central line. This description within the paper helps the reader understand more information on the use of SPC charts. The final central line improved to 40%; however, the results state early discharge improved to 39%. The annotations on the SPC chart should distinguish each PDSA cycle. It is unclear the number of charts reviewed each month during the baseline and prospective chart review. The straight upper and lower control limits indicate the number of monthly charts were the same. According to the graph, it appears the monthly feedback ensured sustainability. It does not indicate who completed the data abstraction sheets.

Balancing measures are a key component to QI initiatives. There are no balancing measures discussed in this article.

The authors pointed out that communication is a key factor for success. They do not indicate other methods to consider for improved results, such as an embedded hand-off tool or including the sheet in the electronic medical record. Additionally, it may be possible to include the night staff team and address some of the deficiencies at that time.

The discussion was specific to the authors institution and would be more helpful to others if it could be more general. It was well reflected and offered suggestions on how things may have gone better.

Conclusions were very concise. However, it also mentioned “to maintain these improvements adjustments will be needed”. This statement may be better in the discussion section and then elaborate on what adjustments could happen in the future to ensure sustainability.

1. Davidoff F, Batalden P, Stevens D, Ogrinc G, Mooney S, Group SD. Publication guidelines for improvement studies in health care: evolution of the SQUIRE Project. Ann Intern Med 2008;149(9):670-6 doi: 10.7326/0003-4819-149-9-200811040-00009[published Online First: Epub Date]|.