DNP 851B Module 4 Run Chart Analysis Over Time Example

Signal Above the Median: A Run Chart Analysis of Action Plan Coverage and Acute Asthma Visits

Student Name

Doctor of Nursing Practice Program, Aspen University

DNP 851B: Project Data Analysis

Instructor Name

Month Day, Year

Signal Above the Median: A Run Chart Analysis of Action Plan Coverage and Acute Asthma Visits

Pre-post tests compare two moments. A run chart shows the whole path between them, and it can reveal whether a change began when the intervention began or was already under way. This paper presents run chart analyses for my DNP project, a nurse-delivered program of written plans and device coaching for children whose asthma is persistent. Written plans are recommended for every patient with asthma (National Asthma Education and Prevention Program, 2007), so plan coverage is a natural clinic-level measure. The paper examines two such measures over 20 months: the share of the 640 children on the asthma registry with a written plan documented in the previous 12 months, and the monthly number of emergency or urgent care visits for asthma among registry children. It explains the rules used to read the charts, reports what they show and states what they add to the pre-post tests.

Why a Run Chart

A run chart plots a measure in time order around its median, so that patterns rather than single points can be judged (Perla et al., 2011). It suits improvement work because it needs no assumptions about the distribution of the data and because clinic staff can read it without statistical training. The tests reported in Module 3 describe the children who enrolled; the run charts describe the clinic as a whole, including children who never enrolled, which makes them a check on whether the program changed the clinic's performance and not only the enrolled children's.

Rules for Reading the Charts

Two rules were used, chosen on evidence rather than habit. A simulation study of run chart rules found that the shift rule, which looks at how many consecutive points sit above or below the median, and the crossings rule, based on how often the line crosses the median, detect real changes while keeping false signals near 5%, and that the trend rule, which looks for consecutive rises or falls, adds almost nothing (Anhøj & Olesen, 2014). The trend rule was therefore not used. Points falling exactly on the median were set aside, as the method requires. For the 16 remaining points, a run longer than seven on one side of the median, or fewer than four crossings, was treated as a signal of non-random variation, following the limits that study derives from the number of useful points. The median was fixed at the 12 baseline months so that project months would be judged against it.

Action Plan Coverage

The table shows the share of registry children with a current written action plan by month. In the 12 baseline months, coverage ranged from 29% to 33%, with a median of 31%. In the eight project months it rose steadily, from 33% to 47%.

With the four points on the median set aside, the eight project months form an unbroken run above it, longer than the limit of seven, so the shift rule signals a non-random change beginning at the start of the project. The crossings rule does not signal on its own: the line crossed the median five times, all during the baseline year, which is above the lower limit for a chart of this length. One signal is enough to conclude that the process changed, and the size of the change is plain: coverage rose by about 16 points in eight months. The 104 enrolled children account for most of that rise, with the remainder coming from plans written at routine visits once the template and the front-desk process were in place.

Acute Care Visits

Monthly acute care visits for asthma among registry children vary strongly with the season, rising in autumn when viruses circulate and schools reopen. A run chart built on a single median would confuse the season with the program. The project months were therefore compared with the same calendar months a year earlier. Across the eight project months there were 74 visits, against 93 in the same months of the previous year, a fall of 20%. Six of the eight months were lower than the year before, one was the same and one was higher. With so few points and strong seasonality, no run chart rule applies, and the result is reported descriptively as consistent with the pre-post finding but not as a separate signal.

What the Charts Add

The charts add two things to the pre-post tests. First, they show that action plan coverage changed when the program began rather than drifting upward beforehand, which makes a pre-existing trend an unlikely explanation. Second, they show that the change reached the clinic's whole registry, not only enrolled children. They cannot rule out other changes that began in the same month, and the log records none of note.

The charts also serve the clinic after the project ends. Each is annotated with the dates of the adaptations recorded during implementation, such as the change in who sends plans to schools, so that staff can see how their work shaped the line. The clinic manager has asked for the coverage chart to be updated monthly as part of the clinic's quality report, which turns a project analysis into a routine measure.

Conclusion

Run chart analysis, using the rules that simulation evidence supports, found a clear signal of improvement in action plan coverage beginning with the project and a fall in acute care visits compared with the same months of the previous year that is consistent with, but weaker than, the enrolled children's results. Together with the pre-post tests, the charts strengthen the case that the program changed care at the clinic.

References

Anhøj, J., & Olesen, A. V. (2014). Run charts revisited: A simulation study of run chart rules for detection of non-random variation in health care processes. PLOS ONE, 9(11), Article e113825. https://doi.org/10.1371/journal.pone.0113825

National Asthma Education and Prevention Program. (2007). Expert panel report 3: Guidelines for the diagnosis and management of asthma (NIH Publication No. 07-4051). National Heart, Lung, and Blood Institute.

Perla, R. J., Provost, L. P., & Murray, S. K. (2011). The run chart: A simple analytical tool for learning from variation in healthcare processes. BMJ Quality & Safety, 20(1), 46-51. https://doi.org/10.1136/bmjqs.2009.037895