NURS 8600 Week 7 DNP project run chart analysis example

Reading the Months: A Run Chart Analysis of Screening, Brief Intervention and Referral Offers Across Launch, Two Improvement Cycles and a Course Correction

Student Name

College of Nursing, Walden University

NURS 8600: DNP Field Experience

Instructor Name

Month Day, Year

Reading the Months: A Run Chart Analysis of Screening, Brief Intervention and Referral Offers Across Launch, Two Improvement Cycles and a Course Correction

Purpose and Method

This analysis presents the project's three process measures as run charts from the baseline months through September and interprets them with the run chart rules. Plotting one measure month by month against its median line, as a run chart does, allows a small set of probability-based rules to separate signals of real change from random variation without requiring the larger datasets that control charts need (Perla et al., 2011). Statistical process control methods, of which run charts are the simplest form, are well suited to improvement work because they show change over time rather than a single before and after comparison (Benneyan et al., 2003).

Each chart plots monthly proportions. A horizontal line marks the baseline median, calculated from the three sampled baseline months. Vertical annotations mark four events: launch on April 1, the first improvement cycle in mid-May and its spread in June, the second improvement cycle in early August and the course correction on September 1. The rules were applied in a fixed order (Provost & Murray, 2011). A shift needs at least six months in a row that sit together above the median or together below it. A trend needs at least five months in a row that each move the same way, up or down. The count of runs is then compared with the number expected by chance, and finally any single month far out of line with the rest is noted. All figures are illustrative values made up for a composite clinic.

Chart 1: Screening Documentation

Table 1 gives the values plotted on the first chart.

Table 1

Screening Documentation by Month

Shift. All six post-launch points lie above the baseline median of 25 percent, meeting the rule for a shift. Trend. The six post-launch points rise every month, meeting the rule for a trend. Interpretation. Screening documentation changed at launch and has continued to rise, although the monthly gains have narrowed since June, suggesting the measure is approaching a plateau in the mid-80s. The change in data source at launch, from a sampled review to a full count, is a caution: part of the jump may reflect better measurement. But a jump from about a quarter to over two thirds of visits is too large to be explained by the change in method alone.

Chart 2: Brief Intervention Documentation

Table 2 gives the values plotted on the second chart.

Table 2

Brief Intervention Documentation for Positive Screens by Month

Shift. All six post-launch points lie above the baseline median of 33 percent. The baseline rests on only seven positive screens, however, so this shift says little. A more useful comparison. Using the median of the first four post-launch months, 53.5 percent, as a working reference, the August and September points both lie above it. Two points do not meet any rule, and it would be premature to claim that the second cycle and the course correction produced a shift. Interpretation. The pattern is consistent with the midpoint finding that brief intervention stalled at about half and with the course correction improving it, but confirmation will need at least four more months above the reference line.

Chart 3: Referral Offers

Referral-eligible visits numbered four or five each month after launch and one or none each baseline month. With so few eligible visits, one patient more or less swings a month's percentage by a fifth or a quarter, and the run chart rules lose their meaning. The chart therefore plots counts, offers made beside eligible visits, rather than proportions: 1 of 4 in April, 2 of 5 in May, 4 of 5 in June, 3 of 4 in July, 3 of 5 in August and 3 of 4 in September. Offers have been made for more than half of eligible visits every month since the first cycle spread, compared with none at baseline. The pattern is encouraging but inconclusive, and the final report will say so.

Balancing Measure: Rooming Time

Median rooming time was 11 minutes at baseline and has been 12 minutes every month since launch. This small increase is real and consistent, and it should be reported rather than explained away. Staff have not described it as a burden, and visit length overall has not changed noticeably, but the clinic should know that the screen costs about a minute per visit. If rooming time rises further, particularly after the course correction adds provider time on phone-coverage days, it would be a signal to revisit the workflow.

Summary for Clinic Leaders

Screening documentation has clearly changed and continues to improve slowly. Brief intervention documentation improved at launch, stalled and may be improving again after the course correction; more months are needed. Referral offers have improved in count but are too few to analyze with run chart rules. Rooming time has risen by one minute. The charts will be updated monthly, with the course correction annotated, and the reference line for brief intervention will be recalculated once six post-correction months are available. The charts will also be shared at the monthly staff meeting, printed large with the annotations, since staff who made the changes should see what followed them. Staff at the August meeting asked why screening kept rising after launch, and the discussion suggested that the declined option and the medical assistants' growing comfort both contributed. Such explanations cannot be proven from the chart, but recording them beside it helps the next team understand the pattern. Keeping the annotations on every chart matters: without them, a reader could not link changes in the measures to the actions taken, and the charts would show only that something changed, not what.

References

Benneyan, J. C., Lloyd, R. C., & Plsek, P. E. (2003). Statistical process control as a tool for research and healthcare improvement. Quality and Safety in Health Care, 12(6), 458-464. https://doi.org/10.1136/qhc.12.6.458

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 and Safety, 20(1), 46-51. https://doi.org/10.1136/bmjqs.2009.037895

Provost, L. P., & Murray, S. K. (2011). The health care data guide: Learning from data for improvement. Jossey-Bass.