How We Will Know: An Evaluation Plan for a Blood Culture Collection Bundle, With Baseline, Measures, Intervals, and Decision Rules
Student Name
Doctor of Nursing Practice Program, Aspen University
DNP860: Evidence-Based Practice for Quality Improvement
Instructor Name
Month Day, Year
How We Will Know: An Evaluation Plan for a Blood Culture Collection Bundle, With Baseline, Measures, Intervals, and Decision Rules
An improvement project succeeds only if its team can tell, with confidence, whether the changes made things better. That requires a baseline, precisely defined measures, a schedule for collecting and displaying data, and rules for interpreting what the data show. This paper sets out the evaluation plan for the emergency department blood culture collection bundle, whose three components are introduced one after another.
Baseline
Baseline data come from 12 months of laboratory records before the first test of change: 9,600 culture sets with a contamination rate of 4.1 percent. The baseline was also examined month by month, because a single annual figure can hide variation. Monthly rates ranged from 3.3 to 4.9 percent with no trend, which suggests a stable process producing contamination at a consistent level. A stable baseline is important: it means that a shift after the intervention is more likely to reflect the change than preexisting drift.
The Outcome Measure
The primary outcome is the blood culture contamination rate, defined as the number of culture sets growing a common skin organism, such as coagulase-negative staphylococci, micrococcus, or diphtheroids, in only one of two or more sets drawn within 24 hours, divided by the total number of sets collected in the emergency department. The definition matches the one used at baseline and is applied by the microbiology laboratory, which reduces the risk that the measure changes during the project. The aim is to reduce the rate below 2 percent within nine months.
Process Measures
Process measures show whether the bundle is being carried out. They are the proportion of culture sets collected by separate venipuncture rather than through a new catheter, recorded in a field added to the collection documentation; the proportion collected using the sterile kit, tracked through kit use against culture orders; and, once introduced, the proportion collected with the diversion device. If the outcome does not improve, process measures reveal whether the problem is the bundle's effectiveness or its adoption.
Balancing Measures
The third set of measures watches for harm the bundle might cause. The first is the proportion of cultures that are true positives, since a change that reduced contamination by reducing the detection of real infection would be harmful; the diversion trial reported no loss of sensitivity (Rupp et al., 2017), but the project will confirm this locally. The second is the time from culture order to antibiotic administration for patients with suspected sepsis, since additional steps could delay treatment. The third is the number of needle sticks per patient, since separate venipuncture adds one for some patients. The fourth is supply cost per culture set.
Intervals and Displays
Weekly contamination rates will be plotted on a run chart for the team's weekly meeting, because weekly data give fast feedback during the testing phase, even though weekly counts are small and noisy. Monthly rates will be plotted on a p-chart, a control chart for proportions, for reporting to leadership and for judging whether change is real. Process measures will be plotted weekly during testing and monthly afterward. Balancing measures will be reviewed monthly. Every chart will be annotated with the date each change was introduced. Annotations will also mark events outside the project that could affect the data, such as a new laboratory instrument or a large group of newly hired nurses.
Decision Rules
Statistical process control distinguishes common cause variation, the ordinary scatter any stable process produces, from special cause variation, which signals that something has changed (Benneyan, 2003). The p-chart will use limits calculated from the 12 baseline months. Special cause will be flagged by the standard rules: any point beyond a control limit, eight or more successive points all falling above, or all below, the average, or six or more points moving steadily in one direction (Provost & Murray, 2011). A run of points below the baseline center line after the bundle is introduced would signal improvement. When a sustained shift is confirmed, the center line and limits will be recalculated to reflect the new process. The rules are written before the data arrive, so that the team cannot choose, after the fact, the interpretation it hoped for.
Who Collects, Reviews, and Acts
Measurement only helps if someone owns each step. The microbiology supervisor will send a weekly file of culture results by collection date, collector, and method each Monday. The DNP student will calculate the rates and update the charts by Tuesday, and the project team will review them at its Wednesday meeting, deciding whether the current cycle continues, changes, or ends. Monthly, the nurse manager will present the p-chart and balancing measures at the emergency department's quality meeting. Should a harm signal appear, such as a rise of more than 15 minutes in the median time to antibiotics for suspected sepsis, the team will pause the current component and investigate before continuing. Data will be stored on the department's secure drive without patient identifiers beyond what the laboratory report requires, and only the team will have access.
Duration and Limitations
Data collection runs on for a full year once all three components are in place, as a test of whether gains hold, and the p-chart will continue to be reviewed quarterly by the department's quality committee afterward. The design has limits: without a control group, other changes, such as new staff or a new laboratory system, could explain improvement. Annotating the charts with such events and comparing with another emergency department in the system, which is not implementing the bundle, will help. An interrupted time series design like the one used in a similar emergency department study, where segmented regression separated the immediate drop after the change from the underlying trend, offers a model for the final analysis (Self et al., 2013).
Conclusion
This evaluation plan specifies a stable baseline, an operational outcome measure, process and balancing measures, intervals and displays for each, and decision rules fixed before data arrive. Together, they will let the team and leadership judge whether the bundle reduces contamination without causing harm, and whether any improvement lasts.
References
Benneyan, J. C. (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
Provost, L. P., & Murray, S. K. (2011). The health care data guide: Learning from data for improvement. Jossey-Bass.
Rupp, M. E., Cavalieri, R. J., Marolf, C., & Lyden, E. (2017). Reduction in blood culture contamination through use of initial specimen diversion device. Clinical Infectious Diseases, 65(2), 201-205. https://doi.org/10.1093/cid/cix304
Self, W. H., Speroff, T., Grijalva, C. G., McNaughton, C. D., Ashburn, J., Liu, D., Arbogast, P. G., Russ, S., Storrow, A. B., & Talbot, T. R. (2013). Reducing blood culture contamination in the emergency department: An interrupted time series quality improvement study. Academic Emergency Medicine, 20(1), 89-97. https://doi.org/10.1111/acem.12057
How this DNP 860 Module 7 example is structured
DNP860 Module 7 papers often build an evaluation plan with baseline, measure and interval. Aspen does not publish module deliverables, so check your classroom for the exact prompt. This example establishes a baseline, defines outcome, process and balancing measures, assigns intervals and displays, fixes decision rules in advance and addresses duration and limitations.
DNP860 Module 7 questions, answered
What does DNP860 Module 7 usually ask for?
The module often asks for an evaluation plan for your improvement project, specifying the baseline, measures, how often data will be collected and how results will be interpreted. Aspen does not publish module deliverables, so your classroom's instructions govern.
What is the difference between outcome, process and balancing measures?
Outcome measures show whether the goal was reached, process measures show whether the changes were carried out, and balancing measures check that the changes did not cause harm elsewhere.
What is special cause variation?
Variation that signals the process has changed, identified on a control chart by rules such as a point outside the control limits or a run of eight or more points on one side of the center line.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official Aspen University document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.