NUR 683 Module 7 Milestone Three Example

Milestone Three: A Measurement Plan for an Interruption and Scanning Bundle Using Statistical Process Control

[Student Name]

Southern New Hampshire University

NUR 683: Patient Safety and Quality Capstone

Module Seven Milestone Three

[Instructor Name]

[Date]

Milestone Three: A Measurement Plan for an Interruption and Scanning Bundle Using Statistical Process Control

The capstone bundle on the Ridgeview telemetry floor has four parts: a protected zone around medication preparation, redirection of calls during the morning pass, scanners and durable wristbands in every room and a just culture response to errors. This milestone explains how the project will know whether the bundle works. It covers the family of measures, how data will be collected, how they will be analyzed with control charts, what will count as a signal and how the data will be governed. The guiding principle is that every measure must be defined, sampled and charted identically in both periods.

Family of Measures

The outcome measures are clinical errors per 100 observed administrations and adverse drug events per 1,000 patient days found by trigger tool review. Process measures are the percentage of administrations interrupted, interruptions per administration and the share of doses where wristband and drug were each scanned. Balancing measures are the time taken to complete the morning medication pass and the time from a call light to a response during the pass, since redirecting interruptions could delay other care. Culture measures are four survey items on fear of blame and speaking up, collected at baseline and six months, and the monthly number of error and near-miss reports.

Table 1. Measures, Sources and Charts

Note. Each measure uses the same definition before and after the bundle.

Observation Sampling

Direct observation remains the core of the plan because it provides a true denominator and detects errors no one reports. Westbrook et al. (2017) used trained observers following nurses with a handheld tool that recorded each interruption and administration, and their trial showed that such an approach can compare wards reliably over time. The capstone adapts that approach on a smaller scale. Two trained observers, the student and a pharmacist, will observe 60 administrations each week: 25 on day shift, including the morning pass, 20 on evenings and 15 on nights. Observation times will rotate through the week, and observers will record interruptions, errors and scanning on a standard form checked against orders afterward.

Before the bundle begins, the two observers will independently code 40 administrations together to check agreement, aiming for at least 85% agreement on error classification. Disagreements will be resolved with a third reviewer, and coding rules will be written down.

Observers will also note the unit census and staffing on each observation day, so that a week with unusually heavy workload can be identified when interpreting the charts rather than being mistaken for an effect of the bundle.

Trigger Tool Review

Observation captures errors in progress but misses harm that appears later. Classen et al. (2011) showed that structured chart review using triggers, such as administration of naloxone, a glucose below 50 or a sudden stop of a medication, uncovered far more adverse events than voluntary reports. The capstone will review 20 randomly selected discharged records from the unit each month using medication-related triggers, and an event confirmed by the review will be counted as an adverse drug event. The monthly rate gives a harm measure independent of observation and reporting.

Analysis with Control Charts

Benneyan (2003) describes statistical process control as a way to ask whether a process is behaving as it always has or whether something new is acting on it. Every process wobbles from week to week; the question is whether a given wobble is larger or more patterned than its usual behavior would produce. A control chart answers that by plotting each value in time order around its average, with boundaries placed roughly three standard deviations above and below, worked out from how much the process itself varies. Values beyond those boundaries, or certain improbable sequences inside them, suggest a new influence. He also emphasizes choosing the chart that matches the data: p charts for proportions, u charts for rates when the number of opportunities varies and g charts for counting opportunities between rare events.

The capstone follows that guidance. Weekly interruption and scanning percentages will be plotted on p charts, weekly errors per 100 administrations and monthly adverse drug events on u charts and the number of administrations between high-alert errors on a g chart, where longer gaps indicate improvement. Limits will be calculated from eight baseline weeks and frozen, so later points are judged against baseline performance. Signals will be judged using standard rules: a point beyond a limit, a stretch of eight successive values that all land higher than the mean, or all lower, or a run of six that keeps climbing or keeps dropping.

Acting on Signals

Signals will prompt action, not just comment. A favorable special cause in a process measure after the bundle begins will be annotated and discussed at the weekly team meeting. An unfavorable signal in a balancing measure, such as the morning pass lengthening beyond its upper limit, will trigger a review within a week and possible adjustment of the call redirection. If error rates show no change after eight weeks despite improved process measures, the team will examine whether interruptions were the main driver or whether other factors, such as staffing, deserve attention.

Data Governance

Observation data will be recorded without nurse names, linked only to shift and date. Trigger tool reviews will use record numbers stored on a secure hospital drive. The project was reviewed by the hospital's quality improvement committee and judged not to require research ethics review, although nurses will be informed about observation and can decline. Charts will be shared at huddles monthly.

Conclusion

The plan combines direct observation, trigger tool review and system reports, analyzed with control charts suited to each type of data, so that the capstone can distinguish a real change from ordinary variation and detect any harm the bundle introduces.

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

Classen, D. C., Resar, R., Griffin, F., Federico, F., Frankel, T., Kimmel, N., Whittington, J. C., Frankel, A., Seger, A., & James, B. C. (2011). 'Global trigger tool' shows that adverse events in hospitals may be ten times greater than previously measured. Health Affairs, 30(4), 581-589. https://doi.org/10.1377/hlthaff.2011.0190

Westbrook, J. I., Li, L., Hooper, T. D., Raban, M. Z., Middleton, S., & Lehnbom, E. C. (2017). Effectiveness of a 'Do not interrupt' bundled intervention to reduce interruptions during medication administration: A cluster randomised controlled feasibility study. BMJ Quality & Safety, 26(9), 734-742. https://doi.org/10.1136/bmjqs-2016-006123