NR 542 Week 4 Reporting Plan Example

Reporting Plan: A Monthly Early ICU Transfer Report Whose Rules Live in the Document Itself, With a Numbered Build Sequence, Reconciliation Checks, a Change Log and a Rule That the Second Run Must Match the First

Student Name

College of Nursing, Chamberlain University

NR 542: Managing Data and Information

Instructor Name

Month Day, Year

Reporting Plan: A Monthly Early ICU Transfer Report Whose Rules Live in the Document Itself, With a Numbered Build Sequence, Reconciliation Checks, a Change Log and a Rule That the Second Run Must Match the First

Purpose

What follows is the method the hospital will use to produce a monthly report of the early ICU transfer measure, a count of moves to intensive care in a patient's first 24 hours on a ward, among adults admitted to hospital medicine through the emergency department. The report replaces the undefined "eight percent" figure quoted at quality meetings. Nobody trusts a report whose second run disagrees with its first, and the fastest way to earn that distrust is to leave the rules somewhere other than the document: in an analyst's head, a spreadsheet formula or an undocumented query. This plan keeps every rule in the report's own documentation, so that anyone rerunning it, or taking it over, gets the same answer.

The Measure, Stated Once

Numerator. Patients in the denominator with a bed movement into an intensive care unit within 24 hours of first arrival on a medical-surgical unit, excluding transfers for scheduled procedures requiring intensive care monitoring.

Denominator. Encounters of adults aged 18 or older, admitted through the emergency department, with an inpatient class, including observation stays later converted, assigned to hospital medicine at first floor arrival, with a first floor arrival in the reporting month, excluding patients with a comfort care order signed before floor arrival.

Reporting. Monthly count and rate, each with its confidence interval, with a rolling 12-month rate, since monthly counts will be small.

The measure is stated once, here, and every other part of the plan refers to it rather than restating it.

The measure was reviewed by the hospital medicine quality lead and the emergency department's medical director before this plan was written, and both signed the definition. That step matters for trust as much as for accuracy: when a figure later surprises someone, the first question is usually whether the definition was fair, and a signed definition answers it.

The Build Sequence

The report is built in seven numbered steps, each a separate, saved query with a comment block repeating the step's rule in words.

Step 1. Select encounters meeting the age, admission source and patient class rules.

Step 2. From the bed movement table joined to the unit table, find each encounter's earliest movement into a unit of type medical-surgical. This time is time zero. Encounters with no such movement are dropped.

Step 3. Keep encounters whose team assignment at time zero is hospital medicine.

Step 4. Drop encounters with a comfort care order signed before time zero.

Step 5. For remaining encounters, find any movement into a unit of type intensive care within 24 hours after time zero.

Step 6. Remove from the numerator any encounter with a scheduled procedure requiring intensive care monitoring within the same 24 hours.

Step 7. Count, calculate the rate and confidence interval, and write the results with the run date, the database refresh date and the version number of the rules.

Reconciliation Checks

Every run includes four checks, and a report is not released if any fails. First, the denominator count must be within 5% of the count of emergency department admissions to hospital medicine in the hospital's official admissions report for the same month; a larger difference means a rule or a table has changed. Second, the numerator must match a hand count of intensive care transfers for five randomly chosen encounters, checked by the nurse informaticist against the bed movement history. Third, the unit type table must be checked against the current list of units, since one unit was reclassified last year without the table being updated. Fourth, a rerun of the previous month using the current version of the rules must reproduce the previously published figure exactly, unless the change log records a rules change.

Kahn and colleagues' harmonized framework separates conformance, completeness and plausibility checks (Kahn et al., 2016). The first and third checks test conformance and plausibility; the second tests correctness against a source; the fourth tests reproducibility, which is what the plan exists to protect.

Change Log and Rules Ownership

The rules have a version number. Any change, such as a new exclusion or a reclassified unit, must be approved by the hospital medicine quality lead and recorded in a change log with the date, the reason, the person approving it and the effect on the most recent figure. When the rules change, the previous 12 months are rerun under the new rules, and the report shows both the old and new figures for one quarter so that readers can see the effect of the change rather than mistaking it for a change in care.

Operational record data are shaped by the purposes and workflows that produce them, and changes in those workflows can change a measure without any change in patients (Hersh et al., 2013). The change log is how the report keeps changes in rules or workflow from being misread as changes in care.

Schedule and Distribution

Each month's run happens once the month-end refresh is finished, normally in the first week, after the reporting database has completed its month-end refresh. It is distributed to the hospital medicine director, the emergency department director and the quality committee, with a one-paragraph note explaining any unusual result and any rules change. The full documentation, including this plan, the queries and the change log, is stored in the quality department's shared folder, not on an analyst's computer.

What the Report Does Not Claim

The report counts transfers; it does not judge whether admission decisions were appropriate. Each monthly note will say so. The report also depends on the accuracy of bed movement times, which the data quality audit will examine. Research on data quality dimensions emphasizes that a measure can be no better than the completeness and correctness of the fields beneath it (Weiskopf & Weng, 2013). The first release will be accompanied by the audit results, so that readers know how much confidence the figure deserves.

The first release will also invite readers to send questions about the definition to the quality lead, and every question received in the first three months will be answered in writing and added to the documentation, so that the rules become clearer each time someone reads them closely.

References

Hersh, W. R., Weiner, M. G., Embi, P. J., Logan, J. R., Payne, P. R., Bernstam, E. V., Lehmann, H. P., Hripcsak, G., Hartzog, T. H., Cimino, J. J., & Saltz, J. H. (2013). Caveats for the use of operational electronic health record data in comparative effectiveness research. Medical Care, 51(8, Suppl. 3), S30-S37. https://doi.org/10.1097/MLR.0b013e31829b1dbd

Kahn, M. G., Callahan, T. J., Barnard, J., Bauck, A. E., Brown, J., Davidson, B. N., Estiri, H., Goerg, C., Holve, E., Johnson, S. G., Liaw, S.-T., Hamilton-Lopez, M., Meeker, D., Ong, T. C., Ryan, P., Shang, N., Weiskopf, N. G., Weng, C., Zozus, M. N., & Schilling, L. (2016). A harmonized data quality assessment terminology and framework for the secondary use of electronic health record data. eGEMs, 4(1), Article 18. https://doi.org/10.13063/2327-9214.1244

Weiskopf, N. G., & Weng, C. (2013). Methods and dimensions of electronic health record data quality assessment: Enabling reuse for clinical research. Journal of the American Medical Informatics Association, 20(1), 144-151. https://doi.org/10.1136/amiajnl-2011-000681