NR 542 Week 2 Data Question Brief Example

Data Question Brief: Turning "Are We Putting Patients on the Floor Who Should Have Gone to the ICU?" Into Two Questions the Record Can Answer, and One It Cannot

Student Name

College of Nursing, Chamberlain University

NR 542: Managing Data and Information

Instructor Name

Month Day, Year

Data Question Brief: Turning "Are We Putting Patients on the Floor Who Should Have Gone to the ICU?" Into Two Questions the Record Can Answer, and One It Cannot

The Request

The director of hospital medicine asked me a question in ordinary language: "Are we putting patients on the floor who should have gone to the ICU?" The question comes from a worry, prompted by a figure quoted at quality meetings that eight percent of medicine admissions go to the intensive care unit within a day, and by two recent patients who deteriorated within hours of arriving on a medical floor. Managers ask about what worries them; a database answers only what it was asked. This brief sits between the two. It translates the director's worry into questions the hospital's record can answer precisely, states the choices made in doing so and names the part of the worry no data query can settle.

What the Worry Contains

The director's question contains at least three distinct questions. The first is descriptive: how often do patients admitted to medical floors move to the ICU soon after admission? The second is comparative: is our rate higher than it should be, or higher than it used to be? The third is about judgment: were those patients too sick for the floor at the moment the admission decision was made? The first two can be answered from the record. The third cannot be fully answered from any database, because it depends on what the admitting team knew and saw, which is only partly recorded.

Question 1: How Often

Population. Adults aged 18 or older admitted as inpatients to a hospital medicine team on a non-ICU bed through the emergency department during the 12 months from July to June. Patients in observation status are included if they were later converted to inpatient status within the same encounter, because they occupy the same beds and are cared for by the same teams.

Time zero. The time the patient arrived on the medical floor, taken from the bed movement record. Time zero is not emergency department arrival or the admission order, because the question is about what happens after the patient reaches the floor, and patients may board in the emergency department for many hours.

Outcome. Transfer to an ICU bed within 24 hours of time zero, taken from the bed movement record.

Exclusions. Transfers to the ICU for planned procedures requiring ICU monitoring, identified from the procedure schedule; patients whose goals of care were limited to comfort care at admission; and transfers to step-down units, which are reported separately.

Measure. The number of patients meeting the outcome divided by the population, reported with the count, the rate and a 95% confidence interval.

A large integrated health system that studied unplanned ICU transfer within 24 hours of admission from the emergency department reported a rate of 2.4% among non-ICU admissions (Delgado et al., 2013). Our definition is close enough to theirs to allow a cautious comparison, and the differences are listed so that readers can judge.

Separating the three questions also helps the director decide what to do with the answers. If early transfers are rare, the worry may be driven by two memorable cases, and the response is case review rather than policy change. If they are common, or rising, the next question is where they concentrate, by time of day, diagnosis or admitting team. The descriptive question has to be answered first, because every other decision depends on its size.

Question 2: Compared With What

The comparative question will be answered two ways. First, against our own history: the same definition applied to the previous two years, to see whether the rate has changed. Second, against the published rate above, with a clear note of differences in population and definition. A single comparison with a rate from another system is weak evidence on its own, because case mix and admission practices differ between hospitals.

What the Record Cannot Answer

The third question, whether patients were too sick for the floor when the decision was made, is the one the director most wants answered. The record can show what was documented, but not everything the team saw, and not whether the decision was reasonable given what was known. Operational health record data are collected for care and billing, not research, and their completeness and timing vary in ways that limit what can be inferred from them (Hersh et al., 2013). Treating a transfer as proof of a wrong initial decision would be a mistake: some patients deteriorate unpredictably.

The record can offer a partial proxy. The hospital calculates a National Early Warning Score from vital signs. A study of this score found that it discriminated well between patients who did and did not go on to cardiac arrest, unanticipated ICU admission or death within 24 hours (Smith et al., 2013). The brief therefore proposes a secondary question: among patients transferred to the ICU within 24 hours, what was the highest early warning score recorded in the four hours before they left the emergency department? A high score would suggest that warning signs were present; a low score would suggest deterioration was not predictable from vital signs. Neither result would settle the question of judgment, and the brief says so.

What the Director Will Receive

The director will receive the rate, count and confidence interval for the past year and the two before it; the comparison with the published rate and its caveats; the distribution of early warning scores among transferred patients; and a short list of transferred patients whose cases could be reviewed by clinicians if the director wants to examine judgment directly. Case review is the only reliable way to address the third question, and the data can make it efficient by pointing to the cases most worth reviewing.

Decisions the Director Should Confirm

Three choices in this brief could reasonably be made differently, and the director should confirm them before the query is run: using floor arrival rather than admission order as time zero, including converted observation patients and excluding step-down transfers from the main rate. Each changes the figure. Writing them down now, before the result is known, prevents anyone from choosing the definition that produces the answer they prefer.

The query will not be run until the director has confirmed these choices in writing.

References

Delgado, M. K., Liu, V., Pines, J. M., Kipnis, P., Gardner, M. N., & Escobar, G. J. (2013). Risk factors for unplanned transfer to intensive care within 24 hours of admission from the emergency department in an integrated healthcare system. Journal of Hospital Medicine, 8(1), 13-19. https://doi.org/10.1002/jhm.1979

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

Smith, G. B., Prytherch, D. R., Meredith, P., Schmidt, P. E., & Featherstone, P. I. (2013). The ability of the National Early Warning Score (NEWS) to discriminate patients at risk of early cardiac arrest, unanticipated intensive care unit admission, and death. Resuscitation, 84(4), 465-470. https://doi.org/10.1016/j.resuscitation.2012.12.016