Category P: Intended and Unintended Consequences of a Quality Improvement Algorithm Involving Electronic and Paper Reminders

1University of Chicago, Chicago, IL
2University of Chicago, Chicago, IL
3University of Chicago, Chicago, IL

Meeting: Hospital Medicine 2008, April 3-5, San Diego, Calif.

Abstract number: 51

Background:

Hospitals are initiating new processes of care to improve compliance with CMS quality measures, often integrating electronic reminders into existing systems of care. Category P is an electronic and paper‐based algorithm to improve timing to antibiotics in the emergency department (ED) for adults with community‐acquired pneumonia (CAP). Electronic cues signal staff to place a paper form in the chart delineating the steps in the algorithm. Although use of the algorithm may improve antibiotic timing for CAP patients, it may also produce unintended consequences.

Methods:

A retrospective chart review of category P patients identified in the ED computer tracking system was performed. Data collected by trained research staff included the time of triage, whether a category P form was in the chart, final diagnosis, time and interpretation of the CXR, and time to antibiotics. Analyses comparing patients with and without the paper form were performed using Fisher's exact test to test for differences in the median for continuous variables and for differences in rates for dichotomous variables.

Results:

From May through September 2007, 232 category P patients were identified, of which 182 charts (78.4%) have been abstracted. The following were their discharge diagnoses: CAP, 37 (20.3%); viral syndrome, 20 (11.0%); asthma or COPD exacerbation, 20 (11.0%); bronchitis, 17 (9.3%); cough, 12 (6.6%); chest pain, 5 (2.7%); dyspnea, 5 (2.7%); other respiratory diagnoses, 2 (1.1%); nonrespiratory diagnoses, 47 (25.8%); and left without being seen, 17 (9.3%). The sensitivity of category P was 41%, with a positive predictive value of 20.3%. For those with the form, median time to CXR and antibiotics was 63.5 and 89 minutes, respectively; 2 patients (9.5%) received antibiotics after more than 4 hours. None received antibiotics prior to CXR. Comparing patients without the form with those with the form, median time to CXR and antibiotics was 98 (P = .10) and 150 (P = .73) minutes; 6 (37.5%) received antibiotics prior to CXR (P = .003), and 3 (18.8%) received antibiotics after more than 4 hours (P = .63).

Conclusions:

Although the differences were not statistically significant, category P patients with CAP who received the form tended to have shorter times to CXR and antibiotics and were less likely to receive antibiotics outside the 4‐hour window. The statistically significant fraction of CAP patients without the form who received antibiotics prior to CXR may indicate the potential for inappropriate use of antibiotics in patients without CAP. Inconsistent use of the form may be improved by linking the form to the electronic triage system to ensure that every category P patient chart has a form.

Author Disclosure:

E. Marlow, none; S. Thomas, none; D. Meltzer, none.

To cite this abstract:

Marlow E, Thomas S, Meltzer D. Category P: Intended and Unintended Consequences of a Quality Improvement Algorithm Involving Electronic and Paper Reminders. Abstract published at Hospital Medicine 2008, April 3-5, San Diego, Calif. Abstract 51. Journal of Hospital Medicine. 2008; 3 (suppl 1). https://www.shmabstracts.com/abstract/category-p-intended-and-unintended-consequences-of-a-quality-improvement-algorithm-involving-electronic-and-paper-reminders/. Accessed September 17, 2019.

« Back to Hospital Medicine 2008, April 3-5, San Diego, Calif.