2.50
Hdl Handle:
http://hdl.handle.net/10755/159754
Type:
Presentation
Title:
Use of POA Coding to Improve the Failure to Rescue Measure
Abstract:
Use of POA Coding to Improve the Failure to Rescue Measure
Conference Sponsor:Midwest Nursing Research Society
Conference Year:2009
Author:Harris, Marcelline
P.I. Institution Name:Mayo Clinic
Contact Address:200 1st St SW, Stabile 11-74, Rochester, MN, 55905, USA
Contact Telephone:507-284-4501
Co-Authors:M. Harris, C. Vanderboom, M. Malinchoc, J. Ransom, A. Hanson, C. Leibson, , Mayo Clinic, Rochester, MN; J. Needleman, , UCLA, Los Angeles, CA; P. Buerhaus, , Vanderbilt University, Nashville, TN;
Background: Failure to Rescue (FTR) is a National Quality Forum (NQF) nursing-sensitive performance measure, selected because it has been shown to be sensitive to nurse staffing variability; FTR is also an Agency for Health Research and Quality (AHRQ) patient safety indicator (PSI). FTR uses inpatient billing data to identify encounters with certain hospital acquired complications that end in death. Six FTR complication types have been defined; for each, a denominator is constructed by applying inclusion and exclusion rules to ICD-9-CM discharge diagnoses and the numerator is the number of deaths. Despite the conservative approach used to develop the original exclusion rules, the FTR measure is criticized for including too many false positives, resulting in problems of both sensitivity and positive predictive validity. Objective: Develop refined exclusion rules by first using Mayo Clinic present on admission (POA) codes to inform construction of the denominator and then validating the exclusion rules using California billing data. Methods: We examined over 200,000 Mayo Clinic hospital discharge abstracts (2003-2006) and over 3 million hospital discharges from California hospitals (2000-2002). Both data sets included POA coding. Analysis included constructing ROCs with increasingly restrictive thresholds for comparing exclusion rules to POA coding and logistic modeling for predicting POA. C-statistics were used to compare the POA informed exclusion rules to the current FTR measure. Results: We have demonstrated significant improvements in the denominator construction. For example, when constructing the pool of persons likely to have hospital acquired acute renal failure, C-statistics improved from .487 (current FTR measure) to > .703 for both the refined exclusion rule and the logistic regression. Improvements were found across the six denominators and the pooled denominator for the FTR measure. Conclusions: We have defined an improved method for constructing the FTR denominator that appears to reduce the noise observed in the current FTR measure. POA information was highly useful for FTR refinements. This study has implications for NQF and PSI updates, use of FTR in hospital level nursing sensitive measures, and quality based reimbursement for hospitals.
Repository Posting Date:
26-Oct-2011
Date of Publication:
17-Oct-2011
Sponsors:
Midwest Nursing Research Society

Full metadata record

DC FieldValue Language
dc.typePresentationen_GB
dc.titleUse of POA Coding to Improve the Failure to Rescue Measureen_GB
dc.identifier.urihttp://hdl.handle.net/10755/159754-
dc.description.abstract<table><tr><td colspan="2" class="item-title">Use of POA Coding to Improve the Failure to Rescue Measure</td></tr><tr class="item-sponsor"><td class="label">Conference Sponsor:</td><td class="value">Midwest Nursing Research Society</td></tr><tr class="item-year"><td class="label">Conference Year:</td><td class="value">2009</td></tr><tr class="item-author"><td class="label">Author:</td><td class="value">Harris, Marcelline</td></tr><tr class="item-institute"><td class="label">P.I. Institution Name:</td><td class="value">Mayo Clinic</td></tr><tr class="item-address"><td class="label">Contact Address:</td><td class="value">200 1st St SW, Stabile 11-74, Rochester, MN, 55905, USA</td></tr><tr class="item-phone"><td class="label">Contact Telephone:</td><td class="value">507-284-4501</td></tr><tr class="item-email"><td class="label">Email:</td><td class="value">harris.marcelline@mayo.edu</td></tr><tr class="item-co-authors"><td class="label">Co-Authors:</td><td class="value">M. Harris, C. Vanderboom, M. Malinchoc, J. Ransom, A. Hanson, C. Leibson, , Mayo Clinic, Rochester, MN; J. Needleman, , UCLA, Los Angeles, CA; P. Buerhaus, , Vanderbilt University, Nashville, TN;</td></tr><tr><td colspan="2" class="item-abstract">Background: Failure to Rescue (FTR) is a National Quality Forum (NQF) nursing-sensitive performance measure, selected because it has been shown to be sensitive to nurse staffing variability; FTR is also an Agency for Health Research and Quality (AHRQ) patient safety indicator (PSI). FTR uses inpatient billing data to identify encounters with certain hospital acquired complications that end in death. Six FTR complication types have been defined; for each, a denominator is constructed by applying inclusion and exclusion rules to ICD-9-CM discharge diagnoses and the numerator is the number of deaths. Despite the conservative approach used to develop the original exclusion rules, the FTR measure is criticized for including too many false positives, resulting in problems of both sensitivity and positive predictive validity. Objective: Develop refined exclusion rules by first using Mayo Clinic present on admission (POA) codes to inform construction of the denominator and then validating the exclusion rules using California billing data. Methods: We examined over 200,000 Mayo Clinic hospital discharge abstracts (2003-2006) and over 3 million hospital discharges from California hospitals (2000-2002). Both data sets included POA coding. Analysis included constructing ROCs with increasingly restrictive thresholds for comparing exclusion rules to POA coding and logistic modeling for predicting POA. C-statistics were used to compare the POA informed exclusion rules to the current FTR measure. Results: We have demonstrated significant improvements in the denominator construction. For example, when constructing the pool of persons likely to have hospital acquired acute renal failure, C-statistics improved from .487 (current FTR measure) to &gt; .703 for both the refined exclusion rule and the logistic regression. Improvements were found across the six denominators and the pooled denominator for the FTR measure. Conclusions: We have defined an improved method for constructing the FTR denominator that appears to reduce the noise observed in the current FTR measure. POA information was highly useful for FTR refinements. This study has implications for NQF and PSI updates, use of FTR in hospital level nursing sensitive measures, and quality based reimbursement for hospitals.</td></tr></table>en_GB
dc.date.available2011-10-26T22:18:13Z-
dc.date.issued2011-10-17en_GB
dc.date.accessioned2011-10-26T22:18:13Z-
dc.description.sponsorshipMidwest Nursing Research Societyen_GB
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.