300 Cost to Correct International Normalized Ratio Among Patients With Warfarin-Associated Major Bleeding in Clinical Practice

Research Forum Abstracts

298

Failure to Correct International Normalized Ratio Following Fresh Frozen Plasma Administration Among Patients With Anticoagulant-Related Major Bleeding in Clinical Practice

pain” order set utilized by physicians and nurses. Our objective was to assess the costeffectiveness of routine coagulation testing of patients with chest pain in the ED. Methods: We conducted a retrospective chart review on a non-consecutive sample of patients evaluated for chest pain in the ED between the dates of August 1, 2010 and October 31, 2010. Charts were reviewed to determine the number and results of coagulation studies ordered, the number of coagulation studies that were appropriately ordered, and the number of patients requiring a therapeutic intervention or change in clinical plan (withholding of antiplatelet/anticoagulant, delayed procedure, or treatment with FFP or vitamin K) based on an unexpected coagulopathy. We considered it appropriate to order coagulation studies on patients with cirrhosis, known/suspected coagulopathy, active bleeding, use of warfarin, or STEMI. We calculated the cost of the coagulation studies ordered on patients without the listed indications. Descriptive statistics were used for analysis. Results: 761 of the 1849 charts for patients evaluated for chest pain during the study period were reviewed. Thirteen patients were excluded due to age ⬍18, and an additional 8 patients left against medical advice before labs were drawn. Of the 740 patients included, 406 (55%) had coagulation studies ordered. Of that 406, 327 (81%) patients with coagulation studies ordered had no indications for testing. 315 patients of the 327 were tested with PT/INR ($19.00) and PTT ($33.50). 11 of the remaining 12 patients were tested with only a PT/INR, and 1 patient was tested with only PTT. One of the 327 patients (0.31%; 95%CI 0.05-1.7%) tested without indication had a clinically significant coagulopathy (INR ⬎1.5, PTT ⬎50 sec), but none (0%; 95%CI 0-1.2%) of the patients with coagulation testing performed without indication required a therapeutic intervention or change in clinical plan. The cost of coagulation testing in these 327 patients was $16,780. By our estimates, routine coagulation testing without indication for patients with chest pain in the York Hospital emergency department cost $165,000 annually. Conclusion: Coagulation testing on chest pain patients in the ED is not costeffective and should not be routinely performed.

300

Cost to Correct International Normalized Ratio Among Patients With Warfarin-Associated Major Bleeding in Clinical Practice

Menzin J, White L, Friedman M, Nichols C, Menzin J, Hoesche J, Jones C/Boston Health Economics, Inc., Waltham, MA; CSL Behring, King of Prussia, PA

Menzin J, Friedman M, Nichols C, Menzin J, Hoesche J, Jones C/Boston Health Economics, Inc., Waltham, MA; CSL Behring, King of Prussia, PA

Study Objective: To assess the frequency of international normalized ratio (INR) remaining uncorrected in patients administered fresh frozen plasma for anticoagulantrelated major bleeding. Methods: A retrospective database analysis was undertaken using electronic medical record data from a large integrated health system. Patients who received fresh frozen plasma between 01/01/2004 and 12/31/2010, and who met the following criteria were selected: major hemorrhage diagnosis the day before to the day after initial fresh frozen plasma administration; INRⱖ2 on the day before or the day of fresh frozen plasma and another INR result up to 1 day after fresh frozen plasma; and supply of warfarin within 90 days prior to hospitalization. INR correction (defined as INR ⱕ 1.3) was evaluated at the last available test up to 1 day following fresh frozen plasma start. Logistic regression analysis was performed to assess probability of remaining uncorrected. Results: 414 patients met selection criteria (mean age 75 years, 53% male, mean baseline Charlson score of 2.5). The majority of patients presented with a gastrointestinal bleed (58%), followed by intracranial hemorrhage (ICH) (38%) and other bleed types (4%). Mean 1st elevated INR was 3.7, and mean INR was 1.5 at the last test. Sixty-seven percent of patients remained uncorrected. Mean time to correction among those achieving correction was 20.6 hours from start of fresh frozen plasma administration. Patients who were older, those with a baseline Charlson score of 4 or greater, and those with non-ICH bleeds were more likely to remain uncorrected within 24 hours of fresh frozen plasma administration. Using an INR threshold of ⱕ 1.5, 39% of patients remained uncorrected. Conclusions: A substantial proportion of patients remain uncorrected following fresh frozen plasma administration, with estimates varying depending on the INR threshold used.

Study Objective: To assess differences in total inpatient costs among patients whose international normalized ratio (INR) remained uncorrected versus corrected following use of fresh frozen plasma in the management of warfarin-associated major bleeding. Methods: A retrospective database analysis was undertaken using electronic medical record data from a large integrated health system. Patients who received fresh frozen plasma between 01/01/2004 and 12/31/2010, and who met the following criteria were selected: major hemorrhage diagnosis the day before to the day after initial fresh frozen plasma administration; INR ⱖ 2 on the day before or the day of fresh frozen plasma and another INR result up to 1 day after fresh frozen plasma; and supply of warfarin within 90 days prior to hospitalization. INR correction (defined as INR ⱕ 1.3) was evaluated at the last available test up to 1 day following fresh frozen plasma start. Costs were evaluated for each patient’s inpatient stay, including those for procedures, medications, physician visits, and facility usage. Statistical differences in costs between patients whose INR remained uncorrected versus corrected were evaluated with the Wilcoxon Rank-Sum test with statistical significance defined as P⬍0.05. All costs were adjusted to 2009 USD using the medical care component of the consumer price index. Results: 414 patients met selection criteria (mean age 75 years, 53% male, mean baseline Charlson score of 2.5). The majority of patients presented with a gastrointestinal bleed (58%), followed by intracranial hemorrhage (38%) and other bleed types (4%). Thirty-three percent of patients were corrected at the last available test up to 1 day following receipt of fresh frozen plasma. Mean volume of fresh frozen plasma, packed red blood cells, platelets, vitamin K, and albumin received were not significantly different among patients whose INR was corrected versus uncorrected. Total inpatient costs (mean⫹SD) were not significantly different among patients whose INR was corrected at 24 hours following receipt of fresh frozen plasma versus patients who remained uncorrected ($18,941 ⫾ $20,806 vs. $21,521 ⫾ $20,767; P⫽0.09). Similarly, total costs were not significantly different by INR correction when stratified by quartiles of hospital length of stay or presenting bleed type. Using an alternative definition of INR correction at ⱕ 1.5, total costs were also not significantly different (P⫽0.10).

299

Cost-Effectiveness of Routine Coagulation Testing in the Evaluation of Chest Pain in the Emergency Department

Kochert E, Goldhahn L, Hughes I, Gee K, Stahlman B/York Hospital, York, PA

Study Objectives: Approximately 5% of all United States emergency department (ED) visits are for chest pain. At York Hospital coagulation tests are part of the “chest

S278 Annals of Emergency Medicine

Volume , .  : October 

Research Forum Abstracts Conclusions: Timely and improved correction of INR values to a target of 1.3 or 1.5 among patients receiving fresh frozen plasma for warfarin-related major bleeding does not appear to be associated with greater total inpatient costs.

301

Under-Recognition of Cervical Neisseria Gonorrhoeae and Chlamydia Trachomatis Infections in Pregnant Emergency Department Patients

Betcher J, Doehrman D, Krivochenitser R, Whalen D, Norman C, Jones JS/Michigan State University College of Human Medicine, Spectrum Health, Grand Rapids, MI; MSU College of Human Medicine, Saint Mary’s Health Care, Grand Rapids, MI; DeVos Children’s Hospital, Grand Rapids, MI

Study Objectives: The Centers for Disease Control and Prevention (CDC) advocates routine testing and liberal antibiotic treatment for pregnant women with suspected Neisseria gonorrhoeae and chlamydia trachomatis infections. However, there are few data describing the frequency of Neisseria gonorrhoeae and chlamydia trachomatis pelvic infections detected by emergency department (ED) bacteriologic testing, but not treated initially because of unconvincing signs or symptoms. The purpose of this study was to: 1) quantify the frequency of underrecognized Neisseria gonorrhoeae and chlamydia trachomatis infections in pregnant women tested in the ED; 2) to describe the characteristics of those not treated during the initial visit; and 3) to determine how many pregnant women with acute cervicitis were lost to follow-up. Methods: This was a retrospective, cohort analysis of consecutive female seen in the ED of 3 academic medical centers during a 24-month study period, with positive results for Neisseria gonorrhoeae/chlamydia trachomatis, and their subsequent follow-up and treatment by the ED and/or county health department. All patients had a positive urine pregnancy test. We excluded those patients with a repeat infection involving the same pathogen, unless adequate treatment was prescribed and an interval culture from our health system or the public health providers was negative for the initial infectious agent. Our key outcome measures were the proportion of pregnant females being untreated in the ED, the time to subsequent treatment, and the proportion lost to follow-up. Descriptive statistics (mean, SD) and frequency tables were used to describe clinical findings and demographic characteristics. Groups were compared using 2-tailed unpaired ttests and Wilcoxon rank sum tests for continuous and ordinal data, while nominal data were analyzed by chi-square tests. Results: During the study period, 735 female patients had positive polymerase chain reaction studies for Neisseria gonorrhoeae and/or chlamydia trachomatis; 179 (24%) were pregnant. Overall, 143 of these pregnant patients with cervicitis (80%) were not treated in the ED. Presenting symptoms included abdominal pain (71%), nausea (45%), vaginal discharge (35%), vaginal bleeding (34%), and urinary complaints (22%). Factors such as age, ethnic background, insurance status, previous STI, or known STI exposure were not associated with antibiotic treatment in the ED. The most common discharge diagnoses were vaginitis (37%), urinary infection (33%), threatened abortion (19%), non-specific abdominal pain (9%), and rule-out ectopic pregnancy (6%). Of the 143 patients with STI not treated in the ED, 114 (80%) were contacted by phone and/or mail. The mean time interval between ED presentation and antibiotic treatment was 4.1 days (range, 2- 37 days). Twenty-nine pregnant patients with STI (20%) were subsequently lost to follow-up. Conclusion: In our study population, the vast majority of pregnant women, who were found to have either Neisseria gonorrhoeae or chlamydia, were not treated in the ED. Further study is warranted to enhance point-of-contact testing, identify better mechanisms for contact and follow-up after ED discharge, and more liberal policies to treat less symptomatic patients empirically.

302

Prediction Method for Fitz-Hugh-Curtis Syndrome in Childbearing-Aged Women With Upper Abdominal Pain

Lee S, Shin J/Seoul National University, Boramae Medical Center, Seoul, Korea, Republic of

Study Objectives: Fitz-Hugh-Curtis syndrome is a perihepatitis associated with pelvic inflammatory disease. Physicians often depend on computed tomography despite the radiation hazard, since it is difficult to diagnose Fitz-Hugh-Curtis syndrome in initial evaluation. Therefore, we developed a new reliable scoring system for screening Fitz-Hugh-Curtis syndrome by using medical history, physical examination, and laboratory findings.

Volume , .  : October 

Methods: We reviewed the registries of childbearing-aged women who presented with upper abdominal pain to urban emergency department from April 2008 to October 2010. We identified the statistically significant predictor variables that affected the Fitz-Hugh-Curtis syndrome, by using multivariable logistic regression analysis, along with 95% confidence interval (95% CI). And then we developed the Fitz-Hugh-Curtis syndrome score for screening models. The performances of each score were evaluated by using the receiver operating characteristics curve. Results: Total 361 eligible childbearing-aged women patients were registered in our study. 50 patients were diagnosed as Fitz-Hugh-Curtis syndrome. The significant factors to affect Fitz-Hugh-Curtis syndrome were as follows: pain onset more than 2 days (OR: 5.988, 95% CI: 2.217-16.177), history of abortion (OR: 2.746, 95% CI: 1.099-6.859), unmarried woman (OR: 3.751, 95% CI: 1.395-10.086), coitus history within 4 weeks (OR: 2.470, 95% CI: 1.005-6.074), migrating pain (OR: 2.979, 95% CI: 1.034-8.584), absence of gastrointestinal symptoms (OR: 4.528, 95% CI: 1.86011.019), absence of urinary symptoms (OR: 3.587, 95% CI: 1.025-12.549) and elevation of C-reactive protein (OR: 13.820, 95% CI: 4.693-40.693). We assigned a proper score to each variable with a range of scores 1 or 2 using the beta-coefficients. The area under the receiver operating characteristics curve of the scoring system were 0.920. Patients were categorized as low risk (score 0⬃3), intermediate risk (score 4⬃6) and high risk (score 7⬃10) group according to their scores. 55.2% (37 patients) of high risk group, 6.2% (13 patients) of intermediate group, and none of low risk group were diagnosed as Fitz-Hugh-Curtis syndrome. Conclusion: We developed the scoring system for screening Fitz-Hugh-Curtis syndrome of childbearing-aged women who had presented with upper abdominal pain. By using the proposed scoring system, it will be easier to screen Fitz-HughCurtis syndrome among other diseases which can cause abdominal pain in women.

303

Can Nursing Staff and Junior Physicians Accurately Predict the Risk of In-Hospital Mortality of Acutely Admitted Medical Patients: An Observational Cohort Study

Brabrand M, Knudsen T, Hallas J/Sydvestjysk Sygehus, Esbjerg, Denmark; University of Southern Denmark, Odense, Denmark

Study Objectives: Most acutely admitted medical patients are initially assessed and treated by junior staff, both nursing staff and as well as physicians. However, at presentation, many decisions have to be made and in order to make the right decisions, the staff will have to decide how ill they think the patient really is. We therefore performed the present study to assess the ability of the staff to accurately estimate the in-hospital mortality of acutely admitted medical patients. Methods: Design: A prospective observational cohort study. Upon contact with the patient, the first member of the nursing staff to assess the patient was asked to complete a form. This form included their initial estimation of the risk of in-hospital mortality (based solely on the information available to them at the present time). The first doctor to treat the patient was asked to complete a similar form. The discriminatory power (ability to discriminate between survivors and non-survivors) of the estimations of the staff was estimated using the area under the receiver-operating characteristics curve (AUROC). Values above 0.8 represent good discriminatory power. The discriminatory power of the nursing staff and physicians were compared using ␹2 test. Calibration (accuracy of the prediction) was assessed using HosmerLemeshow ␹2 goodness of fit test with a p-value above 0.05 indicating acceptable calibration. Setting: The general internal medical and cardiology admission units at a regional teaching hospital.Type of participants: 2,855 consecutively admitted acute medical patients. Median age of the patients was 62 years. Sixty-five patients died while admitted (2.3 % in-hospital mortality). Results: The nursing staff estimated the risk of in-hospital mortality for 1,822 patients (63.8 %). We found a good discriminatory power with an AUROC 0.838. Calculating the goodness of fit, we found a ␹2 (8 degrees of freedom) of 61.6, p ⬍ 0.001.The physicians estimated the risk of in-hospital mortality for 733 patients (25.7 %). We found a good discriminatory power with an AUROC of 0.811. Calculating the goodness of fit, we found a ␹2 (8 degrees of freedom) of 32.6, p ⬍ 0.001. We were only able to include 507 (17.8 %) patients where comparing the discriminatory power of the nursing staff and the physicians. We were not able to show a statistical significant difference as ␹2 was 1.76, p ⫽ 0.18. Conclusion: Both nursing staff and physicians were able to distinguish between survivors and non-survivors with an acceptable discriminatory power. However, both groups were unable to show an acceptable calibration.

Annals of Emergency Medicine S279