Search

Objective Like most real-world data, electronic health record (EHR)–derived data from oncology patients typically exhibits wide interpatient variability in terms of available data elements. This interpatient variability leads to missing data and can present critical challenges in developing and implementing predictive models to underlie clinical decision support for patient-specific oncology care. Here, we sought to develop a novel ensemble approach to addressing missing data that we term the “meta-model” and apply the meta-model to patient-specific cancer prognosis. Materials and Methods Using real-world data, we developed a suite of individual random survival forest models to predict survival in patients with advanced lung cancer, colorectal cancer, and breast cancer. Individual models varied by the predictor data used. We combined models for each cancer type into a meta-model that predicted survival for each patient using a weighted mean of the individual models for which the patient had all requisite predictors. Results The meta-model significantly outperformed many of the individual models and performed similarly to the best performing individual models. Comparisons of the meta-model to a more traditional imputation-based method of addressing missing data supported the meta-model’s utility. Conclusions We developed a novel machine learning–based strategy to underlie clinical decision support and predict survival in cancer patients, despite missing data. The meta-model may more generally provide a tool for addressing missing data across a variety of clinical prediction problems. Moreover, the meta-model may address other challenges in clinical predictive modeling including model extensibility and integration of predictive algorithms trained across different institutions and datasets.

Introduction An important cause of laboratory test misordering and overutilization is clinician confusion between tests with similar sounding names or similar indications. We identified an area of test ordering confusion with iron studies that involves total iron binding capacity (TIBC), transferrin, and transferrin saturation. We observed concurrent ordering of direct transferrin along with TIBC at many hospitals within our health system and suspected this was unnecessary. Methods We extracted patient test results for transferrin, TIBC and other biomarkers. Using these data, we evaluated both patterns of test utilization and test result concordance. We implemented a clinical decision support (CDS) alert to discourage unnecessary orders for direct transferrin. Results Using linear regression, we were able to predict transferrin from either TIBC alone or TIBC with other analytes with a high degree of accuracy, demonstrating that in most cases, direct transferrin in combination with TIBC provides little if any additional diagnostic information beyond TIBC alone. The CDS alert proved highly effective in reducing transferrin test utilization at four different hospitals. Conclusions Concurrent ordering of direct transferrin and TIBC should usually be avoided. Removal of transferrin or TIBC from the test menu or implementation of CDS may improve utilization of these tests.

Objectives To evaluate the use of a provider ordering alert to improve laboratory efficiency and reduce costs. Methods We conducted a retrospective study to assess the use of an institutional reflex panel for monoclonal gammopathy evaluation. We then created a clinical decision support (CDS) alert to educate and encourage providers to change their less-efficient orders to the reflex panel. Results Our retrospective analysis demonstrated that an institutional reflex panel could be safely substituted for a less-efficient and higher-cost panel. The implemented CDS alert resulted in 79% of providers changing their high-cost order panel to an order panel based on the reflex algorithm. Conclusions The validated decision support alert demonstrated high levels of provider acceptance and directly led to operational and cost savings within the laboratory. Furthermore, these studies highlight the value of laboratory involvement with CDS efforts to provide agile and targeted provider ordering assistance.

Objectives Peripheral blood flow cytometry (PBFC) is useful for evaluating circulating hematologic malignancies (HM) but has limited diagnostic value for screening. We used machine learning to evaluate whether clinical history and CBC/differential parameters could improve PBFC utilization. Methods PBFC cases with concurrent/recent CBC/differential were split into training (n = 626) and test (n = 159) cohorts. We classified PBFC results with abnormal blast/lymphoid populations as positive and used two models to predict results. Results Positive PBFC results were seen in 58% and 21% of training cases with and without prior HM (P < .001). % neutrophils, absolute lymphocyte count, and % blasts/other cells differed significantly between positive and negative PBFC groups (areas under the curve [AUC] > 0.7). Among test cases, a decision tree model achieved 98% sensitivity and 65% specificity (AUC = 0.906). A logistic regression model achieved 100% sensitivity and 54% specificity (AUC = 0.919). Conclusions We outline machine learning-based triaging strategies to decrease unnecessary utilization of PBFC by 35% to 40%.

Introduction Laboratory test interferences can cause spurious test results and patient harm. Knowing the frequency of various interfering substances in patient populations likely to be tested with a particular laboratory assay may inform test development, test utilization and strategies to mitigate interference risk. Methods We developed REACTIR (Real Evidence to Assess Clinical Testing Interference Risk), an approach using real world data to assess the prevalence of various interfering substances in patients tested with a particular type of assay. REACTIR uses administrative real world data to identify and subgroup patient cohorts tested with a particular laboratory test and evaluate interference risk. Results We demonstrate the application REACTIR to point of care (POC) blood glucose testing. We found that exposure to several substances with the potential to interfere in POC blood glucose tests, including N-acetyl cysteine (NAC) and high dose vitamin C was uncommon in most patients undergoing POC glucose tests with several key exceptions, such as burn patients receiving high dose IV-vitamin C or acetaminophen overdose patients receiving NAC. Conclusions Findings from REACTIR may support risk mitigation strategies including targeted clinician education and clinical decision support. Likewise, adaptations of REACTIR to premarket assay development may inform optimal assay design and assessment.

Objectives While well-designed clinical decision support (CDS) alerts can improve patient care, utilization management, and population health, excessive alerting may be counterproductive, leading to clinician burden and alert fatigue. We sought to develop machine learning models to predict whether a clinician will accept the advice provided by a CDS alert. Such models could reduce alert burden by targeting CDS alerts to specific cases where they are most likely to be effective. Materials and Methods We focused on a set of laboratory test ordering alerts, deployed at 8 hospitals within the Partners Healthcare System. The alerts notified clinicians of duplicate laboratory test orders and advised discontinuation. We captured key attributes surrounding 60 399 alert firings, including clinician and patient variables, and whether the clinician complied with the alert. Using these data, we developed logistic regression models to predict alert compliance. Results We identified key factors that predicted alert compliance; for example, clinicians were less likely to comply with duplicate test alerts triggered in patients with a prior abnormal result for the test or in the context of a nonvisit-based encounter (eg, phone call). Likewise, differences in practice patterns between clinicians appeared to impact alert compliance. Our best-performing predictive model achieved an area under the receiver operating characteristic curve (AUC) of 0.82. Incorporating this model into the alerting logic could have averted more than 1900 alerts at a cost of fewer than 200 additional duplicate tests. Conclusions Deploying predictive models to target CDS alerts may substantially reduce clinician alert burden while maintaining most or all the CDS benefit.

ObjectivesAn unfortunate reality of laboratory medicine is that blood specimens collected from one patient occasionally get mislabeled with identifiers from a different patient, resulting in so-called “wrong blood in tube” (WBIT) errors and potential patient harm. Here, we sought to develop a machine learning-based, multianalyte delta check algorithm to detect WBIT errors and mitigate patient harm.MethodsWe simulated WBIT errors within sets of routine inpatient chemistry test results to develop, train, and evaluate five machine learning-based WBIT detection algorithms.ResultsThe best-performing WBIT detection algorithm we developed was based on a support vector machine and incorporated changes in test results between consecutive collections across 11 analytes. This algorithm achieved an area under the curve of 0.97 and considerably outperformed traditional single-analyte delta checks.ConclusionsMachine learning-based multianalyte delta checks may offer a practical strategy to identify WBIT errors prior to test reporting and improve patient safety.

Objectives We evaluated changes in the testing menu, volume, and positivity rates for tick-borne illnesses in a New England medical center over an 11-year time frame. Methods Testing data were obtained by a retrospective review utilizing searchable data from a laboratory information system archive. Results Testing for tick-borne infections (TBI) increased 5.3-fold over an 11-year time period and the number of positive test results increased threefold. Annual rates for Lyme serology positivity varied from 14% to 29% and for western blot confirmation from 26% to 48%. Test volumes and the number of positive results increased for all TBI endemic to our region. Conclusions Our results confirm national trends suggesting increasing rates of TBI and substantially increased testing. This may reflect a greater incidence of TBI in our region and/or increased awareness of these infections.

A Man with Leg Weakness, Recurrent Falls, and Anemia A 64-year-old man was admitted to the hospital because of recurrent falls. He had proximal muscle weakness, orthostatic hypotension, and anemia....

Objectives Laboratory-based utilization management programs typically rely primarily on data derived from the laboratory information system to analyze testing volumes for trends and utilization concerns. We wished to examine the ability of an electronic health record (EHR) laboratory orders database to improve a laboratory utilization program. Methods We obtained a daily file from our EHR containing data related to laboratory test ordering. We then used an automated process to import this file into a database to facilitate self-service queries and analysis. Results The EHR laboratory orders database has proven to be an important addition to our utilization management program. We provide three representative examples of how the EHR laboratory orders database has been used to address common utilization issues. We demonstrate that analysis of EHR laboratory orders data has been able to provide unique insights that cannot be obtained by review of laboratory information system data alone. Further, we provide recommendations on key EHR data fields of importance to laboratory utilization efforts. Conclusion We demonstrate that an EHR laboratory orders database may be a useful tool in the monitoring and optimization of laboratory testing. We recommend that health care systems develop and maintain a database of EHR laboratory orders data and integrate this data with their laboratory utilization programs.

ObjectiveA key challenge in clinical data mining is that most clinical datasets contain missing data. Since many commonly used machine learning algorithms require complete datasets (no missing data), clinical analytic approaches often entail an imputation procedure to “fill in” missing data. However, although most clinical datasets contain a temporal component, most commonly used imputation methods do not adequately accommodate longitudinal time-based data. We sought to develop a new imputation algorithm, 3-dimensional multiple imputation with chained equations (3D-MICE), that can perform accurate imputation of missing clinical time series data.MethodsWe extracted clinical laboratory test results for 13 commonly measured analytes (clinical laboratory tests). We imputed missing test results for the 13 analytes using 3 imputation methods: multiple imputation with chained equations (MICE), Gaussian process (GP), and 3D-MICE. 3D-MICE utilizes both MICE and GP imputation to integrate cross-sectional and longitudinal information. To evaluate imputation method performance, we randomly masked selected test results and imputed these masked results alongside results missing from our original data. We compared predicted results to measured results for masked data points.Results3D-MICE performed significantly better than MICE and GP-based imputation in a composite of all 13 analytes, predicting missing results with a normalized root-mean-square error of 0.342, compared to 0.373 for MICE alone and 0.358 for GP alone.Conclusions3D-MICE offers a novel and practical approach to imputing clinical laboratory time series data. 3D-MICE may provide an additional tool for use as a foundation in clinical predictive analytics and intelligent clinical decision support.

In a previous study we reported on the impact of point-of-care testing (POCT) on practice efficiency in an academic primary care practice that was established to develop new models of care delivery. Here we report a follow-on confirmation study in a more typical primary care practice in the community.In this observational study with a retrospective comparison analysis we measured metrics of practice efficiency on two patient cohorts: those that did not receive POCT and those that did.In the patient cohort that received POCT there was a 99% reduction in letters to patients (p0.001), a 75% decrease in calls to patients (not significant due to small numbers), a 50% reduction in follow-up tests per visit (p=0.044) and a 38% reduction in follow-up visits due to abnormal test results (p=0.178). Financial analysis including testing costs, revenues and efficiency gains to the practice demonstrated a net financial benefit of $11.90-14.74 per patient visit.Our data confirm the earlier published findings that POCT can improve metrics of practice efficiency in a primary care practice.

Purpose To quantify the effect of a comprehensive, long-term, provider-led utilization management (UM) program on high-cost imaging (computed tomography, magnetic resonance imaging, nuclear imaging, and positron emission tomography) performed on an outpatient basis. Materials and Methods This retrospective, 7-year cohort study included all patients regularly seen by primary care physicians (PCPs) at an urban academic medical center. The main outcome was the number of outpatient high-cost imaging examinations per patient per year ordered by the patient's PCP or by any specialist. The authors determined the probability of a patient undergoing any high-cost imaging procedure during a study year and the number of examinations per patient per year (intensity) in patients who underwent high-cost imaging. Risk-adjusted hierarchical models were used to directly quantify the physician component of variation in probability and intensity of high-cost imaging use, and clinicians were provided with regular comparative feedback on the basis of the results. Observed trends in high-cost imaging use and provider variation were compared with the same measures for outpatient laboratory studies because laboratory use was not subject to UM during this period. Finally, per-member per-year high-cost imaging use data were compared with statewide high-cost imaging use data from a major private payer on the basis of the same claim set. Results The patient cohort steadily increased in size from 88 959 in 2007 to 109 823 in 2013. Overall high-cost imaging utilization went from 0.43 examinations per year in 2007 to 0.34 examinations per year in 2013, a decrease of 21.33% (P < .0001). At the same time, similarly adjusted routine laboratory study utilization decreased by less than half that rate (9.4%, P < .0001). On the basis of unadjusted data, outpatient high-cost imaging utilization in this cohort decreased 28%, compared with a 20% decrease in statewide utilization (P = .0023). Conclusion Analysis of high-cost imaging utilization in a stable cohort of patients cared for by PCPs during a 7-year period showed that comprehensive UM can produce a significant and sustained reduction in risk-adjusted per-patient year outpatient high-cost imaging volume. © RSNA, 2017.

Objectives We sought to address concerns regarding recurring inpatient laboratory test order practices (daily laboratory tests) through a multifaceted approach to changing ordering patterns. Methods We engaged in an interdepartmental collaboration to foster mindful test ordering through clinical policy creation, electronic clinical decision support, and continuous auditing and feedback. Results Annualized daily order volumes decreased from approximately 25,000 to 10,000 during a 33-month postintervention review. This represented a significant change from preintervention order volumes (95% confidence interval, 0.61-0.64; P < 10-16). Total inpatient test volumes were not affected. Conclusions Durable changes to inpatient order practices can be achieved through a collaborative approach to utilization management that includes shared responsibility for establishing clinical guidelines and electronic decision support. Our experience suggests auditing and continued feedback are additional crucial components to changing ordering behavior. Curtailing daily orders alone may not be a sufficient strategy to reduce in-laboratory costs.

Background:A common challenge in the development of laboratory clinical decision support (CDS) and laboratory utilization management (UM) initiatives stems from the fact that many laboratory tests have multiple potential indications, limiting the ability to develop context-specific alerts. As a potential solution, we designed a CDS alert that asks the ordering clinician to provide the indication for testing, using D-dimer as an exemplar. Using data collected over a nearly 3-year period, we sought to determine whether the indication capture was a useful feature within the CDS alert and whether it provided actionable intelligence to guide the development of an UM strategy. Methods: We extracted results and ordering data for D-dimer testing performed in our laboratory over a 35-month period. We analyzed order patterns by clinical indication, hospital service, and length of hospitalization. Results: Our final data set included 13,971 result-order combinations and indeed provided actionable intelligence regarding test utilization patterns. For example, pulmonary embolism was the most common emergency department indication (86%), while disseminated intravascular coagulation was the most common inpatient indication (56%). D-dimer positivity rates increased with the duration of hospitalization and our data suggested limited utility for ordering this test in the setting of suspected venous thromboembolic disease in admitted patients. In addition, we found that D-dimer was ordered for unexpected indications including the assessment of stroke, dissection, and extracorporeal membrane oxygenation. Conclusions: Indication capture within a CDS alert and correlation with result data can provide insight into order patterns which can be used to develop future CDS strategies to guide appropriate test use by clinical indication.

ObjectivesWe evaluated trends in non-Lyme disease tick-borne disease (NLTBI) testing at a national reference laboratory.MethodsTesting data performed at Quest Diagnostics during 2010 to 2016 were analyzed nationally and at the state level.ResultsTesting and positivity for most NLTBIs increased dramatically from 2010 through 2016 based on testing from a large reference laboratory. The number of positive cases, though not as stringent as criteria for public health reporting, generally exceeds that reported by the Centers for Disease Control and Prevention. The frequency of NLTBI in the US is seasonal but testing activity and positive test results are observed throughout all months of the year. Positive results for NLTBI testing mostly originated from a limited number of states, indicating the geographic concentration and distribution of NLTBIs reported in this study.ConclusionsThis report provides an important complementary source of data to best understand trends in and spread of NLTBI.

Emerging applications of machine learning and artificial intelligence offer the opportunity to discover new clinical knowledge through secondary exploration of existing patient medical records. This new knowledge may in turn offer a foundation to build new types of clinical decision support (CDS) that provide patient-specific insights and guidance across a wide range of clinical questions and settings. This article will provide an overview of these emerging approaches to CDS, discussing both existing technologies as well as challenges that health systems and informaticists will need to address to allow these emerging approaches to reach their full potential.

Objectives We evaluated trends in Lyme disease (LD) testing at a national reference laboratory. Methods LD screening enzyme immunoassay and Western blot testing data performed at Quest Diagnostics during 2010 to 2016 were analyzed nationally and at the state level. Results Overall, 593,800 (11.3%) results were positive of 5,255,636 tests. There was an increase in the rate of positivity over the last 2 years of the study and an increase in the number of positive tests in 2016. Positive tests were observed in all 50 states and the District of Columbia. New York had the most positive tests, whereas Connecticut had the highest positivity rate when normalized to state populations. Some states with historically low rates of LD (eg, Texas, Florida, and California) showed significant increases in testing and positivity rates over time. Conclusions LD testing and positivity have increased in recent years, including in states not historically associated with the disease.

A 66-year-old man presented with malaise, weakness, hypercalcemia, and acute kidney injury. CT of the abdomen and pelvis revealed no evidence of cancer; CT of the chest revealed a sclerotic lesion of the eighth rib. Additional tests were performed, and a diagnosis was made.

Objectives While clinical laboratories report most test results as individual numbers, findings, or observations, clinical diagnosis usually relies on the results of multiple tests. Clinical decision support that integrates multiple elements of laboratory data could be highly useful in enhancing laboratory diagnosis. Methods Using the analyte ferritin in a proof of concept, we extracted clinical laboratory data from patient testing and applied a variety of machine-learning algorithms to predict ferritin test results using the results from other tests. We compared predicted with measured results and reviewed selected cases to assess the clinical value of predicted ferritin. Results We show that patient demographics and results of other laboratory tests can discriminate normal from abnormal ferritin results with a high degree of accuracy (area under the curve as high as 0.97, held-out test data). Case review indicated that predicted ferritin results may sometimes better reflect underlying iron status than measured ferritin. Conclusions These findings highlight the substantial informational redundancy present in patient test results and offer a potential foundation for a novel type of clinical decision support aimed at integrating, interpreting, and enhancing the diagnostic value of multianalyte sets of clinical laboratory test results.

Background Increasing the value of health care delivery is a national priority, and providers face growing pressure to reduce cost while improving quality. Ample opportunity exists to increase efficiency and quality simultaneously through the application of systems engineering science. Objective We examined the hypothesis that Lean-based reorganization of laboratory process flow would improve laboratory turnaround times (TAT) and reduce waste in the system. Methods This study was a prospective, before-after analysis of laboratory process improvement in a teaching hospital emergency department (ED). The intervention included a reorganization of laboratory sample flow based in systems engineering science and Lean methodologies, with no additional resources. The primary outcome was the median TAT from sample collection to result for 6 tests previously performed in an ED kiosk. Results After the intervention, median laboratory TAT decreased across most tests. The greatest decreases were found in "reflex tests" performed after an initial screening test: troponin T TAT was reduced by 33 minutes (86 to 53 minutes; 99% confidence interval, 30-35 minutes) and urine sedimentation TAT by 88 minutes (117 to 29 minutes; 99% confidence interval, 87-90 minutes). In addition, troponin I TAT was reduced by 12 minutes, urinalysis by 9 minutes, and urine human chorionic gonadotropin by 10 minutes. Microbiology rapid testing TAT, a "control," did not change. Conclusions In this study, Lean-based reorganization of laboratory process flow significantly increased process efficiency. Broader application of systems engineering science might further improve health care quality and capacity while reducing waste and cost.

A 49-year-old man was admitted to this hospital after being found unresponsive outdoors; he was next to a half-filled bottle of cloudy liquid with pill fragments. An electrocardiogram showed a QTc interval of 501 msec. A diagnostic test was performed.

We evaluated the effectiveness of an expanded point-of-care (POCT) site inspection checklist over an extended 8-y period.A retrospective review of site inspection deficiency reports in a large academic medical center from 2010 to 2017 (year to date).There was a significant decrease in the number of cited deficiencies per site/inspection from 2010 (3.17) to 2017 (0.27) (p0.001). The percentage of sites without deficiencies steadily increased from 2010 (8.7%) to 2017 (80.7%) (p0.001). The most common citation was documentation of competency assessment followed by results documentation and annual procedure review.Regular inspections of sites performing POCT are necessary to maintain regulatory compliance. Over time significant improvements in compliance are achievable.

Objectives: Point-of-care laboratory testing (POCT) offers reduced turnaround time and may promote improved operational efficiency. Few studies have been reported that document improvements from implementing POCT in primary care. Methods: We measured metrics of practice efficiency in a primary care practice before and after implementation of POCT, including the total number of tests ordered, letters and phone calls to patients, and revisits due to abnormal test results. We performed a cost and revenue analysis. Results: Following implementation of POCT, there was a 21% decrease in tests ordered per patient ( P < .0001); a decrease in follow-up phone calls and letters by 89% and 85%, respectively ( P < .0001 and P < .0001); and a 61% decrease in patient revisits ( P = .0002). Estimated testing revenues exceeded expenses by $6.62 per patient, and potential cost savings from improved efficiency were $24.64 per patient. Conclusions: POCT can significantly improve clinical operations with cost reductions through improved practice efficiency.

Objectives: To assess the frequency, magnitude, and prognostic significance of elevations in cardiac troponin T in noncardiac critically ill patients, including elevations at levels below the limit of detection of traditional assays. Methods: Using a high-sensitivity assay, we measured troponin T (high-sensitivity troponin T [hsTnT]) in 451 unique patients within 12 hours of their admission to a noncardiac medical intensive care unit. Outcomes of patients, grouped by hsTnT level, were compared. Results: Overall, 98% of the study patients had detectable levels of hsTnT (>3 ng/L), and 33% had levels above the diagnostic cutoff of a traditional fourth-generation cardiac troponin T assay. Patient groups with higher hsTnT levels had markedly higher rates of in-hospital mortality ( P < .001) and longer stays in the hospital and intensive care unit ( P < .01). Conclusions: In noncardiac critically ill patients, cardiac troponin T elevations are common but often at levels undetectable by traditional assays. hsTnT elevations predict a more complex clinical course and an increased risk of death.

Information systems provide a critical link between clinical laboratories and the clinicians and patients they serve. Strategic deployment of informatics resources can enable a wide array of utilization initiatives and can substantially improve the appropriateness of test selection and results interpretation. In this article, we review information systems including computerized provider order entry (CPOE) systems, laboratory information systems (LISs), electronic health records (EHRs), laboratory middleware, knowledge management systems and systems for data extraction and analysis, and describe the role that each can play in utilization management. We also discuss specific utilization strategies that laboratories can employ within these systems, citing examples both from our own institution and from the literature. Finally, we review how emerging applications of decision support technologies may help to further refine test utilization, "personalize" laboratory diagnosis, and enhance the diagnostic value of laboratory testing.

Expenditures for heath care in developed economies have been rising steadily over the past decades and have become unsustainable in many countries. This is particularly true in the United States (US) where the country spends more on health care as a percent of gross domestic product that any other country. Changes to the US system of reimbursement for medical services are gradually eroding the traditional fee-for-service model in favor of global payments for episodes of care or even entire populations of patients. In this environment utilization management is gaining an increasingly important role in efforts to control cost and improve the quality of care. This chapter introduces many of the key topics in utilization management and provides a “toolbox” of strategies to support implementation of utilization management initiatives. Many examples of successful initiatives in the literature and in our institution will be described. Further details are provided in the accompanying chapters of this book.

Informatics underlies some of the most effective laboratory utilization management tools. For example, computerized provider order entry systems often offer clinical decision support functionality that utilization management teams can leverage to provide clinicians “just-in-time” test-ordering advice and alerts. Likewise clinical and production data can be extracted from laboratory and clinical information systems and analyzed to identify utilization improvement opportunities, guide utilization management strategies, and monitor the impact of utilization management initiatives. This chapter will review applications of informatics to utilization management. More specifically, the chapter will begin with a discussion of clinical decision support, including testing guidelines and algorithms, and various types of electronic alerts. It will also include some discussion of the information systems that underlie many types of decision support. The chapter will next discuss data analytics and will describe strategies to develop informative utilization metrics and use them to measure the impact of utilization management initiatives. Likewise, it will discuss the use of analytics to identify misutilization, target utilization improvement initiatives, and provide valuable feedback to clinicians. The chapter will conclude with a discussion of practical considerations.

Physician profiling is a technique that can be used to identify variations in practice among physicians. These data have been employed to control costs, improve quality, and assess compliance with practice guidelines. Physician profiling is frequently criticized for producing unreliable or misleading data that fails to reflect the realities of actual clinical practice.

Objectives: The biomarker suppression of tumorigenicity 2 (ST2) is a well-established clinical biomarker of cardiac strain and is frequently elevated in a variety of cardiac conditions. Here, we sought to evaluate the prognostic value of ST2 in critically ill medical intensive care unit (MICU) patients without primary cardiac illness. Methods: We measured ST2 and high-sensitivity troponin T (hsTnT) on plasma specimens collected on 441 patients following admission to a noncardiac MICU and evaluated the prognostic power of ST2 both alone and in multivariate models. Results: Of these critically ill patients, 96% exhibited ST2 concentrations above the reference interval. ST2 concentrations were highly predictive of intensive care unit and hospital length of stay, as well as in-hospital mortality, with high concentrations predicting a poor prognosis. Rates of in-hospital mortality were more than four times higher in patients with ST2 concentrations in the highest compared with the lowest quartile. In multivariate analysis, ST2 remained an important predictor of death after adjustment for age, hsTnT, and common diagnoses. Conclusions: ST2 is increased and predictive of prognosis in critically ill patients without primary cardiac disease, suggesting that critically ill patients may often have unrecognized cardiac injury. Clinical decision support algorithms incorporating ST2 and hsTnT results may be useful in patient risk stratification.

Preanalytic laboratory testing errors are often difficult to identify. We demonstrate how laboratories can integrate statistical models with clinical judgment to develop protocols for preanalytic error detection. Specifically, we developed a protocol to identify spuriously elevated glucose values resulting from improper "line draws" or related phlebotomy errors. Using a decision tree-generating algorithm and an annotated set of training data, we generated decision trees to classify critically elevated glucose results as "real" or "spurious" based on available laboratory parameters. Decision trees revealed that a 30-day patient-specific average glucose concentration lower than 186.3 mg/dL (10.3 mmol/L), a current glucose concentration higher than 663 mg/dL (37 mmol/L), and an anion gap lower than 16.5 mEq/L (16.5 mmol/L) suggested a spurious result. We then used the results from the decision tree analysis to inform the implementation of a clinical protocol that significantly improved the laboratory's identification of spurious results. Similar approaches may be useful in developing protocols to identify other errors or to assist in clinical interpretation of results.

Amphetamines and methamphetamines are part of an important class of drugs included in most urine drugs of abuse screening panels, and a common assay to detect these drugs is the Amphetamines II immunoassay (Roche Diagnostics). To demonstrate that meta-chlorophenylpiperazine (m-CPP), a trazodone metabolite, cross-reacts in the Amphetamines II assay, we tested reference standards of m-CPP at various concentrations (200 to 20,000 g/L). We also tested real patient urine samples containing m-CPP (detected and quantified by HPLC) with no detectable amphetamine, methamphetamine, or MDMA (demonstrated by GC MS). In both the m-CPP standards and the patient urine samples, we found a strong association between m-CPP concentration and Amphetamines II immunoreactivity (r = 0.990 for the urine samples). Further, we found that patients taking trazodone can produce urine with sufficient m-CPP to result in false-positive Amphetamines II results. At our institution, false-positive amphetamine results occur not infrequently in patients taking trazodone with at least 8 trazodone-associated false-positive results during a single 26-day period. Laboratories should remain cognizant of this interference when interpreting results of this assay.