Automated Electronic Health Record-Based Tool for Identification of Patients With Metastatic Disease to Facilitate Clinical Trial Patient Ascertainment.

PURPOSE: To facilitate identification of clinical trial participation candidates, we developed a machine learning tool that automates the determination of a patient's metastatic status, on the basis of unstructured electronic health record (EHR) data. METHODS: This tool scans EHR documents, extracting text snippet features surrounding key words (such as metastatic, progression, and local). A regularized logistic regression model was trained and used to classify patients across five metastatic categories: highly likely and likely positive, highly likely and likely negative, and unknown. Using a real-world oncology database of patients with solid tumors with manually abstracted information as reference, we calculated sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV). We validated the performance in a real-world data set, evaluating accuracy gains upon additional user review of tool's outputs after integration into clinic workflows. RESULTS: In the training data set (N = 66,532), the model sensitivity and specificity (% [95% CI]) were 82.4 [81.9 to 83.0] and 95.5 [95.3 to 96.7], respectively; the PPV was 89.3 [88.8 to 90.0], and the NPV was 94.0 [93.8 to 94.2]. In the validation sample (n = 200 from five distinct care sites), after user review of model outputs, values increased to 97.1 [85.1 to 99.9] for sensitivity, 98.2 [94.8 to 99.6] for specificity, 91.9 [78.1 to 98.3] for PPV, and 99.4 [96.6 to 100.0] for NPV. The model assigned 163 of 200 patients to the highly likely categories. The error prevalence was 4% before and 2% after user review. CONCLUSION: This tool infers metastatic status from unstructured EHR data with high accuracy and high confidence in more than 75% of cases, without requiring additional manual review. By enabling efficient characterization of metastatic status, this tool could mitigate a key barrier for patient ascertainment and clinical trial participation in community clinics.

The development of a method to measure head acceleration and motion in high-impact crashes.

OBJECTIVE: To establish a mechanism to obtain precise measurements of the accelerations of the head in the high-speed racing environment and during crash impacts. The long-term goal is to apply this system to the assessment of head injury in automobile racing drivers and then in participants in other helmeted sports. METHODS: A multidisciplinary team conceptualized, designed, and successfully tested a triaxial accelerometer system capable of measuring head acceleration and motion in high-impact crashes. The system has been implemented successfully in the professional racing environment. RESULTS: Accurate and reproducible data have been obtained from the accelerometer system in tests on manikins and cadavers and in actual racing events. The system has been implemented in two professional racing series in 2003. Information gained from the accelerometer system is currently being entered into a database. Eventually, the data should aid in the development of improved cockpit head protection in racing cars. Improved helmet design not only in motor sports but also in other helmeted sports should benefit from the data collected. These data will also aid the development of improved head injury protection in military aircraft and passenger vehicles. CONCLUSION: Although there has been a significant decrease in the overall rate of injury during the past 25 years, head injury remains a serious safety concern in motor sports and the greatest cause of death. Sports-related head injuries are also of great concern because repeated mild head injury has become an important health issue with potential long-term disability. True human tolerance to brain injury has yet to be established. Our scientific knowledge of brain injury is currently based on results derived from manikin, cadaver, and human volunteer testing, along with animal and computer models. The racing environment represents a venue to ethically measure and evaluate the forces involved in human brain injury.

Physiological responses to high-speed, open-wheel racecar driving.

UNLABELLED: High-speed auto racing has been demonstrated to produce accelerated heart rate (HR) during competition. However, it has not been determined whether the increase in HR was due to physical work efforts or a result of emotional stress. PURPOSE: The purpose of this investigation was to examine the physiological responses associated with open-wheel automobile driving at competitive speeds. METHODS: Oxygen consumption and HR were assessed in seven professional automobile racing drivers during two incrementally paced driving sessions. A portable metabolic analyzer and EKG were directly attached to the subjects as they participated in driving tests on an oval speedway and a roadway course. Maximal physiological responses of the subjects were also determined during a graded treadmill test. RESULTS: During treadmill testing, maximal oxygen consumption (VO2max) ranged from of 42.0 to 59.7 mL x kg(-1) x min(-1) (mean +/- SD = 47.6+/-8.1). The road course and oval speedway testing at competitive speeds elicited mean VO2 values of 38.5 and 21.9 mL x kg(-1) x min(-1), respectively, which correspond to 79% and 45% of VO2max. Road course driving produced mean HR values of 152 beats x min(-1) with 142 beats x min(-1) recorded when driving at competitive speed on the speedway course. CONCLUSIONS: Professional open-wheel race drivers possess cardiorespiratory capacity similar to athletes participating in sports such as basketball, football, and baseball. The VO2 and HR responses to road course driving were similar to those previously reported in traditional sports settings. The findings of this study suggest that professional open-wheel racing drivers should be regarded as athletes that encounter significant physiological stresses.(2)

Routine daily chest radiography in patients with pulmonary artery catheters.

BACKGROUND: Pulmonary artery catheters are widely used invasive monitoring devices in critically ill patients. Clinicians disagree about whether daily chest radiographs are needed or clinical parameters alone are sufficient to verify catheter placement. OBJECTIVES: To determine whether daily chest radiographs are needed to assess migration of pulmonary artery catheters. METHODS: One hundred consecutive patients with pulmonary artery catheters were prospectively evaluated. Clinical criteria for optimal position of the pulmonary artery catheters and findings on chest radiographs were compared. Optimal clinical criteria were (1) amount of air required to measure pulmonary capillary wedge pressure: 1.25 to 1.5 mL and (2) pulmonary artery catheter migrated 1 cm or less from initial position. RESULTS: Three hundred ninety comparisons of clinical criteria and radiographic findings were done. Chest radiographs indicated the catheter required repositioning in 15 (4%) of 390 instances but in only 4 (1%) of 310 instances in which bedside clinical findings indicated adequate catheter position. In 69 (18%) of the 390 cases, the clinical criteria for adequate catheter position were not met, but radiographs showed the catheter in an appropriate position. The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of abnormal clinical criteria were 73%, 82%, 81%, 14%, and 99%, respectively. CONCLUSIONS: Chest radiographs indicated that about 4% of catheters required repositioning. Catheter malposition can be reliably excluded (negative predictive value, 99%) by close observation of specific clinical criteria, so routine daily chest radiographs do not seem justified.

Unexpected findings in trauma patients dying in the intensive care unit: results of 153 consecutive autopsies.

BACKGROUND: The true incidence of missed injuries in trauma-related deaths is unknown, because in only about 60% of injury-related deaths nationwide is an autopsy performed. Few studies have documented the frequency of missed diagnoses leading to deaths specifically in the trauma ICU population. We attempted to evaluate the incidence and nature of missed injuries and complications in trauma- and burn-related deaths in our ICU given an autopsy rate of close to 100%. STUDY DESIGN: The medical records of all trauma- and burn-related deaths in the ICU over a 2-year period were reviewed retrospectively. Missed diagnoses were classified as class 1: major diagnosis that if recognized and treated appropriately might have changed outcomes; class II: major diagnosis that if recognized and treated appropriately would not have changed outcomes; and class III: minor diagnosis. RESULTS: Complete antemortem records were available for 158 patients, of which 153 (97%) underwent autopsy. Mean age was 50 years, and 72% were males. Mean ICU stay was 10 15 days. Four (3%) patients had class I missed diagnoses: bowel infarction, meningitis, retroperitoneal abscess, and bleeding gastric ulcer. Twenty-five (16%) patients had class II diagnoses, and 12 (8%) patients had class III diagnoses. Overall, 81% of 153 patients had either class III diagnoses or no missed injuries or complications. Pneumonia was the most common missed diagnosis. CONCLUSIONS: With an autopsy rate of 97%, 3% of deaths bad missed major diagnoses that might have affected outcomes if recognized antemortem. Autopsy findings can still provide valuable feedback in Improving the quality of care of critically ill trauma patients.

Oral creatine supplementation enhances upper extremity work capacity in persons with cervical-level spinal cord injury.

OBJECTIVE: To examine the effects of short-term creatine monohydrate supplementation on the upper extremity work capacity of persons with cervical-level spinal cord injury (SCI). DESIGN: Randomized, double-blind, placebo-controlled, crossover design study. Consists of 2 treatment phases lasting for 7 days, separated by a 21-day washout period. SETTING: University research laboratory trial. PARTICIPANTS: Sixteen men with complete cervical-level SCI (C5-7). INTERVENTION: Subjects were randomly assigned to 1 of 2 groups and received either 20g/d of creatine monohydrate supplement powder or placebo maltodextrin powder for the first treatment phase; the treatment was reversed in the second phase. Incremental peak arm ergometry tests, using 2-minute work stages and 1-minute recovery periods, were performed immediately before and after each treatment phase (total of 4 assessments). The initial stage was performed unloaded, with power output progressively increased 10 watts/stage until subjects had achieved volitional exhaustion. MAIN OUTCOME MEASURES: Peak power output, time to fatigue, heart rate, and metabolic measurements, including oxygen uptake (VO2), minute ventilation, tidal volume (VT), and respiration frequency. RESULTS: Significantly greater values of VO2, VCO2, and VT at peak effort after creatine supplementation (P <.001). CONCLUSIONS: Creatine supplementation enhances the exercise capacity in persons with complete cervical-level SCI and may promote greater exercise training benefits.