Search PubMed⌕ Search

Biomedical subjects

Josh F Peterson

Publications and source records attributed to Josh F Peterson.

At least 19 recordsLinked to original sources
PubMed2026

An EHR-based framework for modeling growth curves and constructing growth centile charts for genetic disorders.

Growth modeling is central to human genetics, as deviations from typical growth can signal an underlying disorder. In this cohort study, we developed a generalizable framework for generating growth charts across genetic conditions using electronic health records (EHR). Leveraging 22 years of longitudinal EHR data from 452,470 patients across 15 genetic conditions and unaffected individuals, we generated sex- and condition-specific growth charts using Generalized Additive Models for Location, Scale, and Shape, and quantified differences in size, timing, and intensity using SuperImposition by Translation and Rotation (SITAR). SITAR-derived growth parameters showed strong concordance with established annotations in OMIM and Orphanet, and identified previously unreported growth patterns. We stratified cystic fibrosis by CFTR functional class and observed greater growth impairment in individuals with homozygous minimal-function variants compared to those with residual function. This framework provides a generalizable approach for leveraging EHR data to refine genotype-phenotype relationships and enable continuous updating of growth charts across genetic conditions.

Journal Article↗
PubMed2026

Genotype-Guided Antidepressant Prescribing for Patients With Depression: A Randomized Clinical Trial.

IMPORTANCE: The effectiveness of pharmacogenetics to guide prescribing of selective serotonin reuptake inhibitors (SSRIs) for depression remains unclear, despite the well-established association between SSRI pharmacokinetics and genetic variation. OBJECTIVE: To determine whether pharmacogenetic-guided prescribing of SSRIs improves treatment response in patients with depression. DESIGN, SETTING, AND PARTICIPANTS: The ADOPT PGx (A Depression and Opioid Pragmatic Trial in Pharmacogenetics) Depression pragmatic randomized clinical trial was conducted from August 10, 2021, through April 27, 2024, at primary care, psychiatry, or family medicine clinics at enrolling sites throughout the US. Patients were aged 8 years or older and had experienced depression for 3 months or longer. INTERVENTION: Patients were randomized to genotype-guided SSRI prescribing (intervention group) or usual care (control group). Actionable drug metabolism phenotypes were defined as those for which pharmacogenetic clinical guidelines recommend alternative medication selection or dose adjustment. MAIN OUTCOMES AND MEASURES: The primary outcome was change in Patient-Reported Outcomes Measurement Information System (PROMIS) depression T scores at 3 months among patients with the actionable phenotype. Secondary end points included adverse effect severity of SSRIs at 3 months and depression remission (measured with PROMIS depression scores and Patient Health Questionnaire-8 [PHQ-8] scores) at 6 months. RESULTS: This study of 1460 patients included 1239 adults (84.9%) (mean [SD] age, 40.6 [16.7] years) and 221 children (15.1%) (mean [SD] age, 14.6 [1.8] years). Most patients were female (1096 [75.1%]). A total of 692 patients (47.4%) had an actionable phenotype; 351 (50.7%) were assigned to the intervention, and 341 (49.3%) were assigned to usual care. At baseline, 463 of the 692 patients (66.9%) reported having depressive symptoms for more than 2 years, 603 (87.1%) were receiving pharmacologic treatment, and 354 (51.2%) were receiving nonpharmacologic treatment. At 3 months, no significant differences were observed between the intervention and usual care groups in change in PROMIS depression T scores (mean [SD] change, -4.3 [8.4] vs -4.0 [8.1]; P = .68), medication adverse effect burden (mean [SD] change, 8.2 [4.3] vs 7.8 [4.5]; P = .37), or Patient Health Questionnaire-8 score change (mean [SD] change, -3.3 [5.2] vs -2.7 [4.8]; P  = .13). However, at 6 months, the PROMIS depression T-score remission rate (score ≤16) was higher in the intervention group compared with the usual care group (153 of 317 patients [48.3%] vs 122 of 310 patients [39.4%]; P = .02). CONCLUSIONS AND RELEVANCE: In this randomized clinical trial, genotype-guided prescribing of SSRIs did not improve control of depression symptoms at 3 months compared with usual care but was associated with higher depression remission rates at 6 months. These findings suggest a possible longer-term clinical benefit and indicate that future studies should focus on the durability and long-term impact of genotype-guided prescribing in the management of depressive symptoms. TRIAL REGISTRATION: ClinicalTrials.gov Identifiers: NCT04445792 (Master Protocol Research Program platform trial) and NCT05966155 (ADOPT PGx Depression trial).

Humans↗
PubMed2026

Evaluating pregnancy and neonatal outcomes in mothers with genetic disease using electronic health care records.

PURPOSE: The effect of Mendelian disorders on pregnancy and neonatal outcomes is poorly understood because of their rarity and the challenge of compiling complete prenatal and postnatal records. METHODS: Using electronic health records from a single academic center, we developed a retrospective cohort of maternal-infant dyads. Cases were mothers with molecularly confirmed Mendelian disorders paired with live-born infants; controls had no documented genetic disease. Outcomes were evaluated overall, by organ system, and by individual disorder. RESULTS: The cohort included 58,912 dyads, 241 with genetic diagnoses and 58,671 controls. Although overall outcomes were generally favorable, mothers with genetic disorders had higher rates of cesarean delivery and neonatal intensive care unit admission, earlier gestational age, and lower Apgar scores. Neonatal risks were greatest among neurological and cardiovascular disorders. Known associations were replicated, including increased neonatal intensive care unit admission in 22q11.2 deletion syndrome, cesarean delivery in Turner syndrome, and gestational diabetes in cystic fibrosis. We also provided descriptive electronic-health-record-based case reports and case series for 35 disorders previously lacking published pregnancy outcome data. CONCLUSION: This study identifies elevated perinatal risks in specific Mendelian disease groups and demonstrates how electronic-health-record-linked data can support prenatal counseling, risk stratification, and individualized care for individuals with genetic disorders.

Humans↗
PubMed2025

Covering medical care costs for participants in the eMERGE Network: Challenges for equity and implementation.

PURPOSE: To investigate the complexities of covering study-recommended medical care costs for individuals (in order to prevent lack of adherence due to financial reasons), which have received little attention. METHODS: We explored the deliberations, decisions, and challenges faced by the Electronic Medical Records and Genomics (eMERGE) Network during the implementation of a genomic research project recommending clinical care based on high-risk results defined largely by polygenic risk scores. Two surveys were disseminated to eMERGE sites: to identify preferences about payment for specific care recommendations (survey 1) and to understand the operational processes of covering medical care costs (survey 2). RESULTS: Paying for a subset of care recommendations for the funded study duration was identified as the most feasible approach for covering medical care costs for participants who received high-risk genomic results. Each eMERGE site, by necessity, used diverse approaches to pay for medical care costs. CONCLUSION: eMERGE researchers balanced competing concerns about bias, equity, study design, regulatory compliance, and cost in designing a unified approach to cover some of the recommended medical care costs in the study. Many implementation challenges were encountered. Findings can inform researchers and regulatory bodies about the implications and complications of covering medical care costs in translational research studies focused on prevention.

Humans↗
PubMed2025

Characterizing trends in clinical genetic testing: A single-center analysis of EHR data from 1.8 million patients over two decades.

A lack of structural data in electronic health records (EHRs) makes assessing the impact of genetic testing on clinical practice challenging. We extracted clinical genetic tests from the EHRs of more than 1.8 million patients seen at Vanderbilt University Medical Center from 2002 to 2022. With these data, we quantified the use of clinical genetic testing in healthcare and described how testing patterns and results changed over time. We assessed trends in types of genetic tests, tracked usage across medical specialties, and introduced a new measure, the genetically attributable fraction (GAF), to quantify the proportion of observed phenotypes attributable to a genetic diagnosis over time. We identified 104,392 tests and 19,032 molecularly confirmed diagnoses. The proportion of patients with genetic testing in their EHRs increased from 1.0% in 2002 to 6.1% in 2022, and testing became more comprehensive with the growing use of multi-gene panels. The number of unique diseases diagnosed with genetic testing increased from 51 in 2002 to 509 in 2022, and there was a rise in the number of variants of uncertain significance. The phenome-wide GAF for 6,505,620 diagnoses made in 2022 was 0.46%, and the GAF was greater than 5% for 74 phenotypes, including pancreatic insufficiency (67%), chorea (64%), atrial septal defect (24%), microcephaly (17%), paraganglioma (17%), and ovarian cancer (6.8%). Our study provides a comprehensive quantification of the increasing role of genetic testing at a major academic medical institution and demonstrates its growing utility in explaining the observed medical phenome.

Humans↗
PubMed2024

Cost-effectiveness of population-wide genomic screening for Lynch Syndrome and polygenic risk scores to inform colorectal cancer screening.

PURPOSE: Genomic screening to identify individuals with Lynch Syndrome (LS) and those with a high polygenic risk score (PRS) promises to personalize colorectal cancer (CRC) screening. Understanding its clinical and economic impact is needed to inform screening guidelines and reimbursement policies. METHODS: We developed a Markov model to simulate individuals over a lifetime. We compared LS+PRS genomic screening with standard of care (SOC) for a cohort of US adults at age 30. The Markov model included health states of no CRC, CRC stages (A-D), and death. We estimated incidence, mortality, and discounted economic outcomes of the population under different interventions. RESULTS: Screening 1000 individuals for LS+PRS resulted in 1.36 fewer CRC cases and 0.65 fewer deaths compared with SOC. The incremental cost-effectiveness ratio was $124,415 per quality-adjusted life year; screening had a 69% probability of being cost-effective using a willingness-to-pay threshold of $150,000/quality-adjusted life year . Setting the PRS threshold at the 90th percentile of the LS+PRS screening program to define individuals at high risk was most likely to be cost-effective compared with 95th, 85th, and 80th percentiles. CONCLUSION: Population-level LS+PRS screening is marginally cost-effective, and a threshold of 90th percentile is more likely to be cost-effective than other thresholds.

Humans↗
PubMed2024

Diagnostic delay in monogenic disease: A scoping review.

PURPOSE: Diagnostic delay in monogenic disease is reportedly common. We conducted a scoping review investigating variability in study design, results, and conclusions. METHODS: We searched the academic literature on January 17, 2023, for original peer reviewed journals and conference articles that quantified diagnostic delay in monogenic disease. We abstracted the reported diagnostic delay, relevant study design features, and definitions. RESULTS: Our search identified 259 articles quantifying diagnostic delay in 111 distinct monogenetic diseases. Median reported diagnostic delay for all studies collectively in monogenetic diseases was 5.0 years (IQR 2-10). There was major variation in the reported delay within individual monogenetic diseases. Shorter delay was associated with disorders of childhood metabolism, immunity, and development. The majority (67.6%) of articles that studied delay reported an improvement with calendar time. Study design and definitions of delay were highly heterogenous. Three gaps were identified: (1) no studies were conducted in the least developed countries, (2) delay has not been studied for the majority of known, or (3) most prevalent genetic diseases. CONCLUSION: Heterogenous study design and definitions of diagnostic delay inhibit comparison across studies. Future efforts should focus on standardizing delay measurements, while expanding the research to low-income countries.

Humans↗
PubMed2023

The phenotype-genotype reference map: Improving biobank data science through replication.

Population-scale biobanks linked to electronic health record data provide vast opportunities to extend our knowledge of human genetics and discover new phenotype-genotype associations. Given their dense phenotype data, biobanks can also facilitate replication studies on a phenome-wide scale. Here, we introduce the phenotype-genotype reference map (PGRM), a set of 5,879 genetic associations from 523 GWAS publications that can be used for high-throughput replication experiments. PGRM phenotypes are standardized as phecodes, ensuring interoperability between biobanks. We applied the PGRM to five ancestry-specific cohorts from four independent biobanks and found evidence of robust replications across a wide array of phenotypes. We show how the PGRM can be used to detect data corruption and to empirically assess parameters for phenome-wide studies. Finally, we use the PGRM to explore factors associated with replicability of GWAS results.

Humans↗
PubMed2006

Fluconazole for empiric antifungal therapy in cancer patients with fever and neutropenia.

BACKGROUND: Several clinical trials have demonstrated the efficacy of fluconazole as empiric antifungal therapy in cancer patients with fever and neutropenia. Our objective was to assess the frequency and resource utilization associated with treatment failure in cancer patients given empiric fluconazole antifungal therapy in routine inpatient care. METHODS: We performed a retrospective cohort study of cancer patients treated with oral or intravenous fluconazole between 7/97 and 6/01 in a tertiary care hospital. The final study cohort included cancer patients with neutropenia (an absolute neutrophil count below 500 cells/mm3) and fever (a temperature above 38 degrees C or 100.4 degrees F), who were receiving at least 96 hours of parenteral antibacterial therapy prior to initiating fluconazole. Patients' responses to empiric therapy were assessed by reviewing patient charts. RESULTS: Among 103 cancer admissions with fever and neutropenia, treatment failure after initiating empiric fluconazole antifungal therapy occurred in 41% (95% confidence interval (CI) 31%-50%) of admissions. Patients with a diagnosis of hematological malignancy had increased risk of treatment failure (OR = 4.6, 95% CI 1.5-14.8). When treatment failure occurred the mean adjusted increases in length of stay and total costs were 7.4 days (95% CI 3.3-11.5) and $18,925 (95% CI 3,289-34,563), respectively. CONCLUSION: Treatment failure occurred in more than one-third of neutropenic cancer patients on fluconazole as empiric antifungal treatment for fever in routine clinical treatment. The increase in costs when treatment failure occurs is substantial.

Adult↗
PubMed2006

Delirium and its motoric subtypes: a study of 614 critically ill patients.

OBJECTIVES: To describe the motoric subtypes of delirium in critically ill patients and compare patients aged 65 and older with a younger cohort. DESIGN: Prospective cohort study. SETTING: The medical intensive care unit (MICU) of a tertiary care academic medical center. PARTICIPANTS: Six hundred fourteen MICU patients admitted during a process improvement initiative to monitor levels of sedation and delirium. MEASUREMENTS: MICU nursing staff assessed delirium and level of consciousness in all MICU patients at least once per 12-hour shift using the Confusion Assessment Method for the Intensive Care Unit and the Richmond Agitation-Sedation Scale. Delirium episodes were categorized as hypoactive, hyperactive, and mixed type. RESULTS: Delirium was detected in 112 of 156 (71.8%) subjects aged 65 and older and 263 of 458 (57.4%) subjects younger than 65. Mixed type was most common (54.9%), followed by hypoactive delirium (43.5%) and purely hyperactive delirium (1.6%). Patients aged 65 and older experienced hypoactive delirium at a greater rate than younger patients (41.0% vs 21.6%, P<.001) and never experienced hyperactive delirium. Older age was strongly and independently associated with hypoactive delirium (adjusted odds ratio=3.0, 95% confidence interval=1.7-5.3), compared with no delirium in a model that adjusted for other important determinants of delirium including severity of illness, sedative medication use, and ventilation status. CONCLUSION: Older age is a strong predictor of hypoactive delirium in MICU patients, and this motoric subtype of delirium may be missed in the absence of active monitoring.

Age Factors↗
PubMed2005

Intensive care unit delirium is an independent predictor of longer hospital stay: a prospective analysis of 261 non-ventilated patients.

INTRODUCTION: Delirium occurs in most ventilated patients and is independently associated with more deaths, longer stay, and higher cost. Guidelines recommend monitoring of delirium in all intensive care unit (ICU) patients, though few data exist in non-ventilated patients. The study objective was to determine the relationship between delirium and outcomes among non-ventilated ICU patients. METHOD: A prospective cohort investigation of 261 consecutively admitted medical ICU patients not requiring invasive mechanical ventilation during hospitalization at a tertiary-care, university-based hospital between February 2002 and January 2003. ICU nursing staff assessed delirium and level of consciousness at least twice per day using the Confusion Assessment Method for the ICU (CAM-ICU) and Richmond Agitation-Sedation Scale (RASS). Cox regression with time-varying covariates was used to determine the independent relationship between delirium and clinical outcomes. RESULTS: Of 261 patients, 125 (48%) experienced at least one episode of delirium. Patients who experienced delirium were older (mean +/- SD: 56 +/- 18 versus 49 +/- 17 years; p = 0.002) and more severely ill as measured by Acute Physiology and Chronic Health Evaluation II (APACHE II) scores (median 15, interquartile range (IQR) 10-21 versus 11, IQR 6-16; p < 0.001) compared to their non-delirious counterparts. Patients who experienced delirium had a 29% greater risk of remaining in the ICU on any given day (compared to patients who never developed delirium) even after adjusting for age, gender, race, Charlson co-morbidity score, APACHE II score, and coma (hazard ratio (HR) 1.29; 95% confidence interval (CI) 0.98-1.69, p = 0.07). Similarly, patients who experienced delirium had a 41% greater risk of remaining in the hospital after adjusting for the same covariates (HR 1.41; 95% CI 1.05-1.89, p = 0.023). Hospital mortality was higher among patients who developed delirium (24/125, 19%) versus patients who never developed delirium (8/135, 6%), p = 0.002; however, time to in-hospital death was not significant the adjusted (HR 1.27; 95% CI 0.55-2.98, p = 0.58). CONCLUSION: Delirium occurred in nearly half of the non-ventilated ICU patients in this cohort. Even after adjustment for relevant covariates, delirium was found to be an independent predictor of longer hospital stay.

Cohort Studies↗
PubMed2005

Guided prescription of psychotropic medications for geriatric inpatients.

BACKGROUND: Inappropriate use or excessive dosing of psychotropic medications in the elderly is common and can lead to a variety of adverse drug events including falls, oversedation, and cognitive impairment. METHODS: We developed a database of psychotropic medication dosing and selection guidelines for elderly inpatients. We displayed these recommendations to physicians through a computerized order entry system at a tertiary care academic hospital. The system was activated for 2 of 4 six-week study periods in an off-on-off-on pattern. Main outcome measures were agreement with the recommended daily dose for the initial drug order, incidence of dosing at least 10-fold greater than the recommended daily dose, prescription of nonrecommended drugs, inpatient falls, altered mental status as measured by a brief nursing assessment, and hospital length of stay. RESULTS: A total of 7456 initial orders for psychotropic medications were prescribed for 3718 hospitalized elderly patients with a mean +/- SD age of 74.7 +/- 6.7 years. The intervention increased the prescription of the recommended daily dose (29% vs 19%; P<.001), reduced the incidence of 10-fold dosing (2.8% vs 5.0%; P<.001), and reduced the prescription of nonrecommended drugs (10.8% vs 7.6% of total orders; P<.001). Patients in the intervention cohort had a lower in-hospital fall rate (0.28 vs 0.64 falls per 100 patient-days; P = .001). No effect on hospital length of stay or days of altered mental status was found. CONCLUSION: A geriatric decision support system for psychotropic medications increased the prescription of recommended doses, reduced the prescription of nonrecommended drugs, and was associated with fewer inpatient falls.

Accidental Falls↗
PubMed2005

Frequency of potential azole drug-drug interactions and consequences of potential fluconazole drug interactions.

PURPOSE: To assess the frequency of potential azole-drug interactions and consequences of interactions between fluconazole and other drugs in routine inpatient care. METHODS: We performed a retrospective cohort study of hospitalized patients treated for systemic fungal infections with an oral or intravenous azole medication between July 1997 and June 2001 in a tertiary care hospital. We recorded the concomitant use of medications known to interact with azole antifungals and measured the frequency of potential azole drug interactions, which we considered to be present when both drugs were given together. We then performed a chart review on a random sample of admissions in which patients were exposed to a potential moderate or major drug interaction with fluconazole. The list of azole-interacting medications and the severity of interaction were derived from the DRUGDEX System and Drug Interaction Facts. RESULTS: Among the 4,185 admissions in which azole agents (fluconazole, itraconazole or ketoconazole) were given, 2,941 (70.3%) admissions experienced potential azole-drug interactions, which included 2,716 (92.3%) admissions experiencing potential fluconazole interactions. The most frequent interactions with potential moderate to major severity were co-administration of fluconazole with prednisone (25.3%), midazolam (17.5%), warfarin (14.7%), methylprednisolone (14.1%), cyclosporine (10.7%) and nifedipine (10.1%). Charts were reviewed for 199 admissions in which patients were exposed to potential fluconazole drug interactions. While four adverse drug events (ADEs) caused by fluconazole were found, none was felt to be caused by a drug-drug interaction (DDI), although in one instance fluconazole may have contributed. CONCLUSIONS: Potential fluconazole drug interactions were very frequent among hospitalized patients on systemic azole antifungal therapy, but they had few apparent clinical consequences.

Adverse Drug Reaction Reporting Systems↗
PubMed2005

Large-scale implementation of sedation and delirium monitoring in the intensive care unit: a report from two medical centers.

OBJECTIVE: To implement sedation and delirium monitoring via a process-improvement project in accordance with Society of Critical Care Medicine guidelines and to evaluate the challenges of modifying intensive care unit (ICU) organizational practice styles. DESIGN: Prospective observational cohort study. SETTING: The medical ICUs at two institutions: the Vanderbilt University Medical Center (VUMC) and a community Veterans Affairs hospital (York-VA). SUBJECTS: Seven hundred eleven patients admitted to the medical ICUs for >24 hrs and followed over 4,163 days during a 21-month study period. INTERVENTIONS: Unit-wide nursing documentation was changed to accommodate a sedation scale (Richmond Agitation-Sedation Scale) and delirium instrument (Confusion Assessment Method for the ICU). A 20-min introductory in-service was performed for all ICU nurses, followed by graded, staged educational interventions at regular intervals. Data were collected daily for compliance, and randomly 40% of nurses each day were chosen for accuracy spot-checks by reference raters. An implementation survey questionnaire was distributed at 6 months. MEASUREMENTS AND MAIN RESULTS: The implementation project involved 64 nurses (40 at VUMC and 24 at York-VA). Sedation and delirium monitoring data were recorded for 711 patients (614 at VUMC and 97 at York-VA). Compliance with the Richmond Agitation-Sedation Scale was 94.4% (21,931 of 23,220) at VUMC and 99.7% (5,387 of 5,403) at York-VA. Compliance with the Confusion Assessment Method for the ICU was 90% (7,323 of 8,166) at VUMC and 84% (1,571 of 1,871) at York-VA. The Confusion Assessment Method for the ICU was performed more often than requested on 63% of shifts (5,146 of 8,166) at VUMC and on 8% (151 of 1871) of shifts at York-VA. Overall weighted-kappa between bedside nurses and references raters for the Richmond Agitation-Sedation Scale were 0.89 (95% confidence interval, 0.88 to 0.92) at VUMC and 0.77 (95% confidence interval, 0.72 to 0.83) at York-VA. Overall agreement (kappa) between bedside nurses and reference raters using the Confusion Assessment Method for the ICU was 0.92 (95% confidence interval, 0.90-0.94) at VUMC and 0.75 (95% confidence interval, 0.68-0.81) at York-VA. The two most-often-cited barriers to implementation were physician buy-in and time. CONCLUSIONS: With minimal training, the compliance of bedside nurses using sedation and delirium instruments was excellent. Agreement of data from bedside nurses and a reference-standard rater was very high for both the sedation scale and the delirium assessment over the duration of this process-improvement project.

Adult↗
PubMed2005

Adverse drug events occurring following hospital discharge.

OBJECTIVE: To describe the incidence of adverse drug events (ADEs), preventable ADEs, and ameliorable ADEs occurring after hospital discharge and their associated risk factors. DESIGN: Prospective cohort study. SETTING: Urban academic health sciences center. PATIENTS: Consecutive patients discharged home from the general medical service. INTERVENTIONS: We determined posthospital outcomes approximately 24 days following discharge by performing a chart review and telephone interview. Using the telephone interview, we identified new or worsening symptoms, the patient's health system use, and recollection of processes of care. Posthospital outcomes were judged by 2 internists independently. RESULTS: Four hundred of 581 potentially eligible patients were evaluated. Of the 400 patients, 45 developed an ADE (incidence, 11%; 95% confidence interval [CI], 8% to 14%). Of these, 27% were preventable and 33% were ameliorable. Injuries were significant in 32 patients, serious in 6, and life threatening in 7. Patients were less likely to experience an ADE if they recalled having side effects of prescribed medications explained (OR, 0.4; 95% CI, 0.2 to 0.8). The risk of ADE per prescription was highest for corticosteroids, anticoagulants, antibiotics, analgesics, and cardiovascular medications. Risk increased with prescription number. Failure to monitor was an especially common cause of preventable and ameliorable ADEs. CONCLUSION: Following discharge, ADEs were common and many were preventable or ameliorable. Medication side effects should be discussed, and interventions should include better monitoring and target patients receiving specific drug classes or multiple medications.

Boston↗
PubMed2005

Identifying UMLS concepts from ECG Impressions using KnowledgeMap.

Electrocardiogram (ECG) impressions represent a wealth of medical information for potential decision support and drug-effect discovery. Much of this information is inaccessible to automated methods in the free-text portion of the ECG report. We studied the application of the KnowledgeMap concept identifier (KMCI) to map Unified Medical Language System (UMLS) concepts from ECG impressions. ECGs were processed by KMCI and the results scored for accuracy by multiple raters. Reviewers also recorded unidentified concepts through the scoring interface. Overall, KMCI correctly identified 1059 out of 1171 concepts for a recall of 0.90. Precision, indicating the proportion of ECG concepts correctly identified, was 0.94. KMCI was particularly effective at identifying ECG rhythms (330/333), perfusion changes (65/66), and noncardiac medical concepts (11/11). In conclusion, KMCI is an effective method for mapping ECG impressions to UMLS concepts.

Abstracting and Indexing↗
PubMed2003

Electronically screening discharge summaries for adverse medical events.

OBJECTIVE: Detecting adverse events is pivotal for measuring and improving medical safety, yet current techniques discourage routine screening. The authors hypothesized that discharge summaries would include information on adverse events, and they developed and evaluated an electronic method for screening medical discharge summaries for adverse events. DESIGN: A cohort study including 424 randomly selected admissions to the medical services of an academic medical center was conducted between January and July 2000. The authors developed a computerized screening tool that searched free-text discharge summaries for trigger words representing possible adverse events. MEASUREMENTS: All discharge summaries with a trigger word present underwent chart review by two independent physician reviewers. The presence of adverse events was assessed using structured implicit judgment. A random sample of discharge summaries without trigger words also was reviewed. RESULTS: Fifty-nine percent (251 of 424) of the discharge summaries contained trigger words. Based on discharge summary review, 44.8% (327 of 730) of the alerted trigger words indicated a possible adverse event. After medical record review, the tool detected 131 adverse events. The sensitivity and specificity of the screening tool were 69% and 48%, respectively. The positive predictive value of the tool was 52%. CONCLUSION: Medical discharge summaries contain information regarding adverse events. Electronic screening of discharge summaries for adverse events using keyword searches is feasible but thus far has poor specificity. Nonetheless, computerized clinical narrative screening methods could potentially offer researchers and quality managers a means to routinely detect adverse events.

Cohort Studies↗