A scoping review of empathy recognition in text using natural language processing.

OBJECTIVE: To provide a scoping review of studies on empathy recognition in text using natural language processing (NLP) that can inform an approach to identifying physician empathic communication over patient portal messages. MATERIALS AND METHODS: We searched 6 databases to identify relevant studies published through May 1, 2023. The study selection was conducted through a title screening, an abstract review, and a full-text review. Our process followed the PRISMA-ScR guidelines. RESULTS: Of the 2446 publications identified from our searches, 39 studies were selected for the final review, which summarized: (1) definitions and context of empathy, (2) data sources and tested models, and (3) model performance. Definitions of empathy varied in their specificity to the context and setting of the study. The most common settings in which empathy was studied were reactions to news stories, health-related social media forums, and counseling sessions. We also observed an expected shift in methods used that coincided with the introduction of transformer-based models. DISCUSSION: Aspects of the current approaches taken across various domains may be translatable to communication over a patient portal. However, the specific barriers to identifying empathic communication in this context are unclear. While modern NLP methods appear to be able to handle empathy-related tasks, challenges remain in precisely defining and measuring empathy in text. CONCLUSION: Existing work that has attempted to measure empathy in text using NLP provides a useful basis for future studies of patient-physician asynchronous communication, but consideration for the conceptualization of empathy is needed.

Repurposing Clinical Decision Support System Data to Measure Dosing Errors and Clinician-Level Quality of Care.

We aimed to develop and validate an instrument to detect hospital medication prescribing errors using repurposed clinical decision support system data. Despite significant efforts to eliminate medication prescribing errors, these events remain common in hospitals. Data from clinical decision support systems have not been used to identify prescribing errors as an instrument for physician-level performance. We evaluated medication order alerts generated by a knowledge-based electronic prescribing system occurring in one large academic medical center's acute care facilities for patient encounters between 2009 and 2012. We developed and validated an instrument to detect medication prescribing errors through a clinical expert panel consensus process to assess physician quality of care. Six medication prescribing alert categories were evaluated for inclusion, one of which - dose - was included in the algorithm to detect prescribing errors. The instrument was 93% sensitive (recall), 51% specific, 40% precise, 62% accurate, with an F1 score of 55%, positive predictive value of 96%, and a negative predictive value of 32%. Using repurposed electronic prescribing system data, dose alert overrides can be used to systematically detect medication prescribing errors occurring in an inpatient setting with high sensitivity.

Predicting and Planning for Musculoskeletal Service-Connected Disabilities in VA Using Disability for Active Duty OEF/OIF Military Service Members.

INTRODUCTION: Musculoskeletal (MSK) conditions are commonly seen among military service members (SM) and Veterans. We explored correlates of award of MSK-related service-connected disability benefits (SCDB) among SM seeking care in Veterans Affairs (VA) hospitals. MATERIALS AND METHODS: Department of Defense data on SM who separated from October 1, 2001 to May 2017 were linked to VA administrative data. Using adjusted logistic regression models, we determined the odds of receiving MSK SCDB. RESULTS: A total of 1,558,449 (79% of separating SM) had at least one encounter in VA during the study period (7.8% disability separations). Overall, 51% of this cohort had at least one MSK SCDB (88% among disability separations, 48% among normal). Those with disability separations (as compared to normal separations) were significantly more likely to receive MSK SCDB (odds ratio 2.37) as were females (compared to males, odds ratio 1.15). CONCLUSIONS: Although active duty SM with disability separations were more likely to receive MSK-related service-connected disability ratings in the VA, those with normal separations also received such awards. Identifying those at highest risk for MSK-related disability could lead to improved surveillance and prevention strategies in the Department of Defense and VA health care systems to prevent further damage and disability.

Exploring Disparities in Awarding VA Service-Connected Disability for Post-Traumatic Stress Disorder for Active Duty Military Service Members from Recent Conflicts in Iraq and Afghanistan.

INTRODUCTION: We explore disparities in awarding post-traumatic stress disorder (PTSD) service-connected disability benefits (SCDB) to veterans based on gender, race/ethnicity, and misconduct separation. METHODS: Department of Defense data on service members who separated from October 1, 2001 to May 2017 were linked to Veterans Administration (VA) administrative data. Using adjusted logistic regression models, we determined the odds of receiving a PTSD SCDB conditional on a VA diagnosis of PTSD. RESULTS: A total of 1,558,449 (79% of separating service members) had at least one encounter in VA during the study period (12% female, 4.5% misconduct separations). Females (OR 0.72) and Blacks (OR 0.93) were less likely to receive a PTSD award and were nearly equally likely to receive a PTSD diagnosis (OR 0.97, 1.01). Other racial/ethnic minorities were more likely to receive an award and diagnosis, as were those with misconduct separations (award OR 1.3, diagnosis 2.17). CONCLUSIONS: Despite being diagnosed with PTSD at similar rates to their referent categories, females and Black veterans are less likely to receive PTSD disability awards. Other racial/ethnic minorities and those with misconduct separations were more likely to receive PTSD diagnoses and awards. Further study is merited to explore variation in awarding SCDB.

Mental Health Outcomes Among Military Service Members After Severe Injury in Combat and TBI.

INTRODUCTION: Studies examining the mental health outcomes of military personnel deployed into combat zones have focused on the risk of developing post-traumatic stress disorder conferred by mild or moderate traumatic brain injury (TBI). However, other mental health outcomes among veterans who sustained critical combat injuries have not been described. MATERIALS AND METHOD: We examined the associations of moderate and severe TBI and combat injury with the risk for anxiety and mood disorders, adjustment reactions, schizophrenia and other psychotic disorders, cognitive disorders, and post-traumatic stress disorder. We conducted a retrospective cohort study of U.S. military service members critically injured in combat during military operations in Iraq and Afghanistan from February 1, 2002, to February 1, 2011. Health care encounters from (1) the Department of Defense (DoD) Trauma Registry (TR), (2) acute and ambulatory care in military facilities, and (3) civilian facilities are reimbursed by Tricare. Service members who sustained severe combat injury require critical care. We estimated the risk of mental health outcomes using risk-adjusted logit models for demographic and clinical factors. We explored the relationship between TBI and the total number of mental health diagnoses. RESULTS: Of the 4,980 subjects who met inclusion criteria, most injuries occurred among members of the Army (72%) or Marines (25%), with mean (SD) age of 25.5(6.1) years. The prevalence of moderate or severe TBI was 31.6% with explosion as the most common mechanism of injury (78%). We found 71% of the cohort was diagnosed with at least one poor mental health condition, and the adjusted risk conferred by TBI ranged from a modest increase for anxiety disorder (odds ratio, 1.27; 95% confidence interval [CI], 1.11-1.45) to a large increase for cognitive disorder (odds ratio, 3.24; 95% CI, 2.78-3.77). We found TBI was associated with an increased number of mental health diagnoses (incidence rate ratio, 1.52; 95% CI, 1.42-1.63). CONCLUSIONS: Combat-associated TBI may have a broad effect on several mental health conditions among critically injured combat casualties. Early recognition and treatment for trauma-associated mental health are crucial to improving outcomes among service personnel as they transition to post-deployment care in the DoD, Department of Veterans Affairs, or community health systems.

Does one size fit all? An evaluation of the 2018 Leapfrog Group minimal hospital and surgeon volume thresholds for lung surgery.

BACKGROUND: In 2018, the Leapfrog Group set minimum annual lung cancer surgery hospital and surgeon volume thresholds of 40 and 15, respectively. We examined whether outcomes associated with these Leapfrog Group volume thresholds are comparable for patients at the extremes of age and comorbidities. METHODS: We assessed lung cancer patients undergoing lobectomy or pneumonectomy in the New York and Florida State Inpatient Databases for 2007 to 2013. Multivariate logit models evaluated in-hospital mortality, complications, and prolonged length of stay. Median surgeon and hospital volumes were compared between "younger-healthier" (age 18-60 years, Elixhauser Comorbidity Index <1) and "older-sicker" patients (age >77 years, Elixhauser Comorbidity Index >3). RESULTS: The 27,841 patients included 13,277 men (48%). The median patient age was 69 years (interquartile range, 61-77), and mortality was 2.1%. Patients treated by both low-volume surgeons (<15 annual cases) and at low-volume hospitals (<40) had the greatest risk of mortality (2.5%), except for the cohort of younger-healthier patients (mortality <2%). Mortality for older-sicker patients was highest for high-volume surgeons (12%), although higher hospital volume was protective. Increasing hospital volume was associated with decreased mortality (odds ratio [OR], 0.997; 95% confidence interval [CI], 0.995-0.998; P = .0103), complications (OR, 0.998; 95% CI, 0.997-0.999; P < .001), and prolonged length of stay (OR, 0.998; 95% CI, 0.997-1.00; P = .01); similarly, higher surgeon volume was associated with decreased mortality (OR, 0.997; 95% CI, 0.99-1.00; P = .03), complications (OR, 0.997; 95% CI, 0.994-1.00; P = .02), and prolonged length of stay (OR, 0.991; 95% CI, 0.986-0.995; P < .01). CONCLUSIONS: Hospital volume has a greater effect on morbidity and mortality than surgeon volume especially for older-sicker patients, suggesting that Leapfrog Group volume guidelines should emphasize hospital volume over surgeon volume and may be less relevant for younger-healthier patients.

Association Between Specialist Office Visits and Health Expenditures in Accountable Care Organizations.

Importance: Accountable care organizations (ACOs) aim to control health expenditures while improving quality of care. Primary care has been emphasized as a means to reduce spending, but little is known about the implications of using specialists for achieving this ACO objective. Objective: To examine the association between ACO-beneficiary office visits conducted by specialists and the cost and utilization outcomes of those visits. Design, Setting, and Participants: This cross-sectional study obtained data on 620 distinct ACOs from the Centers for Medicare & Medicaid Services Shared Savings Program Accountable Care Organizations Public-Use Files from April 1, 2012, to September 30, 2017. Generalized estimating equation models were used for analysis of ACOs, adjusting for ACO-beneficiary health status, Medicare enrollment groups, ACO size, and proportion of participating specialists. Exposures: Specialist encounter proportion, the percentage of office visits provided by a specialist, was categorized into 7 discrete groups: less than 35%, 35% to less than 40%, 40% to less than 45% (reference group), 45% to less than 50%, 50% to less than 55%, 55% to less than 60%, and 60% or greater. Main Outcomes and Measures: The primary outcome was total expenditures (given in US dollars) per assigned beneficiary person-year. The secondary outcomes were total numbers of emergency department visits, hospital discharges, skilled nursing facility discharges, and magnetic resonance imaging orders. Results: In total, the data set included 1836 ACO-year (number of participation years per ACO) observations for 620 distinct ACOs. Those ACOs with a specialist encounter proportion of 40% to less than 45% had $1129 (95% CI, $445-$1814) lower per-beneficiary person-year spending than did ACOs in the lowest specialist encounter proportion group and had $752 (95% CI, $115-$1389) lower per-beneficiary person-year spending compared with ACOs in the highest specialist encounter proportion group. Monotonic decreases in emergency department visits, hospital discharges, and skilled nursing facility discharges were observed with increasing specialist encounter proportion. Conversely, monotonic increases in magnetic resonance imaging volume discharges were observed with increasing specialist encounter proportion. Conclusions and Relevance: These findings suggest that an ACO's ability to reduce spending may require sufficient involvement in care processes from specialists, who seem to complement the intrinsic primary care approach in ACOs.

Rethinking Thirty-Day Hospital Readmissions: Shorter Intervals Might Be Better Indicators Of Quality Of Care.

Public reporting and payment programs in the United States have embraced thirty-day readmissions as an indicator of between-hospital variation in the quality of care, despite limited evidence supporting this interval. We examined risk-standardized thirty-day risk of unplanned inpatient readmission at the hospital level for Medicare patients ages sixty-five and older in four states and for three conditions: acute myocardial infarction, heart failure, and pneumonia. The hospital-level quality signal captured in readmission risk was highest on the first day after discharge and declined rapidly until it reached a nadir at seven days, as indicated by a decreasing intracluster correlation coefficient. Similar patterns were seen across states and diagnoses. The rapid decay in the quality signal suggests that most readmissions after the seventh day postdischarge were explained by community- and household-level factors beyond hospitals' control. Shorter intervals of seven or fewer days might improve the accuracy and equity of readmissions as a measure of hospital quality for public accountability.

Overdiagnosis of Clostridium difficile Infection in the Molecular Test Era.

IMPORTANCE: Clostridium difficile is a major cause of health care-associated infection, but disagreement between diagnostic tests is an ongoing barrier to clinical decision making and public health reporting. Molecular tests are increasingly used to diagnose C difficile infection (CDI), but many molecular test-positive patients lack toxins that historically defined disease, making it unclear if they need treatment. OBJECTIVE: To determine the natural history and need for treatment of patients who are toxin immunoassay negative and polymerase chain reaction (PCR) positive (Tox-/PCR+) for CDI. DESIGN, SETTING, AND PARTICIPANTS: Prospective observational cohort study at a single academic medical center among 1416 hospitalized adults tested for C difficile toxins 72 hours or longer after admission between December 1, 2010, and October 20, 2012. The analysis was conducted in stages with revisions from April 27, 2013, to January 13, 2015. MAIN OUTCOMES AND MEASURES: Patients undergoing C difficile testing were grouped by US Food and Drug Administration-approved toxin and PCR tests as Tox+/PCR+, Tox-/PCR+, or Tox-/PCR-. Toxin results were reported clinically. Polymerase chain reaction results were not reported. The main study outcomes were duration of diarrhea during up to 14 days of treatment, rate of CDI-related complications (ie, colectomy, megacolon, or intensive care unit care) and CDI-related death within 30 days. RESULTS: Twenty-one percent (293 of 1416) of hospitalized adults tested for C difficile were positive by PCR, but 44.7% (131 of 293) had toxins detected by the clinical toxin test. At baseline, Tox-/PCR+ patients had lower C difficile bacterial load and less antibiotic exposure, fecal inflammation, and diarrhea than Tox+/PCR+ patients (P < .001 for all). The median duration of diarrhea was shorter in Tox-/PCR+ patients (2 days; interquartile range, 1-4 days) than in Tox+/PCR+ patients (3 days; interquartile range, 1-6 days) (P = .003) and was similar to that in Tox-/PCR- patients (2 days; interquartile range, 1-3 days), despite minimal empirical treatment of Tox-/PCR+ patients. No CDI-related complications occurred in Tox-/PCR+ patients vs 10 complications in Tox+/PCR+ patients (0% vs 7.6%, P < .001). One Tox-/PCR+ patient had recurrent CDI as a contributing factor to death within 30 days vs 11 CDI-related deaths in Tox+/PCR+ patients (0.6% vs 8.4%, P = .001). CONCLUSIONS AND RELEVANCE: Among hospitalized adults with suspected CDI, virtually all CDI-related complications and deaths occurred in patients with positive toxin immunoassay test results. Patients with a positive molecular test result and a negative toxin immunoassay test result had outcomes that were comparable to patients without C difficile by either method. Exclusive reliance on molecular tests for CDI diagnosis without tests for toxins or host response is likely to result in overdiagnosis, overtreatment, and increased health care costs.

Comparing patient outcomes of academician-preceptors, hospitalist-preceptors, and hospitalists on internal medicine services in an academic medical center.

BACKGROUND: Patient outcomes with hospitalist care have been studied in many settings, yet little is known about how hospitalist care interacts with trainee care to affect patient outcomes in teaching hospitals. OBJECTIVES: The aim of this study was to compare patient outcomes between hospitalist-preceptors and hospitalists working alone (isolating the effect of housestaff involvement), and between hospitalist-preceptors and academician-preceptors (isolating the effect of attending type, given housestaff involvement). DESIGN: A four-year retrospective cohort study of patients (n = 13,313) admitted to all internal medicine services at an academic medical center from July 2008 to June 2012. MAIN MEASURES: Using generalized estimating equations, we measured readmission within 30 days, hospital length of stay, cost of the index hospitalization, and cumulative cost including readmissions within 30 days. KEY RESULTS: In the adjusted models, 30-day readmission odds were higher for academic-preceptors (OR, 1.14 [95% CI, 1.03 - 1.26]) and hospitalist-preceptors (OR, 1.10 [95% CI, 1.002 - 1.21]) than for hospitalists working alone. Compared with hospitalists working alone, academic-preceptors were associated with shorter length of stay (mean difference, 0.27 days [95% CI, 0.18 - 0.38]), lower index hospitalization costs (mean difference, $386 [95% CI, $192 - $576]), but similar cumulative inpatient costs within 30 days of discharge. Compared with hospitalists working alone, hospitalist-preceptors were associated with shorter length of stay (mean difference, 0.34 days [95% CI, 0.26 - 0.42]), lower index hospitalization cost (mean difference, $570 [95% CI, $378 - $760]), and a trend toward lower cumulative cost (mean difference, $1347 [95% CI, $254 - $2,816]). CONCLUSIONS: Preceptor-led medicine services were associated with more readmissions within 30 days, shorter lengths of stay, and lower index admission-associated costs. However, when considering cumulative hospitalization costs, patients discharged by academician-preceptors incurred the highest cost and hospitalist-preceptors incurred the lowest cost.

Outcomes in patients tested for Clostridium difficile toxins.

Clostridium difficile testing is shifting from toxin detection to C. difficile detection. Yet, up to 60% of patients with C. difficile by culture test negative for toxins and it is unclear whether they are infected or carriers. We reviewed medical records for 7046 inpatients with a C. difficile toxin test from 2005 to 2009 to determine the duration of diarrhea and rate of complications and mortality among toxin-positive (toxin+) and toxin- patients. Overall, toxin- patients had less severe diarrhea, fewer diarrhea days, and lower mortality (P < 0.001, all comparisons) than toxin+ patients. One toxin- patient (n = 1/6121; 0.02%) was diagnosed with pseudomembranous colitis, but there were no complications such as megacolon or colectomy for fulminant CDI among toxin- patients. These data suggest that C. difficile-attributable complications are rare among patients testing negative for C. difficile toxins. More studies are needed to evaluate the clinical significance of C. difficile detection in toxin- patients.

Psoriasis and hypertension severity: results from a case-control study.

BACKGROUND: Epidemiologic studies have provided new insights into the association between psoriasis and cardiovascular diseases. Previous population studies have examined hypertension frequency in psoriasis patients. However, the relationship between severity of hypertension and psoriasis has not been characterized. OBJECTIVE: We sought to investigate whether patients with psoriasis have more difficult-to-manage hypertension compared to non-psoriatic hypertensive patients. APPROACH: We performed a case-control study using the University of California Davis electronic medical records. The cases were defined as patients diagnosed with both psoriasis and hypertension, and controls were defined as patients with hypertension and without psoriasis. In this identified population, 835 cases were matched on age, sex, and body mass index (BMI) to 2418 control patients. KEY RESULTS: Treatment with multiple anti-hypertensives was significantly associated with the presence of psoriasis using univariate (p < 0.0001) and multivariable analysis, after adjusting for diabetes, hyperlipidemia, and race (p < 0.0001). Compared to hypertensive patients without psoriasis, psoriasis patients with hypertension were 5 times more likely to be on a monotherapy antihypertensive regimen (95% CI 3.607.05), 9.5 times more likely to be on dual antihypertensive therapy (95% CI 6.68-13.65), 16.5 times more likely to be on triple antihypertensive regimen (95% CI 11.01-24.84), and 19.9 times more likely to be on quadruple therapy or centrally-acting agent (95% CI 10.58-37.33) in multivariable analysis after adjusting for traditional cardiac risk factors. CONCLUSIONS: Psoriasis patients appear to have more difficult-to-control hypertension compared to non-psoriatic, hypertensive patients.

Mitoxantrone, etoposide, and cytarabine with or without valspodar in patients with relapsed or refractory acute myeloid leukemia and high-risk myelodysplastic syndrome: a phase III trial (E2995).

PURPOSE: To determine whether adding the multidrug resistance gene-1 (MDR-1) modulator valspodar (PSC 833; Novartis Pharmaceuticals, Hanover, NJ) to chemotherapy provided clinical benefit to patients with poor-risk acute myeloid leukemia (AML) and high-risk myelodysplastic syndrome (MDS). PATIENTS AND METHODS: A phase III randomized study was performed using valspodar plus mitoxantrone, etoposide, and cytarabine (PSC-MEC; n=66) versus MEC (n=63) to treat patients with relapsed or refractory AML and high-risk MDS. RESULTS: For the PSC-MEC versus MEC arms, complete response (CR) was achieved in 17% versus 25% of patients, respectively (P=not significant). For patients who had not received prior intensive chemotherapy (ie, with secondary AML or high-risk MDS), the CR rate was increased--35% versus 15% for the remaining patients (P=.018); CR rates did not differ between treatment arms. The median disease-free survival in those achieving CR was similar in the two arms (10 versus 9.3 months) as was the patients' overall survival (4.6 versus 5.4 months). The CR rates in MDR+ (69% of patients) versus MDR- patients were similar for those receiving either chemotherapy regimen (16% versus 24%). The CR rate for unfavorable cytogenetic patients (45% of patients) was 13% compared to the remainder, 28% (P=.09). Population pharmacokinetic analysis demonstrated that the clearances of mitoxantrone and etoposide were decreased by 59% and 50%, respectively, supporting the empiric dose reductions in the PSC-MEC arm designed in anticipation of drug interactions between valspodar and the chemotherapeutic agents. CONCLUSION: CR rates and overall survival were not improved by using PSC-MEC compared to MEC chemotherapy alone in patients with poor-risk AML or high-risk MDS.

Rapid determination of PEGylated liposomal doxorubicin and its major metabolite in human plasma by ultraviolet-visible high-performance liquid chromatography.

A high-performance liquid chromatographic method was developed for the quantification of doxorubicin derived from PEGylated liposomal doxorubicin (Doxil) and its major metabolite in human plasma. This method utilizes Triton X-100 to disperse the liposome, followed by a protein precipitation step with 5-sulfosalicylic acid. Analytes in the resultant supernatant are separated on a Discovery RP amide C(16) column (250 x 3 mm I.D., 5 microm) using an isocratic elution with a mobile phase consisting of 0.05 M sodium acetate (pH 4.0) and acetonitrile (72:28). The retention times for doxorubicin and the internal standard daunorubicin were 4.8 and 10.1 min, respectively. The column eluate was monitored by UV-visible detection at 487 nm. The determination of doxorubicin was found to be linear in the range of 1.0 ng/mL to 25 microg/mL, with intra-day and inter-day coefficients of variation and percent error < or =10%. The recovery of doxorubicin from plasma was >69.3%, with a liposomal dispersion efficiency of >95.7%. Our analytical method for free and PEGylated doxorubicin in human plasma is rapid, avoids organic extractions, and maintains sensitivity for the parent compound and its major metabolite, doxorubicinol.