Building a Learning Health System: Creating an Analytical Workflow for Evidence Generation to Inform Institutional Clinical Care Guidelines.

BACKGROUND: One key aspect of a learning health system (LHS) is utilizing data generated during care delivery to inform clinical care. However, institutional guidelines that utilize observational data are rare and require months to create, making current processes impractical for more urgent scenarios such as those posed by the COVID-19 pandemic. There exists a need to rapidly analyze institutional data to drive guideline creation where evidence from randomized control trials are unavailable. OBJECTIVES: This article provides a background on the current state of observational data generation in institutional guideline creation and details our institution's experience in creating a novel workflow to (1) demonstrate the value of such a workflow, (2) demonstrate a real-world example, and (3) discuss difficulties encountered and future directions. METHODS: Utilizing a multidisciplinary team of database specialists, clinicians, and informaticists, we created a workflow for identifying and translating a clinical need into a queryable format in our clinical data warehouse, creating data summaries and feeding this information back into clinical guideline creation. RESULTS: Clinical questions posed by the hospital medicine division were answered in a rapid time frame and informed creation of institutional guidelines for the care of patients with COVID-19. The cost of setting up a workflow, answering the questions, and producing data summaries required around 300 hours of effort and $300,000 USD. CONCLUSION: A key component of an LHS is the ability to learn from data generated during care delivery. There are rare examples in the literature and we demonstrate one such example along with proposed thoughts of ideal multidisciplinary team formation and deployment.

Ophthalmology Consultation to Detect Endogenous Endophthalmitis: Clinical Characteristics in Consulted Versus Diagnosed Cases Among At-Risk Inpatients.

BACKGROUND AND OBJECTIVE: Predisposing factors for ophthalmology consultations and endogenous endophthalmitis were compared among inpatients with systemic infection. PATIENTS AND METHODS: This was a retrospective cohort study in a tertiary care hospital between January 1, 2010, and December 31, 2014. Multivariable logistic regression was utilized. RESULTS: There were 9,527 encounters identified with systemic infection. The 5-year incidence rate was 8.4% (803/9,527) for consultations and 0.3% (25/9,527) for endophthalmitis. Factors most associated with consultations included positive fungal blood cultures and HIV. Factors most associated with endophthalmitis included positive blood fungal cultures and endocarditis. Four of 25 endophthalmitis patients lacked positive blood cultures; six of 20 endophthalmitis patients with adequate mentation were asymptomatic. CONCLUSIONS: Positive blood fungal cultures were strongly associated with both endophthalmitis and consultations. Endocarditis was strongly associated with endophthalmitis but less associated with consultation and may warrant increased attention. Neither presence of symptoms nor positive cultures may be sufficiently accurate to determine need for consultation. [Ophthalmic Surg Lasers Imaging Retina. 2020;51:159-169.].

Screening Rates for Primary Aldosteronism in Resistant Hypertension: A Cohort Study.

Resistant hypertension is associated with higher rates of cardiovascular disease, kidney disease, and death than primary hypertension. Although clinical practice guidelines recommend screening for primary aldosteronism among persons with resistant hypertension, rates of screening are unknown. We identified 145 670 persons with hypertension and excluded persons with congestive heart failure or advanced chronic kidney disease. Among this cohort, we studied 4660 persons ages 18 to <90 from the years 2008 to 2014 with resistant hypertension and available laboratory tests within the following 24 months. The screening rate for primary aldosteronism in persons with resistant hypertension was 2.1%. Screened persons were younger (55.9±13.3 versus 65.5±11.6 years; P<0.0001) and had higher systolic (145.1±24.3 versus 139.6±20.5 mm Hg; P=0.04) and diastolic blood pressure (81.8±13.6 versus 74.4±13.8 mm Hg; P<0.0001), lower rates of coronary artery disease (5.2% versus 14.2%; P=0.01), and lower serum potassium concentrations (3.9±0.6 versus 4.1±0.5 mmol/L; P=0.04) than unscreened persons. Screened persons had significantly higher rates of prescription for calcium channel blockers, mixed α/ß-adrenergic receptor antagonists, sympatholytics, and vasodilators, and lower rates of prescription for loop, thiazide, and thiazide-type diuretics. The prescription of mineralocorticoid receptor antagonists or other potassium-sparing diuretics was not significantly different between groups (P=0.20). In conclusion, only 2.1% of eligible persons received a screening test within 2 years of meeting criteria for resistant hypertension. Low rates of screening were not due to the prescription of antihypertensive medications that may potentially interfere with interpretation of the screening test. Efforts to highlight guideline-recommended screening and targeted therapy are warranted.

Improving risk stratification in heart failure with preserved ejection fraction by combining two validated risk scores.

Introduction: The Intermountain Risk Score (IMRS) was developed and validated to predict short-term and long-term mortality in hospitalised patients using demographics and commonly available laboratory data. In this study, we sought to determine whether the IMRS also predicts all-cause mortality in patients hospitalised with heart failure with preserved ejection fraction (HFpEF) and whether it is complementary to the Get with the Guidelines Heart Failure (GWTG-HF) risk score or N-terminal pro-B-type natriuretic peptide (NT-proBNP). Methods and results: We used the Stanford Translational Research Integrated Database Environment to identify 3847 adult patients with a diagnosis of HFpEF between January 1998 and December 2016. Of these, 580 were hospitalised with a primary diagnosis of acute HFpEF. Mean age was 76±16 years, the majority being female (58%), with a high prevalence of diabetes mellitus (36%) and a history of coronary artery disease (60%). Over a median follow-up of 2.0 years, 140 (24%) patients died. On multivariable analysis, the IMRS and GWTG-HF risk score were independently associated with all-cause mortality (standardised HRs IMRS (1.55 (95% CI 1.27 to 1.93)); GWTG-HF (1.60 (95% CI 1.27 to 2.01))). Combining the two scores, improved the net reclassification over GWTG-HF alone by 36.2%. In patients with available NT-proBNP (n=341), NT-proBNP improved the net reclassification of each score by 46.2% (IMRS) and 36.3% (GWTG-HF). Conclusion: IMRS and GWTG-HF risk scores, along with NT-proBNP, play a complementary role in predicting outcome in patients hospitalised with HFpEF.

Procalcitonin as a Predictive Marker for Bacteremia in Children With a Central Line and Fever.

OBJECTIVES: Unnecessary use of antibiotics is an increasing problem. In this study, we sought to determine the diagnostic accuracy of procalcitonin in predicting bacteremia in children with a central line and fever, and we sought to determine optimal cutoff values to maximize sensitivity and specificity. This is the largest study to date in which procalcitonin is examined as a predictive marker of bacteremia in pediatric patients with a central line and fever. METHODS: We conducted a retrospective cohort study of children aged 0 to 23 years with a central line and fever of 38°C who had procalcitonin and blood cultures drawn before initiation of antibiotics and had no other identified bacterial infection. Patients were also prospectively monitored via a custom-built electronic medical record dashboard for eligibility. RESULTS: There were 523 patients and >2500 procalcitonin values reviewed for eligibility. Of these, 169 (47%) patients and 335 blood cultures with procalcitonin were included. There were 94 (28%) positive bacterial blood cultures and 241 (72%) negative bacterial blood cultures. In bacteremic cultures, the mean procalcitonin level was 9.96 ± 15.96 ng/mL, and the median procalcitonin level was 4.85 ng/mL (interquartile range 18.5). In nonbacteremic cultures, the mean procalcitonin level was 1.23 ± 10.37 ng/mL, and the median procalcitonin level was 0.3 ng/mL (interquartile range 0.7). A receiver operating characteristic analysis indicated a procalcitonin level of ≥0.6 ng/mL as the best cutoff point that produced a sensitivity of 85.6% and a specificity of 65.7% (area under the curve 0.85). CONCLUSIONS: Procalcitonin is a sensitive biomarker in predicting bacteremia in children with a central line and fever.

Patient vs provider perspectives of 30-day hospital readmissions.

Objective: To compare patients' and providers' views on contributors to 30-day hospital readmissions. Design: Analysis of a qualitative interview survey between 18 May-30 June 2015. Setting: Interviews were conducted during the 30-day readmission hospitalisation at a single tertiary care academic hospital. Participants: We conducted 178 interviews of readmitted patients. Measures: We queried opinions of what factors patients believed contributed to their rehospitalisation and compared this with the perspective of the index admission provider. The primary outcome was the view that the readmission was preventable. A review by a RN (nurse) case manager also provided an assessment based on patient report, provider report and chart review. Results: Patients were more likely to view a readmission as preventable compared with physicians (p<0.0001). Patients identified system issues (defined as factors controlled by the hospital discharge process) as contributors to their readmission in 58% (103/178) of cases while providers identified system issues as the contributor to a patients' readmission in 2% (2/101) of cases. Patients with poor functional status were more likely to feel the cause of their readmission was due to system issues than patients with better functional status (p=0.03). A RN case manager review determined that in 48% (86/178) of cases the system had some amount of contribution to a patient's readmission. There was no significant difference in belief that the readmission was preventable between the RN case manager and the patient (p=0.47). Conclusions: Readmitted patients often feel that the hospital system contributed to their readmission. Providers did not recognise patient and RN case manager identified issues as contributors to hospital readmissions.

Impact of problem-based charting on the utilization and accuracy of the electronic problem list.

Objective: Problem-based charting (PBC) is a method for clinician documentation in commercially available electronic medical record systems that integrates note writing and problem list management. We report the effect of PBC on problem list utilization and accuracy at an academic intensive care unit (ICU). Materials and Methods: An interrupted time series design was used to assess the effect of PBC on problem list utilization, which is defined as the number of new problems added to the problem list by clinicians per patient encounter, and of problem list accuracy, which was determined by calculating the recall and precision of the problem list in capturing 5 common ICU diagnoses. Results: In total, 3650 and 4344 patient records were identified before and after PBC implementation at Stanford Hospital. An increase of 2.18 problems (>50% increase) in the mean number of new problems added to the problem list per patient encounter can be attributed to the initiation of PBC. There was a significant increase in recall attributed to the initiation of PBC for sepsis (ß = 0.45, P < .001) and acute renal failure (ß = 0.2, P = .007), but not for acute respiratory failure, pneumonia, or venous thromboembolism. Discussion: The problem list is an underutilized component of the electronic medical record that can be a source of clinician-structured data representing the patient's clinical condition in real time. PBC is a readily available tool that can integrate problem list management into physician workflow. Conclusion: PBC improved problem list utilization and accuracy at an academic ICU.

Big data modeling to predict platelet usage and minimize wastage in a tertiary care system.

Maintaining a robust blood product supply is an essential requirement to guarantee optimal patient care in modern health care systems. However, daily blood product use is difficult to anticipate. Platelet products are the most variable in daily usage, have short shelf lives, and are also the most expensive to produce, test, and store. Due to the combination of absolute need, uncertain daily demand, and short shelf life, platelet products are frequently wasted due to expiration. Our aim is to build and validate a statistical model to forecast future platelet demand and thereby reduce wastage. We have investigated platelet usage patterns at our institution, and specifically interrogated the relationship between platelet usage and aggregated hospital-wide patient data over a recent consecutive 29-mo period. Using a convex statistical formulation, we have found that platelet usage is highly dependent on weekday/weekend pattern, number of patients with various abnormal complete blood count measurements, and location-specific hospital census data. We incorporated these relationships in a mathematical model to guide collection and ordering strategy. This model minimizes waste due to expiration while avoiding shortages; the number of remaining platelet units at the end of any day stays above 10 in our model during the same period. Compared with historical expiration rates during the same period, our model reduces the expiration rate from 10.5 to 3.2%. Extrapolating our results to the ∼2 million units of platelets transfused annually within the United States, if implemented successfully, our model can potentially save ∼80 million dollars in health care costs.

Patient Outcomes when Housestaff Exceed 80 Hours per Week.

BACKGROUND: It has been posited that high workload and long work hours for trainees could affect the quality and efficiency of patient care. Duty hour restrictions seek to balance patient care and resident education by limiting resident work hours. Through a retrospective cohort study, we investigated whether patient care on an inpatient general medicine service at a large academic medical center is impacted when housestaff work more than 80 hours per week. METHODS: We identified all admissions to a housestaff-run general medicine service between June 25, 2013 and June 29, 2014. Each hospitalization was classified by whether the patient was admitted by housestaff who have worked more than 80 hours per week during their hospitalization. Housestaff computer activity and duty hours were calculated by institutional electronic heath record audit, as well as length of stay and a composite of in-hospital mortality, intensive care unit (ICU) transfer rate, and 30-day readmission rate. RESULTS: We identified 4767 hospitalizations by 3450 unique patients; of which 40.9% of hospitalizations were managed by housestaff who worked more than 80 hours that week during their hospitalization. There was a significantly higher rate of the composite outcome (19.2% vs 16.7%, P = .031) for patients admitted by housestaff working more than 80 hours per week during their hospitalization. We found a statistically significant higher length of stay (5.12 vs 4.66 days, P = .048) and rate of ICU transfer (3.53% vs 2.38%, P = .029). There was no statistically significant difference in 30-day readmission rate (13.7% vs 12.8%, P = .395) or in-hospital mortality rate (3.18% vs 2.42%, P = .115). There was no correlation with team census on admission and patient outcomes. CONCLUSIONS: Patients taken care of by housestaff working more than 80 hours per week had increased length of stay and number of ICU transfers. There was no association between resident work-hours and patient in-hospital mortality or 30-day readmission rate.

Regional right ventricular dysfunction in acute pulmonary embolism: relationship with clot burden and biomarker profile.

Regional right ventricular (RV) dysfunction (RRVD) is an echocardiographic feature in acute pulmonary embolism (PE), primarily reported in patients with moderate-to-severe RV dysfunction. This study investigated the clinical importance of RRVD by assessing its relationship with clot burden and biomarkers. We identified consecutive patients admitted to the emergency department between 1999 and 2014 who underwent computed tomographic angiography, echocardiography, and biomarker testing (troponin and NT-proBNP) for suspected acute PE. RRVD was defined as normal excursion of the apex contrasting with hypokinesis of the mid-free wall segment. RV assessment included measurements of ventricular dimensions, fractional area change, free-wall longitudinal strain and tricuspid annular plane systolic excursion. Clot burden was assessed using the modified Miller score. Of 82 patients identified, 51 had acute PE (mean age 66 ± 17 years, 43% male). No patient had RV myocardial infarction. RRVD was present in 41% of PEs and absent in all patients without PE. Among patients with PE, 86% of patients with RRVD had central or multi-lobar PE. Patients with RRVD had higher prevalence of moderate-to-severe RV dilation (81 vs. 30%, p < 0.01) and dysfunction (86 vs. 23%, p < 0.01). There was a strong trend for higher troponin level in PE patients with RRVD (38 vs. 13% in PE patients without RRVD, p = 0.08), while there was no significant difference for NT-proBNP (67 vs. 73%, p = 0.88). RRVD showed good concordance between readers (87%). RRVD is associated with an increased clot burden in acute PE and is more prevalent among patients with moderate-to-severe RV enlargement and dysfunction.

Association between maintenance fluid tonicity and hospital-acquired hyponatremia.

OBJECTIVE: To evaluate whether the administration of hypotonic fluids compared with isotonic fluids is associated with a greater risk for hyponatremia in hospitalized children. STUDY DESIGN: Informatics-enabled cohort study of all hospitalizations at Lucile Packard Children's Hospital between April 2009 and March 2011. Extraction and analysis of electronic medical record data identified normonatremic hospitalized children who received either hypotonic or isotonic intravenous maintenance fluids upon admission. The primary exposure was the administration of hypotonic maintenance fluids, and the primary outcome was the development of hyponatremia (serum sodium <135 mEq/L). RESULTS: A total of 1048 normonatremic children received either hypotonic (n = 674) or isotonic (n = 374) maintenance fluids upon admission. Hyponatremia developed in 260 (38.6%) children who received hypotonic fluids and 104 (27.8%) of those who received isotonic fluids (unadjusted OR 1.63; 95% CI 1.24-2.15, P < .001). After we controlled for intergroup differences and potential confounders, patients receiving hypotonic fluids remained more likely to develop hyponatremia (aOR 1.37, 95% CI 1.03-1.84). Multivariable analysis identified additional factors associated with the development of hyponatremia, including surgical admission (aOR 1.44, 95% CI 1.09-1.91), cardiac admitting diagnosis (aOR 2.08, 95% CI 1.34-3.20), and hematology/oncology admitting diagnosis (aOR 2.37, 95% CI 1.74-3.25). CONCLUSIONS: Hyponatremia was common regardless of maintenance fluid tonicity; however, the administration of hypotonic maintenance fluids compared with isotonic fluids was associated with a greater risk of developing hospital-acquired hyponatremia. Additional clinical characteristics modified the hyponatremic effect of hypotonic fluid, and it is possible that optimal maintenance fluid therapy now requires a more individualized approach.

A social network of hospital acquired infection built from electronic medical record data.

OBJECTIVE: Social networks have been used in the study of outbreaks of infectious diseases, including in small group settings such as individual hospitals. Collecting the data needed to create such networks, however, can be time consuming, costly, and error prone. We sought to create a social network of hospital inpatients using electronic medical record (EMR) data already collected for other purposes, for use in simulating outbreaks of nosocomial infections. MATERIALS AND METHODS: We used the EMR data warehouse of a tertiary academic hospital to model contact among inpatients. Patient-to-patient contact due to shared rooms was inferred from admission-discharge-transfer data, and contact with healthcare workers was inferred from clinical documents. Contacts were used to generate a social network, which was then used to conduct probabilistic simulations of nosocomial outbreaks of methicillin-resistant Staphylococcus aureus and influenza. RESULTS: Simulations of infection transmission across the network reflected the staffing and patient flow practices of the hospital. Simulations modeling patient isolation, increased hand hygiene, and staff vaccination showed a decrease in the spread of infection. DISCUSSION: We developed a method of generating a social network of hospital inpatients from EMR data. This method allows the derivation of networks that reflect the local hospital environment, obviate the need for simulated or manually collected data, and can be updated in near real time. CONCLUSIONS: Inpatient social networks represent a novel secondary use of EMR data, and can be used to simulate nosocomial infections. Future work should focus on prospective validation of the simulations, and adapting such networks to other tasks.

Discordant aPTT and anti-Xa values and outcomes in hospitalized patients treated with intravenous unfractionated heparin.

BACKGROUND: Both the activated partial thromboplastin time (aPTT) and anti-Xa assay can be used to monitor unfractionated heparin (UFH). Following implementation of an anti-Xa method for heparin dosing protocols in our hospital, we became aware of many patients with discordant aPTT and anti-Xa values. OBJECTIVE: To determine the frequency of discordant aPTT and anti-Xa values in a large cohort of hospitalized patients treated with UFH, as well as the demographics, coagulation status, indication for UFH, and clinical outcomes in this population. METHODS: All aPTT and anti-Xa values from adults hospitalized between February and August 2009 at Stanford Hospital who were treated with UFH were analyzed. All samples were drawn simultaneously. A polynomial fit correlating aPTT and anti-Xa with a 99% confidence limit was designed. Paired aPTT/anti-Xa values were grouped according to whether the paired values fell within or outside of the concordant area. Patients were placed into groups based on concordance status, and clinical outcomes were assessed. RESULTS: A total of 2321 paired values from 539 patients were studied; 42% of data pairs had a high aPTT value relative to the anti-Xa value. Patients with elevated baseline prothrombin time/international normalized ratio or aPTT frequently demonstrated disproportionate relative prolongation of the aPTT. Patients with at least 2 consecutive high aPTT to anti-Xa values had increased 21-day major bleeding (9% vs 3%; p = 0.0316) and 30-day mortality (14% dead vs 5% dead at 30 days; p = 0.0202) compared with patients with consistently concordant values. CONCLUSIONS: aPTT and anti-Xa values are frequently discordant when used to measure UFH in hospitalized patients. A disproportionate prolongation of the aPTT relative to the anti-Xa was the most common discordant pattern in our study. Patients with relatively high aPTT to anti-Xa values appear to be at increased risk of adverse outcomes. Monitoring both aPTT and Xa values may have utility in managing such patients.

Dysferlin overexpression in skeletal muscle produces a progressive myopathy.

OBJECTIVE: The dose-response effects of dysferlin transgenesis were analyzed to determine if the dysferlin-deficient myopathies are good candidates for gene replacement therapy. METHODS: We have generated 3 lines of transgenic mice, expressing low, mid, and high levels of full-length human dysferlin from a muscle-specific promoter. Transgenic skeletal muscle was analyzed and scored for morphological and functional deficits. RESULTS: Overexpression of dysferlin in mice resulted in a striking phenotype of kyphosis, irregular gait, and reduced muscle mass and strength. Moreover, protein dosage correlated with phenotype severity. In contrast to dysferlin-null skeletal muscle, no evidence of sarcolemmal impairment was revealed. Rather, increased levels of Ca(2+)-regulated, dysferlin-binding proteins and endoplasmic reticulum stress chaperone proteins were observed in muscle lysates from transgenic mice as compared with controls. INTERPRETATION: Expression levels of dysferlin are important for appropriate function without deleterious or cytotoxic effects. As a corollary, we propose that future endeavors in gene replacement for correction of dysferlinopathy should be tailored to take account of this.

Single muscle fiber size and contractility after spinal cord injury in rats.

Spinal cord injury (SCI) results in muscle weakness but the degree of impairment at the level of single fibers is not known. The purpose of this study was to examine the effects of T9-level SCI on single muscle fibers from the tibialis anterior of rats. Significant decreases in cross-sectional area (CSA), maximal force (Po), and specific force (SF = Po/CSA) were noted at 2 weeks. Atrophy and force-generating capacity were reversed at 4 weeks, but SF remained impaired. Maximum shortening velocity (Vo) did not change after injury. SCI thus appears to affect various contractile properties of single muscle fibers differently. Normal cage activity may partially restore function but new interventions are needed to restore muscle fiber quality.

Preservation of in vitro muscle fiber function in dermatomyositis and inclusion body myositis: a single fiber study.

Five patients with untreated dermatomyositis, five with inclusion body myositis, and 16 healthy elderly volunteer subjects (controls) underwent open (dermatomyositis and inclusion body myositis) or percutaneous (controls) muscle biopsy. Biopsied muscles included deltoid, biceps and vastus lateralis. Chemically skinned single muscle fibers were activated with Ca(+2); the slack test was performed to determine maximal unloaded shortening velocity (Vo). Parameters measured include single fiber cross sectional area, maximal force, specific force and Vo. 429 Type I and 94 Type IIA fibers were studied. Cross sectional area and maximal force were greater in inclusion body myositis than dermatomyositis or control for Type I and IIA fibers. Specific force of Type I fibers was similar in inclusion body myositis and dermatomyositis but greater than in controls. Vo was greater in Type I, but not IIA, fibers in dermatomyositis compared with inclusion body myositis and controls. The force and velocity generating capacity of single muscle fibers is preserved in patients with dermatomyositis and inclusion body myositis suggesting that dysfunction of the contractile proteins does not contribute to clinical muscle weakness.