CKD Progression Prediction in a Diverse US Population: A Machine-Learning Model.

Rationale & Objective: Chronic kidney disease (CKD) is a major cause of morbidity and mortality. To date, there are no widely used machine-learning models that can predict progressive CKD across the entire disease spectrum, including the earliest stages. The objective of this study was to use readily available demographic and laboratory data from Sonic Healthcare USA laboratories to train and test the performance of machine learning-based predictive risk models for CKD progression. Study Design: Retrospective observational study. Setting & Participants: The study population was composed of deidentified laboratory information services data procured from a large US outpatient laboratory network. The retrospective data set included 110,264 adult patients over a 5-year period with initial estimated glomerular filtration rate (eGFR) values between 15-89 mL/min/1.73 m2. Predictors: Patient demographic and laboratory characteristics. Outcomes: Accelerated (ie, >30%) eGFR decline associated with CKD progression within 5 years. Analytical Approach: Machine-learning models were developed using random forest survival methods, with laboratory-based risk factors analyzed as potential predictors of significant eGFR decline. Results: The 7-variable risk classifier model accurately predicted an eGFR decline of >30% within 5 years and achieved an area under the curve receiver-operator characteristic of 0.85. The most important predictor of progressive decline in kidney function was the eGFR slope. Other key contributors to the model included initial eGFR, urine albumin-creatinine ratio, serum albumin (initial and slope), age, and sex. Limitations: The cohort study did not evaluate the role of clinical variables (eg, blood pressure) on the performance of the model. Conclusions: Our progressive CKD classifier accurately predicts significant eGFR decline in patients with early, mid, and advanced disease using readily obtainable laboratory data. Although prospective studies are warranted, our results support the clinical utility of the model to improve timely recognition and optimal management for patients at risk for CKD progression. Plain-Language Summary: Defined by a significant decrease in estimated glomerular filtration rate (eGFR), chronic kidney disease (CKD) progression is strongly associated with kidney failure. However, to date, there are no broadly used resources that can predict this clinically significant event. Using machine-learning techniques on a diverse US population, this cohort study aimed to address this deficiency and found that a 5-year risk prediction model for CKD progression was accurate. The most important predictor of progressive decline in kidney function was the eGFR slope, followed by the urine albumin-creatinine ratio and serum albumin slope. Although further study is warranted, the results showed that a machine-learning model using readily obtainable laboratory information accurately predicts CKD progression, which may inform clinical diagnosis and management for this at-risk population.

Social Media Engagement at Academic Conferences: Report of the Association of Pathology Chairs 2018 and 2019 Annual Meeting Social Media Committee.

The use of social media at academic conferences is expanding, and platforms such as Twitter are used to share meeting content with the world. Pathology conferences are no exception, and recently, pathology organizations have promoted social media as a way to enhance meeting exposure. A social media committee was formed ad hoc to implement strategies to enhance social media involvement and coverage at the 2018 and 2019 annual meetings of the Association of Pathology Chairs. This organized approach resulted in an 11-fold increase in social media engagement compared to the year prior to committee formation (2017). In this article, the social media committee reviews the strategies that were employed and the resultant outcome data. In addition, we categorize tweets by topic to identify the topics of greatest interest to meeting participants, and we discuss the differences between Twitter and other social media platforms. Lastly, we review the existing literature on this topic from 23 medical specialties and health care fields.

Active Monitoring and Feedback to Improve Blood Culture Fill Volumes and Positivity Across a Large Integrated Health System.

BACKGROUND: The sensitivity of blood cultures increases with the volume of blood collected. However, hospitals face challenges in collecting adequate volume, and underfilled blood bottles are ubiquitous. METHODS: Blood bottle fill volumes were measured using an automated monitoring system across multiples sites (10 hospitals, 3 laboratories) within a large suburban/urban health system. Baseline fill volumes were measured for 4 months. A quality improvement program was then implemented over 36 months. Strategies to improve fill volumes included education, standardized data collection, novel and unblinded information cascades, targeted communication, and bottle markings for blood collectors. RESULTS: A total of 516 201 blood cultures were evaluated over 40 months. In the preimplementation period (January-April 2015), no hospitals collected the recommended 8-10 mL/bottle, and the average system fill volume was 2.3 mL. In the final postimplementation period (January-April 2018), 7 of 10 hospitals achieved ≥8 mL per bottle and the system average increased to 8.6 mL (P < .0001). The positivity rate increased 20%, from 7.39% to 8.85% (P < .001), whereas the contamination rate did not change and remained low. Compared to the preimplementation period, the odds of positive cultures containing potential pathogens increased to 1.18 (95% confidence interval, 1.05-1.32; P = .003). CONCLUSIONS: Here we show that underfilled blood cultures are extremely common but that operational and educational strategies can result in sustained improvements across a complex network of hospitals and laboratories. This leads to increased detection of pathogens, which can have tremendous impact on the management of bloodstream infections and sepsis.

Gross Dissection Time Values of Pathologists' Assistants Using Standardized Metrics.

OBJECTIVES: A validated and objective method to quantify the gross dissection time of pathologists' assistants (PAs) does not exist. We propose a method to calculate standardized work units (dissection time values [DTVs]) to monitor PA productivity. METHODS: The Current Procedural Terminology system specifies six levels of specimen complexity encompassing 176 unique specimen types. Using our institutional dictionary, we designated all specimen types into a priori five levels of complexity based on expected dissection time. We hypothesized that expected time could be matched prospectively with the actual measured dissection time for all specimens. Dissection time data were collected prospectively for 12,775 specimens at two tertiary academic medical centers, and work effort was converted to a numeric DTV equivalent (number of minutes to dissect single specimen/420 minutes in a working day). RESULTS: For 44 of 155 specimen types, measured dissection time for the five "levels" was lower than expected dissection (P < .0001). Accordingly, those 44 specimen types were reclassified to a lower level. CONCLUSIONS: A numeric standard of the work effort for dissection time for 155 specimen types was developed, validated, and then used prospectively to monitor grossing efficiency of PA workforce.

The Role of the Pathologist in Population Health.

CONTEXT.­: As part of its value-based care initiative, the College of American Pathologists has pursued research to better understand the role pathologists can have in population health. OBJECTIVES.­: To answer the following questions: (1) what is the impact of population health and population health management on pathologists; (2) what roles are pathologists playing in population health management; (3) is population health something that pathologists in both larger and smaller settings can engage in; (4) are pathologists in a position to analyze laboratory data for population health, and, if so, what are the key information sources those pathologists must access; and (5) what steps can a pathologist take to become involved in population health? DESIGN.­: We conducted 10 semistructured interviews with pathologists and other medical laboratory leaders who have been active in population health. These interviews were supplemented with a review of the medical literature. RESULTS.­: Pathologists have demonstrated that laboratory data can provide unique value-added contributions to improving the health of populations. These contributions are not limited to pathologists in large, integrated settings. However, pathologists need to be proactive to contribute to health systems' population health efforts and may need to both enhance their own skills and the quality of their data to maximize the value of their contributions. CONCLUSIONS.­: Although not necessarily a definitive summary of the roles that pathologists are playing in population health, this article identifies some of the promising and innovative activities occurring among pathologists and laboratorians.

Impact of Daily Electronic Laboratory Alerting on Early Detection and Clinical Documentation of Acute Kidney Injury in Hospital Settings.

Acute kidney injury, especially early-stage disease, is a common hospital comorbidity requiring timely recognition and treatment. We investigated the effect of daily laboratory alerting of patients at risk for acute kidney injury as measured by documented International Classification of Diseases diagnoses. A quasi-experimental study was conducted at 8 New York hospitals between January 1, 2014, and June 30, 2017. Education of clinical documentation improvement specialists, physicians, and nurses was conducted from July 1, 2014, to December 31, 2014, prior to initiating daily hospital-wide laboratory acute kidney injury alerting on January 1, 2015. Incidence based on documented International Classification of Diseases diagnosis of acute kidney injury and acute tubular necrosis during the intervention periods (3 periods of 6 months each: January 1 to June 30 of 2015, 2016, and 2017) were compared to one preintervention period (January 1, 2014, to June 30, 2014). The sample consisted of 269 607 adult hospital discharges, among which there were 39 071 episodes based on laboratory estimates and 27 660 episodes of documented International Classification of Diseases diagnoses of acute kidney injury or acute tubular necrosis. Documented incidence improved significantly from the 2014 preintervention period (5.70%; 95% confidence interval: 5.52%-5.88%) to intervention periods in 2015 (9.89%; 95% confidence interval, 9.66%-10.12%; risk ratio = 1.73, P < .001), 2016 (12.76%; 95% confidence interval, 12.51%-13.01%; risk ratio = 2.24, P < .001), and 2017 (12.49%; 95% confidence interval, 12.24%-12.74%; risk ratio = 2.19, P < .001). A multifactorial intervention comprising daily laboratory alerting and education of physicians, nurses, and clinical documentation improvement specialists led to increased recognition and clinical documentation of acute kidney injury.

Improving American Healthcare Through "Clinical Lab 2.0": A Project Santa Fe Report.

Project Santa Fe was established both to provide thought leadership and to help develop the evidence base for the valuation of clinical laboratory services in the next era of American healthcare. The participants in Project Santa Fe represent major regional health systems that can operationalize laboratory-driven innovations and test their valuation in diverse regional marketplaces in the United States. We provide recommendations from the inaugural March 2016 meeting of Project Santa Fe. Specifically, in the transition from volume-based to value-based health care, clinical laboratories are called upon to provide programmatic leadership in reducing total cost of care through optimization of time-to-diagnosis and time-to-effective therapeutics, optimization of care coordination, and programmatic support of wellness care, screening, and monitoring. This call to action is more than working with industry stakeholders on the basis of our expertise; it is providing leadership in creating the programs that accomplish these objectives. In so doing, clinical laboratories can be effectors in identifying patients at risk for escalation in care, closing gaps in care, and optimizing outcomes of health care innovation. We also hope that, through such activities, the evidence base will be created for the new value propositions of integrated laboratory networks. In the very simplest sense, this effort to create "Clinical Lab 2.0" will establish the impact of laboratory diagnostics on the full 100% spend in American healthcare, not just the 2.5% spend attributed to in vitro diagnostics. In so doing, our aim is to empower regional and local laboratories to thrive under new models of payment in the next era of American health care delivery.