Building a vertically integrated genomic learning health system: The biobank at the Colorado Center for Personalized Medicine.

Precision medicine initiatives across the globe have led to a revolution of repositories linking large-scale genomic data with electronic health records, enabling genomic analyses across the entire phenome. Many of these initiatives focus solely on research insights, leading to limited direct benefit to patients. We describe the biobank at the Colorado Center for Personalized Medicine (CCPM Biobank) that was jointly developed by the University of Colorado Anschutz Medical Campus and UCHealth to serve as a unique, dual-purpose research and clinical resource accelerating personalized medicine. This living resource currently has more than 200,000 participants with ongoing recruitment. We highlight the clinical, laboratory, regulatory, and HIPAA-compliant informatics infrastructure along with our stakeholder engagement, consent, recontact, and participant engagement strategies. We characterize aspects of genetic and geographic diversity unique to the Rocky Mountain region, the primary catchment area for CCPM Biobank participants. We leverage linked health and demographic information of the CCPM Biobank participant population to demonstrate the utility of the CCPM Biobank to replicate complex trait associations in the first 33,674 genotyped individuals across multiple disease domains. Finally, we describe our current efforts toward return of clinical genetic test results, including high-impact pathogenic variants and pharmacogenetic information, and our broader goals as the CCPM Biobank continues to grow. Bringing clinical and research interests together fosters unique clinical and translational questions that can be addressed from the large EHR-linked CCPM Biobank resource within a HIPAA- and CLIA-certified environment.

Delays in Diagnosis and Treatment of Breast Cancer: A Safety-Net Population Profile.

Background: Timely detection and treatment of breast cancer is important in optimizing survival and minimizing recurrence. Given disparities in breast cancer outcomes based on socioeconomic status, we examined time to diagnosis and treatment in a safety-net hospital. Methods: We conducted a retrospective review of all patients with breast cancer diagnosed between July 1, 2010, and June 30, 2012 (N=120). We limited our analytic sample to patients with nonrecurrent, primary stage 0-III breast cancer (N=105) and determined intervals from presentation to diagnosis, diagnosis to first treatment, last surgery to chemotherapy initiation, and last surgery to start of radiation therapy (RT). Using logistic regression, we calculated unadjusted odds of receiving timely treatment (< median time) versus more delayed treatment (≥ median time) as a function of age, language, ethnicity, insurance, Charlson comorbidity index, disease stage, method of first presentation (screening mammography vs care provider), symptoms at presentation, and type of surgical treatment. Results: Patients aged 55 to 64 years accounted for most of the sample (n=37; 35.2%). Median time from presentation to diagnosis (23 days), time from diagnosis to first treatment, and time from surgery to chemotherapy initiation fell within intervals published in the literature; median time from last surgery to start of RT was greater than recommended intervals. Factors significantly associated with longer intervals than median time included stage, method of presentation, language, surgical treatment, insurance, and ethnicity. Conclusions: Patients in this safety-net setting experienced acceptable diagnosis and treatment intervals, except for time to RT. Focused interventions that help care providers access imaging quickly for their symptomatic patients could improve time to diagnosis. Concentrating additional efforts on non-English-speaking, Hispanic patients and those who need to receive RT could improve time to treatment.

The association of cannabis use on inpatient psychiatric hospital outcomes.

BACKGROUND: The associations between cannabis use and psychosis are well documented in numerous studies. There is a need to evaluate the impact of cannabis use on inpatient psychiatric utilization and outcomes. OBJECTIVES: To evaluate the impact of cannabis use on psychiatric hospital outcomes. METHODS: This study was conducted between April 20, 2015 and October 20, 2015. All patients (n = 120) admitted to Denver Health with psychotic symptoms were administered a urine toxicology screening testing for the presence of 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THC-COOH, the active metabolite of cannabis). Patients with positive tests were compared to those with negative tests on several measures, including length of stay, presence or lack of 30-day readmission, Brief Psychotic Rating Scale (BPRS) score, and use of antipsychotics and/or sedatives/anxiolytics. RESULTS: There were 120 patients. Twenty nine were women and 91 were men. Patients testing positive for THC-COOH had a shorter length of stay compared to patients testing negative for THC-COOH, after adjusting for age, prior psychiatric admissions, history of a psychotic-spectrum disorder, and comorbid additional substance use (p = 0.02). There were no differences in 30-day readmissions, 30-day post-discharge presentation to the Denver Health psychiatric emergency department, BPRS scores, and medication administration. CONCLUSION: Patients presenting with psychotic symptoms and cannabis use require shorter inpatient psychiatric hospitalizations. This study is the first to quantify this observation and highlights the need for future clinical decision-making tools that would ideally correlate cannabis use with the degree of potential need for expensive and scarce mental health resources, such as psychiatric hospitalization.

MENDS-on-FHIR: leveraging the OMOP common data model and FHIR standards for national chronic disease surveillance.

Objectives: The Multi-State EHR-Based Network for Disease Surveillance (MENDS) is a population-based chronic disease surveillance distributed data network that uses institution-specific extraction-transformation-load (ETL) routines. MENDS-on-FHIR examined using Health Language Seven's Fast Healthcare Interoperability Resources (HL7® FHIR®) and US Core Implementation Guide (US Core IG) compliant resources derived from the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) to create a standards-based ETL pipeline. Materials and Methods: The input data source was a research data warehouse containing clinical and administrative data in OMOP CDM Version 5.3 format. OMOP-to-FHIR transformations, using a unique JavaScript Object Notation (JSON)-to-JSON transformation language called Whistle, created FHIR R4 V4.0.1/US Core IG V4.0.0 conformant resources that were stored in a local FHIR server. A REST-based Bulk FHIR $export request extracted FHIR resources to populate a local MENDS database. Results: Eleven OMOP tables were used to create 10 FHIR/US Core compliant resource types. A total of 1.13 trillion resources were extracted and inserted into the MENDS repository. A very low rate of non-compliant resources was observed. Discussion: OMOP-to-FHIR transformation results passed validation with less than a 1% non-compliance rate. These standards-compliant FHIR resources provided standardized data elements required by the MENDS surveillance use case. The Bulk FHIR application programming interface (API) enabled population-level data exchange using interoperable FHIR resources. The OMOP-to-FHIR transformation pipeline creates a FHIR interface for accessing OMOP data. Conclusion: MENDS-on-FHIR successfully replaced custom ETL with standards-based interoperable FHIR resources using Bulk FHIR. The OMOP-to-FHIR transformations provide an alternative mechanism for sharing OMOP data.

MENDS-on-FHIR: Leveraging the OMOP common data model and FHIR standards for national chronic disease surveillance.

Objective: The Multi-State EHR-Based Network for Disease Surveillance (MENDS) is a population-based chronic disease surveillance distributed data network that uses institution-specific extraction-transformation-load (ETL) routines. MENDS-on-FHIR examined using Health Language Seven's Fast Healthcare Interoperability Resources (HL7® FHIR®) and US Core Implementation Guide (US Core IG) compliant resources derived from the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) to create a standards-based ETL pipeline. Materials and Methods: The input data source was a research data warehouse containing clinical and administrative data in OMOP CDM Version 5.3 format. OMOP-to-FHIR transformations, using a unique JavaScript Object Notation (JSON)-to-JSON transformation language called Whistle, created FHIR R4 V4.0.1/US Core IG V4.0.0 conformant resources that were stored in a local FHIR server. A REST-based Bulk FHIR $export request extracted FHIR resources to populate a local MENDS database. Results: Eleven OMOP tables were used to create 10 FHIR/US Core compliant resource types. A total of 1.13 trillion resources were extracted and inserted into the MENDS repository. A very low rate of non-compliant resources was observed. Discussion: OMOP-to-FHIR transformation results passed validation with less than a 1% non-compliance rate. These standards-compliant FHIR resources provided standardized data elements required by the MENDS surveillance use case. The Bulk FHIR application programming interface (API) enabled population-level data exchange using interoperable FHIR resources. The OMOP-to-FHIR transformation pipeline creates a FHIR interface for accessing OMOP data. Conclusion: MENDS-on-FHIR successfully replaced custom ETL with standards-based interoperable FHIR resources using Bulk FHIR. The OMOP-to-FHIR transformations provide an alternative mechanism for sharing OMOP data.

Machine Learning-Based Exploratory Clinical Decision Support for Newly Diagnosed Patients With Acute Myeloid Leukemia Treated With 7 + 3 Type Chemotherapy or Venetoclax/Azacitidine.

PURPOSE: There are currently limited objective criteria to help assist physicians in determining whether an individual patient with acute myeloid leukemia (AML) is likely to do better with induction with either standard 7 + 3 chemotherapy or targeted therapy with venetoclax plus azacitidine. The study goal was to address this need by developing exploratory clinical decision support methods. PATIENTS AND METHODS: Univariable and multivariable analysis as well as comparison of a range of machine learning (ML) predictors were performed using cohorts of 120 newly diagnosed 7 + 3-treated AML patients compared with 101 venetoclax plus azacitidine-treated patients. RESULTS: A variety of features in the two patient cohorts were identified that may potentially correlate with short- and long-term outcomes, toxicities, and other considerations. A subset of these diagnostic features was then used to develop ML-based predictors with relatively high areas under the curve of short- and long-term outcomes, hospital stays, transfusion requirements, and toxicities for individual patients treated with either venetoclax/azacitidine or 7 + 3. CONCLUSION: Potential ML-based approaches to clinical decision support to help guide individual patients with newly diagnosed AML to either 7 + 3 or venetoclax plus azacitidine induction therapy were identified. Larger cohorts with separate test and validation studies are necessary to confirm these initial findings.

Changes in Healthcare Encounter Rates Possibly Related to Cannabis or Alcohol following Legalization of Recreational Marijuana in a Safety-Net Hospital: An Interrupted Time Series Analysis.

OBJECTIVES: Liberalization of marijuana laws in Colorado contributed to increases in cannabis-related adverse events over time. We examined characteristics of patients with healthcare encounters possibly related to cannabis and assessed the temporal association between legalization of recreational marijuana and healthcare encounters possibly attributed to cannabis. METHODS: Annual encounter rates possibly related to cannabis and alcohol were compared using negative binomial regression. Two-time intervals, pre/post-recreational marijuana legalization (January 2009 to December 2013 and January 2014 to December 2015, respectively) were used to examine changes in monthly rates of emergency encounters and hospitalizations possibly related to cannabis. Level and trend changes on encounter rates by legalization period were assessed using interrupted time series analyses. Encounters possibly related to alcohol were used as a comparator group. RESULTS: Most encounters identified during the study period had alcohol-related International Classification of Diseases Diagnosis and Procedural Codes (ICD-9/10-CM) codes (94.8% vs 5.2% for cannabis). Patients with encounters possibly related to cannabis were younger, more likely to be hospitalized and more likely to be admitted to the psychiatric unit than patients with encounters possibly related to alcohol. Initial and sustained effects of encounter rates possibly related to cannabis demonstrated an increased trend in slope before and after recreational marijuana legalization. The slope became more abrupt following legalization with a significant increase in trend during the post-legalization period (ß = 2.7, standard error = 0.3, ρ < 0.0001). No significant change was noted for encounters possibly related to alcohol. CONCLUSIONS: Additional research should identify patients at highest risk of an adverse health event related to cannabis and quantify costs associated with cannabis-related healthcare delivery.

Peer Navigator Intervention for Latinos on Hemodialysis: A Single-Arm Clinical Trial.

Background: Latinos with end-stage renal disease (ESRD) have worse mental and kidney composite health-related quality of life (HRQOL) scores compared to non-Latino ESRD patients. Latino ESRD patients uniquely report that social factors (e.g., lack of care coordination, food insecurity, and low health literacy) and mental health influence their HRQOL. We developed a culturally tailored peer navigator (PN) intervention to improve the HRQOL of Latinos on hemodialysis. Objective: To determine the feasibility of the PN intervention. Design: Single-arm prospective study. The PN provided individualized support with advance care planning, care coordination, and counseling about the importance of diet and mental health. Setting and Participants: Latino with ESRD receiving scheduled outpatient thrice-weekly hemodialysis or reliant on emergency-only hemodialysis in Denver. Main measures: Recruitment, retention rates, data completeness, intervention dose, patient- and staff-reported satisfaction with the intervention. Results: Of 49 eligible patients, 40 (82%) agreed to participate. The majority of participants received scheduled outpatient hemodialysis (75%), 20 were women (50%), with a mean (standard deviation [SD]) age of 56 (13) years. No participants withdrew from the intervention. One participant died. The mean (SD) number of PN visits per participant was 7 (2) and the mean (SD) length of the visits was 97 minutes (49). The majority of visits took place at the hemodialysis facility (59%) and home (27%). The vast majority of participants reported that the PN improved their quality of life as a patient on hemodialysis (95%). Conclusions: The PN intervention achieved feasibility goals and was well received by participants.

Symptom Burden among Latino Patients with End-Stage Renal Disease and Access to Standard or Emergency-Only Hemodialysis.

BACKGROUND: Patients with end-stage renal disease (ESRD) have a high symptom burden and this negatively impacts health-related quality of life. Little is known about the symptom burden of Latinos with ESRD and variable access to hemodialysis. OBJECTIVE: To estimate the symptom burden of Latinos with ESRD and access to standard or emergency-only hemodialysis. DESIGN/SETTING/SUBJECTS: Observational descriptive study of Latino adults with ESRD receiving standard or emergency-only hemodialysis. Patients completed the Edmonton Symptom Assessment System Revised: Renal (ESAS-r:Renal). We used descriptive statistics and propensity score adjustment to conduct the analysis. MEASUREMENT: ESAS-r:Renal. RESULTS: Participants (N = 67) had a mean age of 58 years (standard deviation [SD] ±13) and a mean Charlson Comorbidity Index of 6.6 ± 2.5, and had been on hemodialysis a mean of 42 months (SD ±43). On average, Latinos with ESRD experienced 7 (SD ±3) symptoms with a mean of 5 ± 3 symptoms reported as moderate or severe. After adjusting for propensity score, emergency-only hemodialysis patients reported experiencing more nausea compared to standard hemodialysis patients (odds ratio 8.95, 95% confidence interval: 1.17-68.31, p = 0.03). CONCLUSION: Latinos with ESRD have a high symptom burden and compared to patients with standard hemodialysis, patients who rely on emergency-only hemodialysis report more nausea. A national treatment strategy that provides standard hemodialysis for undocumented immigrants with ESRD is an important next step.

Alcohol misuse and outpatient follow-up after hospital discharge: a retrospective cohort study.

PURPOSE: Patients with alcohol misuse are less likely to receive preventive health services but more likely to utilize emergency health services. However, the association between alcohol misuse and outpatient follow-up after hospitalization is unknown and may depend on whether a patient experiences a critical illness. We sought to determine whether alcohol misuse was associated with lower rates of outpatient follow-up after hospital discharge and whether the magnitude of this association differed in patients who experienced a critical illness. MATERIALS AND METHODS: This was a retrospective cohort study using administrative data from an urban safety net hospital. Patients were included if they were admitted between 2011 and 2015, were between the ages of 18 and 89, resided within the safety net county, were discharged home, and were at moderate to high risk for hospital readmission within the subsequent 30 days. Alcohol misuse was identified using a combination of ICD-9 codes and response to a single screening question. The primary outcome was a combined measure of follow-up with a primary care physician or specialist provider in the 30 days following hospital discharge. Multivariable logistic regression was used to adjust for factors known to be associated with healthcare utilization. RESULTS: Overall, 17,575 patients were included in the analysis; 4984 (28%) had alcohol misuse. In the 30 days following hospital discharge, 46% of patients saw any outpatient provider. In an unadjusted analysis, the association between alcohol misuse and attending any outpatient follow-up was dependent on whether patients had a critical illness (p value < 0.0001) with the highest rates of follow-up in survivors of critical illness without alcohol misuse (53%, 95% CI 51%, 55%) followed by patients without alcohol misuse or critical illness (49%; 95% CI 48%, 50%), patients with alcohol misuse without critical illness (38%; 95% CI 36%, 39%), and patients with alcohol misuse and a critical illness (37%; 95% CI 35%, 40%). Adjusting for factors associated with healthcare utilization, these findings were modestly attenuated but unchanged. CONCLUSIONS: Patients with alcohol misuse who are at moderate to high risk for hospital readmission may benefit from targeted interventions to increase rates of outpatient follow-up after hospital discharge.

Association of Emergency-Only vs Standard Hemodialysis With Mortality and Health Care Use Among Undocumented Immigrants With End-stage Renal Disease.

Importance: Undocumented immigrants with end-stage renal disease have variable access to hemodialysis in the United States despite evidence-based standards for frequency of dialysis care. Objective: To determine whether mortality and health care use differs among undocumented immigrants who receive emergency-only hemodialysis vs standard hemodialysis (3 times weekly at a health care center). Design, Setting, and Participants: A retrospective cohort study was conducted of undocumented immigrants with incident end-stage renal disease who initiated emergency-only hemodialysis (Denver Health, Denver, Colorado, and Harris Health, Houston, Texas) or standard (Zuckerberg San Francisco General Hospital, San Francisco, California) hemodialysis between January 1, 2007, and July 15, 2014. Exposures: Access to emergency-only hemodialysis vs standard hemodialysis. Main Outcomes and Measures: The primary outcome was mortality. Secondary outcomes were health care use (acute care days and ambulatory care visits) and rates of bacteremia. Outcomes were adjusted for propensity to undergo emergency hemodialysis vs standard hemodialysis. Results: A total of 211 undocumented patients (86 women and 125 men; mean [SD] age, 46.5 [14.6] years; 42 from the standard hemodialysis group and 169 from the emergency-only hemodialysis group) initiated hemodialysis during the study period. Patients receiving standard hemodialysis were more likely to initiate hemodialysis with an arteriovenous fistula or graft and had higher albumin and hemoglobin levels than patients receiving emergency-only hemodialysis. Adjusting for propensity score, the mean 3-year relative hazard of mortality among patients who received emergency-only hemodialysis was nearly 5-fold (hazard ratio, 4.96; 95% CI, 0.93-26.45; P = .06) greater compared with patients who received standard hemodialysis. Mean 5-year relative hazard of mortality for patients who received emergency-only hemodialysis was more than 14-fold (hazard ratio, 14.13; 95% CI, 1.24-161.00; P = .03) higher than for those who received standard hemodialysis after adjustment for propensity score. The number of acute care days for patients who received emergency-only hemodialysis was 9.81 times (95% CI, 6.27-15.35; P < .001) the expected number of days for patients who had standard hemodialysis after adjustment for propensity score. Ambulatory care visits for patients who received emergency-only hemodialysis were 0.31 (95% CI, 0.21-0.46; P < .001) times less than the expected number of days for patients who received standard hemodialysis. Conclusions and Relevance: Undocumented immigrants with end-stage renal disease treated with emergency-only hemodialysis have higher mortality and spend more days in the hospital than those receiving standard hemodialysis. States and cities should consider offering standard hemodialysis to undocumented immigrants.

Barriers to Early Hospital Discharge: A Cross-Sectional Study at Five Academic Hospitals.

BACKGROUND: Understanding the issues delaying hospital discharges may inform efforts to improve hospital throughput. OBJECTIVE: This study was conducted to identify and determine the frequency of barriers contributing to delays in placing discharge orders. DESIGN: This was a prospective, cross-sectional study. Physicians were surveyed at approximately 8:00 AM, 12:00 PM, and 3:00 PM and were asked to identify patients that were "definite" or "possible" discharges and to describe the specific barriers to writing discharge orders. SETTING: This study was conducted at five hospitals in the United States. PARTICIPANTS: The study participants were attending and housestaff physicians on general medicine services. PRIMARY OUTCOMES AND MEASURES: Specific barriers to writing discharge orders were the primary outcomes; the secondary outcomes included discharge order time for high versus low team census, teaching versus nonteaching services, and rounding style. RESULTS: Among 1,584 patient evaluations, the most common delays for patients identified as "definite" discharges (n = 949) were related to caring for other patients on the team or waiting to staff patients with attendings. The most common barriers for patients identified as "possible" discharges (n = 1,237) were awaiting patient improvement and for ancillary services to complete care. Discharge orders were written a median of 43-58 minutes earlier for patients on teams with a smaller versus larger census, on nonteaching versus teaching services, and when rounding on patients likely to be discharged first (all P < .003). CONCLUSIONS: Discharge orders for patients ready for discharge are most commonly delayed because physicians are caring for other patients. Discharges of patients awaiting care completion are most commonly delayed because of imbalances between availability and demand for ancillary services. Team census, rounding style, and teaching teams affect discharge times.