Improving Chronic Disease Management With Pieces (ICD-Pieces)

Parkland Center for Clinical Innovation, Parkland Health and Hospital System, Texas Health Resources, Connecticut Center for Primary Care, Dallas VA Research Corporation, G-Health Enterprises, National Institutes of Health (NIH), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Heart, Lung, and Blood Institute (NHLBI)

ICD-Pieces (Parkland Intelligent e-Coordination and Evaluation System) trial is a National Institutes of Health (NIH) Healthcare Systems(HCS) Collaboratory demonstration project to improve management of patients with a triad of Chronic Kidney Disease, hypertension and diabetes with Pieces in four HCS including Parkland, Texas Health Resources (THR), ProHealth Physicians Incorporation and North Texas VA. Pieces is a decision support technology platform created by Parkland Center for Clinical Innovation(PCCI). The primary objective is to test the hypothesis that a collaborative model of primary care and subspecialty care intervention enhanced by Pieces and practice facilitators compared to standard clinical practice will reduce all-cause hospitalizations in patients with coexisting chronic kidney disease, diabetes and hypertension. Secondary objectives are: a)Test if implementation of the collaborative model will reduce 30-day readmissions, emergency room visits, cardiovascular events or deaths and disease-specific hospitalizations; b) Develop and validate risk predictive models for disease-specific hospitalizations, all-cause hospitalizations, 30-day readmissions, emergency room visits, cardiovascular events and deaths for patients with chronic kidney disease, diabetes and hypertension. c) Collect demographic and clinical data to assist phenotyping patients with chronic kidney disease, diabetes and hypertension. d) Obtain safety data including Acute Kidney Injury, progression of chronic kidney disease, electrolyte disturbances and medication errors, and drug toxicity; e) Collect resource utilization information including hospitalizations, emergency room visits, outpatient visits, and diagnostic or therapeutic procedures completed. Candidate patients in selected clinics will be enrolled over a period of 2 years and followed for 12 months. Pieces will ascertain both primary and secondary outcomes from the Electronic Health Record supported with data from the Dallas Fort Worth Hospital Council (DFWHC), Accountable Care Organization (ACO) reports and VA database, and deaths from Social Security Index (SSI) data.

ICD-Pieces is a randomized, pragmatic clinical trial in four large healthcare systems to improve care of patients with coexistent chronic kidney disease, diabetes and hypertension. The investigators hypothesis is that patients who receive care with a collaborative model of primary care-subspecialty care enhanced by novel information technology (Pieces) and practice facilitators will have fewer all-cause hospitalizations, readmissions, disease-specific hospitalizations, Emergency Room visits, cardiovascular events and deaths than patients receiving standard medical care. The investigators will also aim to develop a better understanding of risk predictors in patients with chronic kidney disease, diabetes and hypertension to guide future recommendations of therapies that are tailored to individual patients. The primary objective of the study is to test the hypothesis that a collaborative model of primary care enhanced by novel information technology and practice facilitators will allow to leverage data from electronic health records to identify patients with the triad of chronic kidney disease, diabetes and hypertension using objective and reproducible criteria, and provide clinician support for implementation of best practices of care, monitoring clinical measures, adjusting treatments and reduce 12-month hospitalization rates. In this study disease-specific hospitalizations for chronic kidney disease, diabetes and hypertension include hospitalizations due to cardiovascular complications, congestive heart failure, volume overload, accelerated/malignant/uncontrolled hypertension, acute coronary syndromes, myocardial infarction, stroke, coronary/peripheral revascularization, limb ischemia/amputations, diabetes complications, uncontrolled diabetes, hypoglycemia, acute kidney injury, hyperkalemia, electrolyte disturbances, medication errors, drug toxicity, and infections. Secondary Objectives: The study will test if implementation of the collaborative model of primary care-subspecialty care interventions will reduce 30-day readmissions (for patients who are hospitalized), emergency room visits, cardiovascular events, deaths or disease-specific hospitalizations. In this study disease-specific hospitalizations for chronic kidney disease, diabetes and hypertension include hospitalizations due to cardiovascular complications, congestive heart failure, volume overload, accelerated/malignant/uncontrolled hypertension, acute coronary syndromes, myocardial infarction, stroke, coronary/peripheral revascularization, limb ischemia/amputations, diabetes complications, uncontrolled diabetes, hypoglycemia, acute kidney injury, hyperkalemia, electrolyte disturbances, medication errors, drug toxicity, and infections. 2.) Develop and validate predictive models for risks of hospitalizations, emergency room visits, cardio vascular events and deaths for all patients with coexistent chronic kidney disease, diabetes and hypertension and predict risks of 30-day disease-specific readmissions for patients who are hospitalized. 3.) Capture data (demographic, clinical, medications, laboratories, procedures) to phenotype patients with a triad of chronic kidney disease, diabetes and hypertension. 4.) Obtain important safety data for patients with chronic kidney disease, diabetes and hypertension including adverse safety events, acute kidney injury and progression of chronic kidney disease (even for patients not hospitalized). 5.) Obtain information on resource utilization including not only hospitalizations but also all emergency room visits, outpatient visits and diagnostic and therapeutic procedures. 6.) Evaluate the impact of the collaboratory model of care on patient Reported Outcomes (PROs) including health-related quality of life, patient satisfaction, Patient perspective on quality of their care and measures of patient perception of burden related to care of their chronic kidney disease, diabetes and hypertension. 7.) Evaluate the impact of the collaborative primary care-subspecialty care model on provider satisfaction with resources and ability to manage patients with coexistent chronic kidney disease, diabetes and hypertension. Study Outcome Measures: The primary outcome of this trial is all cause hospitalizations for patients with a triad of chronic kidney disease, diabetes and hypertension. Specifically, the outcome will be hospitalization rates at 12 months for study participants. The investigators will include both regular hospitalizations as defined by Center for Medicaid and Medicare(CMS) and observation status overnight (to avoid variations related to definition of inpatient status/hospitalization using the recent implementation of the "two midnight rule" CMS-1599-F). Study Design: The study will employ a prospective stratified cluster randomization design. The stratum is each of the four large healthcare systems participating in the study. The unit of randomization will be primary care clinics. In some healthcare systems several primary care clinics share the same geographic location and personnel and they will be randomized as a single unit. The cluster design of the study is best suited to detect important differences in outcomes between the intervention and control groups[8,55]. The decision to use primary care clinics as a unit for randomization in the study is based on the ability to implement different models of care in the active intervention sites as compared to the control sites. The collaborative model of care which includes novel information technology, subject identification, facilitation of patient care, monitoring of outcomes and participation from facilitators can be most efficiently applied to the workflow of clinics when they are fully randomized to active intervention. The cluster randomization design with clinics receiving collaborative primary care-subspecialty care versus standard care also limits the risk of cross-contamination between intervention and control groups in the study Primary care practices will be stratified by healthcare systems and randomly allocated to either intervention group or standard medical care group using a randomized permutation block within stratum. Based on the assignment of the clinic where a patient goes, each patient will be assigned either to the intervention group or the standard medical care group. All eligible patients of clinics who are randomized to the study will be included in the comparison of the two intervention groups regardless of intervention compliance (intention-to-treat analysis) to investigate if patients in intervention group have significantly less all-cause hospitalizations than those in the standard medical care group. Evaluation will also be performed to determine treatment effects on disease-specific hospitalizations, emergency room visits, cardiovascular events and deaths. There will be two study groups: active intervention group randomized to the collaborative model of care facilitated by novel information technology and practice facilitators and standard/usual care group. The intervention in the active group is implementation of a collaborative model of care that facilitates delivering best care practices to patients who have coexistent chronic kidney disease, type 2 diabetes and hypertension. The model uses a novel information technology platform called Pieces and practice facilitators with the purpose of allowing for early identification of patients with objective criteria and to implement best practices of care, monitor important clinical measures, adjust treatments and achieve improved outcomes. The intervention will be delivered in the outpatient setting. Data collection for assessment of study objectives will be mainly based on information technology tools to capture data from the electronic health record. Some data fields will require collection of data from Dallas Fort Worth Regional Hospital Council, ProHealth Accountable Care Organization databases, VA of North Texas and Social Security Death Files Index. A Data Safety Monitoring Board (DSMB) has been assembled by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and will oversee study planning and implementation of the study.

Chronic, Risk prediction model, Quality Improvement (QI), Pragmatic trial, Cluster randomization, Collaborative care, clinical informatics, Chronic kidney disease, hypertension, high BP, diabetes

Group to receive Collaborative model of primary care and subspecialty care enhanced by Pieces and Practice Facilitator

Pieces will access Electronic Health Record for all patients receiving care at the participating sites to detect patients with a triad of chronic kidney disease, diabetes and hypertension, facilitate management and monitor outcomes. To maximize successful implementation of care, a Practice Facilitator will be at each site with standardized role training using a curriculum based on the Agency for Healthcare Research and Quality (AHRQ) Practice Facilitation Handbook. Specific interventions are maintaining BP less than 140/90 mmHg, use of angiotensin-converting-enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB), treatment with statins, aiming for glycosylated Hemoglobin (HgA1C) at the recommended target, and avoiding nephrotoxic medications. Additional interventions include chronic kidney disease education for Primary Care Providers (PCP) and patients using National Kidney Disease Education Program (NKDEP) materials.

All cause hospitalizations for patients with a triad of chronic kidney disease, diabetes and hypertension

Hospitalization rates at 12 months for all study participants, capturing all cause unplanned hospitalizations including both regular hospitalizations as currently defined by CMS and observation status overnight. Hospitalizations will be ascertained from Electronic Health Record of each participating healthcare system with assistance from Pieces.

30-day all cause readmissions (for those patients who have an index hospitalization), emergency room visits, cardiovascular events and deaths, and disease-specific hospitalizations for cardiovascular complications, congestive heart failure, volume overload, hypertension complications, acute coronary syndrome, myocardial infarction, coronary/peripheral revascularization, stroke, amputation/limb ischemia, uncontrolled diabetes, hypoglycemia, diabetes complications, acute kidney injury, hyperkalemia, electrolyte disturbances, medication errors, drug toxicity and infections.

Inclusion Criteria CKD Inclusion Criteria (present at least ≥ 3 months apart) There will be two or more Estimated Glomerular Filtration Rate (eGFRs) calculations less than 60ml/minute (corrected for BSA) or Two or more positive tests for albuminuria and/or proteinuria Albuminuria/proteinuria can be defined by quantitative criteria with albumin/creatinine ratio greater than 30mg/g, urine protein creatinine ratio greater than 200mg/g or positive dipstick with protein detection (adjusted for urinary concentration/specific gravity). Diabetes Inclusion Criteria Only patients with type 2 diabetes will be enrolled in this study. Random blood glucose greater than 200mg/dL Hemoglobin A1C greater than 6.5% Use of hypoglycemic agents or Type 2 diabetes included in problem list Hypertension Inclusion Criteria Systolic blood pressure greater than 140 mmHg on two different occasions at least one week apart Diastolic blood pressure greater than 90 on two occasions at least more than one week apart Use of antihypertensive agents except thiazide diuretics or Hypertension included in problem list Exclusion Criteria: Exclusion criteria will be minimal in this pragmatic trial. The collaborative model of care will not be implemented in patients younger than 18 years or older than 85 years of age or patients who have CKD stage 5/End Stage Renal Disease(ESRD. Primary care practitioners have the option of not implementing the intervention on any of their patients if they believe benefit to be minimal or risk too high due to patient comorbidities

We are obliged to share our data within the analytic team of the ICD-Pieces trial. The authors will retain the rights to the de-identified final data until the trial is completed. Interested parties will be able to download information about the predictive model from the PCCI website: www.pccipieces.org Short term data storage of transfer files will occur via encrypted drives between password protected encrypted computers