Identify and monitor clinical variation using machine intelligence: a pilot in colorectal surgery.

Standardized clinical pathways are useful tool to reduce variation in clinical management and may improve quality of care. However the evidence supporting a specific clinical pathway for a patient or patient population is often imperfect limiting adoption and efficacy of clinical pathway. Machine intelligence can potentially identify clinical variation and may provide useful insights to create and optimize clinical pathways. In this quality improvement project we analyzed the inpatient care of 1786 patients undergoing colorectal surgery from 2015 to 2016 across multiple Ohio hospitals in the Cleveland Clinic System. Data from four information subsystems was loaded in the Clinical Variation Management (CVM) application (Ayasdi, Inc., Menlo Park, CA). The CVM application uses machine intelligence and topological data analysis methods to identify groups of similar patients based on the treatment received. We defined "favorable performance" as groups with lower direct variable cost, lower length of stay, and lower 30-day readmissions. The software auto-generated 9 distinct groups of patients based on similarity analysis. Overall, favorable performance was seen with ketorolac use, lower intra-operative fluid use (< 2000 cc) and surgery for cancer. Multiple sub-groups were easily created and analyzed. Adherence reporting tools were easy to use enabling almost real time monitoring. Machine intelligence provided useful insights to create and monitor care pathways with several advantages over traditional analytic approaches including: (1) analysis across disparate data sets, (2) unsupervised discovery, (3) speed and auto-generation of clinical pathways, (4) ease of use by team members, and (5) adherence reporting.

High-level adoption of electronic health records.

The adoption of electronic health records (EHRs) is now a national priority, but evidence shows that the level of typical adoption is low. We describe the goal of "high-level adoption" of an EHR system defined as a community of clinicians using an EHR system to share patient information, and create a summary of a patient's health, through the timely documentation of care, as well as through effective communication among caregivers, with a focus on patient safety and efficient use of resources. There are 10 key principles to achieving high level-adoption. These are: (1) shared medication lists; (2) shared patient problem lists; (3) established guidelines for communication about patient information among clinicians; (4) use of standard terminology; (5) preventive services; (6) integration of external data; (7) computerized provider order entry; (8) regional health information exchange; (9) adoption of clinical decision support; and (10) reuse of data for public health and research.

Reducing the prescribing of heavily marketed medications: a randomized controlled trial.

CONTEXT: Prescription drug costs are a major component of health care expenditures, yet resources to support evidence-based prescribing are not widely available. OBJECTIVE: To evaluate the effectiveness of computerized prescribing alerts, with or without physician-led group educational sessions, to reduce the prescribing of heavily marketed hypnotic medications. DESIGN: Cluster-randomized controlled trial. SETTING: We randomly allocated 14 internal medicine practice sites to receive usual care, computerized prescribing alerts alone, or alerts plus group educational sessions. MEASUREMENTS: Proportion of heavily marketed hypnotics prescribed before and after the implementation of computerized alerts and educational sessions. MAIN RESULTS: The activation of computerized alerts held the prescribing of heavily marketed hypnotic medications at pre-intervention levels in both the alert-only group (adjusted risk ratio [RR] 0.97; 95% CI 0.82-1.14) and the alert-plus-education group (RR 0.98; 95% CI 0.83-1.17) while the usual-care group experienced an increase in prescribing (RR 1.31; 95% CI 1.08-1.60). Compared to the usual-care group, the relative risk of prescribing heavily marketed medications was less in both the alert-group (Ratio of risk ratios [RRR] 0.74; 95% CI 0.57-0.96) and the alert-plus-education group (RRR 0.74; 95% CI 0.58-0.97). The prescribing of heavily marketed medications was similar in the alert-group and alert-plus-education group (RRR 1.02; 95% CI 0.80-1.29). Most clinicians reported that the alerts provided useful prescribing information (88%) and did not interfere with daily workflow (70%). CONCLUSIONS: Computerized decision support is an effective tool to reduce the prescribing of heavily marketed hypnotic medications in ambulatory care settings. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00788346.

Electronic Support for Public Health: validated case finding and reporting for notifiable diseases using electronic medical data.

Health care providers are legally obliged to report cases of specified diseases to public health authorities, but existing manual, provider-initiated reporting systems generally result in incomplete, error-prone, and tardy information flow. Automated laboratory-based reports are more likely accurate and timely, but lack clinical information and treatment details. Here, we describe the Electronic Support for Public Health (ESP) application, a robust, automated, secure, portable public health detection and messaging system for cases of notifiable diseases. The ESP application applies disease specific logic to any complete source of electronic medical data in a fully automated process, and supports an optional case management workflow system for case notification control. All relevant clinical, laboratory and demographic details are securely transferred to the local health authority as an HL7 message. The ESP application has operated continuously in production mode since January 2007, applying rigorously validated case identification logic to ambulatory EMR data from more than 600,000 patients. Source code for this highly interoperable application is freely available under an approved open-source license at http://esphealth.org.

Improving quality of care using a diabetes registry and disease management services in an integrated delivery network.

Disease management for chronic conditions is a call for collaboration among all parties of the health care system. The Caritas Christi Health Care System established a unified American Diabetes Association (ADA) recognized outpatient diabetes self-management education program (DSME) in each of its six hospital communities and has established an Internet data portal with managed care organizations to improve preventive care for thousands of patients with diabetes. This article describes the stepwise process of building the successful Caritas Diabetes Care Program and the central role of the Caritas Diabetes Registry over a 5-year period.

Advancing performance measurement in oncology.

The American healthcare system, including the cancer care system, is under pressure to improve patient outcomes and lower the cost of care. Government payers have articulated an interest in partnering with the private sector to create learning communities to measure quality and improve the value of healthcare. In 2006, the American Society of Clinical Oncology (ASCO) unveiled the Quality Oncology Practice Initiative (QOPI), which has become a key component of the measurement system to promote quality cancer care. QOPI is a physician-led, voluntary, practice-based, quality-improvement program, using performance measurement and benchmarking among oncology practices across the United States. Since its inception, ASCO's QOPI has grown steadily to include 973 practices as of November 2010. One key area that QOPI has addressed is end-of-life care. During the most recent data collection cycle in the fall of 2010, those practices completing multiple data collection cycles had better performance on care of pain compared with sites participating for the first time (62.61% vs 46.89%). Similarly, repeat QOPI participants demonstrated meaningfully better performance than their peers in the rate of documenting discussions of hospice and palliative care (62.42% vs 54.65%) and higher rates of hospice enrollment. QOPI demonstrates how a strong performance measurement program can lead to improved quality and value of care for patients.

Advancing performance measurement in oncology: quality oncology practice initiative participation and quality outcomes.

The American health care system, including the cancer care system, is under pressure to improve patient outcomes and lower the cost of care. Government payers have articulated an interest in partnering with the private sector to create learning communities to measure quality and improve the value of health care. In 2006, the American Society for Clinical Oncology (ASCO) unveiled the Quality Oncology Practice Initiative (QOPI), which has become a key component of the measurement system to promote quality cancer care. QOPI is a physician-led, voluntary, practice-based, quality-improvement program, using performance measurement and benchmarking among oncology practices across the United States. Since its inception, ASCO's QOPI has grown steadily to include 973 practices as of November 2010. One key area that QOPI has addressed is end-of-life care. During the most recent data collection cycle in the Fall of 2010, those practices completing multiple data collection cycles had better performance on care of pain compared with sites participating for the first time (62.61% v 46.89%). Similarly, repeat QOPI participants demonstrated meaningfully better performance than their peers in the rate of documenting discussions of hospice and palliative care (62.42% v 54.65%) and higher rates of hospice enrollment. QOPI demonstrates how a strong performance measurement program can lead to improved quality and value of care for patients.

Factors determining the persistence or recurrence of well-differentiated thyroid cancer treated by thyroidectomy and/or radioiodine in the Boston, Massachusetts area: A retrospective chart review.

OBJECTIVE: To assess predictors of well-differentiated thyroid cancer (WDTC) persistence/recurrence. DESIGN: This was a retrospective chart review of thyroid carcinoma patients seen 1979-2007 in a Boston, Massachusetts-area multispecialty group. Of 1,025 patients, 431 met eligibility criteria. Cox proportional hazards models were used to assess predictors (gender, age, ethnicity, tumor size, surgical histology) of WDTC persistence/recurrence (elevated thyroglobulin levels with negative thyroglobulin-antibodies; or positive imaging). Local extension of disease and lymph node involvement could not be assessed. RESULTS: Mean age at initial surgery (n = 431, 74% women, 79% Caucasian) was 45.8 ± 13.5(SD) years. Mean tumor (papillary, 91%; follicular, 5%; Hurthle cell, 2%; ≥1 type, 2%) size was 2.5 ± 1.6(SD) cm. Most tumors were unifocal (57%) and ≥1 cm (89%). Over 2,600 person-years of follow-up, persistence/recurrence occurred in 52 patients (12%) 4.3 years (median; range 0.2-23.2 years) after surgery. Gender, ethnicity, tumor size, multifocality, and histology were not predictive of persistence/recurrence, while older age was predictive in some models. CONCLUSIONS: In WDTC patients treated by total and near total thyroidectomy and radioiodine and analyzed without consideration of local, locoregional, and distant extent of disease, neither size of tumor nor male gender contribute to disease persistence/recurrence. Age at diagnosis seems to have some positive prognostic value even if only patients older than 21 years at diagnosis are considered. Due to the rare occurrence of follicular (also oxyphilic) histotype, this conclusion refers mainly to patients with papillary thyroid cancer.

Clinician attitudes towards prescribing and implications for interventions in a multi-specialty group practice.

BACKGROUND: Prescribing decisions are subject to a myriad of external forces, including patient requests for advertised medications. Although numerous factors influence prescribing, resources to support unbiased evidence-based prescribing are not widely available. METHODS: To guide future interventions, we surveyed clinicians about influences on prescribing, awareness of pharmaceutical costs and attitudes towards computerized decision support. A 21-item survey was sent to 604 prescribing clinicians in a large multi-specialty group practice that employs a robust electronic medical record. RESULTS: Surveys were returned from 405 clinicians (67%). Most respondents (87%) felt that direct-to-consumer (DTC) advertising prompts patients to request inappropriate medications, and more than one in five clinicians (22%) reported difficulty declining patients' requests for advertised medications. Providers with more clinical sessions per week reported greater difficulty. Although 93% of clinicians felt they have access to the information needed to guide prescribing, only about half (54%) reported they are aware of how much patients pay for prescription medications. Clinicians' awareness of medication costs varied considerably by specialty, with behavioural health clinicians being the most aware. The majority of providers (79%) stated that computerized prescribing alerts are a clinically useful source of information. CONCLUSIONS: Although the majority of clinicians reported that DTC advertising leads many patients to request medications that are inappropriate for their condition, a sizable proportion of clinicians reported difficulty declining these requests, and many are unaware of medication costs. Interventions to support prescribing decisions should provide the busiest clinicians with up-to-date, specialty-specific evidence and cost information.