Digital Dashboards Dig Deep
Digital dashboards and business intelligence come in various flavors. From integrated health systems to critical access hospitals, the term digital dashboard can be retrofitted to meet the needs of many different improvement initiatives. Regardless of the definition, the technology has made great strides in the past five years to be a success story in the IT world and assist CMIOs in tackling a diverse range of tasks from population management to quality reporting.
The dictate has been set: Reimbursement will soon be tied to the quality of performance. Thus, the adoption of technology equipped with business intelligence, such as a digital dashboard, is on the rise as healthcare providers look to tie performance to quality metrics and dive deeper into their facility's processes and operations. Even an August report by healthcare market research firm KLAS noted that 56 percent of those surveyed indicated they plan to move to a single enterprise business intelligence vendor.
In the report, 64 percent of survey respondents said their organizations plan to become an accountable care organization (ACO) in the next three years. "Overall, 60 percent of those planning to become an ACO feel their business intelligence system will be able to provide the necessary analytics," the report said.
From 2010 to 2011, the use of business intelligence increased by about 5 percent, respondents reported. Organizations, such as the Duke University Health System (DUHS) in Durham, N.C., have proven the merits of a business intelligence/digital dashboard approach. The intervention prompted by their initial safety analysis prevented approximately 157 potential cases of Clostridium difficile colitis, a bacteria that causes intestinal disease, according to Jeffrey M. Ferranti, MD, CMIO, vice president for clinical informatics, associate director at the Duke Center for Health Informatics, and colleagues. The group published its findings in the March 2010 Journal of the American Medical Informatics Association (JAMIA).
Using an online cohort generation tool, clinicians identify cohorts of interest, display census-corrected aggregate safety statistics and click on bars within graphs to get into encounter- or event-specific details. "These reports also allow us to dynamically aggregate and disaggregate information to actively identify systemic issues and intervene on the basis of timely, accurate and high-confidence data," the authors wrote.
"Based on available estimates of incremental hospital costs associated with C difficile colitis, this saved about $3,600 per case," says Ferranti. Total savings, using a conservative analysis, was $578,968, according to the article. "By actively leveraging data in support of care quality, you're able to show benefit to the [care of] patients, as well as financial benefits. Business intelligence and health analytics tools help you understand where there are opportunities for improvement in your organization," he says.
However, according to KLAS, there seems to be marginal use of predictive analytics as only 16 percent of the respondents reported using the tool, with 11 percent of the 16 percent responding they use the tool for labor forecasting, 11 percent for readmission predictions and 22 percent for financial impact modeling.
Despite the KLAS findings, facilities are harnessing the power of dashboards to drive predictive analytics. For example, Kay Stewart-Huey, cardiac service line administrator at Children's Healthcare of Atlanta, uses dashboards to measure trends and develop a strategic approach for driving performance in targeted areas based on the dashboard's analytics.
With more than 850 cardiac surgeries performed each year and about 1,300 cardiac catheterization lab cases, Stewart-Huey uses dashboards to monitor the Children's Sibley Heart Center's performance in the areas of staffing, outcomes, growth, finance and service. The dashboard platform, which pulls financial and clinical information from data registries, identifies strong and weak performances and initiates action from the team to improve in these areas. Using the dashboard to report referring physician satisfaction over time, the metrics revealed a need for improvement in communicating with referring physicians on the status and disposition of their patients. By initiating defined steps of communication for relaying timely information on all patients and measuring the communication step interactions, the team was able to improve communication with this particular group of physicians. After the new process was implemented, referring physician satisfaction increased by almost 10 percent from one year to the next.
"The dashboard gives us the opportunity to have a barometer on particular metrics that are vital to our operations," Stewart-Huey says.
Deaconess Health System, a 500-bed acute care facility in Evansville, Ind., is one example of a facility that has improved clinical reporting. Greg Hindahl, MD, CMIO at Deaconess, says that since implementing a dashboard platform more than two years ago, the facility has improved its blood clot prevention compliance in its ICUs from 79 percent last fall to more than 90 percent in July. "There is still significant fluctuation in the data from month to month so this is a work in progress, but the dashboard helps us easily see these fluctuations and gives us an idea of where to focus our attention for ongoing improvements," Hindahl says.
Currently at Deaconess, many ambulatory clinicians are using a dashboard platform, which measures items related to patient satisfaction and throughput. Clinicians are required to manually run a separate report to look at data related to diabetics, including hemoglobin A1C or whether vaccines were administered. Hindahl says that one of the things Deaconess is striving for is to make meaningful clinical and financial data available to clinicians so they can be used to improve patient care in a cost effective way. "Our ultimate goal is a full-blown data warehouse where all the information comes together and is normalized, so it is meaningful and useful for our entire care team," Hindahl says.
The shift from disparate systems to a single data warehouse will likely accelerate in the near future. For example, DUHS has built an internal enterprise data warehouse over the last five years. It contains information on approximately four million patients and includes the clinical revenue cycle, research and operational data. This is the same data warehouse that Ferranti used to maximize C difficile prevention gains, as cited in JAMIA.
DUHS is exploring other business intelligence opportunities, including development of the Duke Enterprise Data Unified Data Content Explorer (DEDUCE), a front end to the data warehouse that allows Duke faculty to create cohorts of patients for use in safety, quality and financial projects. Ferranti hopes this will help clinicians ask the database relevant questions in their specific areas of expertise.
Mount Sinai has ramped up its efforts in the last year through the use of a web-based dashboard to look into different domains including areas of productivity that department chairs can assess through a password-protected portal to gaze into department operations.
Through the portal, a department chair can view data such as length of stay on any faculty members' patient, and apply trending data to see the progression/regression over the last four quarters. Each physician has access to his own individual report; administrators are provided access to their respective aggregated departmental reports at the discretion of their department chairs. For example, the department of medicine's operations can be split from surgery into general surgery or vascular surgery reports.
"It has allowed us to create more accountability around quality metrics," Nash says. "This creates a better means of pinpointing quality metrics and creates a collective responsibility."
Included in this cohort is University of North Carolina (UNC) Healthcare in Chapel Hill. The UNC Health System began piloting dashboards in December 2010 geared towards physicians and physician leaders and allowing them to view how they are performing in a number of key indicators, including case volume, length of stay, mortality, readmission rates and cost per case. With 35 inpatient services across the UNC hospitals, five areas were selected to pilot the dashboard with plans to begin the remaining roll out process in October of this year, says Glen Spivak, vice president, business development, operational efficiency at UNC Hospitals.
Business intelligence tools are fertile ground for efforts such as value-based purchasing and ACO activities. If a clinician is going to be paid for quality, then he or she needs to understand the data. Business intelligence tools, including digital dashboards, can be a sound investment on a CMIO's wish list.
The dictate has been set: Reimbursement will soon be tied to the quality of performance. Thus, the adoption of technology equipped with business intelligence, such as a digital dashboard, is on the rise as healthcare providers look to tie performance to quality metrics and dive deeper into their facility's processes and operations. Even an August report by healthcare market research firm KLAS noted that 56 percent of those surveyed indicated they plan to move to a single enterprise business intelligence vendor.
In the report, 64 percent of survey respondents said their organizations plan to become an accountable care organization (ACO) in the next three years. "Overall, 60 percent of those planning to become an ACO feel their business intelligence system will be able to provide the necessary analytics," the report said.
From 2010 to 2011, the use of business intelligence increased by about 5 percent, respondents reported. Organizations, such as the Duke University Health System (DUHS) in Durham, N.C., have proven the merits of a business intelligence/digital dashboard approach. The intervention prompted by their initial safety analysis prevented approximately 157 potential cases of Clostridium difficile colitis, a bacteria that causes intestinal disease, according to Jeffrey M. Ferranti, MD, CMIO, vice president for clinical informatics, associate director at the Duke Center for Health Informatics, and colleagues. The group published its findings in the March 2010 Journal of the American Medical Informatics Association (JAMIA).
Using an online cohort generation tool, clinicians identify cohorts of interest, display census-corrected aggregate safety statistics and click on bars within graphs to get into encounter- or event-specific details. "These reports also allow us to dynamically aggregate and disaggregate information to actively identify systemic issues and intervene on the basis of timely, accurate and high-confidence data," the authors wrote.
"Based on available estimates of incremental hospital costs associated with C difficile colitis, this saved about $3,600 per case," says Ferranti. Total savings, using a conservative analysis, was $578,968, according to the article. "By actively leveraging data in support of care quality, you're able to show benefit to the [care of] patients, as well as financial benefits. Business intelligence and health analytics tools help you understand where there are opportunities for improvement in your organization," he says.
However, according to KLAS, there seems to be marginal use of predictive analytics as only 16 percent of the respondents reported using the tool, with 11 percent of the 16 percent responding they use the tool for labor forecasting, 11 percent for readmission predictions and 22 percent for financial impact modeling.
Despite the KLAS findings, facilities are harnessing the power of dashboards to drive predictive analytics. For example, Kay Stewart-Huey, cardiac service line administrator at Children's Healthcare of Atlanta, uses dashboards to measure trends and develop a strategic approach for driving performance in targeted areas based on the dashboard's analytics.
With more than 850 cardiac surgeries performed each year and about 1,300 cardiac catheterization lab cases, Stewart-Huey uses dashboards to monitor the Children's Sibley Heart Center's performance in the areas of staffing, outcomes, growth, finance and service. The dashboard platform, which pulls financial and clinical information from data registries, identifies strong and weak performances and initiates action from the team to improve in these areas. Using the dashboard to report referring physician satisfaction over time, the metrics revealed a need for improvement in communicating with referring physicians on the status and disposition of their patients. By initiating defined steps of communication for relaying timely information on all patients and measuring the communication step interactions, the team was able to improve communication with this particular group of physicians. After the new process was implemented, referring physician satisfaction increased by almost 10 percent from one year to the next.
"The dashboard gives us the opportunity to have a barometer on particular metrics that are vital to our operations," Stewart-Huey says.
Drill down on data
The objective of business intelligence efforts, such as digital dashboards, is to provide decision makers with information so they can drill deeper into questions and curiosities with a touch of a button, Stewart-Huey says.Deaconess Health System, a 500-bed acute care facility in Evansville, Ind., is one example of a facility that has improved clinical reporting. Greg Hindahl, MD, CMIO at Deaconess, says that since implementing a dashboard platform more than two years ago, the facility has improved its blood clot prevention compliance in its ICUs from 79 percent last fall to more than 90 percent in July. "There is still significant fluctuation in the data from month to month so this is a work in progress, but the dashboard helps us easily see these fluctuations and gives us an idea of where to focus our attention for ongoing improvements," Hindahl says.
Currently at Deaconess, many ambulatory clinicians are using a dashboard platform, which measures items related to patient satisfaction and throughput. Clinicians are required to manually run a separate report to look at data related to diabetics, including hemoglobin A1C or whether vaccines were administered. Hindahl says that one of the things Deaconess is striving for is to make meaningful clinical and financial data available to clinicians so they can be used to improve patient care in a cost effective way. "Our ultimate goal is a full-blown data warehouse where all the information comes together and is normalized, so it is meaningful and useful for our entire care team," Hindahl says.
The shift from disparate systems to a single data warehouse will likely accelerate in the near future. For example, DUHS has built an internal enterprise data warehouse over the last five years. It contains information on approximately four million patients and includes the clinical revenue cycle, research and operational data. This is the same data warehouse that Ferranti used to maximize C difficile prevention gains, as cited in JAMIA.
DUHS is exploring other business intelligence opportunities, including development of the Duke Enterprise Data Unified Data Content Explorer (DEDUCE), a front end to the data warehouse that allows Duke faculty to create cohorts of patients for use in safety, quality and financial projects. Ferranti hopes this will help clinicians ask the database relevant questions in their specific areas of expertise.
Cleared for take off
"Presenting the data is not the same as improving care," notes Ira S. Nash, MD, CMO, senior vice president for medical affairs at Mount Sinai Hospital. The 1,171-bed provider in New York City with 80,000 emergency room visits and 450,000 outpatient visits annually is part of a growing group of organizations that have started to roll out more enhanced dashboard efforts.Mount Sinai has ramped up its efforts in the last year through the use of a web-based dashboard to look into different domains including areas of productivity that department chairs can assess through a password-protected portal to gaze into department operations.
Through the portal, a department chair can view data such as length of stay on any faculty members' patient, and apply trending data to see the progression/regression over the last four quarters. Each physician has access to his own individual report; administrators are provided access to their respective aggregated departmental reports at the discretion of their department chairs. For example, the department of medicine's operations can be split from surgery into general surgery or vascular surgery reports.
"It has allowed us to create more accountability around quality metrics," Nash says. "This creates a better means of pinpointing quality metrics and creates a collective responsibility."
Included in this cohort is University of North Carolina (UNC) Healthcare in Chapel Hill. The UNC Health System began piloting dashboards in December 2010 geared towards physicians and physician leaders and allowing them to view how they are performing in a number of key indicators, including case volume, length of stay, mortality, readmission rates and cost per case. With 35 inpatient services across the UNC hospitals, five areas were selected to pilot the dashboard with plans to begin the remaining roll out process in October of this year, says Glen Spivak, vice president, business development, operational efficiency at UNC Hospitals.
Dashboards and benchmarks
The potential use for digital dashboards also applies to benchmarking, especially as it relates to population management initiatives, a key category in ACOs. Over time, a rich dataset could help administrators and clinicians understand their practices' trends, as well as their populations.Business intelligence tools are fertile ground for efforts such as value-based purchasing and ACO activities. If a clinician is going to be paid for quality, then he or she needs to understand the data. Business intelligence tools, including digital dashboards, can be a sound investment on a CMIO's wish list.