Having a good theoretical understanding of business intelligence is all well and good, but what really matters is how you apply it.
At a Glance
- To use business intelligence effectively, healthcare organizations should start small, align organizationally, and leverage success.
- Organizations should determine which measures they need and how to present them.
- Organizations should reinvest savings to continually improve.
If you are posed with an important strategic decision, how would you start your response?
a. I feel
b. I think
c. I know
Our current economic reality is contributing to an era where decisions are scrutinized and indecisions are causing missed opportunities. More than ever, it is time to know the answers to today's questions and be proactive in addressing tomorrow's challenges. It is now necessary to defend your positions with solid empirical evidence and timely measurements. With that necessity is an increasing need to manage your way through the minefield of data and organizational noise. Through all of this is an opportunity to lead through better sharing of important information and data across the organization.
Now is the time to go from zero to BI.
When you boil it down, business intelligence (BI) uses clear targets and factual data that are made transparent across the organization to achieve business goals and desired outcomes. Healthcare providers that embrace this management approach have seen dramatic improvements in clinical, operational, and financial performance. However, despite well-documented benefits and the best intentions, few BI programs get past the planning stage. The quest for perfection, more than anything else, is to blame. As in chess, theory is great; however, at some point, you have to put your pieces in play and execute.
To realize an effective system, organizations have to start somewhere. Among the myriad consultants, tools, and "best practices," the most effective approach is to start small, align organizationally, and leverage success.
To start small, you should take an iterative, phased approach; have a clear purpose; start with a manageable scale; and use feedback to retool the strategy.
Iterative, phased approach. Hospitals looking for real-world results are successfully employing a simple iterative process focused on quick implementations. Instead of investing years of IT, management, and consulting resources developing the perfect metrics and scorecard, this approach focuses on creating a usable BI solution today and then refining, improving, and evolving it over time based on experience. Start with measuring a few critical metrics at one facility, and once these have been integrated into daily decision making, roll these proven metrics out to other parts of your organization. Implementing the perfect enterprise scorecard on day one is an unrealistic goal, and a project that rarely gets off the ground.
Clear purpose. Everything starts with and flows from a clear purpose. At the top level, purpose means the organization's strategy; at the lower levels, purpose is defined by department goals and objectives. For example, a department goal to reduce supply costs per case by 15 percent provides a clear goal. If you do not know where you are going, you cannot create the appropriate measures that tell you when you have arrived. Therefore, focus with laser beam precision on those concepts and tools that make the most sense for your organization and develop a meaningful long- and short-term strategy for success.
Manageable scale. Once the strategy has been made clear, it is important to take a phased approach toward implementation that focuses on early wins and refines the appropriate metrics over time. In the early stages, less is more. A common pitfall for BI implementations is attempting to deliver all things to all people at once. Start with three to five meaningful measures and fully integrate them into the process and feedback loop of your hospital. Another way to manage scale is to start with a region, facility, or department and use the knowledge gained in the process when building the enterprisewide scorecard. With an iterative process, you can set your organization up for long-term success by building positive momentum through the delivery of important and meaningful solutions in the near term.
Feedback and retooling. Today's approach and measurements are sufficient only to answer today's performance questions. In a rapidly changing hospital environment, your program can take seemingly endless directions. The challenge is to align your organizational mindset around what is important for your organization and BI program today and be flexible to address future challenges and new variables that can affect your organization tomorrow. By starting small and with a clear purpose, your organization can purposefully manage scale and set itself up for long-term success. In building upon achievement, you are taking away one of the largest barriers to success: trying to do too much too soon and without proper support.
As with the chess example, the best laid BI plans often fail when it comes time to execute. A typical hospital organization has a multitude of disparate data sources, competing priorities, and limited resources. What is a decision maker eager to realize BI success to do? Once the organization's purpose is clear and the iterative approach is confirmed, the focus should turn to building the proper solutions. To be successful, you have to first match the available data to the strategy.
Hospital organizations often have disparate clinical, financial, and operational audiences requiring varying data for effective decision making. These multiple audiences have resulted in numerous reporting systems, disconnected data repositories, and limited distribution of information beyond the executive suite. It is common for healthcare executives, managers, and clinical decision makers to rely heavily upon IT and decision support analysts to manually aggregate data and program complex queries on an ad hoc basis. In most cases, results are delivered in spreadsheets, which are a great tool for many different tasks but are not well suited to managing the business performance of a large organization. Simply put, hospitals have a difficult time getting meaningful and relevant information to business and clinical decision makers in an actionable format. Typical questions that help focus attention and assist in matching the data to the strategy include:
- Which performance measures will get executives and managers fired or promoted?
- What data do they need to investigate these measures?
- What additional data are nice to have?
- How do you make sense of all the information and data collected?
- Are the users visual or numerical thinkers?
It is essential to piece together what is important, what information is necessary, and how to use the information once it has been collected.
Although this step sounds intuitive and relatively basic for experienced decision makers, it is often overlooked. Consequently, the long-term success is left in doubt. This level of preparation is equally important when entering the sometimes overwhelming world of healthcare organization data. Knowing what you want and how it is supposed to look once you get it is often half the battle (see sidebar below and Exhibit 1).
View Exhibit 1
How to Make Measures Actionable
With a usable data infrastructure and scalable implementation plan in place, the next (and arguably most important) steps are to determine the appropriate measures and serve up the results in an appropriate fashion. Fortunately, there is no need to "reinvent the wheel" when selecting best-practice metrics, so your deployment time can be in days rather than months. Various government agencies, clinical quality initiatives, benchmarking organizations, industry associations, and your peer hospitals have all published metrics that have set the standards for how hospitals measure performance.
The second sidebar below lists some measures based on industry research and the experiences of healthcare executives that can be used as a starting point. This list was formulated with a cross-discipline perspective (e.g., CEOs, CFOs, CIOs, and chief medical officers) with experience at both small hospitals and large systems.
Work with your team to determine which measures you should use. As mentioned previously, start with three to five and adjust based upon feedback. In all cases, room for customization for each hospital organization is necessary. This flexibility is critical because strategic goals vary from organization to organization. Furthermore, the measures have to be dynamic. Developing an organizational scorecard is an iterative process, and measures should flex to respond to changes in such areas as strategic goals, external forces, board requirements, and leadership changes.
Although the measures in the sidebar are relevant to the roles of management, additional steps are needed to make them actionable. The best measures for managing performance use fresh data updated as frequently as the organization can monitor. For example, in the current economic climate, CFOs will likely want to have daily visibility into payer reimbursements; whereas, they will want monthly or even quarterly insight into patient demographic shifts.
Equally important, performance measures need to spur action. These measures should be linked to strategy, goals, or objectives. The question to ask when evaluating a metric is, "What action will we take if this measurement shows underperformance (increasing denials, for example)?" If there is an answer (e.g., increase insurance verification staff), keep the measure. If there is no answer, remove the measure from the list.
How to Present the Information
To encourage meaningful action, measures should be put into context according to the scope of recipient's responsibilities. Common mechanisms are interactive scorecards, dashboards, and reports. Typically measures are presented by:
- Scorecards to align and focus the organization to strategic goals
- Dashboards to understand operational execution to tactical targets
- Reports to understand underlying performance
The most successful presentation of performance measures link the three mechanisms together to provide clarity on an individual's performance and its influence on organizational goals. As such, each measure should allow for:
- Comparison of actual against budget, target, prior period, forecast, and/or benchmarks
- Drill down to service lines, departments, or unit, as applicable
- Drill down to patient, clinician, payer, and transaction
The key to this entire process is to get the right information into the right hands at the right time. Scorecards, dashboards, and dynamic reports built on a consolidated source of data are the best mechanism to ensure that everyone in the hospital is focused on the same targets and looking at the latest numbers. These types of capabilities ensure there are no issues with multiple versions and manual maintenance as with spreadsheets. When outliers or areas of interest are identified, it is important to be able to drill down to the underlying detail to perform root-cause analysis. Scorecards, dashboards, and dynamic reports (see Exhibit 2) work together to :
- Proactively identify the existence of a problem
- Operationalize the plan to change behavior and fix the problem
- Provide instant feedback on current performance and trends
- Understand the magnitude and historical context
- Provide personalized key performance indicators
- Provide specific measures for a particular constituency
- Determine the root cause of the problem
- Click and refresh to follow intuitive drill paths
- Move at the speed of thought to create new views and insight
View Exhibit 2
Keep in mind the way in which the decision makers process information. This, more than most factors, will determine how you should present the information. Some decision makers prefer reams of tables with infinite layers of details. Some will prefer visual cues as to what is important in the least amount of time. An effective system will have a comprehensive array of scorecards, dashboards, and dynamic reports to cater to the needs of your constituents.
Growth Through Success
As advocated previously, starting small and employing an iterative approach to BI will create the best chances for success. But the successful leader does not stay small. Growth within your BI program will provide exponential benefits throughout your organization by expanding the breadth and depth of the available analysis.
BI should, at the very least, be a budget-neutral position. Each day, week, and month where a decision maker is able to take action on previously unavailable or unclear information is an
opportunity to reduce costs, increase cash, or improve productivity-all of which provide a tangible ROI. Reinvesting these savings and goodwill into new programs provides a self-funding entity that continually improves the organization.
Working bottom up from department and business offices, leaders can identify additional areas to expand the scope of the BI program. The speed at which an organization expands depends on many factors; however, the scope should follow as closely as possible to the long-term vision that was identified during the planning period with the flexibility to solve new issues.
It may be the expansion of measures from the original three to five measures, or it could move in the direction of adding new data sources to the mix. In either case, the new projects should rely on the same fundamentals that garnered the original success: Have a clear purpose, manageable scale, and constant feedback and retooling. The results will be a pathway for incremental and sustained performance improvements that will lead to best in class in all aspects of the organization.
Sean Kirby is director of product management, MedeFinance, Dallas (firstname.lastname@example.org).
Brian Robertson is COO, MedeFinance, Emeryville, Calif. (email@example.com).
Sidebar 1: Data Sources and Their Uses
Healthcare organization data come from many sources. This sidebar shows some sources of data that are typically available, and the potential benefit of using each source.
- Highlight aged accounts receivable
- Improve timeliness of billing and payment
- Track collector performance
- Improve clean claim submissions
- Sort bill holds by reason code
- Verify patient demographic information
Denials and Adjudication
- Improve denial overturn rates
- Improve denials as percentage of billed charges by payer
- Create an adjudication matrix
- Meet coding and billing compliance
- Automatically apply Recovery Audit Contractors, Office of Inspector General (OIG), and other Centers for Medicare & Medicaid Services (CMS) audit rules
- Create a claims database with audit work flow capabilities
Payroll, Time, and Attendance
- Perform staff productivity and salary expense analysis
- Track nursing hours per patient day
- Track staff versus agency, registered nurse versus licensed practical nurse, and overtime versus regular hours
- Analyze profitability at the service line, physician, payer, and patient level
- Perform physician benchmarking on cost categories and cost/case
- Track clinical resource and cost utilization at the charge level
- Perform financial statement performance analytics
- Monitor days cash on hand, net income trends, supply expense
- Drill down from financial statements into detailed journal entries
Operating Room (OR) Workflow
- Analyze OR efficiency and throughput analysis
- Monitor surgery delays, operating room cycle time, surgeon benchmarking
- Receive time-stamped data for all activities in the OR
Emergency Department (ED) Workflow
- Analyze ED throughput and operational efficiency
- Monitor ED wait times, triage efficiency, and arrival/discharge times of day
- Analyze Time-stamped data for all activities in the ED
- Track payer contract performance and payment variances
- Perform underpayment analysis and contract scenario modeling
- Focus on the analytics; do not use as a repricing engine
Physician Order Entry
- Provide clinical analytics and best-practice clinical protocols
- Analyze utilization of order sets at the physician level
- Pull together laboratory, pharmacy, charge detail, etc. from a clinical perspective
CMS Core Measures
- Measure clinical quality and safety metrics
- Utilize the same extract mandated by CMS
- Track beta blocker at arrival, smoking cessation advice, etc.
Medicare Provider Analysis and Review (MedPAR) Benchmarks
- Compare Medicare activity to national and state benchmarks
- Track average length of stay, diagnosis-related-group utilization, etc.
- View actual hospital data side-by-side with MedPAR statistics
Program for Evaluating Payment Patterns Electronic Report (PEPPER)
- Integrate OIG benchmarks into compliance analysis
- Track complications coding, one day stays, observation status
- Load data from Excel-based PEPPER into BI solutions
U.S. Census Bureau Demographic Data
- Understand demographic trends in surrounding market
- Conduct zip code analysis by age, income, home value, employment, etc.
- Leverage BI solution mapping capabilities
U.S. Census Bureau Financial Data
- Segment all self-pay accounts according to risk
- Identify low-income patients who qualify for charity or Medicaid
- Optimize self-pay accounts on front end and in collections
State Hospital Association
- Create more relevant regional peer groups
- Access data publicly available from most state hospital associations
- Integrate these benchmarks into BI solutions for variance analysis
U.S. Postal Service Address Data
- Resolve address and demographic errors (e.g., social security number)
- Significantly reduce return mail and bad debt
- Identify variances at point of service and again before billing
Sidebar 2: Example of Performance Measures for Healthcare Organizations
These measures can be used to analyze performance by service line, department, unit, inpatient, or outpatient data against budget, prior year, or forecast target.
- Market share percentage
- Average daily census (ADC)
- Patient volume
- Uninsured volume
Case and Efficiency
- Patient satisfaction (HCAHPS)
- Physician satisfaction
- Average time emergency department (ED) door to bed
- Average time ED treatment to release
- Divert hours for ED
- Average length of stay (ALOS)
- Staff turnover rate
- Salary expense per adjusted patient day
- FTE per adjusted patient day
- FTE per ADC
- Supply expense per adjusted patient day
- Joint Commission
- AMI core measure
- Heart failure core measure
- Pneumonia core measure
- Pregnancy and related core measure
- Mortality rate
- Case mix index
- Outlier ALOS
- Complication rate
- Readmit rate
- Hospital stays in last six months of life
- Medication error rate
- Patient falls
- Pressure ulcer rate
- Hospital-acquired infection core measure
- Never events
- Returns to operating room
- Contribution margin
- Profit margin
- Bad debt as a percentage of net revenue
- Charity care as a percentage of net revenue
- Self-pay receivable as a percentage of accounts receivable
- Days cash on hand
- Cash on hand
- Average daily cash collections
- Current ratio
- Accounts receivable (A/R) days
- Percentage of A/R > 90 days
- Discharged not final billed
- Denials rate
- Staff turnover rate
- Salary expense per adjusted patient day
- FTE per adjusted patient day
- FTE per ADC
- Supply expense per adjusted patient day
Publication Date: Thursday, January 01, 2009