Business Intelligence Network - The Vision for BI and Beyond

Using Business Intelligence to Translate Healthcare Quality Measures into Value

Published: October 27, 2005

Although cost is certainly a significant healthcare issue, providers must also consider the value of quality measures.

Cost, cost, cost! It seems that the only thing we hear about in healthcare today is cost. In fact, healthcare costs are rising far faster than the economy as a whole. Everyone was stunned when General Motors reported that they pay more for their workers’ healthcare than for the steel used in the production of their cars. We must do something to stem the tide of this growing beast.

But there is another side to the healthcare equation, which is the value of the healthcare we receive. Contrary to conventional wisdom, quality does not necessarily increase costs for an organization or individual. When designed wisely, a quality program actually pays off. We have seen this in numerous other industries over the past twenty years. During this time, the Japanese taught us a lesson by delivering higher quality goods at lower costs. This basically turned the “quality-costs” argument on its head. Since then, evidence showing that “quality-pays” has materialized in specific industries within the manufacturing, transportation, communication and services sectors.

It is time to apply these lessons from other industries to healthcare. This will require massive changes in organizational practices, policies and structures. Using these lessons will also change the ways that healthcare is bought, sold, performed and evaluated. Even the definition of the product of healthcare will be transformed in many cases.

In the past, healthcare providers focused on delivering “hard” items. Examples of this include medical directives to patients based on specialized knowledge, prescriptions, devices and other tangibles. As our society becomes more information-centric and focused on monitoring individual health, healthcare products will shift to “softer” items, which include advice, evidence, information and wisdom. Providers are grappling with how to make money as well. An example of this is how physicians are spending more time e-mailing patients and discussing treatments over the phone.

Improving the quality of healthcare will require one of the tools that were instrumental in each of the industry transformations mentioned above. This is information about processes and procedures, customers (i.e. patients), markets, products, operations and decisions.

Business intelligence can help organizations improve quality. Because of this, it can improve the care these organizations are delivering. Business intelligence can also help companies understand and justify the costs of that care. Doing this, however, requires knowledge of how quality is measured in healthcare, as well as how these measures can be used to generate benefits for providers, payers, purchasers and patients.

Measuring Healthcare Quality

The Institute of Medicine (IOM) set the stage for measuring healthcare quality with its 2001 report, Crossing the Quality Chasm. This report provided a broad definition of quality, using what it referred to as Six Aims, or characteristics of quality:

• Patient-Centeredness—Ensuring that patient values, needs and preferences guide all clinical actions.
• Timeliness—Reducing waits and sometimes harmful delays for those who receive care and those who provide care.
• Equity—Providing care that does not vary in quality because of a patient’s demographics.
• Effectiveness—Providing care based on evidence, as well as avoiding over use or under use.
• Efficiency—Avoiding waste of equipment, supplies, ideas and energy.
• Patient Safety—Avoiding injuries to patients from care that is intended to help them.

Since this report came out, hundreds of public and private groups have drilled down on these six aims to develop specific measures of quality. The Agency for Healthcare Research and Quality (AHRQ) has collected 718 individual quality measures from 102 of these organizations. The Agency then sorted these measures into five domains:

• Outcome Measures—Measures clinical results, such as mortality, lacerations, number of patients in control for A1c, LDL and other clinical indicators, emergency department encounters, birth traumas, general comfort questionnaire score, physical functional status, ability to perform everyday activities, pain reduction, wound reclosure rates, tobacco cessation, reaction rates, etc.
• Access Measures—Measures utilization of healthcare facilities, such as hospital admissions, visits, examinations, amputation rate, homeless program entry, wait times, etc.
• Process Measures—Measures clinical actions intended to lead to positive health outcomes, such as prescriptions, treatments, tests, frequency of measurements, counseling and discussion, immunizations, mammograms, interventions, screenings, follow-up appointments, instructions given and education provided.
• Structure Measures—Measures clinical healthcare capabilities in place, such as  procedure volume, time to next appointment, competency assessments of clinical approach and methods, healthcare staff skill levels, etc.
• Patient Experience Measures—Measures reflecting the view of the patient, such as satisfaction survey scores, self-reported wellness perceptions, communication perceptions, quick care, responsive care, information clarity and quality, helpfulness of staff, problems getting care, inpatient experience, visit experience and overall perception of organization.

These measures provide various organizations with a wealth of potential standards to judge their own quality, as well as the quality of their partners along the healthcare continuum. Such organizations include providers, payers, purchasers, patients, public healthcare authorities and collaborative groups. It is almost certain that you and your organization will encounter several of these measures in:

• Quality reporting efforts
• Marketing programs
• Pay-for-performance programs
• Internal performance and quality goal-setting, etc.

Thus, measuring quality is very complex. Unless your provider organization can translate it into value, measuring quality can be very costly. Doing this requires the ability to use and manage detailed, historical data. Business intelligence comes in here.

Example Measure:  Blood Sugar Control

Blood sugar (hemoglobin A1c) is widely used as a clinical quality indicator for diabetics. For our purposes, it is a good example for several reasons:

• It is outcome-oriented (i.e. a result of providing care) instead of input-oriented (i.e. the actions and effort of providing care).
• As stated above, it is a widely used measure for critical disease conditions. This has become a national trend.  
• It can be a leading indicator for other measures used to judge the effectiveness of care. For instance, physician groups are measured using A1c control (the result), while health plans are measured using A1c testing (the process leading to the result). In this way, this one measure promotes collaboration between these two types of organizations.
• It has business and economic implications. Blood sugar is measured as a percent with 7.0% being defined as optimal, 7.0% to 9.0% being near-optimal, and greater than 9.0% being poor control. The American Diabetes Association estimates that for every percent reduction of blood sugar, the risk of an expensive hospitalization is reduced by 14-20%.

So how might provider organizations use this measure to improve the value of care for their patients, payers, purchasers and society as a whole?

Using Patient Intelligence for Business Value

Patient intelligence is a specific business intelligence application that organizes clinical, business and operational data for decision-making purposes by healthcare organizations. This data is used to support programs like disease management, outcomes management, clinical performance and process improvement, cost and waste reduction, quality accreditation and predictive analytics. It is also used to provide data for healthcare research.

The general benefits of developing patient intelligence capabilities fall into five categories:

• Clinical—Care that is organized, timely, proactive and based on real-world clinical evidence.
• Business—Better revenue prediction, efficiency improvement and cost control.
• Strategic—Evidence to support product and service-line decisions, location and access decisions, quality accreditation and marketing efforts, as well as structure and price insurance products.
• Operational—Greater efficiency, reduced waste, better staffing and scheduling decisions.
• Educational—Contribution to the larger base of healthcare evidence – clinical, operational and administrative.

Given this framework, how would a provider organization apply patient intelligence to improving its A1c control scores, and therefore translate an important quality measure into value?

• Clinical—Using A1c control to prioritize patients in your diabetic population organizes their care and reduces the chance of serious outcomes in later stages of the disease. Knowing which clinics, physicians and specific patients need help the most is critical.
• Business—Many patients are not in control simply because they have not been tested, or have not been tested often enough. This is lost revenue for a provider organization. In addition, many pay-for-performance programs compensate organizations based on this and other clinical outcome measures. This is also lost revenue.
• Strategic—A1c control is one form of evidence that supports decisions. These decisions regard whether or not your organization is set up to serve chronic care populations. Do you have the right people to serve these populations? Are you conveniently located near these populations?  Are you marketing your capabilities to treat diabetic patients as effectively as you should?
• Operational—Evidence-based prioritization leads to greater efficiency and reduced waste. Measures like A1c control help you make decisions about who to staff based on their skills, as well as what skills to develop in areas where your organization falls short.
• Educational—One provider organization I worked with used their prowess in A1c control to develop an educational program. This program was used for several purposes: to cross-train in-house, collaborate with health plans to drive other measures important to payers (i.e. blood sugar testing within the last year) and as a beacon for presentations at industry conferences. They are even developing a program to sell their educational services in this area, based on the A1c control evidence they have accumulated.

Next Steps

Don’t let cost be the only measure of your service. Doing this will drive both you and your organization nuts, and possibly out of business. It is essential that your programs and efforts focus on quality, the other half of the healthcare value equation. While healthcare quality measures are plentiful, implementing them are complex and data intensive. But it can be done. There is tremendous value in pursuing quality programs for clinical, business, research and marketing value. And this value will flow to patients, payers, providers and the country as a whole.

But you must start somewhere. Look at the quality measures you are currently using, and more importantly, the quality measures that are influencing others’ opinion of your organization and its value. Also consider the difficulties your organization has gathering, managing and using the data needed to actually understand and profit from these measures. This is where business intelligence can make a real difference.

Thanks for reading. I look forward to your comments.