IF A KEY question benefit and risk managers and insurance executives ask is, "How can we provide better quality health care while keeping our bottom line looking good?," the answer is not, "Grab all the data you can get."

The trend to collect more data is good. However, data collection for its own sake won't answer quality health care and bottom-line needs; that data must be organized and actionable. So what is the answer?

The answer is a tool that provides a perspective on data: a tool that offers a view from 60,000 feet, as well as a view from the detailed, micro level. Managers and executives need this to reveal the provider and member health care behavior already resident in claims transaction data.

The health care data warehouse, complete with integrated analysis components that function within a business context, is emerging as the tool of choice for many of those executives and managers. A data warehouse moves computing power and data analysis capabilities to those who can identify fundamental trends affecting cost and quality. The ability to take claims, member and provider data and develop forecasts on which to base health plan offerings is arising as a central requirement for running day-to-day business operations and planning wisely for the future.

Transforming data into knowledge

Before researching data for the knowledge needed to create cost-effective, quality-focused health benefit plans that address employees' specific needs, companies must make sure this data is complete and accessible.

Whether paying claims or contracting with another organization to pay claims, companies need to make certain that detailed information is kept. Information that paints a complete picture of the health plan's population will provide better assistance in designing a plan that serves the members' needs efficiently.

If the contracted organization subcontracts for pharmacy or other benefits claims processing, whenever possible, organizations should keep their data collection parameters parallel, such as consistency among IDs.

If data collection methods don't include these parameters now, the company should definitely be working in that direction for the future. Until that time, links can be established to map data to consistent IDs.

Because much health care data comes from different sources-a company's own database and those of contracted and subcontracted processing organizations-and may run on different operating systems, not to mention different hardware, it is of primary importance that collected data can be assembled as a whole to get a complete picture of the plan population.

Again, this is where a data warehouse is an essential tool.

Using a sophisticated data integration process, data from disparate sources can be married within the warehouse, providing an accessible, easily updated collection of data available for analysis in a non-programming Windows environment.

Along with providing consistency, data placed within the warehouse is scrubbed/cleaned and organized into subsets, such as episodes of care, pharmacy, health risk indicators, readmissions and inpatient stays, data that will help define how services are delivered and how they are paid.

Once data is mapped within a data warehouse, the company has a fundamental tool for making intelligent queries and fine-tuning plans, truly turning data into usable information.

With data complete, organized and accessible, managers and executives can begin the search for knowledge by following the dollars.

Where are plan members spending their money, and who or what are they spending it on?

Look at the plan population for areas of likely high expense. Review the health plan population using parameters such as age, sex and location, and view providers on a case mix adjusted basis.

Then look over their utilization patterns.

Do women 19 to 32 comprise a large part of plan membership? If so, a review of maternity benefits would be beneficial. Are claims relating to hypertension pouring in? Perhaps some education for plan members might be a wise idea. Are many emergency room visits logged in a particular neighborhood? The company might want to consider contracting with an additional provider or network in that region.

For a more structured search, a data warehouse and related analytic tools offer views or templates of data, which include relevant statistics, in order to simplify the analysis.

These views, which should allow users to customize search parameters, can steer management in creating ad hoc reports, enabling managers to analyze information in a way that's useful to reaching a company's own conclusions on health care delivery.

For example, an ad hoc report of interest to plan analysis might cover all professional services denied within the plan year, including what services were denied, the reasons for denial, whether the claims were in or out of network and measurements such as submitted amounts.

HEDIS standards provide a yardstick

Industry standards also give shape to queries. Many organizations use the Health Plan Employer Data and Information Set as a standard for measuring the quality of health care plan performance.

This measurement, developed by The National Committee for Quality Assurance, acts as a report card for health care delivery organizations, providing a consistent standard against which to evaluate quality.

A health care data warehouse that uses a company's administrative data to produce HEDIS measures, including quality indicators, utilization review and membership attributes, would provide an invaluable tool to any organization that wants to perform trend analysis on the performance of the health plan and target areas for improvement or to monitor quality and access to care.

By applying HEDIS standards to health care plan performance, any company can take an important step forward in incorporating measures that respond to employees' needs for a plan that offers value and accountability.

Provider profiles

If a company contracts directly for services with a provider network, those providers obviously play a central part in offering a quality and cost-effective employee health plan.

Who are the physicians? What is the referral rate for each?

Are some physicians treating an above-average number of sicker employees?

How does this affect their performance comparison?

Sound medical management of any plan requires an analytic component for viewing treatment and health care utilization, allowing companies to:

Profile providers using an objective performance measurement.

Compare treatment patterns of plan physicians.

Assess disease-based pharmaceutical use.

Understand physician referral patterns.

Review targeted services, including emergency room and mental health care.

Identify and compare plan hospitals in the context of services provided, such as readmission rates and length of stay.

Claims data alone cannot provide a comprehensive picture of medical services. This is where a data warehouse with analysis capabilities affords the organization the ability to make the best use of all its data.

Membership, provider and capitation/fee-for-service information is stored in different databases. A data warehouse will integrate these data sources into a single place and format so that data originally from one source can expand the value of data originally from another source.

Differences among patients-such as diagnosis, age, complicating conditions and major surgeries-influence treatment, cost and care utilization. In reaching an objective understanding of a provider's practice, these differences should be factored into an overall view.

Industry standard methods for classifying patient illness are episode treatment groups, or ETGs, and diagnosis-related groups, or DRGs.

Both methods offer clinical and statistical adjustments based on factors that influence treatment. Data warehouse analysis components incorporating these methods can assist management in making decisions concerning potential quality or efficiency issues.

Understanding financial factors

As a direct contractor of services, providing a quality employee health plan takes money, but with a data warehouse and analytic resources, there are ways to monitor and control escalation of how much quality care costs.

To manage the financial side of a plan, expense data from claims submission and capitation payments are married with revenue data within the data warehouse, so that analysis will allow managers to:

Develop effective per-member-per-month revenue targets and capitation rates.

Evaluate the effectiveness of the program in meeting health care demands.

Track premiums vs. expenditures, and evaluate critical indicators of plan performance.

Monitor contracts, including stop loss, and withhold account balances.

Though data is the basis for creating a plan that provides quality health care at a positive bottom-line rate, it is not the solution.

The complete solution requires that data be organized and actionable. A health care data warehouse, complete with easy to use, non-programming analysis components, can be the tool companies need to develop and maintain the kind of plan that will satisfy the needs of employees and the employer.

Frank Mohr is director of health information products for Resource Information Management Systems Inc. in Naperville, Ill.