Category Archives: Administration

Healthcare entities must account for hurricane-related disruptions

Hurricanes can have a significant operational and financial effect on healthcare providers, insurers, and payers. Organizations that deliver or finance healthcare services in impacted areas must consider the various outcomes resulting from any disruptions. In this article, Milliman’s Lynn Dong, Scott Jones, and Michael Polakowski highlight a list of short-term and long-term effects for organizations to evaluate.

Tips for ensuring quality encounter data submission in Medicaid managed care programs

Naugle-AndrewOne area of data management within the healthcare industry that is getting new emphasis and interest from regulators is a focus on encounter data. Today, the majority of Americans receiving healthcare services funded through the Medicaid program are enrolled in some form of managed care. Under this scheme, states contract with managed care organizations (MCOs) that take on responsibility for providing Medicaid services in exchange for a fixed capitation payment. This approach is in contrast to the traditional fee-for-service (FFS) program where providers submit claims directly to the state for payment.

It is widely recognized that the FFS approach to provider payment creates perverse incentives for delivery of unnecessary services and uncoordinated care. Medicaid managed care gets away from this by providing a fixed capitation amount to the Medicaid MCOs, giving them incentives to more effectively manage care. The MCOs often pass along those same incentives by paying certain providers using capitation as well.

Notwithstanding the undesirable incentives that FFS creates, one area where FFS excels is the collection of timely and complete data about the services rendered to each patient and information about the price of those services. Because FFS claims are essentially invoices, this approach offers strong incentives for providers to submit claims in a timely manner (for prompt payment) and to ensure those claims are a complete reflection of services rendered (for complete payment).

Under capitation the payment mechanism is decoupled from the data collection process. In lieu of claims MCOs must collect encounter data from their providers and then submit that data to the states. Unfortunately, this takes away the direct financial incentives providers have for timely and complete data submission. As a result many states have struggled to collect credible information about the services delivered under Medicaid managed care. Such data is essential for important activities such as rate setting and program management.

Viewing submission of encounter data as an MCO function, many states have implemented strict contractual requirements coupled with tough performance guarantees and financial penalties to motivate MCOs to improve the quality and timeliness of the encounter data they submit. Failure to perform can have significant consequences for MCOs, including financial penalties, corrective action plans, bad press, and even contract termination. Based on our experience working with states and MCOs to help improve encounter data quality, we have identified a few things that can help improve the results:

1. Evangelize the importance of encounter data among providers. Once decoupled from the payment process it can be hard to convince providers of the importance of collecting and reporting encounter data. Ensuring that these constituents understand the value of encounter data and are submitting complete and timely information is key for MCOs to meet their contractual obligations to the state. Regardless of how strong the processes are within the MCO to ensure complete and timely submission, if the source information coming from providers is incomplete what goes to the state will also be incomplete. In addition to including encounter data requirements in their provider contracts, MCOs should include encounter data as a topic in their provider communication/education plans, and evangelize its importance whenever they can.

2. Develop clear submission requirements, definitions, and data specifications. In any situation where data is being submitted to a third party, lack of agreement and understanding of the actual submission requirements, definitions, and data specifications is a setup for downstream conflicts. Among states the requirements for submission often vary and sometimes change during the term of a contract. States and MCOs should engage in a collaborative process to ensure that all parties are working from the same guidance and interpretation of the submission requirements.

3. Establish an interdisciplinary team. Many MCOs view encounter data submission as one department’s responsibility—typically finance, operations, or information systems. In reality, it takes skilled and knowledgeable resources from throughout the organization to drive a high-quality process and result. MCOs should establish an interdisciplinary team that brings together the experts and makes everyone involved accountable for the outcome. That team may be led by someone from one of those primary departments but requires support from others.

4. Implement automated data validation and reconciliation processes. Submitting encounter data typically involves collecting data from multiple sources, transforming it into a new format, and then submitting it to the state. There are many opportunities for errors along the way. To provide an early warning of potential errors and to facilitate root cause assessment when errors are identified, MCOs should implement data validation and reconciliation processes that run parallel with the encounter submission preparation process. Often states will have internal processes they run on the data when they receive it; mirroring these processes can proactively reduce error rates. In addition, where possible, it makes sense to reconcile financial information against financial systems such as the general ledger, not just what’s in the claim system.

5. Track and resolve errors promptly and completely. When submitting encounter data some errors are unavoidable and thus both MCOs and states must have resources and processes in place to support timely and complete resolution of those errors. Errors can be caused by changes in the way data is collected within the MCO, changes in the state’s internal validation processes, or simply anomalies in the data. Regardless of the reason, each error should be tracked and resolved to completion. Root causes should also be identified to allow for error prevention.

As states have made the transition from FFS to managed Medicaid, the quality of utilization and cost data they receive has eroded. With the majority of Medicaid beneficiaries today receiving healthcare services through managed care plans it is more difficult for states to perform effective oversight of these programs. Multiple encounter data improvement initiatives have emerged as this has become an area of focus for both states and the federal government. MCOs should expect this scrutiny to continue, but should also recognize that through a systematic approach to managing and reconciling data and a collaborative posture with their state partners many of these challenges can be overcome.

This article first appeared at Milliman MedInsight.

Milliman MedInsight announces big data initiative

Milliman has announced the formal launch of a big data initiative. The MedInsight Data Confidence Model is a data management methodology that helps clients leverage concepts of big data for business performance improvement and transparency.

MedInsight is Milliman’s popular healthcare analytic platform used by over 200 health plans, employers, at-risk providers/accountable care organizations (ACOs), state governments, community health coalitions, and third-party administrators. Consistently recognized for its superior data integration and warehousing capabilities, MedInsight has continually developed and perfected its Data Confidence Model since 1997.

The MedInsight Data Confidence Model is applied as a client data warehouse is being created, and it is then leveraged on an ongoing basis for client data management. It includes both automated and manual processes to identify anomalies or irregularities in client data sets. Milliman aggregates data in innovative ways and compares the results against ranges of quality norms Milliman has developed over 15 years. Much like a lab test, each data metric in client data sets has a normal range. Results outside those norms often indicate a need for further investigation and/or remediation. The Data Confidence Model helps with common big data issues such as:

• Inconsistencies in how data was formatted and entered
• Relational integrity problems between components of the data
• Data volume consistency issues
• Duplication of data
• Integration and crosswalking of data from multiple data source providers
• Reasonableness of the metrics created by the system

Successful management of healthcare cost and quality for a single organization requires transparency into cost and quality data. In order to be effective, transparency requires integration of large volumes of high-quality and timely data—key components of the big data movement. The complexity of big data requires advanced data integration and quality assurance techniques in order for the underlying information to be reliable.

“MedInsight’s business has grown dramatically over the past four years, and we now maintain and process more than 15 billion healthcare records for 126 million healthcare members,” said Kent Sacia, Milliman Principal and MedInsight’s founder. “Supporting one of the largest healthcare claim data sets in the U.S. requires sophisticated data management techniques. Our MedInsight Data Confidence Model gives clients the confidence they need to rely on the data and information within their MedInsight data warehouses,” added Sacia.

The MedInsight Data Confidence Model is a critical component to help tackle the most difficult challenge of healthcare analytics—giving companies high confidence in making decisions by ensuring that the data that goes into the system is accurate.

The importance of administrative cost benchmarking

In the late 1990s, online travel agencies revolutionized the airline industry by publishing fares and allowing consumers to search for and purchase tickets. No longer would consumers have to rely on an agent to filter and present options; travelers could search across all vendors and use their own criteria to evaluate their options and purchase a ticket. The individual and small group health insurance markets are poised for the same sort of dramatic change, driven by the now familiar concept of the online marketplace, known in the health insurance industry as the exchange.

Although the operation of a health insurance exchange is quite different from that of an online travel agency, these distribution channels are similar in their impact on price transparency. Under the old travel agent model, consumers would first search for tickets based on convenience factors (e.g., travel dates and times, routes, etc.) and then use price to differentiate among a few options. Likewise, in the individual and small group health insurance markets, price is often presented after the purchaser has already narrowed the options to a few that meet non-price criteria. In both of these situations, price is applied as a deciding factor after the consumer has already narrowed the universe of choices to a subset of similarly appealing options; and the consumer lacks visibility to the prices of choices that were eliminated in that process. Online markets, on the other hand, allow consumers to see the prices of all or most options at the same time, making price a primary determining factor when making a purchase decision. This new presentation format, which allows consumers to choose one product over another based on a small dollar price difference, discourages significant price variation among competitors for similar products.

For most health insurance products, price is comprised of three primary components: benefit expense, administrative expense, and risk margin. Although benefit expense makes up the lion’s share of the premium or price, administrative cost differentials among health insurers can also materially contribute to premium differences. These differences will become more pronounced and may affect consumer purchasing decisions as the benefit expense component of premium is constrained by the medical loss ratio (MLR) requirements of the Patient Protection and Affordable Care Act (PPACA). These rules effectively create a benefit expense floor, requiring that health insurers in the individual and small group markets spend no less than 80% of premium on benefits (85% in the large group market), or pay a rebate to policyholders. It is likely that MLRs for individual and small group products will eventually settle around the 80% level or higher. In this new world, the importance of managing administrative cost will increase as price competition puts pressure on overall premiums and the MLR rules force administrative cost and risk margin into a fixed share of the premium dollar.

Benchmarking is one of the most effective tools available to help health insurers manage administrative expense. For insurers working to achieve MLR targets through administrative cost reduction, a benchmarking assessment can offer a function-by-function comparison of administrative expenses and staffing levels versus competitors and peers. Such an analysis can help organizations figure out where to target their cost reduction initiatives or determine what cost level is appropriate for a given department, cost center, or function.

For insurers that have already achieved the MLR targets, administrative benchmarks combined with a dashboard view can allow for monitoring of administrative expense variation throughout the year. Optimizing administrative cost is not something that can be achieved overnight; it takes time to plan and implement cost management initiatives, and months or years before the benefits accrue to the bottom line. Thus a dashboard coupled with benchmarks can provide management the tools they need to effectively manage their price competitiveness in this new distribution paradigm.

This article first appeared at Milliman MedInsight.

Health plan administrative costs: The 67% rule

Milliman has studied and published health plan administrative cost benchmarks for over 15 years. Our research has included direct observations of over 100 health plans supplemented with the review of many more statutory reports. Of the many trends and changes that we have observed during this time, one general statistical observation has stood out: health plan administrative costs generally increase at a rate of about 67% of the rate of increase in medical cost inflation. This equates to an annual rate of increase of 6% for the period 2006-2011.

Conceptually, administrative costs, over a period of time, should be affected by:

  • Increased efficiency through the use of electronic records and technology
  • General salary and cost of goods inflation
  • Regulation (positively or negatively)
  • Increases or decreases in administrative workload, such as claims, customer services calls, etc.

 

None of the above appear to be all that closely related to medical inflation. Salary and general inflation should be the primary driver of healthcare administrative cost inflation. But with an annual salary and goods inflation level of around 2%, we must look to some other cause of the 6% annual increase. Certainly, as we consumers collectively continue to expect and demand more medical services and prescription drugs, we can end up creating more administrative burden. But shouldn’t any increase in administrative transactions have been more than negated by the proliferation of advanced technologies and electronic transactions?

This article first appeared at Milliman MedInsight.

Optimizing administrative expenses

Administrative costs are often mentioned as a source of waste in healthcare financing.

We asked Andrew Naugle to tell us what is driving spending and where savings may be found.

Q: How much do health plans spend on administration and can they reduce that cost?

Andrew Naugle: If a health plan gets a dollar in the door, it spends about 85 cents of it on benefits, about 12 cents on administration, and it gets to keep about 3 cents as profit or surplus. Those numbers come from a study I did looking at the annual statements of about 900 health plans. Of course there is some variability, but the numbers hold pretty true.

Health plans spend a lot of their time and attention trying to manage benefit costs. That’s not an unreasonable area to focus on if 85% of the premium dollar goes to pay for benefits. However, reducing benefit costs can be a real challenge, as it requires changing member and provider behavior—a tall order.

On the other hand, while administrative expense represents a much smaller piece of the premium pie, health plans tend to be in control of their own administrative spending. Managing administrative cost, therefore, ought to be easier, as plans need only change their own behaviors rather than the behavior of third parties that don’t appreciate being told what to do.

Too bad it’s not that simple, because managing administrative expense can be very painful for organizations for a whole host of reasons. Ultimately, I have seen clients achieve significant improvement in their administrative expense positions. However, real bottom-line impact requires a comprehensive strategy, management buy-in and commitment, and major investments of time and energy.

Continue reading