Health Watch. In late 2008, the health insurance industry faced a number of major challenges. What does ERM mean for. Health Practitioners?

Transcription

1 1 What does ERM mean for Health Practitioners? By Max Rudolph 2 Chairperson s Corner By Susan Pantely 3 Letter from the Editors By Mary van der Heijde and Grady Catterall 7 Health Insurance ERM: One Approach By Jeff Garnett 10 New Member Impact on Medicaid Managed Care Costs By Bela Patel Fernandez and Jeff Smith 13 Soundbites from the American Academy of Actuaries Health Practice Council By Heather Jerbi and Melissa Stevens 16 Health Section Issues New IBNR Study By Cabe Chadick and Steven Siegel 18 Navigating New Horizons an interview with Eduardo Lara di Lauro By Sarah Lawrence 22 Effective Contracting with Pharmacy Benefit Managers: Protecting a plan sponsor s resources By Brian N. Anderson and Robert Cosway 28 Modeling Anti-selective Lapse and Optimal Pricing in Individual and Small Group Health Insurance By Andrew Wei ISSUE 63 FEBRUARY 2010 Health Watch What does ERM mean for Health Practitioners? by Max Rudolph Why Care About Enterprise Risk Management? In late 2008, the health insurance industry faced a number of major challenges. Stock prices of publicly traded companies dropped precipitously. As layoffs occurred, the number of lives covered under employer plans decreased, which placed intense pressure on insurance company overhead expenses. Assets lost significant value, putting surplus at risk. However, companies that proactively had considered the possibility of these risks and studied their potential impacts and interactions maintained a competitive advantage and in the current environment, even the slightest edge can make or break a company. Considering such risks is the essence of enterprise risk management (ERM). During the recent financial crisis, there were numerous cases in which ERM either helped or could have helped. Exactly how it was applied (or would have been applied) remains an open question in some of these cases, but an examination of several prominent companies that ran into difficulties is extremely enlightening. There are companies that manipulated financial statements, assumed complex models were perfect, and generally adopted a culture where anything goes, so long as senior managers could continue to pay themselves well. By contrast, thinking about the big picture over a long-term horizon, encouraging skepticism, and simply using common sense would have served these companies well. Max J. Rudolph, FSA, CERA, is the owner of Rudolph Financial Consulting LLC in Omaha, Neb. He can be reached at max.rudolph@ rudolphfinancialconsulting. com. CONTINUED ON PAGE 4

2 Chairperson s Corner by Susan Pantely Susan Pantely, FSA, MAAA, is principal & consulting actuary at Milliman Inc. in San Francisco, Ca. She can be reached at susan. pantely@miliman.com. Health Care Reform Last fall, leadership of the Society of Actuaries Health Section Council and the American Academy of Actuaries Health Practice Council grappled with ways to provide research and information to congressional staff and the public regarding the proposed health care legislation. We believed there were a multitude of topics where health actuaries could provide unbiased, non partisan expertise. This was no small task for many reasons. First, we needed to decide which topics we felt we should address. Second, our work needed to be completed under tight deadlines as the legislation was constantly in motion. Finally, we relied on teams of volunteers who had other professional and personal obligations. Additionally, the teams of volunteers had no previous experience working with each other. The Technical Report Federal Health Care Reform 2009: Start Up Capital Costs for Health Care Co-ops and a Public Plan, prepared by a joint workgroup of the American Academy of Actuaries and the Society of Actuaries was the result of these efforts. The report was released late October followed by a webcast for congressional staff and media to provide an overview of the report and answer questions. The team developed a model to project the capital requirements of health care co-ops and a public plan under a variety of potential scenarios. The report describes the model, the key drivers of capital requirements, and the amounts of capital required to support health care co-ops or a public plan until they become self sustaining under each scenario. We hope these efforts helped members of Congress, their staffs, and the public to understand possible implications regarding the various health care reform proposals under consideration. Another joint workgroup modeled the implications of the proposed tax on Cadillac plans. Thank you to all the volunteers who contributed to these efforts. New Research Reports A number of methods exist for calculating IBNR reserves though techniques have not been substantially updated in a number of years. In addition, increased scrutiny on financial reporting has led to the increased necessity of being able to provide reliable, accurate results. The desire to address these factors and provide qualitative guidance to health actuaries led to the research report Comparison of Incurred But Not Reported (IBNR) Methods which assesses the accuracy of commonly IBNR methods over a wide range of scenarios. The report was prepared by Cabe Chadick, Wes Campbell, and Finn Knox-Seith. (This issue of Health Watch includes a separate article with more information about this research report.) Health care delivery systems should differentiate themselves by their quality and efficiency achievements. Additionally, quality and efficiency are considered in most health care reform initiatives. Measurement of Healthcare Quality and Efficiency: Resources for Healthcare Professionals reviewed and inventoried the range of quality and efficiency measures currently available. The report serves as a resource on quality and efficiency measures that demonstrate the performance of hospitals and physicians. The report was prepared by Ian Duncan, Greger Vigen, and Sheryl Coughlin. We welcome your input. If you have any suggestions for topics for new research, please feel free to contact me or any member of the section council. n 2 FEBRUARY 2010 Health Watch

3 Letter from the Editors by Mary van der Heijde and Grady Catterall ISSUE 63 FEBRUARY 2010 Health Watch The way to an actuary s heart is through a really nifty IBNR model. Well, perhaps that is a bit of an exaggeration (and reveals what nerds your trusty editors are as well), but after seeing a spirited discussion of new IBNR techniques break out after hours at the Spring Health meeting this year, it s clear that nothing can light up a room full of actuaries like debating the merits of a hybrid chain-ladder method versus a trusty basic developmental factor method. (Intrigued? Make sure to check out the fascinating new IBNR study summarized later in this issue.) Another interesting observation from this IBNR debate was that there were actuaries from multiple countries participating, each with different ideas. This reminds us not only that there are people all over the world who do what we do, but also that we have much more in common with our international counterparts than we might think at first. As we re looking at coping with the notable changes likely to occur in the U.S. health care system, it s interesting to see how other countries with the same challenges have dealt with similar issues. For better or for worse, the U.S. health care system is unique among those used in the world s developed countries. However, we are faced with many of the same challenges as are our international counterparts, such as mitigating high medical cost trends, improving disease and utilization management, and increasing competitiveness within existing regulatory restrictions. Perhaps we could avoid reinventing the wheel by expanding our understanding of how these issues have been dealt with successfully (and unsuccessfully) in other countries. At a more specific level, we can directly learn from other countries in certain technical areas, such as Solvency II (risk-based solvency requirements in the European Union, which are the inspiration for similar requirements in other countries). It s interesting to compare how the roles and responsibilities of a health insurance actuary differ by country. While the role of the actuary typically includes data analysis, benefit design, and premium rate setting, in many areas the domain is less clearly defined. From talking with a colleague in Brazil, we ve learned of a key issue that actuaries there have faced: The lack of clarity about their role has diminished their perceived value, so that the more strategic analyses have been done by economists, accountants, or statisticians instead of by actuaries. However, as you can read in this issue s Navigating New Horizons piece featuring Mexican actuary Eduardo Lara di Lauro, the role of the actuary in Mexico has become both well respected and wide reaching in terms of the scope of important business management decisions that actuaries are invited to address. A colleague working in the German health care market reports that the scope of tasks actuaries are expected to perform has increased notably, and that the actuarial voice has strengthened within their companies. We ve seen both increased diversity in the workplace and increased interaction with actuaries outside of the United States, particularly in the context of multinational firms. We encourage you to learn more about actuarial work done outside of the United States or Canada, and gain from and contribute to the expanding international actuarial community. A great resource is the SOA s International Section page ( We hope this issue helps share some new interesting perspective and inspires you to keep your passport updated! n Mary van der Heijde, FSA, MAAA, is a consulting actuary at Milliman Inc in Denver, Colo. She can be reached at mary.vanderheijde@ milliman.com. Grady Catterall, FSA, MAAA, is a senior staff actuary at Kaiser Permanente in Rockville, Md. He can be reached at Grady.C.Catterall@ KP.org. Published by the Health Section Council of the Society of Actuaries This publication is free to section members. Currentyear issues are available from the Communications Department. Back issues of section publications have been placed in the SOA library and on the SOA Web site: ( Photocopies of back issues may be requested for a nominal fee Section Leadership Susan Pantely, Chairperson Judy Strachan, Vice Chairperson Kevin Law, Secretary & Treasurer Daniel Bailey, Council Member Joan Barrett, Council Member Grady Catterall, Council Member Robert Cosway, Council Member Beth Grice, Council Member Scott Haglund, Council Member Jon Hendrickson, Council Member Ross Winkelman, Council Member Sudha Shenoy, Council Member Grady Catterall, Editor ph: grady.c.catterall@kp.org Mary van der Heijde, Editor ph: mary.vanderheijde@milliman.com SOA Staff Kathryn Baker, Staff Editor ph: f: kbaker@soa.org Sara Teppema, Staff Fellow Health ph: f: steppema@soa.org Jill Leprich, Project Support Specialist ph: f: jleprich@soa.org Julissa Sweeney, Graphic Designer ph: f: Facts and opinions contained herein are the sole responsibility of the persons expressing them and should not be attributed to the Society of Actuaries, its committees, the Health Section or the employers of the authors. We will promptly correct errors brought to our attention. Copyright 2010 Society of Actuaries. All rights reserved. Printed in the United States of America. Health Watch FEBRUARY

4 What does ERM mean FROM PAGE 1 For instance, American International Group (AIG) required a federal bailout because one small division tasked with modeling complex credit default swaps came to be seen as a profit machine for the company and was allowed to grow unchecked. Even though the models contained several major flaws, skepticism was frowned upon. A large number of profitable, well run operations within AIG now bear the scars from these misjudgments. Other examples include Health South, which manipulated its financial statements, and Long- Term Capital Management, which seemed to believe that its models and assumptions were infallible. And then there s Enron, whose financial and cultural collapse was so strong that it destroyed its auditor along with itself. The list goes on. So why should health practitioners care about enterprise risk management? The answer lies in the following question: who is better-suited to consider these issues as they apply to health insurance companies than actuaries? With their advanced mathematical training, broad background in insurance topics, detailed knowledge of the various functional areas of a health insurer, and finely honed skills in contemplating all aspects of risk, actuaries have a unique set of skills that provides a strong basis for managing risks rigorously and holistically. Indeed, when health insurers do develop strong risk management programs, these programs often are managed by actuaries, with many actuaries serving as Chief Risk Officers. Even when the Chief Risk Officer is not an actuary, actuaries often play an important role on risk management committees. We expect to see more actuaries in these positions as the Society of Actuaries Chartered Enterprise Risk Analyst (CERA) designation becomes more prevalent. So far, over 30 health practitioners have earned the CERA designation. Clearly, ERM is a significant growth opportunity for the actuarial profession, and health actuaries are no exception. What is Risk? When formulating a definition of ERM, it is important to start with the R risk. This means different things to different people, based on their perspective and experience. The economist Peter Bernstein, author of Against the Gods: The Remarkable Story of Risk, was quoted in CFA Magazine (March/April 2004) as saying Risk is about the unknown, the inescapable darkness of the future. In terms of a technical definition, one requirement for the presence of risk is uncertainty: if an outcome is known in advance, regardless of whether it is desirable or undesirable, there is no risk. (Flying into space without oxygen is not a risk under this definition you will surely die; there is no uncertainty involved.) The other requirement for risk to be present is exposure: you must actually be exposed to an uncertain event for it to be considered a risk. While these requirements can be applied as a general guide, it is important to emphasize that the assessment of risk is a subjective matter, with no absolute right or wrong. Risk Management Risks are often viewed in terms of either volatility or downside exposure. Some tools for risk management the M in ERM such as the Capital Asset Pricing Model are driven by the measured volatility of a particular metric (in this case, the movement of equity prices). Working along these lines, a publicly traded health insurance company might focus its risk management efforts on potential variations in its GAAP income. As for downside risk exposure, an entity faces it each time it sets goals that might not be achieved. Further illustrations of such exposure range from the personal level (such as the risk that an individual will be unable to retire when desired with a sufficient level of income) to the company wide level (such as the risk that an insurer will become insolvent). When specific risks are identified and managed individually, the result is what we might call silo risk management that is, it resembles a group of grain silos standing next to each other but operating independently, with no interaction between them. The disadvantage of silo risk management is that it can lead to duplication and a lack of overall coordination. More to the point, decisions that make sense for an individual risk will not always be in the organization s overriding best interests. 4 FEBRUARY 2010 Health Watch

5 What does ERM mean Putting the E in ERM: Breaking Down the Silos ERM extends beyond silo risk management by taking a holistic approach, considering all risks in the aggregate rather than individually. The ERM process considers the impact of combinations of risk, both measuring and managing the correlations between all risks. The interaction between individual risks will vary depending on the nature of the risks, and certain risk combinations will not have steady correlations. For example, when times are very good (as well as when times are very bad), many financial risks trend similarly, and their correlations increase. There are a number of mathematical techniques now available to measure these changing risk correlations. ERM combines these quantitative methods with qualitative tools (as discussed below, in the section on Key Risk Indicators) to assess risk. Both types of tools are needed to create an ERM framework. Developing an ERM Framework When considering ERM broadly, a practitioner starts by developing a framework for a consistent process. This is also an opportunity to advance other projects that leverage such processes. Whether it is principle-based approaches to reserves and capital requirements, scenario planning or predictive modeling, many of the techniques involved build off each other. Initial design specifications and later improvements can be incorporated into a base model that is then used for many tasks. Using one base model saves time, and it provides a common thread connecting a range of different projects. Once the model is explained to clients, it becomes easier to explain the various projects for which the model is used. Risk Identification The first step in developing an ERM framework is to identify the risks taken by the insurer, which will vary based on lines of business and investment philosophy. Major categories might include strategic, operational, credit, and interest rate risks. There are several tools used in the industry that can provide a starting sample of risks to consider. Each risk should be assigned to whoever is accountable for managing it. In some instances this will be a com- mittee rather than an individual, but that should be the exception rather than the rule. Don t make this a bigger project than it really is. Develop a relationship with the internal and external audit teams to capture all risks and avoid duplication of effort. The firm s business team probably already knows its risks and just needs to write them down. Sit down with the risk owners to determine the likelihood and severity of each risk, both before and after any mitigation efforts. Each risk should be clearly defined, with current status and any plans to manage it differently documented and updated at least annually. This process will help to prioritize the risks and determine which ones will be discussed at the board level. Key Risk Indicators A leading indicator provides information that allows you to act in advance. For example, when you re on the road, the turn signal of the car ahead of you is a leading indicator. The importance of this indicator is underscored by the sound of your own screeching tires every time the other driver fails to use it before making a turn. Key risk indicators (KRIs) should be developed for each risk. There will be general industry KRIs as well as indicators unique to a particular company. Many will be variants of an existing set of metrics: the lagging indicators collected as part of the reporting process. For example, morbidity is measured by claims paid. Firms need to search for leading indicators that help drive business decisions prior to claims turning out higher than expected. In this case, perhaps leading economic indicators for trend, such as projected CPI and unemployment rates, will help the risk owner improve the decision making process. Carrier-specific data, such as provider contract expiration dates, are also candidates to add value. These metrics will evolve over time, enhancing the firm s ability to make the right decisions. While KRIs for financial risks often use quantitative measures, other risks are better suited to qualitative assessments. For example, issues that fall CONTINUED ON PAGE 6 ERM extends beyond silo risk management by taking a holistic approach, considering all risks in the aggregate rather than individually. Health Watch FEBRUARY

6 What does ERM mean FROM PAGE 5 into the category of reputational risk might best be measured by asking the senior management team to rank each of these related risks in terms of urgency or severity as high, medium or low. Resulting rankings can be trended over time; a sharp increase in the ranking of a particular risk would be cause for concern. Another example that is present in today s environment is the external risk due to health reform legislation. There is definite value to holding interactive sessions with your clients to flesh out risks that are not conducive to quantitative measurement: it helps the practitioner avoid the what you can t measure you can t manage trap. Aggregation Aggregation of risks is an important part of the ERM process, but it has no standardized approach. A risk manager can use the NAIC RBC formula to anticipate marginal impacts of future decisions. If desired, a more conservative assessment can be produced by ignoring diversification benefits. Looking at risks holistically requires you to think about how the risks fit together. For example, including capital considerations as part of an insurer s pricing discipline, and making this part of a broad ERM strategy, allows more consistent decision making and can add value. Communication Communication efforts will determine the success or failure of an ERM framework. And a key part of ensuring the long-term success of the program is the internal message. Don t develop ERM just to meet a rating agency requirement. External stakeholders such as rating agencies and equity analysts can provide some consistent standards to help the risk manager get started, but the primary customers of ERM efforts are internal senior management and the board of directors. To develop your ERM message, think about your current efforts and how you can grow them iteratively over several years. That way you have a story to tell when you meet with the various internal stakeholders, and your game plan can evolve and improve over time. Looking at what other companies are writing in their public documents will provide additional guidance in crafting both your internal and external messages. Most importantly, remember that enterprise risk management is a process, not a project. It enters the culture of the firm. Without a strong risk culture in place, the ERM effort is just busy work. For a firm without such a culture, management is kidding itself (and everyone else) if it claims to have a useful ERM framework in place. What It Means to Have a Risk Culture Once the basics of an ERM framework are in place, you will find that risk owners will have their own ideas about how to improve the process. That is when you know the risk culture is taking hold. In ERM nirvana, everyone would wear a button saying I am a risk manager. This implies that the culture must be both top-down, with strong leadership from the board and senior management, and bottom-up, with entry level employees comfortable with being a part of the risk management process. Advanced ERM techniques involve both sophisticated modeling efforts and a deepened risk culture. Techniques to improve forecasting results can include scenario planning, stress testing, stochastic modeling, and assessing emerging risks. Thinking carefully about potential events (without dwelling on them obsessively) can give an insurer a leg up on its competition. Improved risk culture requires an analytical approach and a skeptical attitude, with business plans questioned and hearty debates encouraged. Such a challenging environment can be uncomfortable at first. But how many insurance executives wish they had fostered these types of discussions prior to entering the long- term care market, or before offering secondary guarantees on variable annuities? Skepticism and Common Sense As noted above in the discussion of KRIs, qualitative efforts are just as important as those with specific metrics. Risk managers should utilize their experience to sniff out risks. Even if a risk can t be measured, it should be evaluated based on volatility considerations and downside risk appraisals. Ask questions such as: What is the worst thing that could 6 FEBRUARY 2010 Health Watch

7 What does ERM mean happen if I accept this risk? Can I live with that? Think of how your peers would react to a front page article in the local newspaper naming you as the person responsible for accepting a particular risk. Would you be comfortable with that? These are the kinds of common sense, gut reaction approaches that you need to incorporate into your ERM efforts. Using ERM to Make Better Decisions Actuaries practice in a variety of industries, from insurance to asset management to sports statistics services. Some say that ERM differs within various practice areas. This is mostly a definitional misunderstanding. The basics of an ERM process or framework do not vary by entity. It is true that the primary risks of a health insurer differ from that of other firms, but the process of identifying risks, developing a risk culture, and making better decisions is common to all firms. This is true for financial services firms as well as non-financial services firms, and for corporations as well as individuals. The similarities between ERM in the health business and ERM in other insurance lines go beyond just the framework. The basics of insurance risk management maintaining sound contracting processes, managing adverse selection, and paying claims correctly apply to all practice areas. Health actuaries just have to be extra vigilant when managing certain risks, such as volatility due to large claims or unanticipated changes in utilization and cost trends. Summary Enterprise risk management is an evolving field. It has been implemented to various degrees at financial services firms such as banks and insurers as well as at companies that focus on manufacturing and services. ERM covers a broad range of qualitative and quantitative techniques, but the first line of defense is common sense: if it doesn t feel right, then it probably is worth a longer look. Firms that encourage skepticism and contrarian thinking rather than penalizing them have a healthy risk culture and likely will have a competitive advantage. Companies that develop key risk indicators and study the way they drive decision making have a better understanding of the risks inherent in their business. These organizations are well on their way to making decisions that optimize value added. n Think of how your peers would react to a front page article in the local newspaper naming you as the person responsible for accepting a particular risk. Health Insurance ERM: One Approach by Jeff Garnett One core discipline underlying all financial firms is asset liability management (ALM). For banks, that discipline resides centrally and coordinates the potentially unrelated activities around deposit capture and lending. For many insurers, ALM is more decentralized reflecting a close relationship between asset and liability that is embedded within pricing, underwriting, and reserving activities. Centralized versus decentralized ALM is a major contributing factor to the early recognition in banking of enterprise risk management (ERM) as a value added activity. Many insurers have taken longer to adopt ERM, particularly where there is a short pricing cycle such as for health insurance. Aetna was an early adopter of ERM within health care, but a relative newcomer to the discipline when compared to banks and life insurers. We believe the timing of our adoption allowed us to learn from CONTINUED ON PAGE 8 Health Watch FEBRUARY

8 Health Insurance ERM FROM PAGE 7 the mistakes of others, such as constructing an ERM program that meets the prescriptive tenets of regulators, rating agencies, consultants, and software vendors, but not the needs of the business. In the earliest stages of ERM, Aetna followed the common path of identifying enterprise wide risks, adopting an agreed upon description of these risks, and prioritizing them based upon a relative system of valuation. At this early stage Aetna also built ERM to meet the emerging requirements of external constituencies. Manage and monitor a comprehensive list of potential enterpriserisks, and reprioritize at least annually Aetna s approach, though refined over the years, remains largely intact as a foundational ERM process. By itself, this prioritization allows us to allocate scarce resources including discussion time at the Board of Directors and Committees, audit activities, and spending on management controls. As importantly, it serves as a comparative description of our business model and environment over time. The risks that are core to our operations have, over the years, risen to the top of the list. Others that are not core to operations still qualify as enterprise wide risks but reside lower in the prioritization. Over time, this natural stratification led us to recognize that ERM would be more effective if it resided within the management process rather than continue as an independent and separate function. A new phase of ERM at Aetna commenced with its integration into the management process specifically the planning and performance management group. From this vantage point, ERM now has a view into the day to day issues within the core management process, as opposed to having to seek them out. It has allowed ERM to recast existing processes to be leading indicators of risk, to leverage management information systems to deliver risk based metrics, and to provide relevant risk related input to management discussions and decision making. 8 FEBRUARY 2010 Health Watch

9 Health Insurance ERM One key change to ERM in this new phase was to adopt the language of Aetna s management process, and to reflect that language in its own unique processes such as risk dashboards. Another change was to leverage ERM resources and tie them to subject matter expertise through the use of risk champions across the management team. The ERM process now adds focus to decisions around management controls that are under consideration, and have the most potential impact on risk. This new phase configuration and approach is tailored to Aetna s management process and culture. There is more than one approach to ERM, and Aetna s experience suggests that ERM can reside and flourish within an existing management process. Jeffrey R. Garnett, CFA, is managing director, Enterprise Risk Management with Aetna. He may be reached at , or GarnettJ@aetna.com. Health Watch FEBRUARY

10 New Member Impact on Medicaid Managed Care Costs by Bela Patel Fernandez and Jeff Smith Bela Patel Fernandez, FSA, MAAA is a director of actuarial services for AmeriChoice in Phoenix, Ariz. She can be reached at Bela_Patel_ Fernandez@uhc.com. Jeff Smith is VP of revenue strategies for AmeriChoice in Phoenix, Ariz. He can be reached at Jeff_L_Smith@uhc.com. The recent economic downturn has caused shifts that cast doubt on some fundamental assumptions used in actuarial pricing. The national unemployment rate is now 9.8 percent 1, which is double the rate in January Increasing unemployment not only causes a lack of income, but also a lapse of medical coverage since the majority of working-age Americans participate in employer sponsored health care coverage. For some, Medicaid is the safety net that provides access to affordable health care in trying times. Medicaid began in 1965 primarily as cash assistance for the indigent, but has now expanded into health and long term care services for low income, disabled, and elderly Americans. In this article, we will explore how the recession has impacted new member enrollment in the various state health care programs and the resulting impact on trend. First, state health care programs cover three main categories of individuals: Temporary Aid to Needy Families (TANF): The majority of the members in the TANF program are pregnant women, mothers and their children. Recently, many states have expanded eligibility to include childless adults who also meet income requirements. Children s Health Insurance Program (CHIP): This program provides health care coverage for all children who meet eligibility requirements. The parents may not qualify for TANF coverage, however if their income falls within CHIP guidelines, the children of the family will be eligible for coverage under this program. Aged, Blind, Disabled (ABD): Individuals in this category are elderly and may have a disability. Some also qualify for Medicare coverage, and are also referred to as Dual Eligibles (Medicare/Medicaid). New additions to the Medicaid rolls are, on average, more costly than their longer duration counterparts. There are two main reasons for this: Time of Enrollment: Medicaid enrollment often occurs when an individual is at the hospital or provider. Pent -up Demand: Care may have been deferred during a lapse in coverage. When eligibility is confirmed, members need to catch up on missing care. To demonstrate this, Ingenix Consulting and AmeriChoice s Finance department undertook a study of new member health care costs to determine the overall financial impact. Ingenix reviewed claims cost data across several health plans that have experienced high rates of membership growth due to unemployment increases in their geographic region (see Table 1). The markets studied included Arizona, Maryland, Tennessee, New Jersey, New York, and Wisconsin. Both medical and pharmacy claims were examined. TABLE 1 Unemployment Rate State Increase Arizona 6.0% 9.1% 3.1% Maryland 4.6% 7.2% 2.6% New Jersey 5.8% 9.8% 4.0% New York 5.8% 8.9% 3.1% Tennessee 6.9% 10.5% 3.6% Wisconsin 4.7% 8.3% 3.6% Source: Bureau of Labor Statistics, Sept 2009, Seasonally Adjusted A member s duration was based on their first month of eligibility. Interruptions in coverage during the study period were not included in total duration. For example, a member who enrolled in January 2009, lost coverage for March and April, and re entered the program in May 2009, would have May recorded as duration month three. Members were assigned to one of three duration classifications depending on the length of their membership (1 6 months, 7 12 months, and 13+ months). Claims cost was converted to a Per Member Per Month (PMPM) basis to enable direct comparison between the first two quarters of 2008 and Findings In addition to showing that new members have, on average, higher health care costs (see Table 2) the study showed that the proportion of short duration members was increasing. However, not all markets are seeing the same rate of growth in new members or the same distribution of costs among the populations of differing duration. 10 FEBRUARY 2010 Health Watch

11 New Member Impact Where this growth is significant, the increase in higher cost short-duration members brings with it increased overall health care cost. As Table 3 shows, short duration members grew as a proportion of total membership through the third quarter of 2009 and are projected to continue growing through Historically, the financial and encounter reports submitted to the states only reflect the costs associated with the durational mix of members during the reporting period ( ). Just as the mix of young and old members can affect health care costs, the mix of short and long duration members can have the same effect. If states wait for this effect to show up in the encounter data used as the base to set capitation rates, the impact of more short-duration members will not be recognized until 2011 due to the lag inherent in claim and encounter submission. To reflect the additional cost associated with the recent greater influx of short-duration members in a more timely fashion, an explicit adjustment to the historical base rate or trend is needed. The change in member mix has had a profound impact on average health care costs. Table 4 illustrates the cost impact of these new members. The relative cost due to the influx of more short duration members rapidly increased during the first part of This impact is expected to peak and then gradually ease as the proportion of longer duration members begins to expand. The rate of this mitigation will depend on a number of factors including the state of the economy and its ability to begin reversing the unemployment trend as well as member persistency. TABLE 2 TOTAL PMPM COSTS BY DURATION TABLE 3 AMERICHOICE AFDC/FHP MEMBERSHIP GROWTH & DURATIONAL MIX FOR HIGH-GROWTH PLANS (AZ,MD,NY,WI) TABLE 4 AMERICHOICE MEDICAL COST INCREASE ATTRIBUTABLE TO MEMBERSHIP DURATIONAL MIX Conclusion Based on this analysis, we can draw the following conclusions about Medicaid programs: New members cost more than existing members, new members comprise a greater proportion of membership than historically reported, given the high unemployment rates we are experiencing, historical data does not readily reflect this greater cost, and additional consideration must be given in pricing capitation rates to reflect this impact. Addressing the issue of health care cost in a Medicaid program is the joint responsibility of two parties the health plan and the state. Health plans must be good stewards of public funds and work to keep inappropriate utilization down and operating costs in check. States must do their part as well by offering actuarially sound rates that encompass emerging cost trends through thorough data analysis and the inclusion of appropriate actuarial assumptions. n Health Watch FEBRUARY

12 12 FEBRUARY 2010 Health Watch

13 Soundbites from the American Academy of Actuaries Health Practice Council by Heather Jerbi and Melissa Stevens What s New Despite some recent uncertainty about the prospects for health care reform, the issue remains on nation s domestic agenda. As such, the Academy s Health Practice Council (HPC) has been actively involved in educating policymakers on the actuarial considerations related to many of the provisions being considered insurance market reform, individual mandate, public plan option, etc. The council and its work groups/ task forces have been engaging policymakers in a number of ways including producing educational policy statements and issue briefs, hosting Hill briefings, visiting with congressional/agency staff, and participating in panels hosted by external policy organizations. We continue to receive and respond to inquiries from policymakers, as well as the media, related to health care reform. Policy statements One ongoing project is in response to feedback received during the course of the annual Capitol Hill visits. The HPC and the Federal Health Committee have developed a series of short policy statements, providing an actuarial perspective on various potential components of health care reform proposals. These statements are called Critical Issues in Health Reform and address the following issues: Actuarial Equivalence Administrative Expenses Gender Considerations in a Voluntary Individual Health Insurance Market Health Insurance Cooperatives Individual Mandate Market Reform Principles Merging the Small Group and Individual Markets Minimum Loss Ratios Public Plan Option Risk Pooling Transitioning into New Markets The HPC has developed a dedicated webpage through the Academy s website in order to highlight these new policy statements, as well as additional materials related to health care reform. The Web page can be found at: issues/health_reform.asp. In addition to this series of policy statements on specific health care reform issues, the HPC and the Society of Actuaries collaborated on two projects, both resulting in public policy statements based on research provided by the Society. The first project was a comment letter 1 developed in July by a joint work group of the Academy s Federal Long Term Care Task Force and the Society s Long Term Care Insurance Section Council regarding the Community Living Assistance Services and Supports Act (CLASS Act). The comment letter, submitted to the Senate Committee on Health, Education, Labor and Pensions, addressed the work group s analysis of the structure and funding approaches proposed in the CLASS Act. It indicated that premiums may exceed affordable levels for those in the intended population of the CLASS Act and that the program (as structured) would be unlikely to cover more than a very small proportion of the intended population. On November 2, another joint work group of the Academy s Health Practice Council and the Society of Actuaries held a webcast 2 for media and congressional staff in conjunction with the release of a detailed technical report 3 on the necessary start-up capital requirements for health insurance cooperatives and a public plan option. Under the modeled scenarios, the work group found that start-up capital requirements ranged from $1.7 billion to $45.6 billion. The wide variation is attributable to three unknowns how many people enroll, the difference between pricing assumptions and actual claims, and average claims. Capitol Hill/Media briefings In addition to the development of policy statements, the Academy has hosted briefings for congressional staff and the media and met with congressional nov09.pdf CONTINUED ON PAGE 14 Health Watch FEBRUARY

14 Soundbites FROM PAGE 13 The SOA will be collecting and analyzing data from volunteer companies, and the Step-Loss Work Group will use the aggregated data to propose an update to the current stop-loss factors. committee staff on a number of occasions. On June 19 and 22, the Academy hosted two briefings related to health care reform. On June 19, Cori Uccello briefed members of the press on the keys to a viable health care system: markets must attract a broad cross section of risks, market competition requires a level playing field, and health spending growth must be reduced. On June 22, Karl Madrecki, Cathy Murphy Barron, David Shea and Dale Yamamoto briefed policymakers on similar material. The briefing offered an extended period of time for the approximately 35 congressional staff and external health policy attendees to ask questions of the actuaries. This briefing was followed by a targeted Hill visit with the health policy team with one of the House committees with jurisdiction for health care reform. Medicare In October, the Medicare Steering Committee released a new issue brief, Medicare Advantage Payment Reform. 4 The issue brief discusses the current Medicare Advantage payment system, provides an overview of the goals of the program, and addresses two alternative mechanisms for reducing payments to Medicare Advantage plans. NAIC activities In September, the Health Practice Financial Reporting Committee released a new practice note, Practice Note on the Revised Actuarial Statement of Opinion Instructions for the NAIC Health Annual Statement Effective December 31, On June 13, the National Association of Insurance Commissioners (NAIC) Blanks Working Group adopted revised instructions for the health annual statement actuarial opinion. The revised instructions are effective for the Dec. 31, 2009 annual statement filing. The revised health actuarial opinion instructions: require a qualified health actuary to be appointed by the board of directors; include recommended language, any deviations or additions to that language must be identified in a check box section; and require a supporting Actuarial Memorandum. The Academy included discussion of these changes at the Life and Health Qualifications Seminar, held in November in the Washington, DC area. The Stop Loss Work Group has partnered with the Society of Actuaries (SOA) to update stop loss factors for health RBC formulas. The SOA will be collecting and analyzing data from volunteer companies, and the Stop Loss Work Group will use the aggregated data to propose an update to the current stop-loss factors. Insurance companies and reinsurers interested in supplying data for the study should contact Barbara Scott at the SOA, at bscott@soa. org, and provide her with your name and contact information. The NAIC also asked the Academy to examine the current health care receivable factors. Currently, all health care receivables use the same factor. In response, the Academy formed the Health Care Receivable Factors Work Group, which will analyze the current factor for each line and make a recommendation to the NAIC. The NAIC is looking to make any necessary changes effective in time for the 2010 year end financial statement. International Experience In light of ongoing efforts to reform the U.S. health care system, members of the Health Practice International Task Force are developing articles on a variety of international health care systems. The November/December edition of Contingencies featured an article on international health reform, Health Care Reform: Learning from Others Netherlands 6, which is the second in a series of articles. The first article appeared in the May/ June edition of Contingencies. In that article, John Berkto interviewed Yair Babad, an Israeli actuary. Both articles can be found on the Contingencies Web site at: pdf 6 nline/ ?sub_id=qxylfphsquij#pg54 14 FEBRUARY 2010 Health Watch

15 Soundbites Ongoing Activities The Academy s Health Practice Council has many ongoing activities. Below is a snapshot of some current projects. Health Practice Financial Reporting Committee (Darrell Knapp, Chairperson) The committee continues to work on updating several practice notes (Small Group Certification, Large Group Medical, and General Considerations). Long Term Care Principles-Based Work Group (Bob Yee, Chairperson) This work group is in modeling phase of their work and will be providing quarterly updates to the NAIC Accident and Health Work Group in Stop Loss Work Group (Eric Smithback, Chairperson) This work group is continuing to update a 1994 report to the NAIC on stop-loss factors, and is currently checking data calculations prior to re-starting the modeling phase of their work. Disease Management Work Group (Ian Duncan, Chairperson) This work group has begun development of a public statement on evaluating wellness programs. Medicare Supplement Work Group (Michael Carstens, Chairperson) This work group has submitted recommended changes to the Medicare Supplement Refund Formula to the NAIC s Medicare Supplement Refund Formula Subgroup, of the Accident and Health Working Group, and continues to work with the NAIC to develop a refund formula. If you want to participate in any of these activities or you want more information about the work of the Academy s Health Practice Council, contact Heather Jerbi at Jerbi@actuary.org or Melissa Stevens at stevens@actuary.org. n Health Watch FEBRUARY

16 Health Section Issues New IBNR Study by Cabe Chadick and Steven Siegel iterations and alternative business situations. (See box on page 17 for a listing of the IBNR methods which were tested.) With the wide variety of policy types, adjudication practices, lag times, and other variables, it is impossible to say that a given method is the appropriate one in all cases. Even in the testing performed, different methods for a particular block performed better under different scenario tests. L&E analyzed the results in an effort to provide some basic guidelines to help health actuaries determine the appropriate method to choose for particular circumstances. up until now, there has been very little by way of comparative information on the approaches. If you were to take a poll of health actuaries on what they consider their bread and butter responsibilities, there is no doubt that estimating Incurred but Not Reported Reserves (IBNR) would rank high. Furthermore, for many health actuaries, working on estimation of IBNR is their first assignment when embarking on their careers. Given the importance of IBNR estimation for the profession, a number of approaches for calculating it have been developed over the years. But, up until now, there has been very little by way of comparative information on the approaches. Recognizing this lack of comparative information, the Heath Section Council of the Society of Actuaries commissioned a research project to assess the accuracy of commonly used IBNR estimation methods over a wide range of scenarios. Lewis & Ellis, Inc. ( L&E ) was awarded the contract to perform this research. The resulting report and supporting material are available on the SOA Web site at: research-ibnr-report.aspx To conduct the study, a stochastic model was constructed to compare and score estimates produced by the IBNR methods that were selected for testing. The testing was done over a significant number of General observations from the study include: 1. Lag methods tend to be the most common methods used by health actuaries; however, the study results consistently show them to have the highest standard deviation in the scenarios that were tested. If lag methods are the desired approach, the more advanced average lag methods and hybrid methods are much better alternatives to the straight average methods. The hybrid methods, especially, produce fairly low mean errors and standard deviations. 2. For rapidly completing blocks in virtually all tested scenarios, the results of the Paid PMPM method appeared the closest to the mean and had the lowest standard deviation. This method was designed to have a lower variance than other methods, and the testing corroborated that design under certain conditions. Subject to caveats mentioned in the study, the Paid PMPM method had the lowest standard deviation in virtually all of the tests on all blocks. The Paid PMPM method, however, also has some weaknesses that are described in the report. Robert Lynch, who has written articles about the Paid PMPM method, indicated in the course of the research that it is under patent protection and proprietary. In light of this, those who would like to use the method may want to consult with legal counsel before doing so for their particular application and circumstances. 3. Seasonality of claims can have a material impact on the mean error. For example, seasonality exhibited by claim concentration early in a calendar year resulted in typical material positive 16 FEBRUARY 2010 Health Watch

17 New IBNR Study errors (i.e., reserve sufficiencies) for the recent incurral months for a number of the IBNR calculation methods, when applied to a year-end valuation date. Seasonality exhibited by claim concentration later in the calendar year (e.g., large deductibles) exhibited the opposite typical results across most IBNR calculation methods. 4. Applying an IBNR method (e.g., Benktander) that relies on premium for its calculations requires adjustment and likely actuarial judgement. 5. There is a material risk to IBNR accuracy when removing large claims from the dataset in which the IBNR calculation method (e.g., traditional lag, Paid PMPM, etc.) is applied. 6. Care should be taken in determining which IBNR method to choose for different medical insurance block types. For example, the self funded block type and the HMO Inpatient block type had the highest standard deviation of error results observed across most IBNR methods tested. The study is intended to serve as a practical guide for the calculation of claim reserves, focusing on the different methodologies available for IBNR calculations. In keeping with this goal, the study does not endorse one method over the other the choice really comes down to what method will work best in a given situation and set of unique circumstances. It is the authors hope that the guide will serve as a starting point for any actuary establishing a new reserving process or wishing to re examine an existing process. This study reviewed the following IBNR methods: Exposure Methods: - Loss Ratio Method - PMPM Method Basic Development Methods: - Straight-Average Lag with Average Periods of 3, 6, 9, 12 Months - Straight-Average Lag without Outliers - Geometric Average Lag - Harmonic Average Lag - Dollar-Weighted Average Lag Cross-Incurral Period Method Hybrid Chain-Ladder Methods - Hybrid Loss Ratio Method with Outliers Removed - Bornhuetter-Ferguson with Straight-Average Lag - Gunnar-Benktander with Straight-Average Lag - Credibility-Weighted with Straight-Average Lag Paid PMPM Method Stochastic Methods More information on each of the methods above, including additional details and formulas, can be found in Appendix B of this study. n Cabe W Chadick, FSA, MAAA, is a consulting actuary and principal with Lewis & Ellis, Inc. in Richardson, Texas. He can be reached at or cchadick@lewisellis. com. Steven Siegel, ASA, MAAA, is a research actuary for the Society of Actuaries, in Schaumburg, Ill. He can be reached at ssiegel@soa.org. Health Watch FEBRUARY

18 Navigating New Horizons an interview with Eduardo Lara di Lauro By Sarah Lawrence The Mexican Market Lara di Lauro s path to becoming an actuary was similar to that of many actuaries in the United States, but there are also key differences in both the training and the market he would be entering.. Mexican health care is primarily split into two sections. Approximately half of the people of Mexico, those who are formally employed by organizations, are covered by Social Security. The other half are covered by another public health insurance option called Seguro Popular (Popular Health Insurance). 18 FEBRUARY 2010 Health Watch Eduardo Lara di Lauro As advances in technology and travel make this world a much smaller place, the world of an actuary is becoming significantly larger. The opportunity to communicate and work with people on an international level not only fosters new business ventures, but also makes it possible to gain the one thing actuaries are constantly seeking more of data. Few actuaries know this as well as Eduardo Lara di Lauro. As principal and director of Milliman s health consulting practice in Mexico and a committee member of the International Association of Actuaries Health Section, Lara di Lauro has had the opportunity to work with actuaries from many different parts of the world. International cooperation means we can join efforts to publish studies and analysis and share them with each other, he said. I think specifically the Society of Actuaries has published a lot of studies that reach us even in Mexico, but we try to look at them in a more broad way. We try to see how we can adapt specific ideas into other markets. An ability and willingness to adapt is one of the most important characteristics for any company to acquire before stepping into the international market, according to Lara di Lauro. While actuaries worldwide share many common goals, learning about a country s specific culture and market is the best first step to forming an effective business plan. The Popular Health Insurance is mainly for the poor and for those without a formal job or Social Security, Lara di Lauro said. But since both sectors have all of the troubles most public insurances have, like long lines and deficient quality in some instances, that s why there are private heath insurance companies. It s what we call a double payment system. We pay for Social Security, but we don t use it. Then, we acquire private health insurance for pretty much the same coverage. Lara di Lauro said the population of Mexico is approximately 108 million and just 8 million have private health insurance. Getting His Start As a child growing up in Mexico City, Lara di Lauro said he was always interested in mathematics. When I had to choose my profession, I was thinking of engineering or math, he said. But I have a sister who is also an actuary and she influenced me to study the actuarial sciences. Lara di Lauro said studying to become an actuary in Mexico is quite different from studying in the United States. He earned his bachelor s degree in actuarial science from the Science Faculty of the Universidad Nacional Autonoma de Mexico after five years of training and study, as well as passing a professional exam. From Mexico s National College of Actuaries, he passed the required exams to become a Certified Actuary in Accidents and Health. He has also earned advanced degrees in applied mathematics and business administration and finance. Lara di Lauro entered the insurance industry in 1986 and from 1988 to 1995 worked in leading and developing the life and health insurance lines of business for three insurance companies, as well as held a position as an actuarial consultant with

19 Navigating New Horizons Wyatt Consultants. In that time he learned why the actuarial profession was a good fit for him. What I really like is the applied mathematics, he said. I like how you can apply the theoretical mathematics into a reality. Create models to describe a risk and then measure that risk in order to protect lives or property. It s a magical thing when you create a good model and you are able to monitor your model and how it is reflecting the reality. Lara di Lauro said he also enjoys sharing his experiences with others. I like how the knowledge can be shared with other professionals, such as physicians and nurses, or people specializing in IT systems, he said. So all together we can produce really good models and really good outcomes which benefit people, because all kinds of insurance but mainly health insurance or life insurance are a social good. You are providing a good to society, and I really like that. Finding His Niche In 1995, Lara di Lauro began working with his specialization in managed care when he was given the opportunity to create a health insurance company from scratch. As managed care director and general manager of TPA for General de Seguros, Lara di Lauro was responsible for adapting the managed care concept to the Mexican market. At that time, health insurance in Mexico was mainly indemnity plans, he said. There was not any kind of managed care tools, and the coverage that companies used to offer were pretty much limited and without any preventive component. Lara di Lauro attended the Arizona Health Science Center of the University of Arizona for two years to learn more about managed care, and went on to create a specialized health insurance company with concepts adapted to the Mexican environment. So through that effort after three and a half years, that company was the first company to receive the authorization from the National Commission of Insurance in Mexico to sell comprehensive health plans, he said. Lara di Lauro said one of the challenges of introducing this U.S. model into the Mexican market included changing people s attitudes and paradigms. The focus was primarily to change the behavior of the provider, and also the attitude of the insurance company toward the providers because insurance companies saw themselves as just in the financing area, but they started to take action on the delivery side, he said. So they started to build strong provider networks and semi-open provider networks. Lara di Lauro said, as a result, insurance companies had to start negotiating contracts with both hospitals and physicians. And it s the same challenge nowadays. There is still the challenge to get a better understanding of the business from the hospital s point of view and physician s point of view. Instead of fighting with the insurance company, it s better to try to make win-win agreements with them. The Next Step In 1999, Lara di Lauro was approached by four branches of Blue Cross Blue Shield Texas, Illinois, Arizona and California with their plan to expand into the Mexican market. As general manager and director of the project, it was Lara di Lauro s responsibility to lead and develop the infrastructure to launch Blue Cross Blue Shield de Mexico as a Specialized Health Insurance Institution. This included creating a business plan assessment, achieving a work plan, making financial projections and obtaining authorization from the National Commission of Insurance. He also opened four branches and hired and trained personnel. Unfortunately, circumstances in the U.S. changed quickly and the business venture was brought to a halt. We created a health insurance company from scratch and after three and a half years we had the authorization to operate this health insurance company, then September 11, 2001 came, Lara di Lauro said. After September 11, they were pretty afraid and decided to stop and cancel the project. We had to properly shut down the company. CONTINUED ON PAGE 20 You are providing a good to society, and I really like that. Health Watch FEBRUARY

20 Navigating New Horizons FROM PAGE 19 Paying attention to the small details of a culture is one of the most important aspects to consider when adapting actuarial business models from one country to another. Consulting for Milliman After Lara di Lauro was able to close the Blue Cross Blue Shield branches he had helped to build, he reached out to a consulting firm in Mexico called Grupo Technica. I already knew that they had some kind of agreement with Milliman to represent Milliman in Mexico, he said. So I reached them and we set out to work together and put together a business plan to offer to the people from Milliman to establish their healthcare consulting practice in Mexico. In 2003, after almost two years of developing that business plan, Lara di Lauro traveled to the United States to make a presentation to Milliman that resulted in the creation of their health care consulting practice in Mexico. He now serves as principal and managing director of that practice. We provide advice to health insurance companies in Mexico, but also a couple of years ago we started to do some life consulting also, he said. We provide comprehensive consulting work with product development and research calculation through claims analysis and strategic planning. We do a lot of research about opportunities in the health insurance market, and also support investors to create new specialized health insurance companies. Branching Out Lara di Lauro said some of this consulting work is being done in Central and South America as well, where actuaries are few and there is often a lack of proper data.. We can provide good insight or good advice to companies in those countries that don t have enough actuaries, he said. In some markets there are just a few actuaries, but they are in really top positions, so they can t really implement or make the calculations. They hire us to make those calculations or teach their people to make those calculations. It s a good opportunity for us to work together in such markets. Gathering the right data is often the biggest challenge, Lara di Lauro said. When some companies don t have enough data, quality is the issue, he said. Some companies are not working hard or not investing or placing adequate importance on the data for their health insurance products, so when you review their databases there is a lack of really important information and fields. Lara di Lauro said helping to gain data in these areas has a positive effect that extends beyond just that company or the country it is in. We can adjust that information and complement that information with information from other countries and then adapt to those specific circumstances, he said. That s an important part of the actuarial work we can do, gathering information from other countries. Advice Lara di Lauro said paying attention to the small details of a culture is one of the most important aspects to consider when adapting actuarial business models from one country to another. I have seen several failures introducing information systems from the U.S. into Mexico and the companies invest a lot of money trying to adapt, but they didn t make an initial analysis asking if the basics of the system will support those differences, he said. Even small differences, such as the average length of a person s name, can throw a wrench in the works if they are not considered. In Mexico we use our father s name and our mother s name, and in the U.S. you just use your father s name, Lara di Lauro said. So in Mexico you need an additional field to put your mother s name and that causes a lot of changes in the system. So even tiny things could push you into failure and you need to consider that all the time. Teaming up with local people to research and understand the local features of an area is the most effective way to avoid such problems. If you want to learn by yourself it s going to take awhile, so it s better to get support or to get advice from local people, Lara di Lauro said. In that way you can make a good team and know if your model or your tool is going to be able to adapt or meet the local requirements. So making that team cooperation is really an important issue to have success. 20 FEBRUARY 2010 Health Watch