Frequently Asked Questions (FAQ) The Harvard Pilgrim Independence Plan SM

Frequently Asked Questions (FAQ) The Harvard Pilgrim Independence Plan SM Plan Year: July 2010 June 2011 Background The Harvard Pilgrim Independence Plan was developed in 2006 for the Commonwealth of Massachusetts Group Insurance Commission (GIC). The GIC, like many other large employer groups, was concerned about the rising cost of health care coverage. As part of its Clinical Performance Improvement (CPI) Initiative, the GIC wanted a health benefit that not only controlled cost but also maintained a comprehensive level of benefits and choice of providers; used information about provider quality and cost-efficiency performance to tier providers; and encouraged members to use this information in making health care decisions and working with their providers to maintain good health practices. To meet these goals, the GIC worked with its six participating health insurance plans to begin offering tiered products. The Harvard Pilgrim Independence Plan 1. What kind of product is the Harvard Pilgrim Independence Plan? The Harvard Pilgrim Independence Plan features a deductible and consumer copay differentials based on a three-tiered provider network. It is a Preferred Provider Organization (PPO) in which members do not have assigned PCPs and can go to any provider for treatment. The product does not require a referral for specialty services. 2. What are the deductible costs? An in-network deductible of $250 for individual coverage, and $750 for family coverage, is applied to all services except: immunizations, prenatal/postpartum care, office visits, mammograms, MH/SA admissions, wigs, and speech therapy. (Ancillary tests and procedures performed at an office visit are subject to the deductible.) The deductible for out-of-network medical services is $400 for individuals, and $800 for families. 3. How were providers assigned to tiers? Physicians were placed into tiers based on their type of practice, primary care or specialty care; the location of their practice, Massachusetts or another state; evaluation of certain specialty areas, see question 5; and for those specialists whose specialties were evaluated, availability of sufficient measurement data. Physician Tier 1(*not tiered, assigned a Tier 1 level copay) Measured MA specialists exceeding thresholds *Primary care providers *Behavioral health providers Physician Tier 2 (*not tiered, assigned a Tier 2 level copay) Measured MA specialists meeting thresholds *Measured MA specialists with insufficient scoring data, including new specialists *Non-measured MA specialists *NH, ME, & RI specialists Physician Tier 3 Measured MA specialists not meeting thresholds Acute care hospitals were placed in tiers based on success on national measures of quality and cost. Hospital Tier 1 Hospital Tier 2 Hospital Tier 3 Measured acute care hospitals that had preferable costs and passed the quality threshold Measured acute care hospitals that had mid-range costs Measured acute care hospitals that had lower costs but did not pass the quality threshold Acute care hospitals with insufficient data to rank Measured acute care hospitals that had higher costs Page 1 of 7 Revised 03/18/10

Specialized acute care hospitals 4. What are the copayment costs across tiers? Member copayment amounts for the period of July 1, 2010 June 30, 2011 are: Tier 1 Copay Tier 2 Copay Tier 3 Copay Office Visit $20 $35 $45 Hospital Inpatient Admission $250, then deductible applies $500, then deductible applies $750, then deductible applies 5. Are members responsible for any other cost sharing amounts? Members are also responsible for copayments related to the following hospital services. These costs are not based on a hospital's tier assignment. Copay Day Surgery To be determined Emergency Department To be determined Outpatient High Technology Radiology To be determined Measured Specialists 6. Which Massachusetts specialists were measured? Physicians in the following specialties were measured and placed into one of three tiers. Allergy/Immunology Gastroenterology Obstetrics/Gynecology Otolaryngology Cardiology (Medical) General Surgery Ophthalmology Pulmonary Disease Dermatology Neurology Orthopedics Rheumatology Endocrinology 7. How did Harvard Pilgrim choose the specialties that were measured? Harvard Pilgrim evaluated the specialties that our members use most often, for which we had the most data to measure; that showed the widest variation in the ways doctors treat similar conditions; and that represented the most meaningful opportunity for cost savings for members. 8. Were all physicians in the 13 specialties measured for quality and cost efficiency? Both quality and cost-efficiency measures were used to tier physicians in cardiology (medical), endocrinology, obstetrics/gynecology, and rheumatology. Physicians in allergy/immunology, dermatology, gastroenterology, general surgery, ophthalmology, neurology, otolaryngology, pulmonary disease, and orthopedic surgery have been rated on cost-efficiency only, as there were not adequate measures, or a sufficient number of physicians meeting the minimum sample size, to evaluate quality for these specialties. 9. How was a provider s specialty determined? Mercer determined a provider s specialty based on designations from the six plans. In cases where a provider s specialty designation differed among the plans, Mercer utilized other sources such as the Board of Registration, or the type of claims the specialist submitted, to determine the specialty designation. 10. On what basis were individual specialists measured? For those specialties where both quality and cost-efficiency measures were used for tiering, each specialist was evaluated first on quality of care delivered. This assessment was received from Resolution Health, Inc. (RHI), who evaluated providers on quality (refer to the following section on Quality Measurement). If the quality designation assigned by RHI was C, the provider was placed in Tier 3. Other providers were evaluated on a cost-efficiency basis after adjusting by HPHC s contractual prices. If a provider s assignment was Tier 2 but the quality designation was A and the price-adjusted cost-efficiency score placement was within 10% of the Tier 1 cut-off point, the provider was re-assigned to Tier 1. Page 2 of 7 Revised 03/18/10

For those specialties where quality designations could not be used, tiers were determined on the basis of price-adjusted cost-efficiency scores only. (See question 26 for more information about costefficiency.) Providers with a minimum of 30 episodes in the Mercer All-Payer data were tiered in all thirteen specialties per the above methodology. 11. Were specialists who practice together scored as a group or individually? Measured specialists were scored individually, using data aggregated across all of the GIC health plans for each physician. 12. Were specialists who practice together assigned to the same tier? Not necessarily. It is possible that specialists in the same practice may have different tier assignments based on their master specialty designations and/or their individual quality and cost-efficiency measurement results. 13. How were PCPs who also practice in a measured subspecialty tiered? The tier assignment for a physician practicing as a PCP with a subspecialty is based on his/her master specialty designation. For example, if a pediatrician with a subspecialty in cardiology was identified as a pediatrician by the physician master specialty designation, the pediatrician was assigned to Tier 1 as a PCP. However, if the analysis done by Mercer identified the pediatrician as a cardiologist, then he/she was evaluated and assigned to Tier 1, Tier 2, or Tier 3, based on measure results. 14. How long is a tier assignment in effect? Tier assignments are evaluated each year, based on the prior three years of claims data. A tier assignment is in effect from July 1 through June 30 of the following year. 15. Are the detailed quality and cost-efficiency data that determined tier results for measured specialists available? Yes, you may email independence_plan@harvardpilgrim.org to request the patient-specific quality results and/or detailed ETG data that were used to determine your tier. Harvard Pilgrim can only provide data that is specific to Harvard Pilgrim members. Specialist Quality Measurement 16. Which quality measures were used to determine quality scores? Resolution Health, Inc, (RHI) identified the specialties for which there were adequate quality measures, and scored physicians based on their performance against those measures when the physician had a sufficient number of observations. The RHI measures are based on established clinical guidelines and were reviewed by the CPI Initiative Physician Advisory Committee. A detailed technical specifications booklet, is available on our Web site, www.havardpilgrim.org/providers. 17. Which specialties had adequate quality measures? Among the thirteen specialties tiered by Harvard Pilgrim, there were adequate quality measures for cardiology, endocrinology, obstetrics/gynecology, and rheumatology. 18. How did the quality component factor into tier results for those specialties where quality measures were used? In determining a specialist s tier assignment, RHI first looked at quality performance, or care effectiveness. RHI evaluated the quality performance data for each physician. Physicians who met the RHI quality threshold, or for whom there was insufficient data to evaluate quality, were then evaluated for cost-efficiency by Harvard Pilgrim. Harvard Pilgrim placed physicians who did not meet the RHI quality threshold in Tier 3. They were not evaluated on cost-efficiency. Page 3 of 7 Revised 03/18/10

19. How much data was considered insufficient to evaluate a physician on quality? Cardiologists, endocrinologists, obstetrician/gynecologists, and rheumatologists with fewer than 30 observations were not scored on quality due to insufficient quality data for the identified measures. These physicians were evaluated on cost-efficiency only. 20. What level of performance was necessary to meet the quality threshold? RHI established a probability distribution for those specialties with adequate quality measures. Physicians with a greater than 75% probability of a C rating were deemed by RHI to have not passed the quality threshold and were placed in Tier 3. Physicians given an A or B rating met or exceeded the quality threshold and were then tiered on cost-efficiency. Physicians with insufficient data were tiered on cost-efficiency only. Additional information about the quality methodology is available on our Web site, www.havardpilgrim.org/providers. Episode Treatment Group (ETG) Software 21. What is ETG methodology? ETGs identify and aggregate historical claims activity into complete episodes of care, which are comprised of all inpatient, outpatient, professional, ancillary and pharmacy claims incurred as a patient moves through the continuum of care. It takes into consideration case mix and comorbidities, by grouping together clinically homogenous illnesses, using a combination of ICD-9 diagnosis codes, CPT-4 procedure codes, UB-92 revenue codes, NDC pharmacy codes and time (episodic specific parameters). More information on ETGs can be found at www.ingenix.com. 22. How is an episode defined? An episode of care encompasses all of the care received by a patient for a single medical condition. There can be multiple providers (physicians, facilities, ancillary providers, pharmacies, etc.) involved in the care of the patient. Claims continue to be added to an episode of care until there is a sufficient gap in the treatment (defined as a clean period ) to deem it to be complete. If, after the clean period is reached, there are indications of more claims for the condition, the ETG grouper software starts another episode for that condition. The length of time (i.e., clean period) that defines the start of another episode differs across clinical conditions. 23. Can patients have concurrent episodes? Yes. In clinical practice, patients often are receiving care for more than one condition at the same time. The ETG grouper recognizes this reality. When the ETG grouper encounters claims for a patient with multiple open episodes of care, it attempts to determine which episode makes the most sense for the particular diagnosis and procedure codes on the claim. For example, if a patient has bronchitis and ankle sprain diagnoses on the same claim, a chest x-ray would probably go to bronchitis and a claim for crutches would go to the ankle sprain. 24. How do ETGs handle chronic disease, which has no beginning and no end? The ETG measures services provided over a period of one year for a chronic condition. 25. Are doctors who treat sicker patients penalized because the care is more extensive? ETG scores are adjusted for comorbidities, patient demographics and whether a health condition is episodic or chronic. Further, outlier episodes are removed from the computation if they would be likely to dominate the measure. 26. How were costs attributed to each physician? Mercer used neutral pricing for all services included in each ETG and attributed an episode to the physician responsible for the most "cluster" dollars if the physician had 25% or more of the episode cost. A cluster is the smallest unit of analysis in the ETG software and an episode can consist of more than one cluster. More information on ETGs can be found at www.ingenix.com Page 4 of 7 Revised 03/18/10

27. How was the cost-efficiency score calculated, and what does it mean? The cost-efficiency score is a performance index calculated as the ratio of the attributed physician's actual costs to average costs for the peer group when treating the same mix of conditions. Mercer compared a physician's actual costs to his/her peer group's (specialty's) average cost for each episode. Mercer then summed the physician s actual costs for all episodes that met inclusion criteria and the peer group average costs for those same episodes. The ratio of those costs is a physician's costefficiency score. So, for example, a cost-efficiency score of 0.85 implies that the physician is 15% more cost-efficient than his/her peers. Mercer calculated efficiency scores based on neutral prices to eliminate differences in unit prices across health plans. Harvard Pilgrim adjusted the neutral costefficiency scores produced by Mercer using contracted rates and rank-ordered these price-adjusted cost-efficiency scores to create tier assignments, as explained in question 9, above. Calculation of cost-efficiency score 28. What time period did this analysis cover? The analysis used claims data with dates of service from January 1, 2006 - December 31, 2008. 29. How were outlier episodes defined? ETGs with fewer than 50 episodes across all physicians were not considered in cost-efficiency calculations. To reduce noise, ETGs identified as being inconsistent with commonly occurring ETGs in a specialty, and ETGs with very high costs (such as transplants and isolated signs and symptoms, and non-specific diagnoses) were also excluded from cost-efficiency calculations. 30. Were high cost outliers and low cost outliers included in the calculation of cost-efficiency scores? Mercer defined "high outliers" as an ETG proxy price amount that exceeded two standard deviations from the mean proxy price for that ETG. "Low outliers" were defined as the bottom 1 st percentile of proxy price amounts for an ETG. High and low outlier episodes were excluded from the analysis. 31. What is a Proxy Price? A proxy (neutral) price algorithm assigns a price to each claim based on information on the claim line and on a standard fee schedule to eliminate differences in unit price across health plans. The proxy price was assigned by Mercer to all claim lines in the all-payer data. 32. Were claims for all time periods given the same weight? Recent claims were given more weight than earlier claims by assigning a weight of 1.00 to episode costs in 2006, 1.25 to episode costs in 2007, and 1.50 to episode costs in 2008. 32. What are component costs? Component costs include Management (face to face interactions with a clinician and includes evaluation and management codes plus other face to face claims); Surgical Charges (does not include anesthesia, surgery suit, and assistant or second surgeon); Facility (Inpatient hospital); and Ancillary Page 5 of 7 Revised 03/18/10

(lab testing, radiology, DME, home care, and non-inpatient hospital charges plus much more). The costs listed in the reports are price neutral, and are derived from Medicare and other sources. 33. Which members were included in the analysis? Mercer used all member claims data from GIC participating plans to conduct this analysis. But catastrophic cases were excluded from cost-efficiency scoring. Catastrophic cases were calculated by summing all proxy price dollars for each member prior to running through the ETG grouper, and episodes for the top 1.5% of members were excluded from analysis. 34. What was the resulting Tier break-out, by percentage, for tiered specialists? Approximately 20% of tiered specialists were assigned to Tier 1, 65% to Tier 2, and 15% to Tier 3. Hospital Evaluation 35. Which hospitals were tiered? For the FY2010 Independence Plan year, all Harvard Pilgrim-contracted acute care hospitals were placed in one of three inpatient admission copayment tiers. 36. What was the basis of hospital tiering? Tiering was based on publicly reported quality performance and on relative cost. 37. How was a hospital s quality performance determined? Quality performance was based on composite clinical scores, patients hospital experience, and patient safety performance. 38. Which quality performance reports were used? Harvard Pilgrim used three resources: Hospital Compare, performance information reported by hospitals to the Center for Medicare and Medicaid Services (CMS) for care provided during the period of January 2008 to December 2008, which was publicly reported by CMS in September 2009. The Leapfrog Group patient safety measures reflecting hospital reporting as of September 2009, and Hospital Consumer Assessment of Healthcare Providers & Systems (HCAHPS) survey results. 39. How were composite clinical scores created? Composite clinical scores were created by dividing the sum of all patients who received the recommended care by the sum of all patients who should have received the recommended care for heart attack (AMI), heart failure, pneumonia, and surgical care improvement. 40. How were patients hospital experiences evaluated? Patients hospital experiences were evaluated based on the hospital s HCAHPS survey results on the overall rating question and eight summary and individual questions, as compared to the national average. 41. How was patient safety performance evaluated? Patient safety performance was evaluated using four Leapfrog measure categories: preventing medical errors, staffing the ICU appropriately, taking steps to avoid harm, and managing serious errors. 42. What determined the quality threshold? A hospital could pass the quality threshold by having shown performance over the 75 th percentile in 2 of 6 categories used in the Harvard Pilgrim Health Care Hospital Quality Report: heart attack, heart failure, pneumonia, surgical care, patient safety (Leapfrog), or patients hospital experience (HCAHPS). For further information go to the Hospital Quality Report at https://www.harvardpilgrim.org/portal/page?_pageid=253,236081&_dad=portal&_schema=portal Page 6 of 7 Revised 03/18/10

43. What was hospital cost efficiency based on? Hospital cost efficiency was based on Harvard Pilgrim s 2008 cost for contracted acute care facilities, using a composite average inpatient and outpatient cost for each facility. 44. Which costs were used to calculate the average cost? 2008 utilization for all contracted facilities was repriced to 2010 Harvard Pilgrim contractual terms in order to calculate a composite inpatient and outpatient cost for each facility. All carveouts, per diem rates, case rates or outlier payments were taken into account when repricing claims to 2010 terms. Transplant and subacute admissions were excluded. 45. How was the relative facility cost calculated? Harvard Pilgrim used a standard inpatient and outpatient benchmark upon which to compare the cost of all in-network facilities. The reimbursement rates for each facility s specific inpatient and outpatient claims for 2008 dates of service were inflated based upon 2009 and 2010 contracted rate increases. The same utilization was also repriced to the applicable benchmark rates. Each facility s inpatient and outpatient inflators over the benchmark were normalized to the network average inflator for each category. Using a standard community hospital mix of services, the normalized scores were averaged into one overall composite relative cost score for each facility. 46. How were hospitals compared? Hospitals were ranked from highest to lowest cost based on their overall composite cost score. 47. Were all hospitals evaluated and ranked? No. Hospitals with fewer than 30 Harvard Pilgrim member admissions in the 2008 calendar year were not assigned a cost ranking due to insufficient data. In addition, specialized facilities, including Children s Hospital, Dana Farber Cancer Institute, Massachusetts Eye and Ear Institute and New England Baptist Hospital, were not ranked. 48. How were hospital copayment tiers established? Three copayment tiers were created: Low, Middle, and High. Hospitals that passed the quality threshold and had lower costs were assigned to the low copayment tier (Tier 1). Hospitals with midrange costs, and hospitals with a lower costs that did not pass the quality threshold, were assigned to the middle copayment tier (Tier 2). Higher cost hospitals were assigned to the high cost tier (Tier 3). 49. Were unranked hospitals assigned to a copayment tier? Yes. All unranked hospitals were assigned to a Tier 2 copayment. 50. What percentages of MA network hospitals were placed in each copayment tier? Approximately 15% of MA network hospitals are in the high copay tier, 54% in the middle copay tier and 31% in the low copay tier. Page 7 of 7 Revised 03/18/10