Excellus Health Policy Reports No. 4 Informing the public about critical health care issues September 2002

Transcription

1 No. Informing the public about critical health care issues September 2002 Average length of stay in Upstate New York hospitals: Opportunities for savings EXECUTIVE SUMMARY How does Upstate New York compare on hospital length of stay? v New York has historically had longer lengths of stay (LOS) in the hospital than in many other parts of the country. The average length of stay in New York State (7 days in 1999) is longer than the national average (5.2 days). It also exceeds lengths of stay in many other areas, including cities with similarly cold climates such as Minneapolis/St. Paul, MN. This report shows the average length of stay for adult medical patients in hospitals across Upstate New York. It does not include surgical, obstetric, pediatric, psychiatric, or alcohol/substance abuse treatment patients. After taking into account the fact that some hospitals take sicker patients than others, there is substantial variation in lengths of stay. sõ average length of stay from mid-2000 to mid-2001 ranged from a little over 3 days to around 7 days. The average for Upstate New York was 5. days, compared to.7 days nationally and.0/.1 days in the best practice benchmarks of Minneapolis/St. Paul, MN, and Sacramento, CA. Upstate hospitals did make substantial progress in reducing lengths of stay during the 1990s. The average length of stay in the 36 counties in the Excellus region that have hospitals declined from 6. days in 199 to.9 days in 2000 (this covers all patients, including maternity). s and other providers should be commended for this accomplishment and encouraged to do more. How long a patient stays in the hospital depends on the availability of hospital beds, as well as the supply of post-hospital care. This includes rehabilitation facilities, home care, and nursing homes. Whether hospital subacute units are needed should be reexamined. However, appropriate management of length of stay is important not only in hospitals but also in post-acute settings. No., September 2002, Page 1

2 Why is length of stay important? v Keeping people in the hospital longer than necessary has both clinical and cost implications: Patients can be exposed to infections often present in hospitals and to the possibility of medical errors. It can reduce access to care by filling beds and causing back-ups in the Emergency Department as patients wait for a hospital bed. The benefits of getting people up and moving around are best realized by moving them to residential environments such as their homes or nursing homes. Keeping people in the hospital longer in the most intensive setting available also raises costs. Reducing the length of stay for adult medical patients in Upstate New York hospitals to the national average would lower hospital costs by up to $7 million. Bringing it closer to best practice benchmarks would save substantially more. Given the relative scarcity of qualified nurses in todayõs environment, it would also allow hospitals to allocate their nursing staff to patients who need them the most right after they enter the hospital. Current challenge v Reducing hospital stays requires a community-wide effort. s, physicians, home care agencies, nursing homes, and insurers need to work together to ensure that patients are not kept in the hospital longer than clinically necessary. No., September 2002, Page 2

3 Days length-of-stay trends in New York State New York State has historically had longer lengths of stay (LOS) in the hospital than many other parts of the country. Between 1995 and 1999, lengths of stay declined faster in New York than nationally, but they are still substantially above the national average Average hospital length of stay in New York State vs. the U.S. average New York The gap between New YorkÕs performance and the states with the lowest lengths of stay is even greater. For example, in 1999 the average length of stay in Oregon was.0 days. LOS Trends in Upstate New York The lengths of stay in the metropolitan New York City area are particularly long and drive up the state average. But lengths of stay in Upstate New York are still substantially above the U.S. average and far above national benchmarks. 7 U.S. 5.2 SOURCE: American Association, Statistics 2001 Why is length of stay important? Length of stay matters for several reasons: Clinical considerations: Some patients are concerned about whether or not they will be discharged too quickly from the hospital. Arranging for adequate post-hospital care is clearly critical. Yet staying in the hospital when it is not clinically necessary is not a good idea either. patients can be exposed to a variety of infections. Some can be dangerous, particularly to people who are not in good health. It increases the possibility of being subject to medical errors. For those patients who are able, getting up and moving around quickens recovery. This is often best accomplished at home or in a nursing home. Most patients prefer to recuperate at home. Given an increasing scarcity of qualified nurses, lowering length of stay would enable hospitals to allocate more of their nursing staff to care for patients when they need it most, usually shortly after they enter the hospital. Cost: Length of stay is also important because of the unnecessary costs associated with extended lengths of stay. Many payers (including Excellus and Medicare) reimburse Upstate New York hospitals the same amount per case regardless of how long the patient stays in the hospital. (The amount paid depends on the patientõs diagnosis and other factors.) This payment system created an incentive for hospitals to reduce lengths of stay. It has not worked sufficiently in many cases. As a result, hospitals face more financial constraints. In the long run, the higher costs are built into local payment systems and the whole community pays. If hospital costs are higher than Medicare payments, then some of those hospital costs may be shifted onto private payers. Overcrowding: Extra days in the hospital may restrict access to care by creating back-ups in hospital Emergency Departments (ED), when there is no room to move emergency patients into regular hospital beds. Crowded EDs with long waits can compromise patient care. No., September 2002, Page 3

4 To compare the length of stay fairly across hospitals, it is important to adjust for possible differences in the sickness of patients in various hospitals. All of the data presented below adjust for these differences using 3MÕs APR-DRGs, which are a widely used and accepted approach. The sources and methods used in this analysis are described in the box on page 11. Using this risk adjuster should also permit comparisons across hospitals that use observation units (that is, units that care for patients for up to 23 hours) to varying degrees. Days Risk-adjusted average hospital length of stay for medical patients, July 2000 June Upstate New York.7 National Average.1.0 Sacramento Minneapolis/ St. Paul SOURCE: The New York data are compiled from the New York Statewide Planning and Research Cooperative System (SPARCS). The benchmarks are for The average length of stay for all medical patients in the Excellus region, which encompasses all of Upstate New York except for the Albany area, is 5. days. This value compares to a U.S. average of.7 days and best practice benchmarks of.0 in Minneapolis/St. Paul, MN, and.1 in Sacramento, CA. This analysis only includes ÒmedicalÓ patients, such as those hospitalized for medical conditions like congestive heart failure, pneumonia, or AIDS. It excludes surgical, obstetric, pediatric, psychiatric, and alcohol/substance abuse treatment patients. This was done out of a concern that comparing these other types of patients required greater examination of their comparability across hospitals. For example, there are significant differences in the types of surgeries done at different hospitals. It may therefore be more valid to compare LOS across hospitals for specific surgeries. This would have made this report too long. There is substantial variation in the average LOS across hospitals in Upstate New York. This is consistent with research showing that there are large variations in physiciansõ practices for many conditions and procedures across different settings (see, for example, the Dartmouth Atlas for Health Care at After controlling for differences in patientsõ sickness, the average LOS ranges from a little over 3 days for several rural hospitals to around 7 days. Among hospitals that care for the most complex patientsñsuch as Strong Memorial, Rochester General, Buffalo General, Erie County Medical Center, and University -SUNY Upstate Medical CenterÑthere is also substantial variation, with average LOS ranging from.8 to 6. days. Strong Memorial has an average LOS for medical patients of about.8 days, which is almost at the national average of.7. This suggests that meeting the national average is achievable in the Upstate New York environment. Data on hospitals in the Excellus service area (except a few small ones) are found on pp Some of the hospitals with low LOS are small rural hospitals. This raised the question of whether this was because they transfer their most complex patients to larger urban hospitals. The analysis was repeated after excluding all transfer cases. The average LOS for most hospitals did not differ substantially. Most of them were the same or changed by only 0.1 day. No., September 2002, Page

5 Special challenges faced in Buffalo The federal Housing and Urban Development Department recently released a report on ÒHealthcare Inpatient Capacity Analysis, Buffalo-Niagara Falls, New York.Ó It was prepared by the Center for Health Policy Studies, Albany, NY. The report reveals substantial excess hospital capacity, as well as a high average length of stay, in Buffalo area hospitals. Among the key findings are the following: Based on the projected decrease in population, the number of acute care hospital beds required in 2005 will decline to 2,660. This level is 1,585 fewer beds than the region has today. As hospital occupancy declines, some hospitals will probably have to close. Patients in Buffalo tend to stay in the hospital longer, but they are not sicker, than patients in the other regions examined in the study. Lowering length of stay could provide cost savings to the Buffalo-Niagara health care system. ÒThe higher length of stay most likely results from historical admission patterns and pressures to keep hospital inpatient beds filledó (emphasis added). The Buffalo health care system as a whole continues to lose substantial amounts of money. More specifically, in 1999, the system could have overcome its $3 million deficit if it had reduced average length of stay from 6.25 days [for all patients] to the Mid-Atlantic region median of 5.9 days. s could thereby have reduced variable costs by $69 million. If the average length of stay had been reduced to the average LOS of.6 days in Cleveland, Detroit, and Pittsburgh, the Buffalo area hospitals might have saved $16 million. This would have substantially reduced the financial pressures on the hospitals. To further illustrate the impact of long LOS, in 1999 the cost per patient day in Buffalo was $653, far lower than the average of $1,029 in eight comparison regions. But because of the long lengths of stay, both the cost and the revenue per discharge were well above the US average. This means employers and insurers paid more for hospital care, while hospital resources were stretched thin across a large number of days for each hospitalization. The staff was spread thinly across the patients in the hospital as well, with 3.3 inpatient FTEs per occupied bed. This level was the lowest among the eight comparison regions, which averaged.19 FTEs. Reduction of hospital stays would have allowed hospitals to focus scarce nursing staff on patients who most require care. No., September 2002, Page 5

6 Which patients have the longest lengths of stay? Older patients (65 and older) account for both the majority of admissions (59%) and the majority of ÒexcessÓ days (79%) that are spent in hospitals in the metropolitan counties in the Excellus region (Erie, Monroe, Onondaga, Oneida, and Broome). Excess days refer to the number of days spent in the hospital beyond the national average for that type of patient. However, the length of stay in Upstate New York hospitals for medical patients is higher than the U.S. average for both older and younger adults. The LOS is 13% higher than the national average for adults aged 18-6 and 30% higher for adults 65 and older. There is therefore a significant number of excess hospital days among the younger population as well. Many of the excess LOS days do occur among Medicare patients. In addition to concerns about clinical issues (see box on p. 3), these long lengths of stay can have broader implications. The federal government sets Medicare payments for inpatient care. The federal Balanced Budget Act (BBA) has moderated the growth in payments during the last few years. If hospitals incur excessive costs for this population, they may shift costs onto commercially insured patients. A corollary of this finding is that the length of stay in Upstate New York is particularly high for sicker patients (but not those with life-threatening illness). APR-DRGs, the system used to adjust for differences in patientsõ level of illness across hospitals (see description in box on page 11), divides patients into four categories based on how sick they are: minor, moderate, major, and severe (that is, life-threatening). The following graph shows how much the LOS for medical patients in Upstate NY exceeds the national average and the benchmark rate in Minneapolis/St. Paul. For example, a value of 13% for patients with major illness compared to Minneapolis/St. Paul means that the length of stay for this type of patients is 3% higher in Upstate New York than in Minneapolis/St. Paul. 150% 10% 130% 120% 110% 100% 90% 80% Upstate New York LOS as a percentage of the US average and best practice benchmark, by APR-DRG severity of illness category 111% 130% 115% 135% 123% 13% 110% How much could be saved by reducing average length of stay? 138% Minor Moderate Major Major Severe Severe Upstate NY compared to US average Upstate NY compared to Minneapolis Compared to the national average length of stay, Upstate New York hospitals in the Excellus region provided about 188,000 extra days of hospital care from mid-2000 to mid-2001, or 1% more. The cost of a hospital day is higher at the beginning of the stay than at the end. Reducing the average length of stay for adult medical patients by one day across the Excellus region would produce about $7 million in savings from hospital care. This value is based on a conservative estimate of the costs associated with the last day of a hospital stay of $250. Some of these potential savings would be offset by higher costs for other providers, such as home care agencies and nursing homes. For example, the cost for those No., September 2002, Page 6

7 Excess days patients who go to nursing homes can be estimated as around $150 per day. But the savings would still be substantial. Additional savings could be gained from reducing lengths of stay for other types of patients, such as those who have surgery, which are not covered in detail in this report. If the average length of stay in Upstate New York (5. days) were reduced to the more aggressive benchmark of Minneapolis/St. Paul (.0 days), 35,000 fewer days (or 26% of total volume) would be used, for a savings in hospital costs of about $86 million. The number of excess days compared to the U.S. average and the associated potential savings are shown below. The greatest potential for savings is found in Western New York, followed by Utica-Watertown, Central New York, the Rochester area, and the Southern Tier. The average lengths of stay across the five upstate regions are similar. Therefore, the difference in the number of excess days is driven primarily by variation in the number of admissions and in the size of the areaõs population. 80,000 60,000 0,000 20,000 0 Excess days* and associated costs in Upstate New York hospitals by Excellus region $16.1 million Western New York $8.0 million Rochester Area $.6 million Southern Tier $8.5 million Central New York $9.9 million Utica- Watertown Upstate New York hospitals have already made substantial progress in reducing length of stay Upstate New York hospitals made significant progress in reducing length of stay during the 1990s and should be commended. This decline was the result of concerted efforts by hospitals, physicians, and other providers. The largest decline occurred in the mid-1990s. For example, between 1995 and 1999, the Syracuse hospitals implemented an effort to reduce hospital lengths of stay through case management and cooperation with their medical staffs. These efforts resulted in a 20% decline in medical-surgical stays and a 17% decline in total stays. Across Upstate New York, between 1998 and 2001 (using January to September data), the average length of stay for medical patients declined 10% from 6.0 days to about 5. days. Data from the New York Department of Health show the average length of stay for medical patients hospitalized in each county. These data are not riskadjusted. Nevertheless, the broad patterns should be valid. They show that the average (unweighted) length of stay across 36 of the counties in the Excellus service area declined from 6. days in 199 to 5.3 days in 1997 and.9 days in 2000 (3 counties were excluded because they have no hospitals). This means that LOS declined 18% between 199 and 1997 and another 7.5% between 1997 and However, a recent newspaper article (Rochester Democrat & Chronicle, 3/2/02) reported that the average length of stay for all patients in Rochester rose between 2001 and These data are difficult to interpret, because of the closure of Genesee and the redistribution of its patients across the community and possibly between inpatient and outpatient care. *Excess days refer to the number of hospital days above the U.S. average for each type of patient. No., September 2002, Page 7

8 What can be done to reduce length of stay? Reducing hospital stays requires a community-wide effort. staff, physicians, home health agencies, nursing homes, and insurers must work together to make the health care system function more effectively in the interest of patients. Communities rightfully expect hospital emergency departments to provide care 2 hours a days/7 days a week. For hospitals to operate more effectively, case managers, physicians, home care agencies, and longterm care services must increase their availability and be more flexible. Their availability on weekends is particularly important, as data from Syracuse show that a disproportionate number of the longer lengths of stay occur among patients who are discharged on Monday. The availability of home care services and nursing home care is critically important. This is not merely a question of the supply of beds or home care aides but especially of how these resources are organized. The necessary services need to be available to care for patients when they leave the hospital, including home care, nursing homes, and rehabilitation services. s in New York have said that their job is made more difficult by the lack of subacute care units, that is, units that take care of patients whose needs fall between those of a hospital bed and a nursing home bed. The state examined this issue several years ago and concluded that a special payment category for subacute units was not warranted. This issue should be reexamined. Whether there is adequate access to all types of post-hospital care at the time they are needed should also be assessed. Clinical and social management of patients in postacute settingsñwhether in hospital-based subacute units or nursing homesñis also critically important. Simply moving patients from one setting to another without improving the management of care is not sufficient. The goal for most patients should be to return them to an appropriate community setting as quickly as clinically possible, to maximize the possibility of their living a more independent life. This is especially important for older adults. The following table shows the percentage of excess days (that is, the number of days exceeding the national average) accounted for by patients in each discharge category. 80% 60% 0% 20% 0% Percentage of excess days for patients in each discharge category, by HSA region 33 % 0 % 8 % 27 % 31 % 25 % 22 % Western New York 19 % 23 % 16 % This chart shows that the longest excess stays occur among patients discharged to nursing homes, followed in most regions by home care and self-care. 52 % 62 % 0 % 2 % 0 % 0 % Finger Lakes Central New York (incl. Utica) Binghamton Self-care Home care Nursing home Other No., September 2002, Page 8

9 Forty-seven percent of the excess days occur among patients who are transferred to nursing homes. The ÒotherÓ category consists primarily of patients who are transferred or die. These data do not indicate whether the delays are due to the hospitalõs slowness in getting patients ready for discharge or, for example, because the nursing home is not accepting patients at the right time or does not handle more complex cases. This can be ascertained by a closer examination of these issues by the providers involved. The regional distribution of days beyond the severityadjusted benchmarks indicates that there are different combinations of issues that generate extended stays. Discharges to nursing homes account for the largest numbers of excess days in all four regions. However, the impact of self-care and home care varies substantially. In the Buffalo area, excess days for self-care and home care combined exceed those of nursing homes. In the other regions, a larger percentage of the excess days occurred among those discharged to nursing homes. The importance of cooperation in reducing stays from physicians, hospital staff, and long-term care providers also needs to be underlined. Discharges to home care and nursing homes depend on cooperation from long-term care providers, but they also need proactive efforts by hospital staff and physicians to identify patients. Traditionally, hospitals have also assumed leadership for moving patients through the system. The hospitals in Syracuse have recently taken steps to address the availability of post-acute care. Because of the difficulties they sometimes encounter in placing patients in nursing homes, they have applied for permission from the state to build new nursing home beds. In response, the Syracuse nursing homes have agreed to work with the hospitals to develop the capacity to care for some of the most complex patients ready for nursing home placement, such as those on ventilators or needing intravenous medication. They have also agreed to work toward moving some of their patients to less intensive settings, when that is appropriate. The reduction in hospital lengths of stay is a challenging area that requires creative efforts by a variety of providers of care. Some Upstate New York hospitals have achieved notable reductions in length of stay under the current system, including hospitals that care for complex patients. s throughout the nation, including some in Upstate New York, have successfully applied a number of strategies. These include the following: Beginning discharge planning as soon as a patient enters the hospital. This requires a proactive approach, which involves the patient, family, physician, care management, and postdischarge services at the beginning of the stay, rather than the day before discharge. Addressing patientsõ needs simultaneously. Treating individual diagnoses in sequence may extend hospital stays unnecessarily. No., September 2002, Page 9

10 Stimulating changes in physician practice patterns, which can be challenging. One approach that has shown some success is providing physicians with feedback on their own performance compared to their peers. s need to develop data concerning hospital stays, physician practice, and the use of home care and nursing home services, if they have not already done so. These data need to be shared with all of those involved in the process to ensure that care is both effective and efficient. The Advisory Board, Inc., is a think tank/consulting firm in Washington, DC, that works closely with hospitals. A major area of focus for them this summer is improving patient throughput. The Advisory Board suggests a number of micro-level tactics to reduce length of stay including the following: Employing a physician discharge advisor. Bottom line s in Upstate New York have made considerable progress in reducing patientsõ length of stay. This is important because it reduces costs and, when done properly, can be beneficial clinically. Yet comparisons to the national average and to best practices in relatively similar areas such as Minneapolis/ St. Paul show that substantial opportunities for improvement remain. The potential financial savings from reducing length of stay are substantial. The reallocation of funds from unnecessarily long hospital stays can focus health care resources, such as nursing, on patients who are most in need of treatment. The monies saved could be used to bolster the financial health of hospitals whose services are needed in our communities. They can also help control the growth in health care premiums, which have a significant impact on access to care and on the viability of businesses and other employers in our communities. Providing weekend support services. Greater use of hospitalists who care for patients while they are in the hospital. No., September 2002, Page 10

11 Where do the length of stay data come from? The length-of-stay data were calculated from discharge data submitted by hospitals to the New York Department of HealthÕs Statewide Planning and Research Cooperative System (SPARCS). Data from mid-2000 to mid-2001 were used. s with fewer than 300 medical discharges were excluded. Some of these are facilities that care for special groups of patients, such as rehabilitation and alcohol abuse patients. Surgical patients were excluded out of concern that the substantial variation in the complexity of procedures performed at different hospitals might bias the comparison. Admissions for alcohol/substance abuse treatment, obstetrics, pediatrics, and psychiatry were excluded, as well as patients under 18 years old. The national average was obtained from National Discharge Survey, distributed by the 3M Health Information Systems, Wallingford, CT, and is for The data on Minneapolis/St. Paul were obtained from Minnesota and Healthcare Partnership, Minneapolis, MN, and are for Minneapolis/St. Paul was chosen because of its relatively low length of stay, generally high quality of care in the community, and a climate that is similar to Upstate New YorkÕs. The data on Sacramento were obtained from the Office of Statewide Health Planning and Development, Sacramento, CA, and are for Length of stay is affected by the severity of patientsõ medical condition. To take into account possible differences in patientsõ health when comparing data from different regions, lengths of stays are adjusted for patientsõ diagnoses and severity of illness using APR-DRGs (All Patient Refined Diagnosis-Related Groups) from 3M. APR-DRGs are a risk adjustment methodology used to compare data about hospitalizations between hospitals. APR-DRGs use discharge abstracts to extract base DRGs and add complexity sub-scales (1=minor, 2=moderate, 3=major, =extreme) to reclassify them into 1,528 sub-classes. Data are reported for the Excellus region, which includes all of Upstate New York except for Albany. For some measures, separate data are presented for Western New York (Buffalo area), the Rochester area, the Southern Tier, Central New York (Syracuse area), and Utica-Watertown. The hospitals included in each area are listed in the tables at the end of this report. The data for this report were provided by the Executive Council, the cooperative planning organization in Syracuse for Community-General, Crouse, St. JosephÕs Health Center, and University -SUNY Upstate Medical University. The CouncilÕs Executive Director, Ronald Lagoe, PhD, coauthored this report. Copies of more detailed data on each hospital, with information on the length of stay for each severity category (minor, moderate, major, and severe), can be obtained by contacting Excellus Health Plan, Inc. The contact information is listed on the last page. No., September 2002, Page 11

12 Risk-Adjusted Length of Stay (July 2000 June 2001) Western New York Bertrand Chaffee Westfield Memorial Tri-County Memorial MINNEAPOLIS/ST PAUL (1999) Jones Memorial Children's Sisters of Charity Kenmore Mercy US AVERAGE (1999) Brooks Memorial Woman's Christian Assoc Millard Fillmore Suburban Hosp Degraff Memorial Lake Shore Inter-Community Regional Hosp Medina Memorial Olean General Roswell Park Cancer Inst Mercy St. Joseph's Mount St. Mary's Millard Fillmore Erie County Med Ctr Lockport Memorial Niagara Falls Mem Med Ctr Sheehan Memorial Buffalo General United Memorial SOURCE: SPARCS, compiled by the Executive Council in Syracuse. The sources of the national average and benchmark data are listed on p. 11. The data were risk-adjusted using 3M s APR-DRGs. The best practice benchmark is Minneapolis/St. Paul, MN. DRGs for surgery, alcohol/substance abuse treatment, obstetrics, pediatrics, and psychiatry, as well as patients under 18 years old were excluded. s with fewer than 300 adult medical admissions were also omitted. No., September 2002, Page 12

13 Risk-Adjusted Length of Stay (July 2000 June 2001) Rochester Myers Community MINNEAPOLIS/ST PAUL (1999) Soldiers & Sailors Mem Clifton Springs NH Noyes Memorial US AVERAGE (1999) Strong Memorial Geneva General FF Thompson Park Ridge 5.3 Highland 5. Wyoming County Community 5.6 Southern Tier St. James Mercy 3.7 Rochester General Newark-Wayne Community Lakeside Memorial Genesee Ira Davenport Memorial MINNEAPOLIS/ST PAUL (1999) Schuyler.5 Chenango Memorial.7 US AVERAGE (1999).7 Corning 5.1 Wilson 5.2 Our Lady of Lourdes 5. Binghamton General St. Joseph s Arnot Ogden Med Ctr SOURCE: SPARCS, compiled by the Executive Council in Syracuse. The sources of the national average and benchmark data are listed on p.11. The data were risk-adjusted using 3M s APR-DRGs. The best practice benchmark is Minneapolis/St. Paul, MN. DRGs for surgery, alcohol/substance abuse treatment, obstetrics, pediatrics, and psychiatry, as well as patients under 18 years old were excluded. s with fewer than 300 adult medical admissions were also omitted. No., September 2002, Page 13

14 Risk-Adjusted Length of Stay (July 2000 June 2001) Central New York MINNEAPOLIS/ST PAUL (1999) Cayuga Medical Center.2 US AVERAGE (1999).7 St. Joseph's Hosp Health Ctr 5.1 Crouse 5. Cortland Memorial University -- SUNY UMU Community General Oswego 6.1 Utica-Watertown AL Lee Memorial Auburn Memorial Moses-Ludington Lewis County General Canton-Potsdam MINNEAPOLIS/ST PAUL (1999) Mary Imogene Bassett.1 EJ Noble St. Lawrence. AO Fox Memorial.6 Community Memorial.7 Massena Memorial.7 US AVERAGE (1999).7 Amsterdam Memorial.8 Alice Hyde Memorial 5 St. Elizabeth's 5 Nathan Littauer 5.1 Rome Memorial 5.1 Adirondack Medical Center 5.3 Carthage Area 5. Hepburn Medical Center 5. Oneida Health Systems 5.7 St. Mary's Amsterdam 5.9 Faxton 5.9 Samaritan Medical Center 6.1 Champlain Valley Physician's 6.1 St. Luke's Memorial 6.2 Little Falls SOURCE: SPARCS, compiled by the Executive Council in Syracuse. The sources of the national average and benchmark data are listed on p.11. The data were risk-adjusted using 3M s APR-DRGs. The best practice benchmark is Minneapolis/St. Paul, MN. DRGs for surgery, alcohol/substance abuse treatment, obstetrics, pediatrics, and psychiatry, as well as patients under 18 years old were excluded. s with fewer than 300 adult medical admissions were also omitted. No., September 2002, Page 1

15 Excellus Health Policy Reports This report is the fourth in a series that is being developed in response to recommendations from local leaders and the public to educate the community on the issues of cost, access, and quality of health care in Upstate New York. These reports are intended to address relevant health policy issues facing our communities. No. 1. Uninsured rates in Upstate New York: Lower than U.S. but rising health care premiums pose a threat No. 2. Drivers of premium increases in upstate New York: Benchmarking against national performance No. 3. Insurance coverage for experimental treatments: New YorkÕs External Review Law should encourage enrollment in randomized controlled clinical trials. More detailed information on this topic is available in the Excellus Health Policy White Paper, A Rational Clinical Trial Payment Policy: Lessons from the Use of Bone Marrow Transplantation for Metastatic Breast Cancer (No. 1, May 2001). No.. Average length of stay in Upstate New York hospitals: Opportunities for savings Copies of these reports can be downloaded from the Excellus Web site at If you would like to be added to the mailing list for future Excellus Health Policy Reports or if you have any comments, please contact Health Policy Director, Excellus Health Plan, Inc., 165 Court Street, Rochester, NY 167, EHPR@excellus.com. No., September 2002, Page 15

16 Excellus, Inc. Excellus, Inc., headquartered in Rochester, is a holding company that finances and delivers health care services across 5 counties in Upstate New York. With $ billion in annual revenues and more than 6,700 employees, the Excellus companies insure 2.15 million people. Through our subsidiaries, we provide long-term care insurance across the United States, serve as a regional leader in employee benefits consulting and administrative services, and provide medical or home health care to 90,000 people in five Rochester-area counties. The Excellus name stands for dignified access to high-quality, affordable coverage. Excellus Companies Contact Information BlueCross BlueShield of Central New York 165 Court Street BlueCross BlueShield of the Rochester Area Rochester, New York 167 BlueCross BlueShield of Utica-Watertown (585) Excellus Benefit Services MedAmerica Genesee Region Home Care Lifetime Health RMSCO Support Services Alliance (SSA) Univera Healthcare of Western New York No., September 2002, Page 16