Randomization and treatment of study patients. Adapted from Lederle et al.5 AAA indicates abdominal aortic aneurysm.
Bootstrap replications show the differences in costs, life-years, and quality-adjusted life-years (QALYs) on the cost-effectiveness plane between patients randomized to endovascular or open repair of abdominal aortic aneurysm (AAA). Differences are calculated as the endovascular repair group finding minus the open repair group finding. The large dot indicates the point estimate from the study. Solid diagonal line indicates willingness to pay $50 000 per life-year or QALY; dashed diagonal line, willingness to pay $100 000 per life-year or QALY.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Lederle FA, Stroupe KT, Kyriakides TC, Ge L, Freischlag JA, for the Open vs Endovascular Repair (OVER) Veterans Affairs Cooperative Study Group. Long-term Cost-effectiveness in the Veterans Affairs Open vs Endovascular Repair Study of Aortic Abdominal Aneurysm: A Randomized Clinical Trial. JAMA Surg. 2016;151(12):1139–1144. doi:10.1001/jamasurg.2016.2783
Copyright 2016 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
How do elective open and endovascular repair of abdominal aortic aneurysm (AAA) compare in cost-effectiveness?
In this multicenter randomized clinical trial with follow-up to 9 years, survival, quality of life, costs, and cost-effectiveness did not differ between elective open and endovascular repair of AAA.
For patients with AAA who are candidates for both procedures, selection of either procedure remains reasonable and can be guided by patient and physician preference.
Because of the similarity in clinical outcomes after elective open and endovascular repair of abdominal aortic aneurysm (AAA), cost may be an important factor in choosing a procedure.
To compare total and AAA-related use of health care services, costs, and cost-effectiveness between groups randomized to open or endovascular repair.
Design, Setting, and Participants
This unblinded randomized clinical trial enrolled 881 patients undergoing planned elective repair of AAA who were candidates for open and endovascular repair procedures. Patients were randomized from October 15, 2002, to April 15, 2008, at 42 Veterans Affairs medical centers. Follow-up was completed on October 15, 2011, and data were analyzed from April 15, 2013, to April 15, 2016, based on intention to treat.
Main Outcomes and Measures
Mean total and AAA-related health care cost per life-year and per quality-adjusted life-year (QALY).
A total of 881 patients (876 men [99.4%]; 5 women [0.6%]; mean [SD] age, 70 [7.8] years) were included in the analysis. After a mean of 5.2 years of follow-up, mean life-years were 4.89 in the endovascular group and 4.84 in the open repair group (P = .68), and mean QALYs were 3.72 in the endovascular group and 3.70 in the open repair group (P = .82). Total mean health care costs did not differ significantly between the 2 groups (endovascular group, $142 745; open repair group, $153 533; difference, −$10 788; 95% CI, −$29 796 to $5825; P = .25). Costs related to AAA, including the initial repair, constituted nearly 40% of total costs and did not differ significantly between the 2 groups (endovascular group, $57 501; open repair group, $57 893; difference, −$393; 95% CI, −$12 071 to $7928; P = .94). Lower costs due to shorter hospitalization for initial endovascular repair were offset by increased costs from AAA-related secondary procedures and imaging studies. The probability of endovascular repair being less costly and more effective was 56.8% when effectiveness was measured in life-years and 55.4% when effectiveness was measured in QALYs for total costs and 31.3% and 34.3%, respectively, for AAA-related costs.
Conclusions and Relevance
In this multicenter randomized clinical trial with follow-up to 9 years, survival, quality of life, costs, and cost-effectiveness did not differ between elective open and endovascular repair of AAA.
clinicaltrials.gov Identifier: NCT00094575
Each year, elective procedures to repair abdominal aortic aneurysm (AAA) are performed in 40 000 individuals in the United States.1 Perioperative deaths occur in approximately 1250 of these AAA repairs, which constitutes the second most common cause of death among patients undergoing vascular or general surgery procedures after colectomy.2 Endovascular repair, introduced in the 1990s as a less invasive alternative to open repair, therefore generated great interest. Several randomized clinical trials were undertaken to compare the 2 types of repair. These trials found longer-term outcomes to be similar for both procedures, although perioperative mortality was generally reported to be lower with endovascular repair.3-5 With no clear preference between procedures based on long-term clinical outcomes, economic costs become more important, especially considering that the cost of endovascular grafts is sufficiently high to have caused most previous investigators3,6,7 to conclude that endovascular repair is not cost-effective.
The Department of Veterans Affairs (VA) Open vs Endovascular Repair (OVER) trial was one of several multicenter randomized clinical trials that compared the 2 methods of elective AAA repair and the only one, to our knowledge, conducted in North America. Previous reports compared clinical outcomes to 9 years after open and endovascular repair5 and total cost-effectiveness to 2 years.8 In this report, we compare total and AAA-related use of health care resources, costs, and cost-effectiveness for the randomized groups to the end of the trial with 9 years of follow-up.
The OVER trial design, clinical outcomes, and the methods used to assess use of health care resources, costs, and cost-effectiveness have been described previously.5,8,9 Briefly, 881 patients who had planned elective repair of AAA and were candidates for both procedures were randomized to endovascular or open repair from October 15, 2002, to April 15, 2008, in 42 VA medical centers (listed at the end of the article) and followed up until October 15, 2011. Follow-up visits were scheduled 1 month after aneurysm repair, 6 and 12 months after enrollment, and then yearly. The trial protocol (available in the Supplement) specified that follow-up visits after endovascular repair included computed tomography and plain radiography of the abdomen, whereas after open repair, only computed tomography at 1 year and at the end of the study was stipulated. This study was approved by central human rights committee and the institutional review boards at each participating center. All patients provided written informed consent.
For this analysis, performed from April 15, 2013, to April 15, 2016, we linked the OVER clinical trial database to several VA and Medicare databases to examine the use of health care resources, costs, and outcomes in life-years and quality-adjusted life-years (QALYs). Secondary therapeutic procedures were defined as in the clinical reports5,9 as separate trips to the procedure suite and could occur during the admission for the initial procedure. All costs were adjusted to 2008 US dollars with the Consumer Price Index.10 Methods for calculating costs and assessing cost-effectiveness have been described previously.8 The present study reports use of health care resources, costs, and cost-effectiveness to the end of follow-up (October 15, 2011). Because of the substantial amount of health care received that was unrelated to AAA repair in this long-term analysis, we report total and AAA-related use of resources and costs. The AAA-related events included (1) AAA repair and secondary therapeutic procedures from the previous clinical reports,5,9 (2) procedure codes indicating abdominal computed tomography or ultrasonography, (3) hospitalizations and outpatient visits specified by site investigators as AAA related during the trial, and (4) encounters associated with diagnostic and procedure codes related to aortic conditions in the VA and Medicare databases.
Effectiveness was measured in life-years after randomization and in QALYs, which incorporate health-related quality of life and medical outcomes into a single measure. Health-related quality of life was assessed using the EuroQol 5-dimensions questionnaire, a brief questionnaire designed for this purpose, as previously described.8
The number of events per 100 person-years of follow-up was compared between randomized groups, assuming a Poisson distribution, and 95% CIs were calculated. We calculated 95% CIs around differences in mean costs using bias-corrected accelerated bootstrapping to adjust for skewing in the data. We estimated P values using unpaired 2-tailed t tests.
Of the 881 patients (876 men [99.4%]; 5 women [0.6%]; mean [SD] age, 70 [7.8] years), 843 patients (95.7%) had undergone the specified repair, and vital status was confirmed for all patients at the end of the study.1 Randomization and analysis are depicted in Figure 1. Of 589 patients who were alive at the end of the study, 546 (92.7%) came to the study clinic during the final year, and databases and telephone calls were used to obtain follow-up for secondary outcomes on another 36 (6.1%). During the study, 4278 of 5046 possible EuroQol 5-dimensions questionnaires (84.8%) were completed. As previously reported,1 the primary outcome of all-cause mortality occurred in 146 patients in each group, and the hazard ratio for endovascular to open procedures was 0.97 (95% CI, 0.77-1.22; P = .81). Mortality was significantly lower with endovascular repair at 30 days after surgery and 2 and 3 years after randomization but not thereafter.1
At the end of the study, after a mean of 5.2 years of follow-up, mean life-years were 4.89 in the endovascular group and 4.84 in the open repair group (P = .68), and mean QALYs were 3.72 in the endovascular group and 3.70 in the open repair group (P = .82). Total mean health care costs did not differ significantly between the 2 groups (endovascular group, $142 745; open repair group, $153 533; difference, −$10 788; 95% CI, −$29 796 to $5825; P = .25) (Table). Costs related to AAA, including the initial repair, constituted nearly 40% of total costs and also did not differ significantly between the 2 groups (endovascular group, $57 501; open repair group, $57 893; difference, −$393; 95% CI, −$12 071 to $7928; P = .94). As noted previously2 and presented in the Table, costs for the hospitalization for the initial repair were significantly lower for the endovascular repair group; this reduction was related to the shorter hospital stay and occurred despite the high cost of the endovascular grafts. By the end of the study, however, the endovascular group had more AAA-related outpatient visits, secondary procedures, and imaging studies and, thus, higher costs for these items and especially for the last two, resulting in nearly identical total AAA-related costs (Table).
We conducted bootstrap analyses to characterize the precision of our estimates of cost-effectiveness using cost-effectiveness planes as in previous reports.7,8 Two thousand samplings (with replacement) of the 881 trial participants were performed, as shown in Figure 2. The probability that endovascular repair would be less costly and more effective was 56.8% when effectiveness was measured in life-years (Figure 2A) and 55.4% when it was measured in QALYs (Figure 2B) for total costs. These proportions dropped to 31.3% and 34.3%, respectively, for AAA-related costs (Figure 2C and D).
The observations below the diagonal lines favor endovascular repair if one is willing to pay $50 000 or $100 000 in total costs per life-year or per QALY. Using total costs (Figure 2A and B), if one is willing to pay $50 000, endovascular repair is preferred in 88.4% of observations using life-years and 88.8% using QALYs; if one is willing to pay $100 000, the proportions are 84.8% and 85.5%, respectively. Using AAA-related costs (Figure 2C and D), if one is willing to pay $50 000, endovascular repair is preferred in 59.9% of observations using life-years and 60.7% using QALYs, and if one is willing to pay $100 000, the proportions are 62.2% and 64.4%, respectively.
In this multicenter randomized clinical trial, survival, quality of life, and costs were not significantly different between elective open and endovascular AAA repair after a mean of 5.2 years of follow-up. The lower cost of the hospitalization for initial repair in the endovascular repair group was offset by higher follow-up costs for AAA-related imaging studies and secondary procedures, so that by the end of the study, AAA-related costs were nearly identical in the 2 groups.
A 2014 cost-effectiveness analysis based on the randomized clinical trials of open vs endovascular repair11 emphasized the difference between the OVER trial and the 3 European trials. In the European trials, costs associated with the longer hospital stay for open repair did not offset the high price of the endovascular grafts. All of these costs are higher in the United States.
Our study has several limitations that have been described previously.8 Primarily, our VA accounting methods may not apply directly to other populations, and our protocol required some deviations from usual practice, such as computed tomography 1 year after open repair.
Endovascular repair has lower perioperative mortality, shorter recovery time, and, at least in the US VA system, lower initial costs, despite the high cost of the grafts. However, we found that these advantages did not persist during the next 3 to 5 years. On the other hand, endovascular repair requires more long-term follow-up and carries a small risk for late rupture.1,5,12 Our results indicate that for patients who are candidates for both procedures, selection of either procedure remains reasonable and can be guided by patient and physician preference.
Corresponding Author: Frank A. Lederle, MD, Department of Medicine (III-0), Veterans Affairs Medical Center, One Veterans Drive, Minneapolis, MN 55417 (email@example.com).
Accepted for Publication: May 27, 2016.
Published Online: September 14, 2016. doi:10.1001/jamasurg.2016.2783
Author Contributions: Drs Lederle and Stroupe had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Lederle, Stroupe, Kyriakides.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Lederle, Stroupe, Freischlag.
Critical revision of the manuscript for important intellectual content: Lederle, Stroupe, Kyriakides, Ge.
Statistical analysis: Stroupe, Ge.
Obtaining funding: Lederle, Kyriakides.
Administrative, technical, or material support: Kyriakides, Freischlag.
Study supervision: Lederle.
Conflict of Interest Disclosures: Principal investigators are employees of their respective VAMCs. Site coordinators are paid by the study. No other disclosures were reported.
Funding/Support: This study was supported by the Cooperative Studies Program of the Department of Veterans Affairs Office of Research and Development.
Role of the Funder/Sponsor: The funding source approved the design, established procedures for the conduct of the study, and approved submission of the manuscript for publication. It had no role in the collection, management, analysis, and interpretation of the data and preparation or review of the manuscript.
Group Information: The Open Vs Endovascular Repair (OVER) Veterans Affairs (VA) Cooperative Study Group includes the following investigators and site coordinators at participating VA Medical Centers (VAMCs): Erika R. Ketteler, MD, Darra D. Kingsley, MD, John M. Marek, MD, Richard J. Massen, MD, Brian D. Matteson, MD, J. David Pitcher, MD, Mark Langsfeld, MD, John D. Corson, MD, James M. Goff Jr, MD, Karthnik Kasirajan, MD, Christina Paap, RN, and Diane C. Robertson, RN (Albuquerque, New Mexico); Atef Salam, MD, Ravi Veeraswamy, MD, Ross Milner, MD, Karthikeshwar Kasirajan, MD, and Jane Guidot, RN (Atlanta, Georgia); Brajesh K. Lal, MD, Steven J. Busuttil, MD, Michael P. Lilly, MD, Melita Braganza, Kea Ellis, RN (Baltimore, Maryland); Mark A. Patterson, MD, William D. Jordan, MD, David Whitley, MD, Steve Taylor, MD, Marc Passman, MD, Donna Kerns, RN, Cindy Inman, RN, and Jennifer Poirier, RN (Birmingham, Alabama); James Ebaugh, MD, Joseph Raffetto, MD, David Chew, MD, Subhash Lathi, MD, Christopher Owens, MD, and Kathleen Hickson, RN (Boston, Massachusetts); Hasan H. Dosluoglu, MD, and Karen Eschberger, RN (Buffalo, New York); Melina R. Kibbe, MD, Henry M. Baraniewski, MD, Jon Matsumura, MD, Michelle Endo, Anna Busman, RN, Wendy Meadows, RN, and Mary Evans, RN (Chicago, Illinois); Joseph S. Giglia, MD, Hosam El Sayed, MD, Amy B. Reed, MD, Madeline Ruf, RN, and Stephanie Ross, RN (Cincinnati, Ohio); Jessie M. Jean-Claude, MD, Gilles Pinault, MD, Preet Kang, MD, Nadine White, RN, Matthew Eiseman, RN, and Reba Jones, RN† (Cleveland, Ohio); Carlos H. Timaran, MD, J. Gregory Modrall, MD, M. Burress Welborn III, MD, Jorge Lopez, MD, and Tammy Nguyen, MT (ASCP) (Dallas, Texas); John K. Y. Chacko, MD, Kenneth Granke, MD, Angela G. Vouyouka, MD, Erin Olgren, Prakash Chand, MD, Brenda Allende, Michael Ranella, and Claudia Yales, RN (Detroit, Michigan); Thomas A. Whitehill, MD, William C. Krupski, MD†, Mark R. Nehler, MD, Stephen P. Johnson, MD, Darrell N. Jones, PhD, Pamela Strecker, RN, and Michelle A. Bhola, RN (Denver, Colorado); Cynthia K. Shortell, MD, John L. Gray, MD, Jeffrey H. Lawson, MD, Richard McCann, MD, Mark W. Sebastian, MD, Jean Kistler Tetterton, NP, Carla Blackwell, RN, Patricia A. Prinzo, BS, and Nina Lee, RN (Durham, North Carolina); Frank T. Padberg Jr, MD, Joaquim J. Cerveira, MD, Brajesh K. Lal, MD, Robert W. Zickler, MD, and Karen A. Hauck, RN (East Orange, New Jersey); Scott A. Berceli, MD, W. Anthony Lee, MD, C. Keith Ozaki, MD, Peter R. Nelson, MD, Anne S. Irwin, RN, and Randy Baum, RN (Gainesville, Florida); Bernadette Aulivola, MD, Heron Rodriguez, MD, Fred N. Littooy, MD, Howard Greisler, MD, and Mary T. O’Sullivan, RN (Hines, Illinois); Panagiotis Kougias, MD, Peter H. Lin, MD, Ruth L. Bush, MD, Gene Guinn, MD, Carlos Bechara, MD, Catherine Cagiannos, MD, George Pisimisis, MD, Neal Barshes MD, Sherilyn Pillack, RN, and Barbara Guillory, RN (Houston, Texas); Dolores Cikrit, MD, Stephen G. Lalka, MD, Gary Lemmon, MD, Ryan Nachreiner, MD, Mitzi Rusomaroff, RN, and Elaine O’Brien, RN (Indianapolis, Indiana); Joseph J. Cullen, MD, Jamal Hoballah, MD, W. John Sharp, MD, Jeanne L. McCandless, RN, and Vickie Beach, RN (Iowa City, Iowa); David Minion, MD, Thomas H. Schwarcz, MD, Joy Kimbrough, RN, Laura Ashe, Anna Rockich, MS, and Jill Warner-Carpenter, RN (Lexington, Kentucky); Mohammed Moursi, MD, John F. Eidt, MD, and Sandra Brock, RN (Little Rock, Arkansas); Christian Bianchi, MD, and Vicki Bishop, RN (Loma Linda, California); Ian L. Gordon, MD, Roy Fujitani, MD, Stephen M. Kubaska III, MD, Mina Behdad, MD, Reza Azadegan, MD, Christine Ma Agas, MPH, and Kathy Zalecki (Long Beach, California); John R. Hoch, MD, Sandra C. Carr, MD, Charles Acher, MD, Margaret Schwarze, MD, Girma Tefera, MD, Matthew Mell, MD, Beth Dunlap, RN, and Janice Rieder, RN (Madison, Wisconsin); John M. Stuart, MD, Darryl S. Weiman, MD, Omran Abul-Khoudoud, MD, H. Edward Garrett, MD, Sandra M. Walsh, MA, and Karen L. Wilson, RN (Memphis, Tennessee); Gary R. Seabrook, MD, Robert A. Cambria, MD, Kellie R. Brown, MD, Brian D. Lewis, MD, Susan Framberg, RN, and Christa Kallio, RN (Milwaukee, Wisconsin); Roderick A. Barke, MD, Steven M. Santilli, MD, Alexandre C. d’Audiffret, MD, Nancy Oberle, RN, Catherine Proebstle, NP, and Lauri Lee Johnson, RN (Minneapolis, Minnesota); Glenn R. Jacobowitz, MD, Neal Cayne, MD, Caron Rockman, MD, Mark Adelman, MD, Paul Gagne, MD, Matthew Nalbandian, MD, and Leah J. Caropolo, BS (New York, New York); Iraklis I. Pipinos, MD, Jason Johanning, MD, Thomas Lynch, MD, Holly DeSpiegelaere, RN, and Georgia Purviance, RN (Omaha, Nebraska); Wei Zhou, MD, Ronald Dalman, MD, Jason T. Lee, MD, Bassem Safadi, MD, Sheila M. Coogan, MD, Sherry M. Wren, MD, Doghdoo D. Bahmani, Deanna Maples, NP, and Shawna Thunen, RN (Palo Alto, California); Michael. A. Golden, MD, Marc E. Mitchell, MD, Ronald Fairman, MD, and Sally Reinhardt, RN (Philadelphia, Pennsylvania); Mark A. Wilson, MD, Edith Tzeng, MD, Satish Muluk, MD, Nina M. Peterson, RN, and Maria Foster, RN (Pittsburgh, Pennsylvania); James Edwards, MD, Gregory L. Moneta, MD, Gregory Landry, MD, Lloyd Taylor, MD, Richard Yeager, MD, and Eleanor Cannady, RN (Portland, Oregon); Gerald Treiman, MD, Stephanie Hatton-Ward, RN, and Barbara Salabsky, RN† (Salt Lake City, Utah); Nikhil Kansal, MD, Erik Owens, MD, Melanie Estes, RN, Beth A. Forbes, RN, and Cinda Sobotta, RN (San Diego, California); Joseph H. Rapp, MD, Linda M. Reilly, MD, Sandra L. Perez, NP, Kimberly Yan, Rajaabrata Sarkar, MD, Shelley S. Dwyer, RN, Sandra Perez, NP, and Karen Chong (San Francisco, California); Ted R. Kohler, MD, Thomas S. Hatsukami, MD, David G. Glickerman, MD, Michael Sobel, MD, Thomas S. Burdick, MD, Kimberly Pedersen, RN, and Patricia Cleary, NP (Seattle, Washington); Nikhil Kansal, MD, Erik Owens, MD, Melanie Estes, RN, Beth A. Forbes, RN, and Cinda Sobotta, RN (San Diego, California); Martin Back, MD, Dennis Bandyk, MD, Brad Johnson, MD, Murray Shames, MD, Rebecca L. Reinhard, RN, and Sandra C. Thomas, RN (Tampa, Florida); Glenn C. Hunter, MD, Luis R. Leon Jr, MD, Alex Westerband, MD, Robert J. Guerra, MD, Macario Riveros, MD, John L. Mills Sr, MD, John D. Hughes, MD, Andrea M. Escalante, RN, Shemuel B. Psalms, and Nancy N. Day, RN (Tucson, Arizona); Robyn Macsata, MD, Anton Sidawy, MD, Jonathan Weiswasser, MD, Subodh Arora, MD, and Brenda J. Jasper, MEd (Washington, DC); Alan Dardik, MD, Vivian Gahtan, MD, Bart E. Muhs, MD, Bauer E. Sumpio, MD, Richard J. Gusberg, MD, Marcelo Spector, MD, Jeffrey Pollak, MD, John Aruny, MD, E. Lynne Kelly, MD, James Wong, MD, Penny Vasilas, RN, and Carmelene Joncas, RN (West Haven, Connecticut); and Hugh A. Gelabert, MD, Christian DeVirgillio, MD, David A. Rigberg, MD, and Loretta Cole, RN (West Los Angeles, California); Executive Committee: Frank A. Lederle, MD (cochair), Julie A. Freischlag, MD (cochair), Ted R. Kohler, MD, Elizabeth Latts, MSW, Jon Matsumura, MD, Frank T. Padberg Jr, MD, Tassos C. Kyriakides, PhD, and Kathleen M. Swanson, MS, RPh; Cooperative Studies Program Coordinating Center VA Connecticut Healthcare System, West Haven: Gary R. Johnson, MS (acting director), Peter Guarino, MPH, PhD, Peter Peduzzi, PhD, Margaret Antonelli (assistant director of operations), Cindy Cushing (programmer), Elizabeth Davis (research coordinator), Lynn Durant (quality assurance officer), Shirley Joyner, MHA (research coordinator), Alice Kossack† (research coordinator), Tassos C. Kyriakides, PhD (biostatistician), Mary LeGwin, MS, Vanessa McBride (research coordinator, project manager), Terry O’Connor, PhD (biostatistician), June Poulton, RN (research coordinator), Susan Stratton† (project manager), and Scott Zellner, PhD (research coordinator); Regulatory Affairs and Clinical Compliance Section, Cooperative Studies Program Clinical Research Pharmacy Coordinating Center: Amanda J. Snodgrass, PharmD, Julie Thornton, BS, and Kathleen M. Swanson, MS, RPH; VA Research Site Management and Review Team: Clair M. Haakenson, RPH, MS; Kevin T. Stroupe, PhD (health economist), Center for Management of Complex Chronic Care, VAMC, Hines; Yvonne Jonk, PhD, Center for Chronic Disease Outcomes Research, VAMC, Minneapolis; Data and Safety Monitoring Board: John (Jeb) W. Hallett, MD, Norman Hertzer, MD, Jonathan Towne, MD, David A. Katz, MD, Theodore Karrison, PhD (biostatistician), and John P. Matts, PhD (biostatistician); Human Rights Committee, West Haven: Richard Marottoli, MD, Stanislav Kasl, PhD, Rajni Mehta, MPH, Richard Feldman, William Farrell, Heather Allore, PhD, Edward Perry, MD, James Niederman, MD, Sister Frances Randall, Reverend Michael Zeman, and Donald Beckwith†; Central Administration, Cooperative Studies Program, VA Central Office: Timothy J. O’Leary, MD, PhD, Grant D. Huang, MPH, PhD; and National Project Coordinators: Elizabeth Latts, MSW†, and Marilyn Bader, VAMC, Minneapolis.
Create a personal account or sign in to: