VA's Own Peer Review Process Finds that Failure to Supervise Residents was Responsible for 63 Malpractice Cases From 1997-2002

At 4 VA medical centers (St. Louis, Birmingham, Nashville, and New Orleans) we attempted to locate attending physicians in 29 clinics to determine if there was adequate resident supervision. We found that the attending physicians were not present to supervise the residents’ treatments of patients in 6 of the 29 clinics (see page 25). One resident told us that the attending physician would occasionally come to the clinic to see how things were going, but usually would not stay. Another resident said that the attending physician rarely came to the clinic, but made a brief appearance a couple of weeks ago. Generally, the attending physician returned pager messages within an hour. Another resident told us that the attending physician did not come to the clinic unless there was a case the resident decided the attending needed to review, but the attending was always available by phone or pager.
The attending physicians at one VA medical center told us that attending physicians saw every patient at the university because Medicare and private medical insurance carriers would not pay for care provided only by residents. The residents at another VA medical center told us they ran clinics at the VA medical center but at the university they only ran clinics for the indigent population.
From FY 1997 through the second quarter of FY 2002, the Federal Government paid, on behalf of VA, at least $21 million for 63 malpractice cases where VA’s peer review panel found that the attending VA physicians provided substandard resident supervision. Based on our review of available documentation, the attending physicians were not present to supervise the residents during the performance of a procedure or the provision of a treatment to a veteran in at least eight cases resulting in malpractice settlements totaling $4.7 million. An additional pending case involves an attending surgeon who could not provide needed assistance to a VA medical center patient because he was operating on a non-veteran patient at the affiliated medical school.

Armed Forces Institute of Pathology Report on Malpractice at the VA

From the Office of the Associate Chief Medical Director for Quality Management and the Office of the General Counsel, Department of Veterans Affairs, and the Department of Legal Medicine, Armed Forces Institute of Pathology
DEPARTMENT OF VETERANS AFFAIRS - ANALYSIS OF MEDICAL MALPRACTICE CLAIMS - AN INITIAL REPORT
by RICHARD L. GRANVILLE, MD, JD, MICHAEL GREENBLATT, MPA, DOUGLAS BRADSHAW, JD, GALEN BARBOUR, MD, and FRANK T. FLANNERY, M.D., J.D., LTC, MC, USA

    The authors wish to acknowledge the special assistance o f Mr. Dennis Snyder and Ms. Loretta Hunter in compiling the data for this article.
    The Department of Veterans Affairs (VA) operates one of the nation’s largest medical systems. It is composed of 172 hospitals and more than 350 outpatient facilities. Teaching affiliations are maintained with 160 medical and dental schools nationwide. Professional providers within the VA include 7,324 full-time and 5,921 part time staff physicians, 15,020 resident physicians, and 39,355 nurses.1,2 In FY 1991, there were 951,112 inpatients and 21,932,426 outpatients treated at VA medical facilities.

VA TORT CLAIM INFORMATION SYSTEM

    As with other large medical systems, recent years have witnessed increasing attention directed toward the VA medical malpractice claims experience. In 1985, the VA Office of the Inspector General (OIG) released a report recommending that the VA Department of Medicine and Surgery perform an extensive analysis of those medical conditions and surgical procedures that had resulted in malpractice claims. In March 1987, a subsequent OIG audit noted that the VA had initiated a number of monitors regarding medical malpractice claims but, due to resource constraints, a comprehensive analysis of the characteristics of those claims had not yet been accomplished. In March 1988, at a Congressional oversight committee hearing, Senator John Glenn (DOhio) emphasized the importance of utilizing data generated from the analysis of malpractice claims for purposes of monitoring and improving VA medical care. At that time, the Office of the General Counsel of the VA and that of the Medical Inspector combined efforts to develop the Tort Claims Information System (TCIS).3 Initial procedures regarding TCIS were promulgated in August 1988.4 Each VA medical center was directed to forward reports regarding any medical malpractice claim filed to the Office of the Medical Inspector. These reports were to include a summary of the incident involved, identification of responsible providers, relevant medical records, and available quality assurance documents. The Medical Inspector’s staff was to collect these reports, conduct individual case reviews, and assemble aggregate malpractice data for the respective medical centers and for the system, on a regional and a national basis.

     At first, these case reviews were restricted to the analysis of malpractice claims that were legally closed or finalized because there was no statutory protection for the confidentiality of the peer review information contained within submitted reports. By May 1990, when TCIS gained statutory protection for that confidentiality as medical quality assurance (38 U.S.C. § 3301), 1700 legally finalized malpractice cases had been reviewed. Thereafter, VA medical facilities were directed to report malpractice claims filed and to include a completed peer review analysis of each case.5 Responsibility for TCIS within the VA was transferred from the Office of the Medical Inspector to that of the Associate Deputy Chief Medical Director in November 1990. Tort claims analysis was later deemed an inherent risk management activity within the agency’s quality assurance system, and its responsibility wastransferred in March 1991 to the Office of the Associate Chief Medical Director for Quality Management. In 1992, the VA entered into a sharing agreement, regarding analysis of medical malpractice claims experience, with the Department of Legal Medicine (DLM) at the Armed Forces Institute of Pathology. DLM has engaged in medico legal consultation, research and education within DoD for more than two decades. DLM maintains a registry of all administrative claims, along with relevant medical and legal documents, that concern allegations of negligence in Department of Defense (DOD) medical treatment facilities.

    These claims are subjected to a series of trends analyses. DLM also collates and analyzes other risk management data submitted electronically by the Offices of the Surgeons General of the military services and provides regular summary reports regarding that information to those contributors and to the Assistant Secretary of Defense - Health Affairs (ASD-HA). DLM has been tasked by ASD-HA to conduct these recurring analyses, along with other research regarding those submissions, to assist that office in the derivation and implementation of policies calculated to improve the quality of health care within DoD.6 The sharing agreement of June 1992 calls for DLM to provide the VA with similar data analyses and educational programs. Under the agreement, DLM analyzes the nationwide VA tort claims experience, collates data, and provides periodic reports to assist the VA in identifying any high risk practice areas and developing responsive, remedial quality improvement endeavors.

    To assist the reader in understanding the VA Tort Claim Information System, certain terms and procedures that control the pursuit of a medical malpractice claim against the United States government must be delineated. A petitioner initiates this process by filing a claim, utilizing a designated federal standard form (SF-95), with a legal representative of the federal agency involved. For claims involving the VA, an SF-95 is filed with the respective District Counsel. By regulation, claims remain in an administrative phase for six months, while the federal agency involved is charged with investigating the allegations and attempting a final resolution, either by formal denial or settlement. If the claim is officially denied, the petitioner may seek administrative reconsideration by the VA General Counsel. Alternately, the claimant may then elect to file suit against the government in United States District Court. As a named defendant, the federal government is represented in litigation by the U.S. Department of Justice, and trial or settlement thereafter is ultimately controlled by that agency. It should be noted that if the six month administrative phase of this process passes with no federal agency action, neither settlement nor denial, the petitioner is permitted to consider the claim denied and elect to file suit.

TCIS DATA

        As originally implemented, all malpractice claims filed against the VA were entered into TCIS. The entry for each claim was later updated, without creating a second record, if litigation ensued. Information was extracted from every claim, and data was entered into a computer protocol. Database entries from VA District Counsel offices across the nation included the following: 1) claimant’s name; 2) injured person’s name; 3) VA medical facility; 4) VA legal office; 5) court; 6) amount claimed; 7) fiscal year claim opened;8) disposition; 9) payment or closure date; 10) amount paid; 11) date filed; 12) location of injury; 13) incident
date; 14) hospital service; 15) specialty; 16) alleged negligence code; 17) alleged injury; 18) injury severity; and 19) description of alleged negligence VA District Counsel offices also submitted patient information concerning service-connected injury or
service-connected disability.
    Each VA medical facility submitted a Peer Review Analysis and data entries regarding providers. Provider names, specialties, and assignment categorization within VA were included. Entries derived from peer review included whether: 1) claim incident was preventable; 2) death was preventable; 3) negligence affected clinical outcome; 4) clinical outcome represented normal risk; 5) standard of care was met; 6) patient incident report was indicated; 7) on site investigation was required Further determinations of as many as five separate errors within the provision of the medical care reviewed could be amended to this section of the report, and an opportunity for comment was provided. In May 1992, the Offices of the VA Associate Chief Medical Director for Quality Management, the VA General Counsel and DLM surveyed selected data within TCIS. DLM evaluated only that portion of the TCIS entered by local District Counsel offices. Tables published within this article derive from entries in the TCIS database. The totals reported in different tables vary, and this may simply result from clerical errors. More likely, as has been encountered in the analysis of DOD malpractice data, this variance may reflect the inherent difficulty in securing a complete collection of data from both medical and legal sources within a large system.7 As of the survey date, TCIS contained entries regarding 3,796 malpractice claims. When TCIS began in 1988, only active or recently closed cases were added to the database. The VA has annually received approximately 700 claims since 1989, when TCIS became fully operational. This experience is similar to that of DOD, which maintains 168 medical treatment facilities world-wide and receives approximately 900 claims per year.8

Table 1 indicates the status of 3,796 malpractice cases within TCIS. Approximately 30 percent of the cases are “open”, i.e., not presently resolved. Of open cases, the majority (739 or 62 percent of 1,195) are administrative claims managed by different District Counsel offices. The remainder are either in litigation or their responsibility has been transferred to the VA General Counsel. Approximately 70 percent of the TCIS entries regard malpractice disputes that were closed either in the administrative or the litigation phase. The majority of these, 1,478 cases, were closed by District Counsel offices. Within the 3,796 cases reported in the database, 1,029 or 27 percent had been closed by denial. District Counsel offices settled 449 or 12percent of the cases. The VA General Counsel settled 157 (4 percent) of the cases reported within TCIS and had denied 88 (2 percent). The database includes entries regarding 828 cases (22 percent) that proceeded to litigation. United States Attorney offices settled 477 (12 percent) of these disputes. Of those that were tried, there were 108 (3 percent) that resulted in judgments for the United States and 64 (2 percent) with judgments for the plaintiff.

Table 2 indicates the injury coding available for 3,771
reports. Alleged injuries were related to surgery,
diagnosis, and treatment in 30 percent, 26 percent
and 23 percent of that total, respectively. This is
similar to the experience of St. Paul Fire and Marine
Insurance Company, one of the nation’s largest private
carriers of medical professional liability insurance.10
Their report concerned 7,233 claims filed against
insured physicians in 1989 and 1990. Alleged injuries
were related to surgery, diagnosis and treatment in 25
percent, 27 percent, and 27 percent of those claims,
respectively.
District Counsel offices submitted reports regarding
severity of injury in 3,719 claims, as indicated in
Table 3. The injury was death in 32 percent of
reports. Permanent significant injury or permanent
major injury constituted 18 percent and 14 percent of
the total, respectively. These figures would appear to
suggest that patients who experience relatively severe
adverse clinical outcomes are more likely to file claims.
Table 4 regards the location within a treatment facility
where alleged negligence occurred. Twenty-seven
percent of cases occurred in the operating suite, 19
percent in patients’ rooms, and 22 percent either in
the outpatient clinic or emergency/admitting department.
St. Paul reported that 34 percent of their
claims occurred in the insured physician’s office or
clinic while 64 percent arose in the hospital and 27
percent in surgery.11
Table 5 (next page) reports the hospital service where
the alleged negligence occurred. Of the 3,994 entries,
medicine constituted 31 percent and surgery 32 percent.
Seven percent (266 cases) occurred in ambulatory care.
Table 6 (next page) reflects specialty determinations
within the reports of the District Counsel offices. A
weakness regarding this entry was that only a single
specialty could be reported when, in many cases, there
may have been several specialties involved. Certain
general trends, however, can be detected. Orthopedic
surgery accounted for 157 (13 percent) of these entries.
Internal medicine and general surgery each accounted
for approximately 10 percent, while psychiatry
accounted for 9 percent. Other specialties were
represented less frequently. Specialty areas may well
deserve more attention in any future studies.
Table 7 (next page) indicates the type of medical
negligence alleged in 1,574 cases. Errors in diagnosis
and improper surgical/medical procedures each constitute
25 percent of this total. Allegations regarding
defects in planning or executing treatment occurred
in 336 or 21 percent of the cases reported.
TCIS TODAY
TCIS has recently undergone a major revision.
Coding changes have been applied to a number of
the data collection elements, and additional elements
have been added for the reports of the District Counsel
offices. There have also been changes in the sections
for provider information and peer review determinations.
Additionally, DLM will now review every
malpractice claim brought against the VA, collect
additional data and integrate it with the existing
matrix.

Since the revision, when an administrative malpractice
claim is received, the District Counsel office enters
initial information taken primarily from the claim form
alone. Within 30 days, the reporting District Counsel
office sends a copy of the claim, a TCIS printout, and
a Provider Information and Peer Review Form to the
VA medical facility involved. The office also sends a
copy of the claim and a TCIS printout to DLM.
Within 60 days, the medical facility conducts peer
review and completes the Provider Information and
Peer Review Form (see discussion below). Copies of
these forms and pertinent medical records are then
forwarded by the facility to one of four principal VA
regional offices: the Eastern Region in Baltimore,
Maryland; the Central Region in Lansing, Michigan;
the Southern Region in Jackson, Mississippi; and the
Western Region in San Francisco, California. Lastly,
the medical facility returns a completed copy of the

TABLE 7
Provider Information and Peer Review Form to the District Counsel office. Regional VA offices submit medical
records and the Provider Information and Peer Review Form to DLM. If, at the regional level, any additional peer reviews are conducted, these will similarly be forwarded. Upon closure of an administrative claim by payment or by denial with no ensuing litigation, the District Counsel office sends DLM any medical expert interviews, witness statements, and an updated TCIS printout regarding the claim. Similar entries are filed by either the District Counsel offices or the VA General Counsel regarding those cases that proceed to litigation. DLM reviews the submitted information as received and submits clinical corrections where indicated. DLM also creates a separate database from entries in the Provider Information and Peer Review Forms. Annual reports are submitted to the VA Associate Chief Medical Director for Quality Management each January with regard to data collected the prior fiscal year. Quarterly reports to the VA regions are planned, along with periodic administrative submissions to the VA Associate Chief Medical Director for Quality Management. Additionally, “Open File”, a continuing medical education publication of DLM regarding clinical risk management, is furnished to all VA facilities for full-time physicians and quality management coordinators. The TCIS data collection form has been modified to facilitate the acquisition of complete data and its subsequent analysis. With the new form, a total of three hospital services and three medical specialties involved in a case can be reported. The codes for alleged negligence have been changed, and a number of medical specialties have been added to the database. The Act or Omission Code of the Harvard Risk Management Foundation will be utilized, as adopted within the National Practitioner Data Bank. Codes for alleged injury were also modified to reflect severity coding systems employed in other quality assurance programs. A new Provider Information and Peer Review Form has been developed to replace the prior document. This new form is considered privileged and confidential, by statutory exemption, and therefore not subject to disclosure to third parties. Under the provider information section, up to three providers can be designated with their position, service and status. The peer review section has also been modified to provide information concerning specific components of care, such as diagnosis and treatment, as well as monitoring of the patient. DLM will review all materials and then collect additional data elements which will comprise the Tort Claim Analysis System.
    The specifics of each malpractice case, as stated above, are reviewed, pertinent information extracted and data entered into an automated database. A coding system to designate the presenting symptom has been created, and up to five presenting symptoms or complaints can be added to this system. The working diagnoses, as well as the final diagnoses, are also collected using the ICD-9 clinical modification three-digit code system. Similarly, the disease process is recorded, and up to four diagnoses and four disease processes can be included.
    The area of practice or specialty is also collected, as well as the organ system involved. Up to four organ systems and four specialties or areas of practice can be collected on the same case. An additional coding system for procedures involved in medical malpractice cases has been developed, and up to three procedures can be entered into the database. A medical injury taxonomy has been devised so that the specific injuries can be cataloged as subsets of cases. In addition, conclusions of both government and plaintiff experts are collated. If the health care provider’s opinion is provided, this is also entered into the database, which concludes with the judicial opinion in those cases that end with judgment for either the plaintiff
or defense.
    Various risk management issues can be identified through case review. These diverse issues include the completeness of the medical record, communication problems among the staff and with patients, problems with resident supervision, autopsy issues, laboratory errors, patient follow-up problems, staffing problems, and the failure to consult specialists. Finally, a case synopsis is created in a memo field. This synopsis includes the facts, allegations, alleged errors and related patient injuries, along with the outcome of the case.
    This system should be useful for VA post-graduate educational purposes aimed at identifying high risk practice areas and facilitating quality improvement efforts. Analysis of the data will be performed by both VA officials Office of the Associate Chief Medical Director for Quality Management) and the DLM staff. This data will be used to assist other VA quality improvement programs, such as the Patient Incident Review System. Improving TCIS represents a significant achievement. This has been accomplished through the efforts of many VA officials, both at the central and regional offices, who appreciate the unique aspects of their agency and its mission. With these improvements, information derived from TCIS and the Tort Claim Analysis System will augment the existing VA program for quality management. The targeting of preventable sources of medical negligence should ultimately improve delivery of medical care to the veteran.
1. Department of Veterans Affairs, Office of the Assistant Secretary for Finance and IRM, Employment Data on T.38 Physicians, Dentists and Nurses. Washington, DC. March 31, 1992; Table 2A; 2.

VA Confirms that the VA Director of Research Improperly Spent $1,700,000 on meals, travel, conferences and equipment

Veterans Affairs Memorandum
Date: March 22, 2004
From: Assistant Inspector General for Investigations (51)
Subj: Administrative Investigation – Use of Government Funds, Travel, Personnel, Impartiality, and Management Issues, Research and Development Office, Veteran Health Administration, Report No. 03-03053-115 (Case IQ-0179)
To: Under Secretary for Health (10)
1. Attached is our final report of an administrative investigation into allegations against Dr. Nelda P. Wray, the Chief Research and Development Officer in the Department of Veterans Affairs (VA) Veterans Health Administration (VHA). Complainants alleged that Dr. Wray misused funds provided to VA by pharmaceutical companies; misused Government travel funds; unfairly hired, promoted, and managed staff; and did not act impartially or reasonably when approving and disapproving Research and Development Office projects.
2. We substantiated that, between January 2003 and October 2003, Dr. Wray and certain members of her staff were responsible for improperly spending nearly $1.7 million provided to VA primarily by pharmaceutical companies. The funds were maintained and administered by Friends Research Institute, Inc. (FRI), a private nonprofit corporation. While the pharmaceutical companies provided these funds for VA’s use in conducting specific cooperative research studies, the money was used for purposes unrelated to the projects specified, such as costly research equipment for an unrelated study; consultant and other management services; conference facilities and meals; local and out-of-town restaurant expenses; and other business-related and personal items. In effect, this spending constituted an illegal augmentation of the Department’s appropriations, and a misuse of position. Dr. Wray’s predecessors acted similarly in misspending over $537,000 of these funds during calendar year 2002. The purchases improperly made should have been paid for either from appropriated funds or personally by the Research and Development Office staff.
3. ·(b)(6)· · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · participated in approving the 2002 expenditures, and initially offered the FRI-administered funds to Dr. Wray, advising her that the agreements with pharmaceutical companies allowed the use of those funds for other purposes. In the summer of 2003, when he identified an agreement that required VA to return the funds, he did not thoroughly review the remaining agreements for similar provisions. Five of the 15 agreements we reviewed specifically required that any unused funds at the completion of the study be returned, and an additional 5 agreements clearly indicated the funds provided were to be spent only on the study referenced in the agreement. Neither Dr. Wray nor Mr. John Bradley, the Research and Development Office’s Chief Financial Officer, attempted to determine how the FRI-administered funds should properly be used, even after a General Counsel attorney raised questions to Dr. Wray about the corporation. Mr. Bradley, as the Chief Financial Officer, in particular, should have questioned the use of the money.
VA FORM
MAR 1989 2105
i
4. In spending these Government funds, neither Dr. Wray nor anyone on her staff had authority to enter into contracts on behalf of the Government, and they did not adhere to basic Federal acquisition regulations, such as preparing written contracts and seeking competition. Use of FRI-administered funds appears to have been an expedient way for the Research and Development Office staff to procure goods and services, with no concern their requests would be denied.
5. We recommended that the Under Secretary for Health ensure that the Deputy Under Secretary for Health takes appropriate administrative action against Dr. Wray, ·(b)(6)· · · · · · · · · · ·, and Mr. Bradley; and educates Research and Development Office staff regarding the proper use of money provided by pharmaceutical companies for VA cooperative research studies. We also recommended that the Chief Research and Development Officer be directed to immediately cease spending FRI-administered funds, and that several actions be taken to correct the misuse of funds, including transferring them to VA affiliated nonprofit research corporations or the General Post Fund and properly disposing of excess funds in accordance with the agreements between VHA and the pharmaceutical companies. Finally, we recommended that bills of collection be issued to Dr. Wray and others responsible for approving the use of FRI-administered funds for their own or others’ personal benefit. The Under Secretary concurred with the recommendations, noting that he would rely on advice from the General Counsel regarding whether the bills of collection can be issued.
6. Regarding Dr. Wray’s travel, we substantiated that she traveled unnecessarily to Houston, took circuitous routes through Houston, claimed lodging expenses above the allowable limits, used expensive ground transportation, and claimed other improper expenses. We also identified days Dr. Wray should not have claimed meals and incidental expenses, and days she should have charged annual leave while away from her duty station. Her travel vouchers document a pattern of questionable trips to and through Houston at Government expense. They appear to be a pretext for her to make weekend visits there for personal reasons at Government expense. In total, we identified $9,737 improperly claimed on Dr. Wray’s vouchers, and 6 days she should have charged annual leave but did not. A staff assistant to Dr. Wray generally made her travel arrangements and prepared her travel vouchers. Dr. Wray told us she signed the vouchers once they were completed, but did not review them first. Mr. Bradley and ·(b)(6)· · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · told us that one or the other of them reviewed and approved Dr. Wray’s vouchers after she signed them. They both told us they generally did not question the appropriateness of her claims. On another travel matter, we found that, at Dr. Wray’s request, two staff from the Houston VA Medical Center incurred over $30,000 in temporary duty expenses when they traveled to Washington, DC, to assist her in transitioning to the Chief Research and Development Officer position. We questioned the necessity of these temporary duty assignments, as they appear to have been primarily for Dr. Wray’s personal convenience.
7. We recommended that the Under Secretary for Health ensure that the Deputy Under Secretary for Health takes appropriate administrative action against Dr. Wray, Mr. Bradley, and ·(b)(6)· · · · · · · · · · · · · · · · · · · · · · · · · ·, and provides detailed training on Federal
ii
and VA travel regulations to Dr. Wray’s staff assistant. We also recommended that a bill of collection be issued to Dr. Wray to recoup the cost of travel she took that was not officially necessary, or that was otherwise improperly claimed; and that she be charged 6 days of annual leave for time spent away from her duty station without official necessity. Finally, we recommended that the travel vouchers of those staff who routinely traveled with Dr. Wray be reviewed to determine if similar irregularities exist in their claims. The Under Secretary concurred with the recommendations.
8. Regarding the allegation that Dr. Wray unfairly hired and promoted staff, we found that she gave improper preference to four individuals. Dr. Wray told others she planned to promote two of the employees to positions in the Research and Development Office even before the job announcements had been issued. She granted an improper preference to a third employee and to an applicant for employment by asking her staff to promote/hire them. While examining these and other personnel actions, we found that, historically, a large number of employees working in the Research and Development Office were improperly appointed to their positions without competition under the “Schedule B” authority. Additionally, the Office historically circumvented the limits imposed by the Congressional appropriation for VA Central Office employee salaries by using VHA field-based employees, physically locating them in Washington, DC, to carry out the work of the Office. Finally, we concluded that Dr. Wray’s management style regarding her handling of perceived staff performance issues compromised the staff’s ability to carry out the mission of the Office. The Under Secretary for Health concurred with our recommendations to take appropriate administrative action against Dr. Wray; review the propriety of all Research and Development Office staff appointments made under Schedule B authority; review all positions appointed by field facilities and determine if any of the employees should be returned to the field; and review Dr. Wray’s actions to transfer the duties of three of her senior managers.
9. We substantiated that Dr. Wray violated the Standards of Ethical Conduct for Employees of the Executive Branch when she approved four projects involving participation by a colleague of hers at the Baylor College of Medicine. Considering particularly that Dr. Wray had a close prior professional relationship with this individual, and is still an employee of Baylor College of Medicine, she gave the appearance of favoritism towards him. The Under Secretary for Health concurred with our recommendation to take appropriate administrative action against her.
10. We further substantiated that Dr. Wray did not act reasonably when she re-evaluated and re-scored 130 research proposals that had previously earned fundable scores, less than a month before their effective funding date, and disapproved 15 of them. While VHA policy provides that an investigator’s proposal will be evaluated based on his or her productivity, specific measures of productivity were not previously used. The proposals had completed the merit review process and investigators had been notified of the results before Dr. Wray assumed her position as Chief Research and Development Officer. They were expecting their projects to be funded based on the priority scores they received, in accordance with applicable policy. Dr. Wray’s decision to not fund the projects was contrary to VA policy and to good management iii
iv
practice. We made no recommendation on this matter, but brought it to the attention of the Under Secretary for Health for whatever action he deemed appropriate.
11. We substantiated that Dr. Wray misused a Government purchase card to pay for a meeting that could readily have been convened in Research and Development Office workspace. The Under Secretary for Health concurred with our recommendation to take appropriate administrative action against Dr. Wray for this violation.
12. The Under Secretary’s full response is in the appendix to this report. Regarding the second paragraph of his March 16, 2004 memorandum, in which he references potential criminal prosecution, it should be noted that certain matters discussed in this report were presented to the Department of Justice on March 15, 2004, and prosecution was declined. We will follow up to ensure the actions proposed in response to our recommendations are taken.
(original signed by:)
DANIEL R. PETROLE
Attachment

VA OIG Report March 22, 2005

VA's Own Peer Review Process Finds that Failure to Supervise Residents was Responsible for 63 Malpractice Cases From 1997-2002

At 4 VA medical centers (St. Louis, Birmingham, Nashville, and New Orleans) we attempted to locate attending physicians in 29 clinics to determine if there was adequate resident supervision. We found that the attending physicians were not present to supervise the residents’ treatments of patients in 6 of the 29 clinics (see page 25). One resident told us that the attending physician would occasionally come to the clinic to see how things were going, but usually would not stay. Another resident said that the attending physician rarely came to the clinic, but made a brief appearance a couple of weeks ago. Generally, the attending physician returned pager messages within an hour. Another resident told us that the attending physician did not come to the clinic unless there was a case the resident decided the attending needed to review, but the attending was always available by phone or pager.
The attending physicians at one VA medical center told us that attending physicians saw every patient at the university because Medicare and private medical insurance carriers would not pay for care provided only by residents. The residents at another VA medical center told us they ran clinics at the VA medical center but at the university they only ran clinics for the indigent population.
From FY 1997 through the second quarter of FY 2002, the Federal Government paid, on behalf of VA, at least $21 million for 63 malpractice cases where VA’s peer review panel found that the attending VA physicians provided substandard resident supervision. Based on our review of available documentation, the attending physicians were not present to supervise the residents during the performance of a procedure or the provision of a treatment to a veteran in at least eight cases resulting in malpractice settlements totaling $4.7 million. An additional pending case involves an attending surgeon who could not provide needed assistance to a VA medical center patient because he was operating on a non-veteran patient at the affiliated medical school.

VA cardiac surgery units suffer higher death rates
Wednesday, January 31, 2001 - By JOAN MAZZOLINI
PLAIN DEALER REPORTER  
 

The Department of Veterans Affairs clings to its heart surgery programs even though its patients die more frequently than heart patients in private and public hospitals. The VA acknowledges that some of those programs don’t do enough surgeries each year to guarantee proficiency. Some of those hospitals just don’t have enough heart surgeons willing to work for them, the VA says. This is costing taxpayers millions of dollars each year and risking the lives of veterans. Many of them qualify for Medicare coverage and could go elsewhere if they knew their VA hospital had a troubled heart surgery program. A Plain Dealer investigation found that: More than one-third of the 42 veterans hospitals performing heart surgery don’t do at least 150 heart surgeries a year, the minimum the VA requires and experts recommend. Each year for the last five years, one-quarter to one-third of the hospitals had death rates above what the VA says they should have, considering the ages of the patients and the severity of their illnesses. The VA, as of last June, had monitored nine hospitals in the past 24 months because of high death rates and other quality concerns. The hospitals were: Birmingham, Ala.; Hines, Ill.; West Los Angeles; Augusta, Ga.; Ann Arbor, Mich.; Dallas; Little Rock, Ark.; Tampa; and Salt Lake City. The VA’s head of surgical services defended the programs and said the veterans are sicker patients for whom private hospitals would rather not care. "It’s the best system in the country," said Dr. Gerald O. McDonald, director of surgical services for the VA. "It’s a fantastic national resource. "McDonald said that because most VA hospitals work with nearby teaching hospitals, veterans are, in fact, being treated by some of the best surgeons in the country. "You can go to some community hospitals and find a hell of a lot more problems than our VA system," he said. McDonald also questioned the requirement that VA programs do at least 150 heart surgeries a year, saying a study using data from April 1987 through September 1992 showed that 100 open-heart surgeries a year was an adequate number. But Dr. Shukri Khuri, director of the West Roxbury VA heart-surgery program, said he had encouraged the VA to increase the minimum number to 250 a year. West Roxbury has the largest program, with 473 heart surgeries performed during the 12 months ending September 1999. And its 2.7 percent death rate is among the lowest in the VA. "I hate to brag, but we have lower death rates than the other [private] Boston programs," Khuri said. Khuri declined to comment on whether the VA should get out of the cardiac surgery business, but he did say it would be tough to kill the programs." As you know, Congress people would never want to see a VA close or close a cardiac program," Khuri said. "For them, it’s a source of pride, it’s a source of employment for people and it’s usually the politicians you have to convince before you convince anybody else. "Nonetheless, decades of studies have shown that some of the programs aren’t as good as they should be. Excessive patient deaths and other problems have prompted VA officials to close several programs around the country. At one time, there were 51 veterans hospitals that offered heart surgery; today there are 42.In 1999, more than 7,200 heart surgeries were performed at VA hospitals, and the average death rate was 4.8 percent. In comparison, the average death rate for bypass surgery that year was about 3 percent at the more than 500 hospitals that participated in a Society for Thoracic Surgeons survey. (The vast majority of the 7,200 VA surgeries were bypass operations. Some of the remaining procedures were riskier than bypass surgery.) Deaths mount The Hines VA Hospital outside Chicago has had what VA officials call "excess" deaths for 10 years. Yet its doctors continued performing heart surgery on veterans. Members of the VA’s cardiac surgery committee visited Hines in December of 1999, the fifth time in 10 years. That visit came about because the death rate of veterans undergoing heart surgery was more than 10 percent - twice the national average of 4.7 percent for veterans hospitals. And Hines’ death rate had been high for the previous three years. "The committee expressed serious concern about the lack of adequate nursing support to the cardiac surgery patients and the overall morale problems," the committee’s report said. In a common arrangement between VA hospitals and nearby university centers, Loyola University Medical Center’s cardiac surgery chairman oversees the heart program at Hines. The report pointed out that the chairman believed he’d improved the program in part by discharging "two ... cardiac surgeons working at the VA that he felt were problems and replaced them. "In the end, the committee opted to give the Hines program another chance, with the recommendation that there be "very close monitoring" of every patient death. Loyola’s desire to keep the nearby VA heart program operating was an important factor in the decision to keep the Hines program open. "Loyola considers the continued operation of the Hines VA Medical Center ... to be critical to its educational and research mission," committee members wrote. Death rates at Hines have improved, to slightly higher than the VA national average for the 12 months ending September 1999."I’m not saying things are perfect," said John Fears, acting director at Hines. "If I thought things were out of hand, I’d shut it in a minute. "Problem years Over the last six years, nearly half of the veterans hospitals with cardiac programs have been either monitored or put on probation. Monitoring entails reviewing each patient death. Probation indicates more serious problems that if not corrected quickly could prompt the VA to end the program. The VA shut down the Lexington, Ky., program in late 1996, after death rates reached more than 10 percent. The VA also shut down programs in Brooklyn, East Orange, N.J., and Long Beach, Calif., in the late 1980s.The Little Rock, Ark., program stopped doing heart surgeries in 1997 when its death rate reached nearly 10 percent. While Little Rock is listed as a VA cardiac center, veterans needing heart surgery are sent to the University of Arkansas hospital, said McDonald. He added that the decision to stop heart surgery didn’t come from VA headquarters. Death rates for veterans operated on at the Little Rock university hospital have improved somewhat. "Sometimes there are blips in the ... mortality rates," McDonald said, when asked about the number of programs being monitored. "If there’s a real problem, then there is a site visit."From June 1998 through June 2000, 12 of the 42 heart surgery programs had site visits. West Los Angeles was visited by the quality team in late 1999."Review of the cases showed a variety of potential problems including operative judgment, technical judgment in the timing of operations and the selection of patients, with some patients being extremely high risk, and also in postoperative care," the reviewers found. McDonald was a member of the site visit team. He said he threatened to close the program and that UCLA, the nearby teaching hospital, was dismayed. His threat, he said, "made a lot of difference. "The findings noted that one of the doctors "is a very senior, very experienced, nationally well-known surgeon. ... He is over committed ... time-wise and [has] very little physical presence at the VAMC. "McDonald said that doctor almost never "went to the VA. That was the problem. Now that’s been corrected. I think we got it pretty well worked out. "Though McDonald and others don’t believe the VA will give up heart surgery, the VA awarded a research grant more than two years ago to study whether to keep the heart programs open. The study isn’t completed, but one of the researchers has concluded that the low volume of heart surgeries at VA facilities may have led to "poorer patient outcomes, in terms of both cost to the VA health system and the quality of care provided. "Shipping heart patients to other hospitals, he wrote, "may be a beneficial and cost-effective strategy.

1995 GAO Report on Malpractice Claims at the Veterans Administration

Armed Forces Institute of Pathology Report on Malpractice at the VA

From the Office of the Associate Chief Medical Director for Quality Management and the Office of the General Counsel, Department of Veterans Affairs, and the Department of Legal Medicine, Armed Forces Institute of Pathology
DEPARTMENT OF VETERANS AFFAIRS - ANALYSIS OF MEDICAL MALPRACTICE CLAIMS - AN INITIAL REPORT
by RICHARD L. GRANVILLE, MD, JD, MICHAEL GREENBLATT, MPA, DOUGLAS BRADSHAW, JD, GALEN BARBOUR, MD, and FRANK T. FLANNERY, M.D., J.D., LTC, MC, USA
The authors wish to acknowledge the special assistance o f Mr. Dennis Snyder and Ms. Loretta Hunter in compiling the data for this article.
The Department of Veterans Affairs (VA) operates one of the nation’s largest medical systems. It is composed of 172 hospitals and more than 350 outpatient facilities. Teaching affiliations are maintained with 160 medical and dental schools nationwide. Professional providers within the VA include 7,324 full-time and 5,921 part time staff physicians, 15,020 resident physicians, and 39,355 nurses.1,2 In FY 1991, there were 951,112 inpatients and 21,932,426 outpatients treated at VA medical facilities.
VA TORT CLAIM INFORMATION SYSTEM
As with other large medical systems, recent years have witnessed increasing attention directed toward the VA medical malpractice claims experience. In 1985, the VA Office of the Inspector General (OIG) released a report recommending that the VA Department of Medicine and Surgery perform an extensive analysis of those medical conditions and surgical procedures that had resulted in malpractice claims. In March 1987, a subsequent OIG audit noted that the VA had initiated a number of monitors regarding medical malpractice claims but, due to resource constraints, a comprehensive analysis of the characteristics of those claims had not yet been accomplished. In March 1988, at a Congressional oversight committee hearing, Senator John Glenn (DOhio) emphasized the importance of utilizing data generated from the analysis of malpractice claims for purposes of monitoring and improving VA medical care. At that time, the Office of the General Counsel of the VA and that of the Medical Inspector combined efforts to develop the Tort Claims Information System (TCIS).3 Initial procedures regarding TCIS were promulgated in August 1988.4 Each VA medical center was directed to forward reports regarding any medical malpractice claim filed to the Office of the Medical Inspector. These reports were to include a summary of the incident involved, identification of responsible providers, relevant medical records, and available quality assurance documents. The Medical Inspector’s staff was to collect these reports, conduct individual case reviews, and assemble aggregate malpractice data for the respective medical centers and for the system, on a regional and a national basis. At first, these case reviews were restricted to the analysis of malpractice claims that were legally closed or finalized because there was no statutory protection for the confidentiality of the peer review information contained within submitted reports. By May 1990, when TCIS gained statutory protection for that confidentiality as medical quality assurance (38 U.S.C. § 3301), 1700 legally finalized malpractice cases had been reviewed. Thereafter, VA medical facilities were directed to report malpractice claims filed and to include a completed peer review analysis of each case.5 Responsibility for TCIS within the VA was transferred from the Office of the Medical Inspector to that of the Associate Deputy Chief Medical Director in November 1990. Tort claims analysis was later deemed an inherent risk management activity within the agency’s quality assurance system, and its responsibility was
transferred in March 1991 to the Office of the Associate Chief Medical Director for Quality Management. In 1992, the VA entered into a sharing agreement, regarding analysis of medical malpractice claims experience, with the Department of Legal Medicine (DLM) at the Armed Forces Institute of Pathology. DLM has engaged in medico legal consultation, research and education within DoD for more than two decades. DLM maintains a registry of all administrative claims, along with relevant medical and legal documents, that concern allegations of negligence in Department of Defense (DOD) medical treatment facilities. These claims
are subjected to a series of trends analyses. DLM also collates and analyzes other risk management data submitted electronically by the Offices of the Surgeons General of the military services and provides regular summary reports regarding that information to those contributors and to the Assistant Secretary of Defense - Health Affairs (ASD-HA). DLM has been tasked by ASD-HA to conduct these recurring analyses, along with other research regarding those submissions, to assist that office in the derivation and implementation of policies calculated to improve the quality of health care within DoD.6 The sharing agreement of June 1992 calls for DLM to provide the VA with similar data analyses and educational programs. Under the agreement, DLM analyzes the nationwide VA tort claims experience, collates data, and provides periodic reports to assist the VA in identifying any high risk practice areas and developing responsive, remedial quality improvement endeavors.

To assist the reader in understanding the VA Tort Claim Information System, certain terms and procedures that control the pursuit of a medical malpractice claim against the United States government must be delineated. A petitioner initiates this process by filing a claim, utilizing a designated federal standard form (SF-95), with a legal representative of the federal agency involved. For claims involving the VA, an SF-95 is filed with the respective District Counsel. By regulation, claims remain in
an administrative phase for six months, while the federal agency involved is charged with investigating the allegations and attempting a final resolution, either by formal denial or settlement. If the claim is officially denied, the petitioner may seek administrative reconsideration by the VA General Counsel. Alternately, the claimant may then elect to file suit against the government in United States District Court. As a named defendant, the federal government is represented in litigation by the U.S. Department of Justice, and trial or settlement thereafter is ultimately controlled by that agency. It should be noted that if the six month administrative phase of this process passes with no federal agency action, neither settlement nor denial, the
petitioner is permitted to consider the claim denied and elect to file suit.

TCIS DATA

    As originally implemented, all malpractice claims filed against the VA were entered into TCIS. The entry for each claim was later updated, without creating a second record, if litigation ensued. Information was extracted from every claim, and data was entered into a computer protocol. Database entries from VA District Counsel offices across the nation included the following: 1) claimant’s name; 2) injured person’s name; 3) VA medical facility; 4) VA legal office; 5) court; 6) amount claimed; 7) fiscal year claim opened;8) disposition; 9) payment or closure date; 10) amount paid; 11) date filed; 12) location of injury; 13) incident
date; 14) hospital service; 15) specialty; 16) alleged negligence code; 17) alleged injury; 18) injury severity; and 19) description of alleged negligence VA District Counsel offices also submitted patient information concerning service-connected injury or
service-connected disability.
    Each VA medical facility submitted a Peer Review Analysis and data entries regarding providers. Provider names, specialties, and assignment categorization within VA were included. Entries derived from peer review included whether: 1) claim incident was preventable; 2) death was preventable; 3) negligence affected clinical outcome; 4) clinical outcome represented normal risk; 5) standard of care was met; 6) patient incident report was indicated; 7) on site investigation was required Further determinations of as many as five separate errors within the provision of the medical care reviewed could be amended to this section of the report, and an opportunity for comment was provided. In May 1992, the Offices of the VA Associate Chief Medical Director for Quality Management, the VA General Counsel and DLM surveyed selected data within TCIS. DLM evaluated only that portion of the TCIS entered by local District Counsel offices. Tables published within this article derive from entries in the TCIS database. The totals reported in different tables vary, and this may simply result from clerical errors. More likely, as has been encountered in the analysis of DOD malpractice data, this variance may reflect the inherent difficulty in securing a complete collection of data from both medical and legal sources within a large system.7 As of the survey date, TCIS contained entries regarding 3,796 malpractice claims. When TCIS began in 1988, only active or recently closed cases were added to the database. The VA has annually received approximately 700 claims since 1989, when TCIS became fully operational. This experience is similar to that of DOD, which maintains 168 medical treatment facilities world-wide and receives approximately 900 claims per year.8

Table 1 indicates the status of 3,796 malpractice cases within TCIS. Approximately 30 percent of the cases are “open”, i.e., not presently resolved. Of open cases, the majority (739 or 62 percent of 1,195) are administrative claims managed by different District Counsel offices. The remainder are either in litigation or their responsibility has been transferred to the VA General Counsel. Approximately 70 percent of the TCIS entries regard malpractice disputes that were closed either in the administrative or the litigation phase. The majority of these, 1,478 cases, were closed by District Counsel offices. Within the 3,796 cases reported in the database, 1,029 or 27 percent had been closed by denial. District Counsel offices settled 449 or 12percent of the cases. The VA General Counsel settled 157 (4 percent) of the cases reported within TCIS and had denied 88 (2 percent). The database includes entries regarding 828 cases (22 percent) that proceeded to litigation. United States Attorney offices settled 477 (12 percent) of these disputes. Of those that were tried, there were 108 (3 percent) that resulted in judgments for the United States and 64 (2 percent) with judgments for the plaintiff.

Table 2 indicates the injury coding available for 3,771
reports. Alleged injuries were related to surgery,
diagnosis, and treatment in 30 percent, 26 percent
and 23 percent of that total, respectively. This is
similar to the experience of St. Paul Fire and Marine
Insurance Company, one of the nation’s largest private
carriers of medical professional liability insurance.10
Their report concerned 7,233 claims filed against
insured physicians in 1989 and 1990. Alleged injuries
were related to surgery, diagnosis and treatment in 25
percent, 27 percent, and 27 percent of those claims,
respectively.
District Counsel offices submitted reports regarding
severity of injury in 3,719 claims, as indicated in
Table 3. The injury was death in 32 percent of
reports. Permanent significant injury or permanent
major injury constituted 18 percent and 14 percent of
the total, respectively. These figures would appear to
suggest that patients who experience relatively severe
adverse clinical outcomes are more likely to file claims.
Table 4 regards the location within a treatment facility
where alleged negligence occurred. Twenty-seven
percent of cases occurred in the operating suite, 19
percent in patients’ rooms, and 22 percent either in
the outpatient clinic or emergency/admitting department.
St. Paul reported that 34 percent of their
claims occurred in the insured physician’s office or
clinic while 64 percent arose in the hospital and 27
percent in surgery.11
Table 5 (next page) reports the hospital service where
the alleged negligence occurred. Of the 3,994 entries,
medicine constituted 31 percent and surgery 32 percent.
Seven percent (266 cases) occurred in ambulatory care.
Table 6 (next page) reflects specialty determinations
within the reports of the District Counsel offices. A
weakness regarding this entry was that only a single
specialty could be reported when, in many cases, there
may have been several specialties involved. Certain
general trends, however, can be detected. Orthopedic
surgery accounted for 157 (13 percent) of these entries.
Internal medicine and general surgery each accounted
for approximately 10 percent, while psychiatry
accounted for 9 percent. Other specialties were
represented less frequently. Specialty areas may well
deserve more attention in any future studies.
Table 7 (next page) indicates the type of medical
negligence alleged in 1,574 cases. Errors in diagnosis
and improper surgical/medical procedures each constitute
25 percent of this total. Allegations regarding
defects in planning or executing treatment occurred
in 336 or 21 percent of the cases reported.
TCIS TODAY
TCIS has recently undergone a major revision.
Coding changes have been applied to a number of
the data collection elements, and additional elements
have been added for the reports of the District Counsel
offices. There have also been changes in the sections
for provider information and peer review determinations.
Additionally, DLM will now review every
malpractice claim brought against the VA, collect
additional data and integrate it with the existing
matrix.

Since the revision, when an administrative malpractice
claim is received, the District Counsel office enters
initial information taken primarily from the claim form
alone. Within 30 days, the reporting District Counsel
office sends a copy of the claim, a TCIS printout, and
a Provider Information and Peer Review Form to the
VA medical facility involved. The office also sends a
copy of the claim and a TCIS printout to DLM.
Within 60 days, the medical facility conducts peer
review and completes the Provider Information and
Peer Review Form (see discussion below). Copies of
these forms and pertinent medical records are then
forwarded by the facility to one of four principal VA
regional offices: the Eastern Region in Baltimore,
Maryland; the Central Region in Lansing, Michigan;
the Southern Region in Jackson, Mississippi; and the
Western Region in San Francisco, California. Lastly,
the medical facility returns a completed copy of the

TABLE 7
Provider Information and Peer Review Form to the District Counsel office. Regional VA offices submit medical
records and the Provider Information and Peer Review Form to DLM. If, at the regional level, any additional peer reviews are conducted, these will similarly be forwarded. Upon closure of an administrative claim by payment or by denial with no ensuing litigation, the District Counsel office sends DLM any medical expert interviews, witness statements, and an updated TCIS printout regarding the claim. Similar entries are filed by either the District Counsel offices or the VA General Counsel regarding those cases that proceed to litigation. DLM reviews the submitted information as received and submits clinical corrections where indicated. DLM also creates a separate database from entries in the Provider Information and Peer Review Forms. Annual reports are submitted to the VA Associate Chief Medical Director for Quality Management each January with regard to data collected the prior fiscal year. Quarterly reports to the VA regions are planned, along with periodic administrative submissions to the VA Associate Chief Medical Director for Quality Management. Additionally, “Open File”, a continuing medical education publication of DLM regarding clinical risk management, is furnished to all VA facilities for full-time physicians and quality management coordinators. The TCIS data collection form has been modified to facilitate the acquisition of complete data and its subsequent analysis. With the new form, a total of three hospital services and three medical specialties involved in a case can be reported. The codes for alleged negligence have been changed, and a number of medical specialties have been added to the database. The Act or Omission Code of the Harvard Risk Management Foundation will be utilized, as adopted within the National Practitioner Data Bank. Codes for alleged injury were also modified to reflect severity coding systems employed in other quality assurance programs. A new Provider Information and Peer Review Form has been developed to replace the prior document. This new form is considered privileged and confidential, by statutory exemption, and therefore not subject to disclosure to third parties. Under the provider information section, up to three providers can be designated with their position, service and status. The peer review section has also been modified to provide information concerning specific components of care, such as diagnosis and treatment, as well as monitoring of the patient. DLM will review all materials and then collect additional data elements which will comprise the Tort Claim Analysis System.
    The specifics of each malpractice case, as stated above, are reviewed, pertinent information extracted and data entered into an automated database. A coding system to designate the presenting symptom has been created, and up to five presenting symptoms or complaints can be added to this system. The working diagnoses, as well as the final diagnoses, are also collected using the ICD-9 clinical modification three-digit code system. Similarly, the disease process is recorded, and up to four diagnoses and four disease processes can be included.
    The area of practice or specialty is also collected, as well as the organ system involved. Up to four organ systems and four specialties or areas of practice can be collected on the same case. An additional coding system for procedures involved in medical malpractice cases has been developed, and up to three procedures can be entered into the database. A medical injury taxonomy has been devised so that the specific injuries can be cataloged as subsets of cases. In addition, conclusions of both government and plaintiff experts are collated. If the health care provider’s opinion is provided, this is also entered into the database, which concludes with the judicial opinion in those cases that end with judgment for either the plaintiff
or defense.
    Various risk management issues can be identified through case review. These diverse issues include the completeness of the medical record, communication problems among the staff and with patients, problems with resident supervision, autopsy issues, laboratory errors, patient follow-up problems, staffing problems, and the failure to consult specialists. Finally, a case synopsis is created in a memo field. This synopsis includes the facts, allegations, alleged errors and related patient injuries, along with the outcome of the case.
    This system should be useful for VA post-graduate educational purposes aimed at identifying high risk practice areas and facilitating quality improvement efforts. Analysis of the data will be performed by both VA officials Office of the Associate Chief Medical Director for Quality Management) and the DLM staff. This data will be used to assist other VA quality improvement programs, such as the Patient Incident Review System. Improving TCIS represents a significant achievement. This has been accomplished through the efforts of many VA officials, both at the central and regional offices, who appreciate the unique aspects of their agency and its mission. With these improvements, information derived from TCIS and the Tort Claim Analysis System will augment the existing VA program for quality management. The targeting of preventable sources of medical negligence should ultimately improve delivery of medical care to the veteran.
1. Department of Veterans Affairs, Office of the Assistant Secretary for Finance and IRM, Employment Data on T.38 Physicians, Dentists and Nurses. Washington, DC. March 31, 1992; Table 2A; 2.

Armed Forces Institute of Pathology Report on Malpractice at the VA

From the Office of the Associate Chief Medical Director for Quality Management and the Office of the General Counsel, Department of Veterans Affairs, and the Department of Legal Medicine, Armed Forces Institute of Pathology
DEPARTMENT OF VETERANS AFFAIRS - ANALYSIS OF MEDICAL MALPRACTICE CLAIMS - AN INITIAL REPORT
by RICHARD L. GRANVILLE, MD, JD, MICHAEL GREENBLATT, MPA, DOUGLAS BRADSHAW, JD, GALEN BARBOUR, MD, and FRANK T. FLANNERY, M.D., J.D., LTC, MC, USA

    The authors wish to acknowledge the special assistance o f Mr. Dennis Snyder and Ms. Loretta Hunter in compiling the data for this article.
    The Department of Veterans Affairs (VA) operates one of the nation’s largest medical systems. It is composed of 172 hospitals and more than 350 outpatient facilities. Teaching affiliations are maintained with 160 medical and dental schools nationwide. Professional providers within the VA include 7,324 full-time and 5,921 part time staff physicians, 15,020 resident physicians, and 39,355 nurses.1,2 In FY 1991, there were 951,112 inpatients and 21,932,426 outpatients treated at VA medical facilities.

VA TORT CLAIM INFORMATION SYSTEM

    As with other large medical systems, recent years have witnessed increasing attention directed toward the VA medical malpractice claims experience. In 1985, the VA Office of the Inspector General (OIG) released a report recommending that the VA Department of Medicine and Surgery perform an extensive analysis of those medical conditions and surgical procedures that had resulted in malpractice claims. In March 1987, a subsequent OIG audit noted that the VA had initiated a number of monitors regarding medical malpractice claims but, due to resource constraints, a comprehensive analysis of the characteristics of those claims had not yet been accomplished. In March 1988, at a Congressional oversight committee hearing, Senator John Glenn (DOhio) emphasized the importance of utilizing data generated from the analysis of malpractice claims for purposes of monitoring and improving VA medical care. At that time, the Office of the General Counsel of the VA and that of the Medical Inspector combined efforts to develop the Tort Claims Information System (TCIS).3 Initial procedures regarding TCIS were promulgated in August 1988.4 Each VA medical center was directed to forward reports regarding any medical malpractice claim filed to the Office of the Medical Inspector. These reports were to include a summary of the incident involved, identification of responsible providers, relevant medical records, and available quality assurance documents. The Medical Inspector’s staff was to collect these reports, conduct individual case reviews, and assemble aggregate malpractice data for the respective medical centers and for the system, on a regional and a national basis.

     At first, these case reviews were restricted to the analysis of malpractice claims that were legally closed or finalized because there was no statutory protection for the confidentiality of the peer review information contained within submitted reports. By May 1990, when TCIS gained statutory protection for that confidentiality as medical quality assurance (38 U.S.C. § 3301), 1700 legally finalized malpractice cases had been reviewed. Thereafter, VA medical facilities were directed to report malpractice claims filed and to include a completed peer review analysis of each case.5 Responsibility for TCIS within the VA was transferred from the Office of the Medical Inspector to that of the Associate Deputy Chief Medical Director in November 1990. Tort claims analysis was later deemed an inherent risk management activity within the agency’s quality assurance system, and its responsibility wastransferred in March 1991 to the Office of the Associate Chief Medical Director for Quality Management. In 1992, the VA entered into a sharing agreement, regarding analysis of medical malpractice claims experience, with the Department of Legal Medicine (DLM) at the Armed Forces Institute of Pathology. DLM has engaged in medico legal consultation, research and education within DoD for more than two decades. DLM maintains a registry of all administrative claims, along with relevant medical and legal documents, that concern allegations of negligence in Department of Defense (DOD) medical treatment facilities.

    These claims are subjected to a series of trends analyses. DLM also collates and analyzes other risk management data submitted electronically by the Offices of the Surgeons General of the military services and provides regular summary reports regarding that information to those contributors and to the Assistant Secretary of Defense - Health Affairs (ASD-HA). DLM has been tasked by ASD-HA to conduct these recurring analyses, along with other research regarding those submissions, to assist that office in the derivation and implementation of policies calculated to improve the quality of health care within DoD.6 The sharing agreement of June 1992 calls for DLM to provide the VA with similar data analyses and educational programs. Under the agreement, DLM analyzes the nationwide VA tort claims experience, collates data, and provides periodic reports to assist the VA in identifying any high risk practice areas and developing responsive, remedial quality improvement endeavors.

    To assist the reader in understanding the VA Tort Claim Information System, certain terms and procedures that control the pursuit of a medical malpractice claim against the United States government must be delineated. A petitioner initiates this process by filing a claim, utilizing a designated federal standard form (SF-95), with a legal representative of the federal agency involved. For claims involving the VA, an SF-95 is filed with the respective District Counsel. By regulation, claims remain in an administrative phase for six months, while the federal agency involved is charged with investigating the allegations and attempting a final resolution, either by formal denial or settlement. If the claim is officially denied, the petitioner may seek administrative reconsideration by the VA General Counsel. Alternately, the claimant may then elect to file suit against the government in United States District Court. As a named defendant, the federal government is represented in litigation by the U.S. Department of Justice, and trial or settlement thereafter is ultimately controlled by that agency. It should be noted that if the six month administrative phase of this process passes with no federal agency action, neither settlement nor denial, the petitioner is permitted to consider the claim denied and elect to file suit.

TCIS DATA

        As originally implemented, all malpractice claims filed against the VA were entered into TCIS. The entry for each claim was later updated, without creating a second record, if litigation ensued. Information was extracted from every claim, and data was entered into a computer protocol. Database entries from VA District Counsel offices across the nation included the following: 1) claimant’s name; 2) injured person’s name; 3) VA medical facility; 4) VA legal office; 5) court; 6) amount claimed; 7) fiscal year claim opened;8) disposition; 9) payment or closure date; 10) amount paid; 11) date filed; 12) location of injury; 13) incident
date; 14) hospital service; 15) specialty; 16) alleged negligence code; 17) alleged injury; 18) injury severity; and 19) description of alleged negligence VA District Counsel offices also submitted patient information concerning service-connected injury or
service-connected disability.