Chairman Filner, Congressman Buyer, members of the House Committee on Veterans’ Affairs, good morning.  Thank you for the opportunity to appear here today to discuss VA Cooperative Study Program No. 519, our pharmacy benefits management program, and our work in protecting the health and well being of veterans who volunteer to participate in our research studies.  We share a common goal; to provide the best health care available anywhere for our nation’s veterans; and all of VA’s employees and volunteers work hard every day to ensure that this goal is met.

The Purpose of CSP-519

For more than 60 years, VA’s research program has improved lives through innovation and discovery.  VA researchers played key roles in developing the cardiac pacemaker, the CT scanner, radioimmunoassays, and improvements to artificial limbs.  The first liver transplant in the world was performed by a VA clinician investigator.  Clinical trials established the effectiveness of new treatments for tuberculosis, schizophrenia and high blood pressure.  The Seattle Foot allows people with amputations to run and jump.

Recently, there have been questions and concerns with regard to VA research programs; in particular, with the VA Cooperative Study entitled “Integrating Clinical Practice Guidelines for Smoking Cessation into Mental Health Care for Veterans with Post-traumatic Stress Disorder (PTSD)” (CSP-519).  This study is designed to determine whether integrating smoking cessation and PTSD therapies is more effective in stopping smoking than smoking cessation therapies delivered separately through a smoking cessation clinic, the usual way care is provided at VA. 

Entering patients into the study began in November 2004, and ended in December 2007.  The patients who entered the study all had PTSD, and all wanted to quit smoking.  Patients in the study were allowed to receive medications from their health care providers to help them quit smoking.  This is not a drug study. In many cases, the drugs patients are taking are prescribed by health care providers who are not at all associated with the study.  Whether or not patients were enrolled in this study, all prescribing decisions were made by health care providers in one-on-one consultations with their patients, with those providers deciding which approach was most likely to work for those patients.

The study is being conducted at 10 medical centers, and 945 patients have been enrolled in the study.  Patients participating in the study were randomly assigned into one of two study arms, and they were equally divided between those receiving integrated smoking cessation and PTSD therapies and those receiving smoking cessation therapies delivered separately through a smoking cessation clinic,.  All smoking cessation medications that were FDA-approved and on the VA formulary were made available to providers in both arms.  Every patient enrolled in the study signed an informed consent form.

On June 20, 2008, VA’s Chief Ethics in Health Care Officer provided me with her review of CSP-519.  She concluded that the study aim is consistent with VA’s mission to improve the health and well-being of veterans; the scientific design of the study was appropriate to address the research question and to yield useful data; the study does not expose patients to undue risk; the information about varenicline provided in the informed consent document was appropriate; the study’s subject selection is appropriate; the study protocol adheres to ethical standards for privacy and confidentiality; the plan for monitoring the research is appropriate in terms of timeliness and thoroughness; the protocol reflects consideration and implementation of special safeguards to protect the rights and welfare of research subjects who may be vulnerable to coercion; and the remuneration offered for participation is modest, appropriate, and not coercive. 

On June 25, 2008, I asked for a comprehensive review of the study – looking at whether protocols and safeguards were followed and met to ensure that our patients were receiving proper notice and quality care.   I will discuss this in additional detail later in my remarks.

Importance of the Study

Every year in the United States, smoking accounts for approximately 440,000 deaths.  Premature deaths from smoking rob more than five million years from the potential lifespan of those who have died.  Forty years after the first Surgeon General Report outlined the health effects of smoking, smoking remains the leading cause of preventable death and disease in the United States.  Smoking is a chronic, relapsing disorder, and even smokers who are highly motivated to quit may attempt to quit multiple times before they are finally successful.

Between 33 percent and 45 percent of smokers will die of smoking-related illnesses.  The risk of dying from lung cancer is more than 23 times higher among men who smoke cigarettes.  Smoking is associated with at least 14 other types of cancers, including cancer of the stomach, oral cavity, pharynx, larynx, esophagus, pancreas, and nasal cavity.

Cigarette smokers are 2-4 times more likely to develop coronary heart disease than nonsmokers.  Smoking is also a major cause of cerebrovascular disease, chronic bronchitis, and emphysema, and is associated with gastric ulcers.

Smokers who quit before the age of 50 cut their risk of dying in the next fifteen years in half.  Smokers who quit have a slower rate of decline in lung function and a lower incidence of bronchitis, emphysema and other respiratory conditions than persons who continue to smoke.  Quitting smoking reduces the risk for further congestive heart disease morbidity and mortality.  Smokers with cancer who continue smoking during treatment decrease treatment effectiveness, overall survival prognosis and quality of life, and increase the risk for new morbidities.  Smoking itself has psychiatric consequences.  A recent study on smoking and suicide (Bronsich, et. al., Smoking predicts suicidality: Findings from a prospective community study, in the Journal of Affective Disorders 108 (2008)) found that suicide ideation and suicide attempts were strongly associated with occasional and regular smoking and nicotine dependence, with odds ratios from 1.4 suicides to 5.8 suicides among smokers to one among non-smokers.

Studies have shown that individuals with PTSD are more than twice as likely to smoke as the general population.  While the rate of smoking among VA enrollees in general is approximately 30%, (Miller, D.R., et al., Health Behaviors of veterans in the VHA: Tobacco use: 1999 large health survey of VHA enrollees, VHA Office of Quality and Performance, 2001.) the rate of smoking among veterans with PTSD under VA care is 53% to 60%. (Beckham, JC et al, Prevalence and correlates of heavy smoking in Vietnam veterans with chronic posttraumatic stress disorder, Addictive Behaviors, September-October 1997.; Beckham, JC et al, Smoking in Vietnam combat veterans with post-traumatic stress disorder, Addictive Behaviors, September-October 1997.)  Veterans with PTSD are more likely to be heavy smokers and are only half as likely to quit as are smokers without PTSD in the general population.  VA believes it is our responsibility to help this population and all veterans to quit smoking, and we are continually working to find ways to do so.

The Use of Medications in CSP-519

CSP-519 is not a drug study. It is not a test of Chantix® or any other medication.  The study was proposed, funded, and initiated before Chantix® ever came into existence.  Instead, CSP-519 is a study comparing the effectiveness of two different methods for delivering standard, evidence-based treatments for tobacco use for veterans with PTSD.  These treatments consist of behavioral counseling to stop smoking, which is required for study participation, plus recommended but optional medications for smoking cessation. Subjects could participate in the study without ever taking any medications for smoking cessation.

It is important to note that all medications have possible side effects.   Attempting to regulate body chemistry by using medicines can have both beneficial and harmful effects.  Two people who take the same medicine can have very different experiences.  Physicians must always assess the risk of side effects against the expected benefits of any medication. 

The side effects for any smoking cessation medication can be significant.   Nicotine patches may cause headache, dizziness, lightheadedness, drowsiness, stomach upset, nausea or facial flushing. Patients wearing nicotine patches can experience more serious effects including breathing difficulties, chest pain, irregular heartbeat, nervousness, anxiety, tremors.

Buproprion, marketed as Wellbutrin® or Zyban®, can cause abdominal pain, constipation, decrease in appetite, dizziness, dry mouth, increased sweating, nausea or vomiting, trembling or shaking, trouble sleeping, weight loss, blurred vision, change in sense of taste, drowsiness, feeling of fast or irregular heartbeat, frequent need to urinate, unusual feeling of well being, agitation, anxiety, tinnitus, skin rash, hives, itching, confusion, extreme distrust, hallucinations, seizure, and trouble concentrating.  Overdoses can result in fast heartbeat, hallucinations, loss of consciousness, nausea, seizures, and vomiting.   The FDA has required a “black box warning” for Buproprion stating that the medication, like all antidepressants, may increase the risk of suicide in persons younger than 25.

Varenicline, marketed as Chantix®, has been described by “The Medical Letter” publication as the most effective FDA approved smoking cessation medication available.  During premarketing development of Chantix®, more than 4500 individuals were treated with the drug.  Side effects of varenicline based on this clinical trial include nausea, which is fairly common; headache, difficulty sleeping, and abnormal dreams.  Rarer side effects include a change in taste, vomiting, abdominal pain, flatulence, and constipation. 

On November 20, 2007, the FDA issued an “Early Communication” based on post-marketing reports from users of the drug.   In that message, they wrote that Pfizer, Inc., the manufacturers of Chantix®, had recently submitted to FDA postmarketing cases describing suicidal ideation and occasional suicidal behavior among Chantix® users.  They also wrote that “Chantix®’ role is not clear,” and made several recommendations on more closely monitoring patients for behavior and mood changes.

On February 1, 2008, FDA issued a “Public Health Advisory” notifying health care providers that there may be an association between Chantix® and serious neuropsychiatric symptoms.  It asked that patients and providers be made aware of this finding and stated that it appeared “increasingly likely” there may be an association between Chantix® and serious neuropsychiatric symptoms.  To date, FDA has not asked that varenicline be removed from the market; has not issued a “black box warning” for the medication; and the drug continues to be FDA-approved.

Varenicline is one of approximately 62 FDA-approved drugs which, in their labeling, have been associated with or have concerns related to adverse effects that include suicidal ideation or suicidal behavior.  These include a number of drugs with well-known brand names, including Neurontin®, Topamax®, Depakote®, Sustiva®, Cipro®, Accutane®, Lariam®, Reglan®, Provigil®, Abilify®, Clozaril®, Zyprexa® and Risperdon®.  If VA were to withhold these medications from our patients with mental health issues, we would have great difficulties in treating them at all.

Approximately 6 million Americans have received prescriptions for varenicline.  This figure includes 70,000 VA patients who have received prescriptions for varenicline since VA approved the drug in January 2007 for its formulary.  Nearly 33,000 VA patients are currently taking the medication.  Approximately 6,500 patients now taking varenicline have been diagnosed with PTSD.  2,012 of the 70,000 patients who have taken varenicline, including 400 of those with PTSD, are veterans of Operation Enduring Freedom or Operation Iraqi Freedom.

Two-hundred forty-one patients have been prescribed varenicline at some time during the course of the study, either by VA physicians or by physicians not associated with VA.  As of June 25, 2008, VA is aware of 40 study subjects who are currently taking this medication.

A review of Serious Adverse Effect data among CSP-519 participants indicates that, from January 1, 2007 through June 25, 2008, of the 241 patients prescribed varenicline, 75 had a total of 114 significant adverse effects.  Nineteen of those seventy-five had 22 psychiatric significant adverse effects, including 11 patients who had 12 episodes of suicidal ideation.  There was one suicide attempt in that group, and no suicide completions.

Of the 704 patients who were not prescribed varenicline, between January 1, 2007 and June 25, 2008, 124 unique patients had 171 significant adverse effects during the study.  Twenty-eight of those patients had 36 psychiatric significant adverse effects, including 11 patients who had 14 episodes of suicidal ideation.  Four patients who were not prescribed varenicline attempted suicide, and one committed suicide.  There was one intentional drug overdose, not yet classified as suicide.  It is important to note that adverse effects occur during studies which are unrelated to the study itself.  Throughout the entire course of the study, there were two deaths, unrelated to the study, in the group of patients taking varenicline.  There were also 25 deaths in the group that did not take varenicline, which have not been analyzed.

The patients participating in this particular study, or any VA study, are under the care of physicians who are closely monitoring and evaluating them and changing their treatment if necessary.  The care of our patients is our number one concern whether the veteran is participating in a study or not. 

Patient Awareness of Issues Related to Varenicline Use

On the national level, VA quickly responded to FDA communications about varenicline.  FDA’s November “Early Communication” was distributed to all VA facilities the day after it was issued.  Once VA received the preliminary message, we aggressively searched for events which might signal a problem among our patient population.  This was done not only through an evaluation of the voluntary reporting of adverse drug events throughout our system, but also through the use of VHA’s integrated medication databases to search for any potential safety issues.

On January 18, 2008, VA issued guidelines to providers stating that varenicline “should be reserved for veterans who have not been successful with nicotine replacement therapy and/or buproprion, or for whom buproprion is contraindicated.”  It also stated that before starting varenicline treatments, VA health care providers should educate veterans about the possibility of changes in behavior and mood, and they should be carefully monitored, and that veterans using varenicline should be warned that it can cause drowsiness and should use caution while driving or operating machinery.

On February 24, 2008, after testing at three medical centers, VA provided updates for local medical centers to their prescription software.  The updates provided additional labeling on prescriptions for varenicline.  It stated “Call your doctor immediately if you have mental/mood changes like confusion; new/worsening feelings of sadness/fear; thoughts of suicide, or unusual behavior.”  By April 1, all VA facilities except one had completed this update.  VA also provides patient medication information sheets on all new and renewed prescriptions.  Information on those sheets was updated in the same way.

Institutional Review Boards (IRBs) at all CSP-519 study sites were notified of the FDA “Public Health Advisory” on February 5.  On February 13, the study coordinators sent all study leaders a new patient consent form for patients in the study to sign who were taking varenicline.  They also sent a draft letter for patients informing them of the FDA’s warning.  The draft letter was written by a team of psychiatrists and psychologists who felt the issue of suicide should be discussed in a clinical setting, not a mass mailing, when patients reported to their study coordinators for regular follow ups.  It did explicitly advise patients that they should inform their doctor or the study staff immediately if they noticed any changes in their mood or behavior, or if they would like to stop using the medication.

The cover letter was never intended to serve as a stand-alone document that would duplicate the consent addendum, which was attached.  Instead, the purpose of the cover letter was to provide a brief and concise introduction to the addendum—an addendum that explicitly listed all the potential side effects identified by the FDA’s warning, including suicidal ideation and suicidal behavior.

The timing of mailings of the letter and the consent form addendum were left to the individual IRBs.  VA’s agreement with study participants indicated that if any new specific information became available related to the study we would inform them, and study leaders felt that the FDA “Public Health Advisory” met that standard. 

This plan was formulated by the CSP-519 study team, which included a physician; a pharmacist; a psychologist; a social worker; and other representatives of the CSP Pharmacy Coordinating Center, the CSP Coordinating Center, and the study Chair’s office.  The plan was approved by the CSP Human Rights Office and the CSP-519 Executive Committee, and was overseen by the CSP-519 Data Safety Monitoring Board.

Letters to patients were sent or hand-delivered between February and June.  I am concerned about the time that elapsed at a number of study sites between the receipt of the letters and consent addendums by IRBs, and when they were received by veterans.  I am also concerned about the lack of follow-up by study coordinators to ensure that their directions were carried out.  There is a clear need for improved follow up in this area.

Since March 2007, the VA Center for Medication Safety has monitored varenicline through national pharmacovigilance efforts such as collecting and analyzing spontaneous reports of adverse events.  As a result of their data, FDA’s Public Health Advisory, and the Federal Aviation Administration’s banning of the use of varenicline in airline pilots and air traffic controllers, VA issued a National Pharmacy Benefits Management Services Bulletin on May 30 to all practitioners informing them of the new warnings.  They also sent a Patient Letter for Formulary Leaders and Pharmacy Chiefs to provide to their patients.  Following news reports on the issues at this hearing, I sent a letter June 20 to all 33,000 patients with a current VA prescription for varenicline, offering to find another way to help them quit smoking if they were concerned about taking varenicline or had experienced any side effects.

VA has consistently and continually provided FDA with information on our experience with varenicline, and is continually looking for evidence to indicate whether its use should be continued in our patient population.  To date, however, we have found no evidence that would cause us to discontinue the use of this drug in our patients.  No other health care system approaches the vigilance VA has demonstrated in educating its providers and patients on the possible newly reported risks of varenicline.  The care, treatment, and health of our patients is our number one concern. The research we do benefits them as well as patients throughout the world. 

Safeguards and Protections in VA’s Human Research Program

VA research, done with patient consent, benefits veterans; all Americans; and all citizens of the world. VA gratefully acknowledges the contributions of veterans who volunteer to participate in our research and we take our responsibilities for their care very seriously.   In the past seven years alone, VA employees have authored or co-authored more than 46,000 scientific articles.  Nearly 900 were published in the most eminent of the Nation’s scientific journals, including Science; the New England Journal of Medicine; and the Journal of the American Medical Association.

VA’s Cooperative Study Program (CSP) specializes in designing, conducting, and managing multi-site clinical trials and epidemiological research.  These include studies establishing the cornerstone for hypertension treatment; the long term effects of coronary artery bypass surgery; a study showing that aspirin reduces deaths and heart attacks in patients with unstable chest pain; and a study demonstrating the efficacy of a shingles vaccine.  Just last week, the New England Journal of Medicine published the latest CSP contribution, which compared intensive versus standard therapy in patients with acute renal failure.  The VA Cooperative Studies Pharmacy Coordinating Center is ISO 9000 compliant, and has received many awards for its effective management of clinical trials.

VA policies and procedures in this area are among the best in the Nation. Every VA research facility must have an IRB, the local committee charged with the oversight of all research activities involving the use of human subjects. IRBs must have at least five members with varied backgrounds to promote complete and adequate review of research activities; at least one member whose primary expertise is scientific and one whose primary expertise is non-scientific; and at least one member not otherwise affiliated with the VA medical center.  No IRB may consist entirely of members of one profession.

IRBs must approve, require modifications to, or disapprove all research activities.  Before approval, they must determine that the following requirements are satisfied: risk must be minimized; there must be a reasonable risk to benefit ratio; subjects must be equitably selected; informed consent forms must be valid; the informed consent process for patients must be documented; safety must be monitored; privacy and confidentiality must be maintained; vulnerable subjects must be protected; conflicts of interest must be managed, reduced or eliminated; and investigators must meet education requirements for the protection of human subjects in research.  All IRBs are required to fully document their activities.

VA’s CSP goes above and beyond what other organizations do in the area of human subjects protection through our Human Rights Committees, which determine whether protections of patients’ rights and welfare in VA research studies are adequate. These committees are composed of individuals from the community and VHA with the interest and background required to consider the ethical and legal issues involved in the participants of human subjects in research.  They are also responsible for ensuring that patients’ rights and welfare are protected during studies, and for talking directly to patients to ensure that human rights aspects of cooperative studies are receiving proper attention.

VA has established numerous other safeguards for our patients in VA research studies.  We are one of sixteen federal agencies who have adopted the Common Rule for the protection of human subjects in research.  In 1999, we established an independent office of research compliance and assurance.  This office was succeeded, in 2003, by the Office of Research Oversight (ORO), VA’s primary office responsible for overseeing the responsible conduct of research throughout our system.  In 2003, its first year of operations, ORO made 19 site visits to facilities.  In 2004, that number increased to 22 visits; by 2007, it had tripled, to 67. 

In 2001, VA published a brochure to help veterans understand their rights as research volunteers and to decide whether they want to be research participants.  This was followed, in 2003, by the publication of a handbook describing the procedures all our research facilities must use to implement our agreement to follow the Federal Policy for the Protection of Human Subjects.

In addition, VA has developed a program to accredit all VA research programs by the Association for the Accreditation of Human Research Programs.  In 2003, we required all VA employees involved in human research support programs (except secretaries) to undergo annual training in good clinical practice and the ethical principles of human research protection.  More than 15,000 employees completed the training within 90 days of its establishment.  Since 2003, we have created more than 15 different training programs in human research protection for our employees.

Also, we have established a Central Institutional Review Board to preclude some of the variability we have seen in execution of multi-site projects such as CSP-519.

Actions Underway

Though VA research is regularly reviewed by many organizations, we have directed that each VA human subject protection program must seek accreditation through the Association for the Accreditation of Human Research Protection Programs.  At present, 112 of 115 VA facilities conducting human subject research have submitted applications for accreditation to this organization; all will be reviewed by the end of Calendar Year 2008.  Of these 112 facilities, 57 have already received full accreditation.  In some cases, this supplants previous accreditations received from the National Committee for Quality Assurance (NCQA).  VA leads all federal agencies in accreditation of human research protection programs. 

VA’s National Research Advisory Council, which is composed of internationally recognized medical scientists, has consistently recognized VA’s research program for its success in meeting its obligations to taxpayers and to veterans.

While we have procedures and safeguards in place for the protection of our patients, I am committed to making sure we are doing everything we can on behalf of our Nation’s veterans.  On June 25 I directed the Under Secretary for Health to conduct four evaluations:

First, I requested a comprehensive review of CSP-519, through VA’s Office of Research Oversight, with results to be reported to me within 30 days and with an action plan to address recommendations to be completed not more than 10 days later.

Second, despite the fact that CSP-519 is not a drug study, I directed that there be Institutional Review Board reviews of all PTSD drug protocols in our system to ensure that there is appropriate sensitivity to the study population in the context of FDA alerts and warning.  I also directed a review of the risks of medications that are likely to be used in the study population, and that there has been proper subject notification of associated risks.  The Office of Research Oversight  is to report results to me within 45 days, and the Under Secretary for Health will provide me with an action plan 10 days later.

Third, I tasked the Office of Research and Development  and the Office of Pharmacy Benefits Management to conduct a review of our adverse event reporting system to ensure that there is, in fact, timely reporting and analysis of data, and that the system supports the appropriate escalation of reporting and sensitive issues for subject safety. I expect their results within 20 days, with action plans provided 10 days later.

And fourth, I required the Office of Pharmacy Benefits Management to review VHA’s medication notification system to ensure the system’s policies support timely communications to patients and providers, including those in research programs.  Results here are to be reported within 20 days and action plans provided 10 days later.

In addition, the Inspector General is investigating human subject protections in CSP-519 at the Washington VA Medical Center.  He will testify on what he has learned later in this hearing.  He and his staff have discussed his findings with me.  I appreciate the speed with which he prepared his report, and have appreciated his consideration on clarification of some factual matters.

I have tremendous sympathy for any veteran who has been distressed by reports about this study.  At the same time, VA has an outstanding research program, and has been innovative in its ability to protect its research subjects.  I commit to veterans, and to you, that I will be comprehensive and thorough in my investigations of how this study has been, and is being, conducted.  I am determined that VA will remain a leader in the protection of human subjects and in veteran-centric research.  Thank you again for the opportunity to discuss the important work we were doing for and with the help of veterans. 

Date:  June 23. 2008

From:  Secretary (00)

Subj:  Evaluation of Research

To:  Under Secretary for Health (10)

1. In response to recent allegations I direct you to conduct evaluations of VHA Research and related support functions as outlined below:

1. Part I: The Office of Research Oversight (ORO) will continue a comprehensive review of Cooperative Studies Protocol #519 (Integrating Practice Guidelines for Smoking Cessation into Mental Health Care for Posttraumatic Stress Disorder [PTSD]) consistent with the plan outlined by the Chief Research Oversight Officer (see attachment). ORO should report results to me within 30 days. I expect an action plan to address ORO's recommendations within 10 days after the report has been completed;
2. Part II: ORO will oversee Institutional Review Board reviews of all PTSD protocols to ensure there is: appropriate sensitivity to the study population in the context of FDA alerts and warnings; a review of the risks of medications likely to be used in the study population; and proper subject notification of associated risks. ORO should report results to me within 100 days, I expect an action plan to address those recommendations within 10 days of the report's completion;
3. Part III: The Office of Research and Development (ORD) and the Pharmacy Benefits Management (PBM) Strategic Healthcare Group will conduct a review of the adverse event reporting system in investigational and clinical settings to ensure there is timely reporting and analysis of data and that the system supports the appropriate escalation of reporting and sensitivity issues for subject safety. The review should also address triggers for reporting or escalation of reporting. ORD should report results to me within 20 days, I expect action plans to address the recommendations within 10 days afterwards; and
4. Part IV: The PBM will conduct a review of the Veterans Health Administration medication notification system to ensure that the system's policies support timely communication to patients and providers, including those in research programs. Recommendations from the review will be coordinated with Operations, ORO and ORD to ensure policy coordination. PBM should report results to me within 20 days. I expect an action plan and revised policy to be completed within 10 days of the report.

2. If you have any questions about these tasks, please let me know.

James B. Peake, M.D.

Attachment