TESTIMONY PROVIDED TO THE

NATIONAL COMMITTEE ON VITAL HEALTH STATISTICS

SUBCOMMITTEE ON PRIVACY

“PRIVACY AND HEALTH INFORMATION TECHNOLOGY”

Len Lichtenfeld, MD, FACP
American Cancer Society

Washington, DC

February 24, 2005

Mr. Chairman, members of the subcommittee and guests:

Good morning. My name is Dr. Len Lichtenfeld, deputy chief medical officer of the American Cancer Society. Today, it is my privilege to appear before you on behalf of the Society to discuss the internet, its impact on our organization, our collaborators, and those we serve.

My discussion will address several areas. First, I will review our internet-based information strategy, and some of our specific programs. Next, I will discuss the National Cancer Database, which is operated by the American College of Surgeons and supported by the American Cancer Society. Finally, I will present additional comments regarding electronic health records and related technologies.

My goal is to share with you how privacy issues impact our activities, how we address privacy concerns, and some observations as to how privacy is viewed by those we serve.

Beginning in 1996, the American Cancer Society made a decision to reach out to its constituents nationwide with a broad-based information strategy. In 1997, we established a call center which is available anytime to anyone who has any question about any aspect of cancer. In mid-1997, we initiated our efforts to strengthen Cancer.org, our internet-based information service where many of the same questions can be addressed.

We have been fortunate to experience incredible growth and acceptance of both of these services. I am pleased to share with you that we now have about 1.5 million visitors to our website every month and almost 100,000 callers to our call center. We take pride in this accomplishment, and devote considerable organizational resources to these efforts. We have made a commitment that the information we provide is as accurate and evidence-based as possible.

From the beginning, when it came to privacy, we realized that it was not sufficient to meet the letter of the law and the expectations of those who turn to us for help and information. We needed to exceed those expectations.

The Society realized that we had obligations that extended beyond HIPAA, and that we had to make our own commitment to protect the medical privacy of our constituents. To that end, we developed “Health Information Confidentiality Principles” which were finalized in September 2000, and guide our privacy activities to this day.

In those principles, the American Cancer Society established strict guidelines for its staff, and reassurance to the community, that we took privacy concerns very seriously and had every intent of incorporating those principles into our operational objectives.

It is one thing to say what you are going to do, and another to actually do it. I would now like to take you on a brief tour of some of the products we offer through Cancer.org, and share with you some observations regarding the information people will actually provide online, which may be a surprise to some of us in this room today.

Cancer.org is our flagship website which provides considerable resources on various topics relevant to cancer.

As I mentioned previously, we host over 1.5 million visitors to Cancer.org every month, and the numbers continue to grow. We do have a privacy statement on the site, but the practical privacy policy is very simple: we collect no information, and require no information, about or from visitors to Cancer.org. Beyond using “cookies” to help direct people to areas of individual interest on the site, visitors enjoy “free” access.

Unless someone wishes to make a donation to the Society, there is no data request, no data accumulation, no demographic analysis, and no data entry into our customer relation management system.

As you might expect, we do not receive any feedback—positive or negative—regarding privacy issues on Cancer.org. We do get some feedback from others within our organization who would like to have information about visitors to our site, but the decision has been made to keep the site free of any obligations for these visitors.

Within Cancer.org, we do provide other services to our constituents that may require more information sharing, but only if they seek them.

We have a calendar and reminder service, but that has had limited usage. In comparison, we have had considerable success with our interactive Cancer Survivors Network. This on-line community, which requires 2 ½ FTEs to maintain and operate, has developed into a very effective support mechanism for its members. Since its launch in July, 2000, CSN has had over 1.4 million visitors, and has registered 54,000 members of the community. New member registrations range from 1200 to 4500 monthly.

I wish that time permitted me to go into detail about this wonderful network and its accomplishments. Cancer Survivors Network offers support to its members in a variety of ways, including message boards, private and secure internal CSN email, and personal web pages, to use in whatever way they see fit. There are few rules and regulations, and we don’t permit members to provide medical advice. But, it’s ok to say, in your words, what you are experiencing and feeling, either during treatment or as a survivor, caregiver or both.

This network is built on trust and respect and, although monitored by the Society for violations of site terms and conditions, is in its own way is self-policing. If someone wants to see what the network has to offer, they are able to view unrestricted areas of the site through Cancer.org without a specific registration. But if someone wants to be an active participant and view restricted areas of the site where the chat rooms, email, and web pages are located, they must register. Once again, registration information is limited: a screen name, a password, a valid email address, and a zip code. They can also provide personal profile information about themselves, such as their disease and its status. This information actually lets patients find others similar to themselves by whatever criteria they choose—for the sole purpose of sharing experiences.

It is important to note the Society does not use identifiable information gathered on this site or Cancer.org for its internal activities. We do gather aggregate information for the purposes of tracking and trending. In the opinion of staff, it is this absolute commitment to protect the community participants that has led to the success of this venture.

What is interesting to those who work regularly with the network is the degree to which people make their information available to others through the network. CSN members can remain completely anonymous. However, we have found that names, addresses, and personal interests are regularly provided openly by the members on the network to other members of the community. Movement in the information age requires caution on everyone’s part, especially since posting information on unrestricted areas is searchable. Anecdotally at least, there have not been serious breaches as a result of this information “freedom”.

Our observation about the exchange of information on this network, as stated by one of our staff, is that “By and large, people determine they are all in the same boat, and they share a lot, as little or as much as they want to say. It’s like being in a relationship.”

Two other programs on Cancer.org have significant requirements for disease specific information: our treatment profiler, in cooperation with NexCura, and our clinical trials matching service, in cooperation with EmergingMed.

Both of these programs require specific data entry. The treatment profiler uses an “opt-out” strategy for further products or responses based on data entry; the matching service uses an “opt-in” approach. The matching service privacy policy has been created in close collaboration between EmergingMed and the American Cancer Society; the treatment profiler privacy policy has been created by a company in cooperation with other well-recognized not-for-profit voluntary health organizations

As simple as “opt-in vs. opt-out” may seem, there are significant differences in their implications and what can be done with the information as a result of the approach utilized. Whether consumers truly appreciate the difference is uncertain, but we suspect they do not.

With the “opt-out” strategy, consumers give the organization (in this case, not the American Cancer Society) the opportunity to maintain a database and use that information to advise patients of various opportunities regarding medications, treatments or clinical trials. These messages are, not infrequently, commercially supported and go beyond the initial intent of entering the data into the system, which is to determine the best treatment options for a specific type of cancer in a particular stage or circumstance.

Although one might suspect that such sharing of information with commercial entities, resulting in emails or potentially other forms of contact, would be of concern to patients with cancer, there is nothing in the Society’s experience which would suggest that to be the case. In fact, if anything, it is the impression of our staff that patients and families actually look to the information as a welcome, useful addition to their knowledge.

In comparison, the clinical trials matching service is an “opt-in” service, where the data remains completely under the control of the patient or family member who provides it. This service has been designed from its inception to be certain that patients have control of their data, and that it is used for no purpose other than linking them to clinical trials through matching algorithms. In this program, the need to proactively protect privacy is paramount. Privacy was a crucial element of the negotiations between EmergingMed and the Society when the program was designed.

What we have discovered about providing personal health related information on the internet is once again surprising: almost all patients who pursue information regarding clinical trials on our website are willing to provide detailed information about themselves and their disease. We have also learned that there must be telephone backup to complete the process. As noted by an executive from EmergingMed, most people who seek us on the web are email literate, but not internet-literate. It’s one thing to email a grandchild; it’s another to order from Amazon.com. Since cancer is primarily a disease of older persons, with a median age at diagnosis of 67, this is a distinction worth noting. It also means we must provide choices to meet the needs of our constituents effectively.

In another area of interest, the American Cancer Society works closely with the American College of Surgeons and their Commission on Cancer (or CoC), which operates the National Cancer Database, or NCDB.

The experience of this database highlights an entirely different issue, namely the ability to effectively gather and utilize data to improve the quality of care for patients with cancer.

For many years, the CoC has accredited cancer programs at hospitals throughout the United States. Presently, the there are 1425 accredited cancer programs, all of whom submit data to the NCDB on a regular basis. Here we have an outstanding opportunity to obtain information about patients with cancer diagnosed nationwide, their treatment, the effectiveness of those treatments, and the survivals of those patients entered into this data system.

Although data is submitted to NCDB electronically, it is gathered pretty much the old fashioned way, by having registrars going through charts and relying on sources outside of the hospital to provide periodic follow-up information. From a research and reporting perspective, the delay in obtaining that information limits the usefulness of the data. With the clear trend to more diagnoses being made, and treatments being offered, outside the traditional hospital setting, there are additional limits to the quality and value of the information obtained.

The Commission on Cancer and the American Cancer Society are well aware of these limitations, and have established goals to improve the ease, completeness and timeliness of data collection, reporting, and analysis. All data collected by the NCDB has been de-identified and reported in aggregate, and will always remain so. But we need to have a better way to find the data, wherever it may reside, whether in the hospital, the independent laboratory, the doctor’s office, or elsewhere. Working with Intel, a 3-phase initiative has been proposed, and funding is being sought to implement the plan to improve data processes.

Will we ever be able to be as efficient as we need to be to accurately collect, collate and report the data? Will we ever be able to provide valuable feedback to hospitals and practitioners in something close to real-time? As we speak, NCDB is engaged in a project to obtain real-time information in cooperation with a hospital system, to impact the quality of care by providing rapid feedback to physicians.

We know we can develop the technology—but with new technology comes new challenges. The NCDB experience serves as a reminder of those challenges.

The NCDB systems are already HIPAA compliant. The creation of a health information network, with the data element standards in place, would be a big step forward. Currently, to function properly, the NCDB has negotiated separate privacy agreements with each of their cancer programs. They had to take into account the varying rules regarding privacy protection in all 50 states. They are working to establish a link to the National Death Index to get information on whether a patient has died, and determine the official causes of death. To date, this remains a work in progress. When federal and state law conflict on privacy protection, if state law is more strict, then state law prevails.

As this discussion demonstrates, We may be able to develop the systems, but we also need a nationwide commitment to make them operational and effective. We need to strike a balance between the various institutional, legislative, regulatory and legal frameworks that exist on a local, state, and federal basis to protect patient privacy and our need to have access to information necessary to improve the quality of care we provide our patients.

All of us will have to work together to address these broad legislative, regulatory, legal and ethical issues in order to achieve the important goals envisioned by many of us here today. We look forward to a system that can transmit information, accelerates our ability to understand what is happening with our patients, and provides adequate protection for individual privacy. We must constantly be mindful of addressing concerns about who will have access to an individual’s medical information, and must always act to avoid compromising an individual’s privacy, or allowing the information to be used for discriminatory purposes.

We need to advance clinical and laboratory science-based programs that improve our health, while not compromising respect for the individual. We need clearly understood rules to avoid institutional paranoia that could inhibit valid research and quality improvement undertakings. We need guidance that permits us to share information across state borders.

Imagine what we could accomplish when we are successful. With access to better data, we would be able to remind people that they need to be screened for cancer at the appropriate intervals. We could develop quality algorithms to monitor the diagnosis and treatment of cancer, and assess patients’ quality of life after they receive their treatment. We could learn to analyze medical records and inform patients if they are eligible for clinical trials. We could find a way to quickly determine which new cancer treatments work, and which do not.

The American Cancer Society hopes that, as we enter this new information age, policies will be developed that support our mission to educate the public, to collect data to improve the quality of cancer care, and allow the conduct of research using the internet.

However, the Society does not believe we have to start at square one to convince cancer patients, their families and their caregivers of the importance of obtaining and using this information. As one senior researcher said to me, “Cancer patients are different. They really are willing to cooperate with people they trust. They view their participation as an opportunity for themselves and those who are helping them. They don’t want to be protected to death.”

Our experience at the American Cancer Society supports that viewpoint. We enjoy enormous trust around the nation and throughout the world. Our attention to privacy must mirror that trust, and it does. In exploring the development of the National Health Information Network, we believe that patients, families and caregivers of cancer patients are well aware of the value equations. Perhaps cancer patients are different, and perhaps they do “get it” better than most. After all, they are dealing in many cases with a potentially life threatening disease. To them, information is crucial, time is of the essence, and accuracy is critical.

They trust us to do the right thing, and we respect that trust. We are optimistic that through public and private partnerships creating effective information systems and data-gathering opportunities that we will be able to provide them with what they expect from us.

In closing, I would like to acknowledge the many people who assisted in the preparation of this testimony.

Thank you again for this opportunity to speak with you today.