Defining Quality of Life: Psychologist's Questionnaires Play Growing Role in Cancer Research

Photo courtesy of Northwestern University/Feinberg School of Medicine

David Cella, PhD

David Cella never intended to work in oncology.

At the New York Hospital-Weill Cornell Medical Center, with dreams of going into pediatric or family practice, the PhD candidate hoped to spend his predoctoral fellowship in a childhood psychiatry program. Instead, he was assigned to a one-year rotation at Memorial Sloan- Kettering Cancer Center.

As it turned out, the change in plans was a lucky break for both the doctor and the oncology community.

Putting to work the strategies he learned in graduate school to help people improve their coping and communications skills, Cella developed a questionnaire to scientifically measure the quality of life of patients being treated for cancer. Today, the Functional Assessment of Chronic Illness Therapy (FACIT) Measurement System is widely used and has grown to include 91 questionnaires.¹

“Doing research on quality of life in the mid-1980s, I realized I had been trained in something that oncology needed and needed badly—how to develop and validate a questionnaire that applies to people with cancer,” recalled Cella, 56, who serves as professor and chair of the Department of Medical Social Sciences at the Northwestern University Feinberg School of Medicine in Chicago, Illinois. “The existing questionnaires had all been developed by psychologists like me for applications in mental health, so people were applying these scales to cancer patients, and they weren’t measuring the right things. I thought, ‘I really should do this,’ so I did it. It led to FACIT, and it has become this worldwide thing.”

While most of the questionnaires are dedicated to the experience of living with cancer, its symptoms, and the side effects of related treatments, FACIT and its creator have also branched out into the study of other chronic diseases, including rheumatology, gastroenterology, nephrology, surgery, and transplantation.

Sponsors of clinical trials often voluntarily employ FACIT to determine whether a drug might improve quality of life by keeping disease progression at bay for longer—sometimes enough reason for the FDA to approve a cancer treatment, even if the therapy does not increase overall survival, said Cella.

A recent example arose when Cella oversaw the patient-reported outcomes arm of the phase III AXIS study that led to the approval of axitinib (Inlyta; Pfizer) as a second-line therapy for metastatic renal cell carcinoma (RCC). The drug demonstrated median progression-free (PFS) survival of 6.7 months versus 4.7 months with sorafenib, and the quality-of-life study that Cella conducted indicated that the PFS benefit translated into a 25% risk reduction in the worsening of advanced RCC symptoms.²

As the field goes forward, a next step will be to use FACIT scores to better characterize patient phenotypes, Cella said.

“An example is that some drugs cause extreme fatigue, but in a subset of patients only,” he said. “Now that we’ve done a good job of measuring fatigue, we can help identify patients at risk for this side effect.”

David Cella, PhD, At a Glance

• Is 56 years old
• Is married and has five children ranging in age from 8 to 27. “It keeps me very busy. My kids are my main hobby,” he said.
• Was thrilled, in 1998, to co-coach his son’s basketball team with another dad who was also a renowned professor at Northwestern University, the late Kenneth I. Howard, PhD. “He really was a faculty person I looked up to when I took his course in 1974,” Cella said, “and I was now a peer with this man who was one of my idols.”
• Enjoyed the movie Memento, which he described as “a psychological thriller.”
• Remembers a job as a construction laborer one winter during high school as the worst he ever had. “I was working outside in zero-degree weather carrying wood and dumping garbage,” he recalled. “We were in the middle of the demolition of a building, and I was the low guy on the totem pole. But it paid pretty well, and it helped me get through school.”
• While earning his undergraduate degree, spent one year studying at Loyola University Chicago’s John Felice Rome Center in Rome, Italy.
• Has won numerous awards, starting with the Carleton Blunt Distinguished Scholar Award during his undergraduate years and spanning every stage of his education and career. They include the Cornelia Perry Memorial Award to the Outstanding Postdoctoral Fellow at Memorial Sloan-Kettering Cancer Center in 1986-87, and the President’s Award and recognition for Article of the Year from the International Society for Quality of Life Research in 2008.
• Is a member of numerous scientific trade organizations, including the American Psychological Association, the American Psychosocial Oncology Society, and the American Society of Clinical Oncology. In 2011, Cella served on the Patient Reported Outcomes and Person Centered Care in Mental Health Scientific and Advisory Board, a project of the International Center of Mental Health Policy and Economics.
• Has generated more than 300 publications in the area of quality-of-life measurement in clinical trials and clinical practice.
• Lives by the philosophy “Loose lips sink ships.” It’s better to keep information close to the vest when necessary, he said, than to try to explain it prematurely.

Patients’ Feelings Count in Value-Based Equations

At Northwestern, where he earned his BS in psychology in 1976 and then joined the faculty 15 years ago, Cella leads a measurement lab he describes as “dry,” equipped with computers rather than test tubes, sinks, and centrifuges, and staffed by statisticians, analysts, and survey methodologists.

The scientist spends about 20% of his time on the administrative duties involved with chairing his department, and splits a larger chunk of time between his own research and providing education and training, including lectures both inside and outside the university.

Those off-campus activities typically involve consulting with pharmaceutical companies, provider organizations, oncologic healthcare delivery systems, hospitals, physician groups, and professional societies. More and more often, those groups ask Cella to advise them on how to research or keep records related to patient-reported outcomes of treatment.

“There’s an increasing view that patientreported outcomes should be front and center when determining whether or not a given intervention applied to a group of people, like surgery or a drug, has value to people’s lives,” Cella said. “Their goal is to determine that the outcome is worth paying for. For instance, Medicare is looking at health outcomes from patients’ reports, and wants those to be part of the way they structure reimbursement levels.”

As he did with the AXIS study, Cella also regularly participates in clinical trials. He took part in measuring patient-reported outcomes in a phase III study of everolimus in patients with metastatic RCC, with results reported last year.³ He also tracked patient-reported peripheral neuropathy in patients treated for advanced endometrial cancer, reporting results in 2010.⁴

Currently, Cella is serving as principal investigator on a contract to develop content for patient symptom questionnaires for the clinical trials supported by the National Institute of Neurological Disorders and Stroke. But Cella said he still spends at least half his time immersed in cancerrelated research.

Research Snapshot: Next-Generation Patient Outcome Reporting

David Cella, PhD, is serving as chairman of the steering committee and principal investigator of the statistical center for a national effort to create a next-generation system for measuring patient-reported outcomes. Funded by the National Institutes of Health, the project launched in 2004 and entered its second phase in 2009. It is known as the PROMIS (Patient-Reported Outcomes Measurement Information System) initiative, and involves 150 scientists at 15 institutions.

The program has created about 30 item banks—groups of questions about specific symptoms, such as pain and fatigue—designed to elicit precise, reliable data from patients about how they are feeling and functioning. The system allows the flexibility to choose relevant questions from item banks, and provides comparable measurements across all disease states.

Its goals are to help physicians and scientists assess the effectiveness of existing or proposed treatments, and to aid healthcare providers in designing treatment plans for patients.

In a video that the National Institutes of Health posted online last year, and in a recent conversation with OncologyLive, Cella expressed his excitement about the PROMIS initiative and discussed details of the effort.

On how PROMIS works:

Cella: This is a new way to approach measurement. We could be bold and say it’s like 21st-century measurement. It’s a lot of fun, and it has cancer implications.

We have taken the first-generation lessons we learned with the FACIT [Functional Assessment of Chronic Illness Therapy] Measurement System, which we still know and love, and now migrated to a new way of approaching measurement that measures at the domain or concept level.

Item banks allow you to choose questions and still represent scores on a common scale. Taking fatigue as an example, rather than just getting a FACIT measurement, you can use any of the questions from a 95-item bank to measure fatigue—not just in cancer, but across chronic conditions and the general population. It lets you compare yourself to any other person or group with common metrics.

It’s like visiting a lab and, instead of providing blood or saliva or a tissue sample for an assay, you provide your answers to questions.

On how PROMIS was devised:

We conducted 79 focus groups in that first couple of years with patients from various disease populations and struggling with various symptoms and functional deficits. And that information led to a framework that we put together, and it starts with basic physical, mental, and social self-reported health. Physical health we divide into symptoms and function; mental health we divide into affect, behavior, and cognition; and social health into social relationships and social function.

So how did we take this simple domain framework but then incorporate all that had been learned previously? Taking as examples the categories of physical function, fatigue, and depression/anger/anxiety, we identified, in the aggregate, more than 10,000 questions that were used within our field at the time to measure these functions.

We went through a thorough, rigorous scholaring of each of these questions, looking at their quality, their readability, their translatability into other languages, and their comprehensibility. Through that process, we identified, in these three examples, anywhere from one to 300 questions that were good, either because they started [that way] or because we modified them, and this became the base for the PROMIS item banks we built.

On the use of PROMIS to date:

From the beginning, we wanted [PROMIS] to be a tool for clinical researchers. As we’ve been successful, there really is a tremendous interest from people who are not necessarily doing what we might call research who want to use PROMIS because of this comparability, the standards, and the reliability. We get multiple requests from providers large and small to integrate PROMIS with their electronic health record. There’s interest from payers—United HealthCare, Blue Cross, the Veterans Administration, the Canadian government— in order to get a level playing field for patient-reported outcomes as they move into this increasing accountability in healthcare.

The Gretzky Group, which is working with these Beacon Communities—17 large communities funded by the US Department of Health and Human Services whose goal is to integrate their individual electronic health records with a large, master electronic health record—want to include PROMIS in their system. This Beacon group is likely to be a precursor that people point to in the future of accountable care organizations going forward. So there’s a lot to be excited about.

For more information, visit the PROMIS website at http://www.nihpromis.org. The National Institutes of Health podcast is available at nih.gob.

Desire to Help Leads to Psychology Focus

Busy in so many other areas, the psychologist rarely sees patients anymore.

“Over the first 10 years of my career, I started to convince myself that I could help groups by publishing data about how people cope, and what kinds of treatments are good for quality of life and what kinds are bad,” Cella remembered. “I like to think I switched from helping people one at a time to helping, indirectly, many at a time.”

Born and raised in Evergreen Park, Illinois, a suburb of Chicago, Cella was inspired by his father, a pediatrician, to take up a service-oriented profession. But he was less captivated by the mechanics of the body than by human behavior—the way people differed from each other and responded individually to situations.

“I wanted to help people,” he said, “and I thought I could do that through psychology in a way that was interesting.”

Cella earned a BS in psychology from Northwestern and then both a master’s degree and a doctorate in clinical psychology from Loyola University Chicago. He completed a predoctoral internship and a predoctoral fellowship at The New York Hospital-Weill Cornell Medical Center, during which he worked in Cornell’s Payne Whitney Psychiatric Clinic, and then at Memorial Sloan-Kettering Cancer Center.

During a postdoctoral fellowship, Cella also worked at Payne, serving as chief psychology intern. He completed his fellowship years in psycho-oncology at Sloan-Kettering, where he conducted research on psychosocial aspects of cancer survivorship and on quality-of-life evaluation during cancer.

Early on, he found he was drawn to his work with cancer patients while also feeling deeply troubled by it.

“A common experience I went through is that I began to believe I had cancer,” Cella said. “I remember when I was 27 or 28, noticing a lump on my neck or groin and saying, ‘Oh my God, I’ve got lymphoma!’ That was anxiety and identifying with the people I was working with, and with the help of my instructors I was able to use that to draw a line between my personal and professional life—while not removing my emotional realness from the professional side, which is what patients needed from me. Ultimately, it was enriching, rewarding, and inspiring to meet people day in and day out who were courageously facing a life-threatening situation, and to help them to navigate their way through it.”

Cella so enjoyed his work that, in 1987, he became a clinical assistant psychologist at Sloan-Kettering. Shortly thereafter, he went home to Chicago, where he spent 10 years as director of the Division of Psychosocial Oncology at the Rush Cancer Institute.

He returned to his alma mater, Northwestern, in 1997.

“Anytime I walk through the campus, I have flashbacks to when I was 18,” Cella said. “I loved college, and I have fond memories as I walk through the same campus now as a professor and department chair.”

Efforts to Refine FACIT Hone In on Subjectivity

Upon his return to Northwestern, Cella expanded his research on quality of life during chronic illness into a broader study of outcomes in healthcare. In addition to quality-of-life measurement in clinical trials, his focuses include cross-cultural equivalence of quality-of-life measurement, efficacy of psychosocial interventions in oncology, and medical outcomes.

At the same time, the psychologist has continued to refine his FACIT Measurement System.

That can be a complex process, Cella said, because “it’s not easy to get standardization. The biggest problem that is controllable is asking questions that mean different things to different people, so we try to be simple about the questions— ‘How’s your pain? How’s your fatigue? Are you sad, depressed, or nervous?’ We make the questions as unambiguous as possible, because ambiguity is an invitation for individuality, and we don’t want that kind of individuality, where people are filtering the question with their own value system and affecting the way they respond.”

Even after taking those measures, though, there’s a certain amount of subjectivity about the way patients answer the questions on the form.

“You can’t look at a person and guess at what their pain score would be,” Cella said. “It’s possible that somebody who has external behavioral indications of pain, who is writhing and limping, will say his pain is at a level of 3. Another stoic person may have no outward sign of pain, but they say they’re at an 8, 9, or 10. We don’t really know why; we just have to take it out of faith that if they say they’re a 3, then they’re a 3.

“To some extent, that introduces error into what we measure, but there’s error in everything scientists measure. When you get a white blood count, there are air bars around the estimate, and when you image a tumor, there may be some error in the measurements around dimensionality. You deal with it by generating confidence intervals. All these measurements have a margin of error, and we have reduced that by half in the time I’ve been involved in the work.”

Murkier still, Cella said, is the question of what constitutes quality of life. It’s something he’s determined to help clarify as time goes on, an issue that’s at the heart of his research.

“I’m dedicated to doing a better job of documenting what’s real quality—what does it mean, and how can we prove it,” Cella said. “It’s not easy to pin down, because quality means different things to different people, and they have different values and preferences. If we could pin that down, we’d place more value on it, and doctors would be more comfortable recommending treatments based on quality of life as opposed to purely survival.”

References

1. Questionnaires. FACIT.org website. http://www.facit.org/FACITOrg/Questionnaires. Accessed May 1, 2012.
2. Cella D, Escudier B, Rini BI, et al. Patient-reported outcomes (PROs) in a phase III AXIS trial of axitinib versus sorafenib as second-line therapy for metastatic renal cell carcinoma (mRCC). J Clin Oncol. 2011;29 (suppl; abstr 4504).
3. Beaumont JL, Butt Z, Baladi J, et al. Patientreported outcomes in a phase III study of everolimus versus placebo in patients with metastatic carcinoma of the kidney that has progressed on vascular endothelial growth factor receptor tyrosine kinase inhibitor therapy [published online ahead of print April 1, 2011]. Oncologist. 2011;16(5):632-640.
4. Cella D, Huang H, Homesley HD, et al. Patientreported peripheral neuropathy of doxorubicin and cisplatin with and without paclitaxel in the treatment of advanced endometrial cancer: results from GOG 184 [published online ahead of print September 21, 2010]. Gynecol Oncol. 2010;119(3):538-542.