Skip to Content


Survivorship Issues: New Tools for Assessing Symptom Distress

Network - Summer 2007

Gathering symptom-related data to improve survivors’ daily lives is the central focus of M. D. Anderson’s Department of Symptom Research. This ongoing series of articles, begun in the last issue, tells of the steps being taken to collect solid scientific evidence that can be used to design interventions to relieve the symptom burden caused by cancer and cancer treatments.

Understanding the symptoms of primary brain tumors

Terri Armstrong had been working in the Department of Neuro-Oncology for only a short time when she realized she was hearing the same issues from patients with primary brain tumors. That made her acutely aware that an aspect of the disease wasn’t being addressed: Brain tumor patients have an overwhelming symptom burden.

“It is common for people to go from normal, working lifestyles and doing fine to having a sudden event like a seizure and being thrust into this world of brain cancer, often with residual deficits,” says Armstrong, D.S.N., an advanced practice nurse. “Problems with strength or sensation or ongoing seizures can be devastating.”

Interested in the impact of brain cancer on quality of life, Armstrong began a literature review. She was amazed to discover that what was known about the symptom burden of this patient population was from studies done with data from as early as 1924, while health care had obviously changed a good deal in the intervening years.

At the same time, she decided to work toward a doctorate degree in nursing at The University of Texas Health Science Center at Houston, School of Nursing. Her advisor and mentor, Marlene Cohen, Ph.D., also a professor at M. D. Anderson, introduced her to the work of Charles Cleeland, Ph.D., chair of the Department of Symptom Research.

“At first, I just wanted to go in and do things to help people, but then I realized we don’t really understand their problems,” she says. “It was hard to step back, but that’s where Dr. Cleeland’s group is so helpful. Through them, I learned we have to understand the issues first so we can target interventions.”

Gathering the data

While she knew that the core M. D. Anderson Symptom Inventory developed by Cleeland’s group was an important tool for measuring the severity and impact of cancer-related symptoms in general, she also knew it didn’t address many of the specific symptoms of those with primary brain tumors.

“I wanted to add something to the existing instrument and try to capture the symptoms that were important to this population,” she says.

Somewhat the same as in the development of the core MDASI and MDASI-Thy for patients with thyroid cancer (see spring 2007 Network), Armstrong put together panels with members from both inside and outside M. D. Anderson. These included people from nursing, social work, neuro-psychology, physical therapy, radiation, neuro-oncology and neurosurgery, as well as patients and their caregivers.

“We learned a lot,” she says. “One aspect that was striking to me was the whole issue of pain. Many of the physicians said that the patients didn’t complain about pain. But what we found is that patients on the panel used other words. They would say, ‘I have pressure in my head’ or ‘I have a sensation in my head.’”

She also learned what she was wrong about. For example, she didn’t think this patient population had shortness of breath, but learned that 20% do.

“One of the most important things we learned was something we had suspected but had no way until now to verify,” she says. “People’s symptoms can predict when their tumor is growing.”

Developing the MDASI-BT

The final inventory for patients with primary brain tumors was narrowed to 18 symptoms divided into three categories:

  • Eight focal symptoms — weakness on one side of the body, changes in body sensation and in vision, difficulty understanding, speaking, remembering, starting or completing tasks, walking
  • Five generalized symptoms — headache or pressure in the head, seizures, change in seizure activity, difficulty concentrating, sleepiness
  • Five treatment and medication-related symptoms — changes in appearance, appetite and bowel pattern, irritability, fear

Currently, patients in certain clinical trials complete a questionnaire listing these symptoms on a scale of 0-10, 0 meaning not affected at all, 10 as bad as you can imagine.

“These instruments are such a good thing because they force us to address these questions, instead of practicing a sort of ‘don’t ask, don’t tell’ policy. If we use Dr. Cleeland’s instrument and this module, that makes us ask: How severe is it? How does it impact a patient’s daily life,” Armstrong says. “It facilitates communication.”

Gathering postoperative data on lung cancer patients

Communication also is important to Ara Vaporciyan, M.D., a surgeon in the Department of Thoracic and Cardiovascular Surgery. He has always tried to help his patients understand the side effects their surgery may have. But until now much of what he could tell them was anecdotal, not evidence-based. The ongoing challenge has been to gather symptom-related data in real time.

Now, a clinical trial being carried out by him and his colleagues in conjunction with Xin Shelley Wang, M.D., principal investigator on the study and an associate professor in the Department of Symptom Research, is to test the effectiveness of an interactive voice response system on postoperative symptom control.

The IVRS is a computer- and telephone-based assessment system that allows patients to report distress, sleep disturbance, shortness of breath, constipation and pain — side effects for which they can offer interventions.

All patients enrolled in this randomized study receive a weekly phone call from the IVRS. In turn, they use the keypads of their telephones to rate (on the same 0-10 scale mentioned above) and record symptoms as they occur.

If any of the targeted symptoms exceeds the threshold value chosen by the investigators, one of two things happens: For the control group, the information is recorded and archived for study at a later date. However, for members of the study group, information about symptoms that exceed the threshold is forwarded to the health care team to review and potentially to act upon.

They are currently monitoring 35 patients: 17 in the study arm and 18 in the control group. In addition, another part of the study accrues patients who are receiving minimally invasive surgery, a thorascopy, which can be performed on patients with early stage lung cancer when there is no lymph node involvement.

Learning what to tell patients

“This study is giving us a way of collecting data that allows me to have a greater understanding of what my patients are going through so I can better prepare future patients for surgery,” Vaporciyan says. “It also allows us to determine if IVRS could be helpful in symptom management during this acute postoperative phase. We’ve learned that pain is highest after surgery, then decreases, but after a month increases again only to drift back to baseline after four or five months.”

He speculates that this is due to a combination of factors. When patients come in for their one-month check-up, they are told they can begin to resume normal activity. At the same time, their pain medications run out and they are transitioned to ibuprofen.

“Simply knowing that the pain will most likely increase and conveying it to the patients helps us alert them to move a little slower and avoid the unnecessary pain,” he says.

Wang adds that this might have greater implications, beyond those dealing with thoracic surgery, “Learning the feasibility and primary effect of using IVRS for postoperative symptom control could be important in allowing us to extend use of this method to help lower more patients’ symptom burden from cancer and aggressive treatments.”

© 2015 The University of Texas MD Anderson Cancer Center